key: cord- -d tbf a authors: ungurean, ioan; gaitan, nicoleta cristina title: a software architecture for the industrial internet of things—a conceptual model date: - - journal: sensors (basel) doi: . /s sha: doc_id: cord_uid: d tbf a the internet of things (iot) is an emerging concept that has revolutionized the use of new technologies in everyday life. the economic impact of iot becoming very important, and it began to be used in the industrial environment under the name of the industrial internet of things (iiot) concept, which is a sub-domain of iot. the iiot changes the way industrial processes are controlled and monitored, increasing operating efficiency. this article proposes a software architecture for iiot that has a low degree of abstraction compared to the reference architectures presented in the literature. the architecture is organized on four-layer and it integrates the latest concepts related to fog and edge computing. these concepts are activated through the use of fog/edge/gateway nodes, where the processing of data acquired from things is performed and it is the place where things interact with each other in the virtual environment. the main contributions of this paper are the proposal and description of a complete iiot software architecture, the use of a unified address space, and the use of the computing platform based on soc (system on chip) with specialized co-processors in order to be able to execute in real-time certain time-critical operations specific to the industrial environment. the internet of things (iot) [ , ] is an emerging concept that changes the interactions of people with things in everyday life. iot [ ] allows the connection of ubiquitous objects/things to the internet in order to provide innovative services that can lead to time-and money-saving in the daily activity of the individuals and that can increase the quality of life [ ] . the iot concept is applied in a wide variety of applications such as smart building [ , ] smart transport [ ] , smart cities [ ] , smart healthcare [ , ] , and smart living with the aim to provide new services and an efficiency of the operating costs. in the design and development of iot applications, existing technologies can be used to activate the iot concept, but new technologies can be developed to help develop the specific applications. initially, the iot concept uses only wireless technologies for connection of the things to the internet but now it can use any available technologies, wired and wireless. in fact, iot can reuse wired or wireless communication technologies that have been used in other types of application in which various devices can connect to a computing system directly or through a gateway. the things/objects form everyday life are brought into the virtual environment through sensors or other methods of acquiring real-data from their environment, such as human-machine interfaces. acquired data or virtual things are transmitted to computing platforms where they are processed, they interact between them, and decisions are made, decisions that can be transmitted to the execution sensors , , of in the design and development of iiot solutions, the main issues and challenges are: • the largest number and great diversity of fieldbuses and devices used in the industrial environment; • the compliance with real-time requirements specific to the industrial applications; • interoperability between fieldbuses and devices; • interoperability between iiot systems. in the specialized literature, several reference architectures for iot and iiot are proposed, but these are abstract models which do not deal with how to integrate things from the industrial environment, especially as in this environment are used communication systems (fieldbuses) with specific capabilities such as real-time monitoring and control of time-critical operations. practical examples are also presented in the specialized literature, but these solutions are designed and developed for a certain type of network and a small number of device types. this paper proposes a software architecture for iiot that integrates the concepts of fog/edge computing and fieldbuses specific to the industrial environment. the proposed architecture is organized in layers and it exploits the performances of the data processing at the edge of the network to reduce the bandwidth used for the cloud connection but also to obtain a shorter response time when dealing with time-critical operations. fog/edge computing can integrate features such as low latency, predictability, hard real-time that are very important for monitoring and controlling time-critical operations specific to the industrial environment. furthermore, the architecture contains a virtual environment used for interaction and data exchange between the virtual and/or real objects/things. unlike the iiot architectures presented in the specialized literature, the proposed iiot architecture aims to integrate many fieldbuses into the system. in addition, by using a device description language, they can be added to the system through the plug and play method. the proposal and description of complete iiot software represents the main contribution of this paper. the conceptual model for the iiot system contains some operational and implementation details that are useful in developing an iiot solution based on the proposed architecture. the strengths of the proposed architecture are: • versatility-it can be adapted to the requirements and configuration of different industrial environments; • real-time capabilities-using system on chip (soc) systems with specialized processors for the design and development of the fog nodes; • the integration of several fieldbuses at the same time; • interoperability between different fieldbuses by building a unified address space; • interoperability between iiot systems by using a standardized middleware system. this paper is structured as follows: section presents some iiot architectures presented in the specialized literature and some companies, and the motivation of the proposed iiot architecture. section describes the proposed iiot architecture with some discussions related the implementation. section presents some discussions related to the security issues addressed by the proposed iiot architecture. section presents a comparison between proposed architecture and the architectures analyzed in section . the conclusions are drawn in section . an iiot reference architecture is the abstractization of an iiot architecture that allows the identification of general components and implementation challenges. in the specialized literature, more reference architectures are proposed for iiot. in june , the industrial internet consortium (iic) released version . of the reference architecture for the iiot [ ] . it is focused on different viewpoints (business, usage, functional, and implementation viewpoints). the implementation viewpoint presents the system components and the technologies that can be used to implement the functionalities described in the functional viewpoint. furthermore, the implementation viewpoint contains a technical description of the architecture components, including interfaces, protocols, behaviors, and other properties [ ] , but it does not contain information related to the integration of the fieldbuses and devices from the industrial environment. another important iiot reference architecture is the reference architectural model industry . (rami . ) [ , ] . this is based on a -d model where the axes are live cycle, value stream, and hierarchy levels, and it is service-oriented architecture. the hierarchy level of the architecture is organized in the following levels: product, field devices, control devices, station, work centers, enterprise, and connected world. the value stream level of the architecture is related to the different component functionalities, it includes a communication layer and it is organized the following layers with a high level of abstraction: asset, integration, communication, information, functional, and business. other iot reference architectures are organized on layers, such that defined by the international telecommunication union (itu) that consists of four layers: devices, network, service support and application support, and application [ ] . the device layer includes the device capabilities for interaction for the communication network and gateway capabilities for supporting multiple communication networks. the network layer includes networking and transport capabilities, service support, and the application support layer includes the generic and specific support for the application layer. in [ ] , the authors proposed a case study on the growth of big data in iiot systems, a classification of key concepts, and a presentation of key frameworks and continued with the presentation of future technologies, opportunities, and challenges. these researches concluded that augmented reality, iot devices, cyber-physical systems and industry . big data and analytics (bda) platforms are at an early stage in iiot systems and that solutions should be found in developing new standards that allow interoperability between various platforms but also processing capability of the end-to-end applications for concentric computing systems. a solution to improve the blockchain scalability of iiot systems by guaranteeing system security, latency but also decentralization, that was proposed in the article [ ] have led to performance optimization with a deep reinforcement learning technique (drl). the authors obtained results that demonstrated they can obtain a better efficiency compared to the basic parameters of the system. in another research paper [ ] , the authors concluded based on conducted researches that the success of iiot can be hindered for different reasons such as challenging collaboration between various heterogeneous iiot systems, efficient data management, large, solid, and flexible data technologies, iiot protocols, operating systems, reliable iiot systems but also the coexistence of wireless technologies. in addition to a survey of existing definitions of iiot in [ ] , the authors also present a proposal for a new definition of what the iiot means. the authors also present an analysis of a framework for iiot devices based on security-related issues surrounding iiot but as well as an analysis of the relationships between cyber-physical systems and industry . . a secure fog-based iiot architecture by suitably plugging a number of security features that reduce the trust and burden on the cloud and resource-constrained devices, but also that reduce latency in decision making that improves the performance, is presented in another research paper from the specialized literature [ ] . also, the authors demonstrated that by offloading several computationally intensive tasks to the fog nodes, the battery life of the resource-constrained of end devices is greatly saved. the validation of the architecture was demonstrated through theoretical analyses, practical experiments, but also through simulation and testbed implementation. in [ ] , an iiot application for a sewage treatment plant is proposed. this iiot solution uses control station systems based on the stmicroelectronics stm microcontrollers to update the automation system and to activate the iiot concept. basically, the stm -based control stations act as gateways between field devices connected to fieldbuses and their connection within the local network and further to the internet for remote monitoring and cloud connections. the monitoring operations can be performed remotely from pcs or mobile devices such as smartphones and tablets. the authors concluded that this solution provide the real-time performances and reliability required for monitoring and controlling the sewage treatment plant. in [ ] , the authors define an architecture organized on four layers: sensing layer, networking layer, service layer, and interface layer. in [ ] an architecture is proposed that contains the physical layer, transport layer middleware layer, and application. an architecture with three layers (perception/sensing layer, network layer, and service/application layer) is proposed in [ , ] . all these reference architectures have a high level of abstraction and do not provide details related to the development of the iiot solutions based on these reference architectures. a trust framework for automated guided vehicles (agvs) in smart factories is presented in [ ] . the authors defined a trust measure for avg, designed the trust framework, and presented a set of experiments based on this framework, but they did not provide details related to the communication with the avgs and how the data are acquired from the avgs and integrated into the proposed framework. a software architecture for iiot is presented in [ ] and updated in [ ] . the proposed architecture is organized on levels: things, data provider, middleware, and applications. the interaction between things is performed at the application level, the upper level in the architecture. in this paper, we propose an evolution of the architecture presented in [ , ] by including new technologies such as fog/edge computing concepts. furthermore, now things interact with each other at a lower layer, and at the application/service layer the services necessary for the development of applications are provided, such as monitoring and control specific to the industrial environment or applications for manufacturing tracing and monitoring. in addition, the proposed architecture provides implementation details, not being an abstract architecture as are most of reference architectures from the specialized literature. regarding soc systems that can be used for iot and iiot, in [ ] the authors present a comparison between several single-board computers that can be used for iiot solutions, namely raspberry pi , beaglebone black, banana pi, and odroid-c . of these, the most popular system is the raspberry pi, but in terms of the peripherals provided, the beaglebone black is the best. in [ ] , the authors propose a soc architecture for the industrial internet of things. the soc architecture is implemented on an field programmable gate array (fpga) system with two arm processors that execute the linux operating system. the fpga gates are used to implement real-time operations specific to the industrial environment. a detailed survey related to the security in iiot is presented in [ ] , where the authors performed a systematic review of the literature and identified the security requirements for iiot. these security requirements are related to the confidentiality, integrity, availability (cia) triad, authentication, access control, maintainability, resilience, data security and data sharing, security monitoring, and network security. additionally, the authors included a description of how fog computing can address these requirements and identify some research opportunity for the use of the fog computing for iiot security, namely fog-enabled authentication, fog-enabled access control, fog-enabled maintainability, fog-enabled resilience, and fog-enabled data security and data sharing. a survey for the security of the iiot protocols is presented in [ ] . they focused on the main protocols and standards used in iiot (from middleware systems to fieldbuses) and identify the security vulnerability with the common vulnerability scoring system (cvss). they proposed a vulnerability analysis framework (vaf) that can be used to analyze vulnerabilities for iiot systems. data security in edge-based iiot is investigated in [ ] . authors identify four main challenges reliable data storage, convenient data usage, efficient data search, and secure data deletion, and they proposed a cloud-fog-edge device storage framework for iiot that addresses the challenges identified. in this section, we propose an iiot software architecture that can be used in the development of the practical iiot solutions. before describing the proposed architecture, we will discuss the main challenges in the design and development of the iiot architectures. the main requirements for the development of an iiot architecture are represented by modularity, scalability, and interoperability between the different technologies used in the industrial environment. while iot is human-centered, iiot is machine-oriented [ ] and is based on existing technologies and devices for reliability. the iiot systems are a special class of iot systems that have additional requirements related to latency, real-time capabilities, security, reliability, and safety. these requirements result from the specific nature of the industrial environment where time-critical activities can be carried out. perhaps the most important challenge in developing an iiot architecture is data security because most data in the industrial environment are not public. for this reason, there must be a clear policy regarding restricted access to data only from trusted entities. another challenge in developing an iiot system is the large number of industrial networks (fieldbuses) used in the industrial environment, each with its own characteristics related to the latency, real-time capabilities, transmission rate, and line protocol. all these fieldbuses should coexist in an iiot system. furthermore, in the industrial environment, there are many heterogeneous systems (ex. devices, sensors, programmable logic controllers, and human-machine interface) that must collaborate within an iiot system. the heterogeneous nature of the systems results from the data source data, data stream, data storage, and processing requirements. another challenge is the performance related to soft and hard real-time capabilities. usually, in the industrial environment, critical activities are carried out for missions and safety with imposed requirements regarding timing and reliability. the iiot architecture proposed in this article addresses all these challenges. the proposed architecture is organized on layers and includes the edge/fog computing paradigm applied for the iiot and support for the real-time and low latency requirements specific to the industrial field. figure presents the proposed software architecture organized on four layers: sensing/things, data provider, fog computing, and applications/service. in addition, fog computing and application/service layers can be connected to a cloud server to retrieve and save data. each layer includes services for management and security through which the user can configure the functionality and security implementation at the layer-level. in figure the management and security capabilities are presented vertically because they transcend all layers of architecture. sensors , , x for peer review of development of an iiot architecture are represented by modularity, scalability, and interoperability between the different technologies used in the industrial environment. while iot is human-centered, iiot is machine-oriented [ ] and is based on existing technologies and devices for reliability. the iiot systems are a special class of iot systems that have additional requirements related to latency, real-time capabilities, security, reliability, and safety. these requirements result from the specific nature of the industrial environment where time-critical activities can be carried out. perhaps the most important challenge in developing an iiot architecture is data security because most data in the industrial environment are not public. for this reason, there must be a clear policy regarding restricted access to data only from trusted entities. another challenge in developing an iiot system is the large number of industrial networks (fieldbuses) used in the industrial environment, each with its own characteristics related to the latency, real-time capabilities, transmission rate, and line protocol. all these fieldbuses should coexist in an iiot system. furthermore, in the industrial environment, there are many heterogeneous systems (ex. devices, sensors, programmable logic controllers, and human-machine interface) that must collaborate within an iiot system. the heterogeneous nature of the systems results from the data source data, data stream, data storage, and processing requirements. another challenge is the performance related to soft and hard real-time capabilities. usually, in the industrial environment, critical activities are carried out for missions and safety with imposed requirements regarding timing and reliability. the iiot architecture proposed in this article addresses all these challenges. the proposed architecture is organized on layers and includes the edge/fog computing paradigm applied for the iiot and support for the real-time and low latency requirements specific to the industrial field. figure presents the proposed software architecture organized on four layers: sensing/things, data provider, fog computing, and applications/service. in addition, fog computing and application/service layers can be connected to a cloud server to retrieve and save data. each layer includes services for management and security through which the user can configure the functionality and security implementation at the layer-level. in figure the management and security capabilities are presented vertically because they transcend all layers of architecture. an important aspect of the proposed architecture is the integration of fog/edge computing concepts. thus, in the proposed architecture there are fog/edge/gateway nodes that integrate the fog/edge and data provider layers presented in figure . there may be more such nodes in the iiot architecture implementation. in this context of fog/edge/gateway nodes, another perspective of the proposed architecture is presented in figure (without application/service layer). in the iiot architecture, there can be several nodes, each node connects to one or more fieldbuses, and these nodes implement the fog and data provider layers for these fieldbuses. these layers are software packages implemented for a specific device based on a microprocessor or a soc (system on chip) that runs on a high-level operating system such as windows, linux, or android. these devices have peripherals such as usb (universal serial bus), can (controller area network) bus, uart (universal asynchronous receiver-transmitter), spi (serial peripheral interface), and i c (inter-integrated circuit) that allows the interfacing with fieldbuses used in the industrial environment. for example, through usb we can connect to canopen fieldbus using a usb-can interface, the can bus can be used to connect to canopen fieldbus, the uart can be used to connect to profibus or modbus fieldbus via a uart-rs interface, and spi/i c can be used to connect to canopen fieldbus via an spi-can or i c-can interface. proposed architecture is presented in figure (without application/service layer). in the iiot architecture, there can be several nodes, each node connects to one or more fieldbuses, and these nodes implement the fog and data provider layers for these fieldbuses. these layers are software packages implemented for a specific device based on a microprocessor or a soc (system on chip) that runs on a high-level operating system such as windows, linux, or android. these devices have peripherals such as usb (universal serial bus), can (controller area network) bus, uart (universal asynchronous receiver-transmitter), spi (serial peripheral interface), and i c (inter-integrated circuit) that allows the interfacing with fieldbuses used in the industrial environment. for example, through usb we can connect to canopen fieldbus using a usb-can interface, the can bus can be used to connect to canopen fieldbus, the uart can be used to connect to profibus or modbus fieldbus via a uart-rs interface, and spi/i c can be used to connect to canopen fieldbus via an spi-can or i c-can interface. these fog/edge/gateway nodes can be implemented on various computing systems such as beaglebone ai, raspberry pi , industrial pcs or other types of computing system that allow connection to industrial networks and sufficient computing power for data processing. for the internet connection, g modules can also be used to have a high bandwidth regardless of the physical location of the node. the g is a technology that is beginning to be adopted by most gsm (global system for mobile communications) service providers. the g modules are the hardware support for the internet connection and, for this reason, we will not insist on this aspect. from the software point of view, how the connection to the internet is made is transparent. the following sections will be described in detail for each layer of the four layers in which the proposed architecture is organized. these fog/edge/gateway nodes can be implemented on various computing systems such as beaglebone ai, raspberry pi , industrial pcs or other types of computing system that allow connection to industrial networks and sufficient computing power for data processing. for the internet connection, g modules can also be used to have a high bandwidth regardless of the physical location of the node. the g is a technology that is beginning to be adopted by most gsm (global system for mobile communications) service providers. the g modules are the hardware support for the internet connection and, for this reason, we will not insist on this aspect. from the software point of view, how the connection to the internet is made is transparent. the following sections will be described in detail for each layer of the four layers in which the proposed architecture is organized. at this level, there are physical devices, sensors, actuators, and plcs (programmable logic controllers) specific to the industrial environment. these devices are connected to a specific fieldbus and can send and/or receive data from this fieldbus. examples of such fieldbuses are canopen, modbus, profibus, profinet, ethercat, and wirelesshart. at the upper layer, there must be different adapters/interfaces for communication with devices connected to fieldbuses. for example, usb-can interfaces or soc-based systems with a can port can be used for a canopen fieldbus. also, at the upper layer, there must be extensions through which the devices connected to the network can be configured, depending on the characteristics of each fieldbus. due to the nature of the industrial environment, the complexity, and the requirements for the execution of time-critical activities, some configuration operations of the field devices must be performed manually. once this operation is performed, the fieldbus can operate without major configuration changes. the purpose of this architecture is not to go into detail, it is considered that the networks are configured and at the top layer drivers specific to each fieldbus and the soc on which the fog/edge/gateway node is implemented can be designed and developed. this layer is defined in the proposed iiot architecture only from the hardware point of view, from the software point of view, at the upper level there must be modules through which the fieldbuses and devices connected to these fieldbuses can be configured. the purpose of this layer is to acquire data from devices connected to fieldbuses and store it in a buffer memory in order to be transmitted to the next layer or to send data from the memory buffer to the fieldbuses when data is received from the upper layer. it must be specified that the data flow is bidirectional, in the sense that the values from different sensors/devices can be acquired from the fieldbuses or commands that can be sent to the actuators connected on the fieldbuses. as we specified previously, this layer together with the fog/edge layer is designed and developed to be executed on fog/edge nodes. these nodes can be represented by pcs or soc-based computing systems that have features and can help to achieve real-time and reliable communication with fieldbuses. a system with sufficient computing power to process the data and peripherals needed to connect to the fieldbus that is monitored by the fog node must be chosen. the software architecture of a fog/edge/gateway node is presented in figure . it can be observed from figure that the data provider level connects directly to the fieldbuses, and a fog/edge node can connect to one or more fieldbuses depending on the iiot application and connection capabilities of the fog/edge node. due to the diversity of fieldbuses that can be used in the industrial environment, this layer contains a driver for each type of fieldbus that can be integrated into the proposed iiot architecture. these drivers implement the fieldbus protocol stacks for which it is designed and developed, and the connection method/way for fieldbuses must be considered, using an interface or directly throughout a peripheral (e.g., for a canopen fieldbus we can use a can port, a can-ethernet interface, or a can-usb interface, depending on the connectivity provided by the soc system used). from figure , we can see that the data provider layer includes the drivers for the fieldbuses. a node can instantiate one or more drivers depending on the capabilities of the fog system and the configuration of the industrial environment that is included in the virtual industrial environment internet of things. we can design and develop drivers for mature networks that exist in the industrial environment for a long time or on new communication systems such as narrowband iot or lte cat-m based on g technology. in order to meet the real-time requirements specific to the industrial environment, the fog nodes can be designed and developed on soc systems with specialized coprocessors for the communication implementation with fieldbuses. for example, we can use soc systems that have one or more cores based on the arm cortex ax architecture for the execution of the complex operations and specialized processors (e.g., based on arm cortex mx architectures or other optimized architecture as proposed in [ ] ) for the communication in real-time with the fieldbuses. thus, the tasks with specific real-time requirements for time-critical operations specific to the industrial environment are executed on the specialized coprocessor where these requirements can be fulfilled and guaranteed. the basic idea is to use multiprocessor soc systems that have processors on which linux embedded can be executed for the execution of non-critical operations in the best-effort (soft real-time) manner and specialized coprocessors on which a real-time operating system [ ] can be executed for the implementation of the protocol stack and the execution of critical time operations. with this solution, low and predictable latencies can be achieved at the occurrence of events or at the execution of the critical operations by decoupling the software part with real-time requirements from the software part that does not have strict real-time requirements. in the case of these soc systems, the drivers have a part that runs on specialized coprocessors (real-time cores) and a part that runs on the main processors (application cores). the software part from the specialized coprocessors implements the protocol stack, and the software part from the main processors works as a wrapper that saves the unit data in a buffer memory from where it will be taken and processed by other software modules. in terms of implementation, this solution involves the development of software for asymmetric multi-processing, and great care must be taken at the implementation of the communication between processors in order not to introduce performance bottlenecks. fog nodes can also be developed on computing systems without specialized co-processors, but in this case, we can obtain at most soft real-time capabilities due to the limitations of windows and linux operating systems [ ] (including real-time extensions such as rtai (real-time application interface) or preempt_rt). the specialized co-processors provide support for the development of small software modules with strict hard real-time capabilities and predictable latencies. each fieldbus will have an associated address space that defines the things connected to the fieldbus. one of the main goals of the drivers is to build the address space in a unitary way, hiding the specific details of each fieldbus and connected devices, such as the addressing mode or the in order to meet the real-time requirements specific to the industrial environment, the fog nodes can be designed and developed on soc systems with specialized coprocessors for the communication implementation with fieldbuses. for example, we can use soc systems that have one or more cores based on the arm cortex ax architecture for the execution of the complex operations and specialized processors (e.g., based on arm cortex mx architectures or other optimized architecture as proposed in [ ] ) for the communication in real-time with the fieldbuses. thus, the tasks with specific real-time requirements for time-critical operations specific to the industrial environment are executed on the specialized coprocessor where these requirements can be fulfilled and guaranteed. the basic idea is to use multiprocessor soc systems that have processors on which linux embedded can be executed for the execution of non-critical operations in the best-effort (soft real-time) manner and specialized coprocessors on which a real-time operating system [ ] can be executed for the implementation of the protocol stack and the execution of critical time operations. with this solution, low and predictable latencies can be achieved at the occurrence of events or at the execution of the critical operations by decoupling the software part with real-time requirements from the software part that does not have strict real-time requirements. in the case of these soc systems, the drivers have a part that runs on specialized coprocessors (real-time cores) and a part that runs on the main processors (application cores). the software part from the specialized coprocessors implements the protocol stack, and the software part from the main processors works as a wrapper that saves the unit data in a buffer memory from where it will be taken and processed by other software modules. in terms of implementation, this solution involves the development of software for asymmetric multi-processing, and great care must be taken at the implementation of the communication between processors in order not to introduce performance bottlenecks. fog nodes can also be developed on computing systems without specialized co-processors, but in this case, we can obtain at most soft real-time capabilities due to the limitations of windows and linux operating systems [ ] (including real-time extensions such as rtai (real-time application interface) or preempt_rt). the specialized co-processors provide support for the development of small software modules with strict hard real-time capabilities and predictable latencies. each fieldbus will have an associated address space that defines the things connected to the fieldbus. one of the main goals of the drivers is to build the address space in a unitary way, hiding the specific details of each fieldbus and connected devices, such as the addressing mode or the message format. thus, it is possible to obtain the aggregation and processing of the data acquired from different fieldbuses to which the fog node is connected through a unique software interface, ensuring interoperability between different fieldbuses. the structure of this address space is organized in a tree and it is stored in extensible markup language (xml) files that are loaded when the software module that implements the data provider layer is instantiated. it is considered that a fieldbus can have several devices/sensors/actuators connected, and each device can have several things such as a temperature, a pressure, a level, and so on. each thing is characterized by several attributes such as data type, quality, timestamp, and access type. the data type can be numeric, string, and logic, and the quality can be good or bad. the timestamp represents the time and date of the last update of the things, and the access type can be read-only, write-only, and read-write. a sensor will have the read-only access type and an actuator can have write-only or write-read access type. an example of a space address for a device is presented in figure . thus, we have a device that has n analog inputs and n digital outputs. analog inputs can be connected to sensors such as thermocouples, a resistance thermometer, pressure transducers, etc. and digital outputs that can be connected to execution elements via relays. the address space for this device built in the fog node contains n things for the n analog inputs and n things for the n digital outputs, as illustrated in figure . sensors , , x for peer review of message format. thus, it is possible to obtain the aggregation and processing of the data acquired from different fieldbuses to which the fog node is connected through a unique software interface, ensuring interoperability between different fieldbuses. the structure of this address space is organized in a tree and it is stored in extensible markup language (xml) files that are loaded when the software module that implements the data provider layer is instantiated. it is considered that a fieldbus can have several devices/sensors/actuators connected, and each device can have several things such as a temperature, a pressure, a level, and so on. each thing is characterized by several attributes such as data type, quality, timestamp, and access type. the data type can be numeric, string, and logic, and the quality can be good or bad. the timestamp represents the time and date of the last update of the things, and the access type can be read-only, write-only, and read-write. a sensor will have the read-only access type and an actuator can have write-only or write-read access type. an example of a space address for a device is presented in figure . thus, we have a device that has n analog inputs and n digital outputs. analog inputs can be connected to sensors such as thermocouples, a resistance thermometer, pressure transducers, etc. and digital outputs that can be connected to execution elements via relays. the address space for this device built in the fog node contains n things for the n analog inputs and n things for the n digital outputs, as illustrated in figure . the data acquired from the fieldbuses are saved in a buffer memory, where they will be retrieved and transmitted to the upper layer. the buffer memory is allocated according to the definition found in the xml configuration files associated with each fieldbus. the user has the possibility to configure the communication parameters on the network as well as the introduction or elimination of new devices. this address space is used by the upper layer to read/transmit data to the thing. a thing can be accessed through a path in the form of "network name. device name. thing name". each device will have associated an xml file that defines all the things and attributes of the things it makes available. these files contain generic names for things that can be replaced with aliases in the network setup process. each fieldbus will have an associated xml file in which network-specific configurations (such as communication speed) and the list of devices connected to the fieldbus along with the associated xml files are stored. a fog/edge node has an associated xml file in which the list of fieldbuses to which it connects with the associated xml files are stored. these files are automatically updated during the configuration of fieldbuses to which the fog node connects and loaded when the software module that implements the functionalities of the fog node is executed. the data acquired from the fieldbuses are saved in a buffer memory, where they will be retrieved and transmitted to the upper layer. the buffer memory is allocated according to the definition found in the xml configuration files associated with each fieldbus. the user has the possibility to configure the communication parameters on the network as well as the introduction or elimination of new devices. this address space is used by the upper layer to read/transmit data to the thing. a thing can be accessed through a path in the form of "network name. device name. thing name". each device will have associated an xml file that defines all the things and attributes of the things it makes available. these files contain generic names for things that can be replaced with aliases in the network setup process. each fieldbus will have an associated xml file in which network-specific configurations (such as communication speed) and the list of devices connected to the fieldbus along with the associated xml files are stored. a fog/edge node has an associated xml file in which the list of fieldbuses to which it connects with the associated xml files are stored. these files are automatically updated during the configuration of fieldbuses to which the fog node connects and loaded when the software module that implements the functionalities of the fog node is executed. an important part of the xml file associated with a device is the part that describes how the value of a thing is read or written. this is translated into messages that can be transmitted and read on the fieldbuses, depending on the characteristics of each network. this means that the xml file must also be interpreted by the driver to take over the commands/messages specific to each thing. depending on the type of network, an acquisition cycle can be defined by which data in the buffer are updated periodically and prioritized depending on the criticality of the monitored data and the user configuration. an example of such an acquisition cycle is presented in [ ] . at this layer, the local processing of the data acquired from the industrial fieldbuses is performed and it is the place where things interact with each other. practically at this level, the operations specific to the edge and fog computing concepts are performed. thus, in order to interact between the things, it is proposed to use the publisher-subscriber paradigm, in the sense that a thing publishes its value periodically or when an event occurs, and other things subscribe to this value. there are two types of thing: physical things that are associated with things defined at the data provider layer and virtual things that can be defined at this level and represent the result of processing data from physical and/or virtual things. the virtual things subscribe to one or more values published by the other things, perform certain processing on the read values (mathematical operations, mathematical functions, aggregation functions, etc.) and the obtained result is published as a new value. from the functional point of view, no distinction is made between virtual things and physical things. things can connect to one or more things in a virtual communication environment that is based on the publish-subscribe paradigm and that performs some processing on data and can publish its value. a graphical representation of the virtual publishing-subscriber environment is shown in figure . some things only publish data, for example, things associated with a sensor such as temperature sensor, pressure sensor, and things that subscribe to one or more values but do not publish any values, such as things that are associated with actuators from the industrial environment. internally, things perform a wide range of data processing tasks such as mathematical operations, mediation, aggregation, or other mathematical functions. there may be cloud things that subscribe to certain data and save it to a remote or local cloud server, or things that connect to data and save it to a remote or local database. these types of thing perform certain cloud services on the fog node, such as storage services. the interaction between things can be undertaken at the level of a fog/edge node, but we also want the interaction between things from different fog/edge nodes. in order to achieve requirements related to the interaction between things, a middleware system that provides support for publish-subscribe communication can be used. at the moment, there are several middleware systems that operate on the publisher-subscriber paradigm, such as the advanced message-queuing protocol (amqp) [ ] , message-queuing telemetry transport (mqtt) [ ] , and data distribution service (dds) for real-time systems [ ] . we focused on middleware systems that are based on the publish-subscribe paradigm because the iiot system will be much more versatile. there are other important middleware systems that are widely used in the development of iot systems [ ] such as xmpp (extensible messaging and presence protocol) or coap (constrained application protocol) but these middleware systems have not been considered because they are not based on the publish-subscribe paradigm. from the middleware systems that are based on the publish-subscribe paradigm, we suggest the use of dds for real-time systems is chosen because it was designed specifically for real-time systems. in order to argue for the selection of the dds protocol, the study by [ ] can be analyzed, where the http (s), coap, mqtt, amqp, xmpp, and dds are identified as the main middleware systems based on messages that can be used in the development of the iot systems. from these middleware systems, mqtt and amqp use a broker that can be a bottleneck point. the dds is the only one that uses a bus-like architecture (mqtt uses tree architecture and amqt uses star architecture), has quality of service (qos) levels (mqtt and amqp uses qos levels), is data-centric, and implements a security mechanism like tls, dtls, and dds security. also, the data distribution is very versatile with the possibility to achieve to , to n, and n to [ ] . by using a standard middleware system, interoperability can be ensured between the iiot systems and with other systems could take data from the iiot system. shown in figure . some things only publish data, for example, things associated with a sensor such as temperature sensor, pressure sensor, and things that subscribe to one or more values but do not publish any values, such as things that are associated with actuators from the industrial environment. internally, things perform a wide range of data processing tasks such as mathematical operations, mediation, aggregation, or other mathematical functions. there may be cloud things that subscribe to certain data and save it to a remote or local cloud server, or things that connect to data and save it to a remote or local database. these types of thing perform certain cloud services on the fog node, such as storage services. thus, a dds domain is defined for the architecture, which is divided into several dds partitions, a dds partition corresponding to a fog/edge node. dds defines a unit of information as a topic. thus, for each thing defined at the data provider level, a topic will be associated, which will have the unique name of the access path defined at the previous level. in addition to these things, virtual things are also defined, things that can subscribe to different dds topics, and different processing is performed on the received data. all these things are created by the user in the configuration process based on predefined things that are instantiated and that subscribe to one or more topics. perhaps the most important predefined virtual thing is the expression thing in which certain mathematical operations and functions can be applied to the data with which it is subscribed. thus, the user can create virtual things according to his needs and in accordance with the application of the industrial environment that is integrated into the iiot architecture. due to the solution adopted for the interaction between things, the architecture is very versatile, being able to be used in a wide range of industrial applications. in the case of using the dds middleware system, the publishing-subscription environment from figure can be represented as in figure . the connections between things are not made only locally, but in the dds domain, which is defined at the level of the iiot system, things from the dds domain can be from the same node or from remote fog/edge nodes in a local network or on the internet. as can be seen from figure , in the case of dds middleware, each topic is associated with different qos policies. these policies can be related to data availability, data delivery, data timeliness, and resources. for data availability, we can control how data is available within a domain through the durability, lifespan, and history policies. for data delivery, we can control how data is delivered and exclusive rights for data through the presentation, reliability, partition, destination_order, and ownership policies. for data timeliness, we can control the timeliness for the data through the deadline, latency_budget, and transport_priority policies. through the resources policies, we can control the computing and network resources. all of these policies are described in the dds specifications and they are included in the implementations of these specifications [ ] [ ] [ ] . as can be seen from figure , in the case of dds middleware, each topic is associated with different qos policies. these policies can be related to data availability, data delivery, data timeliness, and resources. for data availability, we can control how data is available within a domain through the durability, lifespan, and history policies. for data delivery, we can control how data is delivered and exclusive rights for data through the presentation, reliability, partition, destination_order, and ownership policies. for data timeliness, we can control the timeliness for the data through the deadline, latency_budget, and transport_priority policies. through the resources policies, we can control the computing and network resources. all of these policies are described in the dds specifications and they are included in the implementations of these specifications [ ] [ ] [ ] . the application/services layer is the place where industry-specific applications can be developed, such as those for remote monitoring and control. we can develop a supervisory control and data acquisition (scada) type application by subscribing and publishing in the dds domain where the things interact with each other. at this layer, it can be designed and developed software solutions/applications that can monitor things in the industrial environment and send commands to these things. also, based on the data saved on the fog nodes or in the cloud, different reports and graphical representations can be made regarding the evolution of the data over time and of the industrial process. at this layer and depending on the capabilities of the fog nodes, on each fog node a web server can be developed that allows the visualization of the values published by the fog node in the dds domain. the web server can also have an extension that provides the facility to configure the fog/edge node remotely, especially since fog/edge nodes can operate without the graphical interface. at this layer, an hmi application can be designed and developed like that presented in [ ] and which can be used to graphically visualization of data acquired from the industrial environment and send commands in the industrial environment through the graphical interface. the application/services layer is the place where industry-specific applications can be developed, such as those for remote monitoring and control. we can develop a supervisory control and data acquisition (scada) type application by subscribing and publishing in the dds domain where the things interact with each other. at this layer, it can be designed and developed software solutions/applications that can monitor things in the industrial environment and send commands to these things. also, based on the data saved on the fog nodes or in the cloud, different reports and graphical representations can be made regarding the evolution of the data over time and of the industrial process. at this layer and depending on the capabilities of the fog nodes, on each fog node a web server can be developed that allows the visualization of the values published by the fog node in the dds domain. the web server can also have an extension that provides the facility to configure the fog/edge node remotely, especially since fog/edge nodes can operate without the graphical interface. at this layer, an hmi application can be designed and developed like that presented in [ ] and which can be used to graphically visualization of data acquired from the industrial environment and send commands in the industrial environment through the graphical interface. security is an important issue in any application distributed over the internet. in the case of iiot, it is even more important because non-public data is transmitted over the internet, and access to this data must be granted on the basis of clear security policies. at the level of things/sensing, security is ensured by the capabilities of each fieldbus. usually, in these fieldbuses, the security mechanisms are not implemented because not everyone has physical access to them. from a security point of view, the risks at this level are small, especially if physical access to them is restricted. in addition, devices connected to networks can have different security mechanisms in place. from the point of view of the current iiot architecture, the security at this level depends on the configuration made at the level of each fieldbus and filed device used, and no new security mechanisms are implemented because this level does not involve the development of software modules. at higher layers, the security mechanisms must be implemented at the fog node level. remote access to fog nodes must be restricted, thus avoiding any security holes that may occur. the operating system must also be kept up to date with the latest security updates. configuration xml files can be encrypted. this does not degrade the performance because they are decrypted when starting the software modules to implement the data provider and fog/edge computing layers, and they are saved in encrypted mode following the configuration operations performed by the user. the biggest security issue can be when publishing data in the publisher-subscriber environment provided by the dds domain defined for the current iiot architecture because those messages can be intercepted when they are transmitted over the internet. in , the latest security specifications for the dds protocol were published. these specifications are also implemented in the opendds implementation, an open-source implementation that can be used for the design and development of fog/edge nodes. the specifications provide the plugins for authentication, access control, logging, data tagging, cryptography, and certificates [ ] . through these facilities, the data are transmitted encrypted over the internet ensuring a high degree of security regarding confidentiality, integrity, and availability. table presents a comparison between the proposed architecture and those described in the related work section. a more detailed quantitative or qualitative analysis between the proposed iiot architecture and the existing ones requires more information than those presented in the literature. moreover, we propose a conceptual model and the experimental results will be implemented and presented in future research articles. for example, we can use beaglebone black for implementation. this is based on the sitara™ am x soc with arm cortex a processors and two programmable real-time unit subsystem and industrial communication subsystems (pru-icss). the arm cortex a processors use a linux-based operating system. one pru-icss can be used to design and implement a driver for the modbus rtu (remote terminal unit) fieldbus and the other pru-icss can be used to design and implement a driver for the canopen fieldbus. the data provider, fog computing, and applications/service layers can be designed and implemented to be executed on the arm cortex a processors under linux. by using many of beaglebone black systems, we can build a complex iiot system that can be geographically distributed. the proposed iiot software architecture the use of the soc-based computing platform with specialized co-processors in order for fog/edge nodes. four layers: sensing/things, data provider, fog/edge, applications/services fog/edge nodes yes. it is proposed an unified address space and a driver for each field bus integrated this paper proposed a conceptual model software architecture for the iiot. the proposed architecture is a conceptual model with a low level of abstraction, being oriented towards the design and development of real-time applications for the iiot. among the strengths of the proposed architecture, we can mention: the integration in the system of many fieldbuses used in the industrial environment, addressing the specific requirements of the industrial environment such as real-time and low-latency capabilities, and the integration in the architecture of the concepts of fog and edge computing. also, the use of the soc-based computing platform with specialized co-processors is proposed in order to be able to execute in real time certain time-critical operations specific to the industrial environment. the specialized co-processors provide support for the development of small software modules with strict hard real-time capabilities and predictable latencies while the main processors provide support for the development of software modules with soft real-time capabilities. we can conclude that the conceptual model is feasible for a combination of hard and soft real-time systems with hard real-time constraints provided by the specialized co-processors. the fog/edge computing can integrate features such as low latency, predictability, and hard real time that are very important for monitoring and controlling time-critical operations specific to the industrial environment. furthermore, the architecture contains a virtual environment used for the interaction and data exchange between the virtual and/or real objects/things located on the same fog / edge node or on different fog/edge nodes. as future work, we intend to implement the proposed architecture on beaglebone black and beaglebone ai systems. privacy enhancing technologies in the internet of things: perspectives and challenges internet of things forensics: recent advances, taxonomy, requirements, and open challenges industrial internet of things: challenges, opportunities, and directions energy efficiency in smart buildings: iot approaches monitoring and control system for smart buildings based on opc ua specifications technology selection for iot-based smart transportation systems smart city iot services creation through large-scale collaboration an internet of things framework for remote monitoring of the healthcare parameters speech analysis for medical predictions based on cell broadband engine making internet of things real the growth in connected iot devices is expected to generate . zb of data in a smart manufacturing service system based on edge computing, fog computing, and cloud computing edge computing in the industrial internet of things environment: software-defined-networks-based edge-cloud interplay structuring reference architectures for the industrial internet of things customer expectation from industrial internet of things (iiot) deploying fog computing in industrial internet of things and industry . literature review of industry . and related technologies industrial internet of things: recent advances, enabling technologies and open challenges industrial-iot-iiot-market-worth- - -billion-by- -exclusive-report-covering-pre-and-post-covid- -market-analysis-and-forecasts-by-meticulous-research.html the industrial information technology handbook industrial communication systems and their future challenges: next-generation ethernet, iiot, and g narrowband internet of things: a comprehensive study analysis of the cyber-security of industry . technologies based on rami . and identification of requirements an architecture based on rami . to discover equipment to process operations required by products internet protocol aspects and next-generation networks, next generation networks-frameworks and functional architecture models, overview of the internet of things the role of big data analytics in industrial internet of things performance optimization for blockchain-enabled industrial internet of things (iiot) systems: a deep reinforcement learning approach the industrial internet of things (iiot): an analysis framework das bit, s. a secure fog based architecture for industrial internet of things and industry . renovation of automation system based on industrial internet of things: a case study of a sewage treatment plant internet of things in industries: a survey efficient naming, addressing and profile services in internet-of-things sensory environments. ad hoc netw rfid technology and its applications in internet of things (iot) an overview of the internet of things for people with disabilities a trust-based team formation framework for mobile intelligence in smart factories an iot architecture for things from industrial environment a middleware based architecture for the industrial internet of things, ksii transactions on internet & information systems analysis of single-board computers for iot and iiot solutions in embedded control systems smart sensor: soc architecture for the industrial internet of things a systematic survey of industrial internet of things security: requirements and fog computing opportunities a survey of iiot protocols toward data security in edge intelligent iiot improving interrupt handling in the nmpra timing comparison of the real-time operating systems for small microcontrollers hard real-time computing systems a flexible acquisition cycle for incompletely defined fieldbus protocols industrial shop-floor integration based on amqp protocol in an iot environment vulnerabilities and limitations of mqtt protocol used between iot devices dds-based protocol-compatible communication platform for mining power system the infrastructure used for this work was partially supported by the project "integrated center for research, development and innovation in advanced materials, nanotechnologies, and distributed systems for fabrication and control", contract no. / . . , sectoral operational program for increase of the economic competitiveness co-funded from the european regional development fund. the authors declare no conflict of interest. the funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results. key: cord- -j hphoi authors: schaar, peter title: datenschutz und internet – es ist kompliziert! date: - - journal: informatik spektrum doi: . /s - - - sha: doc_id: cord_uid: j hphoi angesichts der zunehmenden bedeutung des internets für die private, geschäftliche und staatliche kommunikation werden dessen datenschutzdefizite systemrelevant. die damit einher gehenden risken lassen sich durch datenschutzrechtliche vorschriften allein nicht beherrschen. sie müssen ergänzt werden durch informationstechnische sicherungen, die bereits beim systemdesign greifen (privacy by design). zudem müssen das verbraucher- und wettbewerbstrecht und die haftungsregelungen der neuen situation angepasst werden. "datenschutz und internet passen nicht zusammen". leider ist an dieser weit verbreiteten ansicht mehr als ein körnchen wahrheit. das internet hat sich rasant zu einem medium entwickelt, bei dem Überwachung und das sammeln von daten über seine nutzerinnen und nutzer immer mehr raum einnimmt. bei vielen business-modellen steht die kommerzielle verwertung der nutzerdaten sogar im vordergrund. der jeweils angebotene service wird so gestaltet, dass dabei möglichst viele daten generiert werden, die sich auswerten und monetarisieren lassen. die finanzierungsmechanismen der vordergründig "kostenlosen" dienstleistungen sind für die nutzerinnen und nutzer weitgehend intransparent, genauso wie der konkrete beitrag, den ihre daten dabei leisten. personenbezogene daten sind die wichtigste währung, in der wir für die inanspruchnahme digitaler dienste bezahlen, aber es fehlt das preisschild. anbieter berufen sich auf betriebs-und geschäftsgeheimnisse, wenn wir wissen wollen, wie sie mit unseren daten umgehen und welchen mehrwert sie daraus schöpfen. spätestens seit den enthüllungen von edward snowden im jahr ist zudem allgemein bekannt, dass staatliche stellen, allen voran geheimdienste, das netz zur globalen, umfassenden Überwachung nutzen (vgl. [ ] [ ] [ ] ). sie können dabei nahtlos auf die privatwirtschaftliche datenmaximierung aufsetzen und gelangen zu tiefen einblicken in unser verhalten und unsere persönlichkeit. der von shoshana zuboff geprägte begriff "Überwachungskapitalismus" [ ] beschreibt zutreffend unsere heu-peter schaar schaar@eaid-berlin.de europäische akademie für informationsfreiheit und datenschutz, berlin, deutschland tige, zugleich von kommerzialisierung, Überwachung und individuellem kontrollverlust geprägte gesellschaftsordnung. rein technisch stand schon in den frühzeiten des internets nicht die vertraulichkeit im vordergrund. schon in den er-jahren wurde der nachrichtenaustausch über das netz mit dem versenden einer postkarte verglichen, bei der jeder am transport beteiligte, etwa der postbote, nicht nur die absender-und empfängeradressen, sondern auch den inhalt mitlesen kann. und viele erinnern sich noch an die veröffentlichte karikatur, in der einem vor einem bildschirm sitzenden hund die worte in den mund gelegt werden: "on the internet, nobody knows you're a dog" [ ] . weder die vertraulichkeit, noch die integrität und authentizität der über das internet übertragenen informationen waren gesichert. erst in den späten er-jahren öffnete sich das netz allmählich auch nichtuniversitären einrichtungen. dies eröffnete viele chancen für eine technikaktive protestszene, deren ausläufer bis in die gegenwart reichen -wie etwa den deutschen chaos computer club. es dauerte aber nicht lange, bis sich das internet immer weiter von einem nutzerorientierten zu einem von geschäftlichen interessen geprägten medium wandelte. immer stärker in den vordergrund rückt seither die wirtschaftlich motivierte bereitstellung von digitalen informations-und serviceangeboten. deren finanzierung unterschied sich zunächst nicht so sehr von ihren analogen vorbildern, bei denen die kunden für die jeweils in anspruch genommene leistung zu zahlen hatten. das entwickelte hypertext transfers protocol (http) ermöglichte es, physisch verteilte informationen so zu bündeln, dass sie für den nutzer einheitlich auf einer webseite präsentiert werden. damit entstand die möglichkeit, von externen dienstleistern bereitgestellte werbung einzublenden. diese dienstleister -etwa das gegründete und seit zum google-imperium gehörende unternehmen doubleclick -waren damit nicht nur in der lage, anbieterübergreifend werbebotschaften zu platzieren, sondern können darüber hinaus die datenabrufe einzelnen geräten und nutzern der jeweiligen webseiten zuordnen. mithilfe von cookies konnten aus unterschiedlichen bereichen stammende nutzerinformationen zusammengeführt und in profilen gespeichert werden. heute ist die exzessive sammlung aller möglichen daten über unser verhalten nicht mehr die ausnahme, sondern die regel. sie beschränkt sich nicht mehr auf die virtuelle welt sondern erfasst zunehmend auch unser wirkliches leben. bisher analoge geräte werden digitalisiert und vernetzt. der bereich, in dem unser verhalten nicht erfasst, vermessen und beurteilt wird, schrumpft in atemberaubendem tempo. es ist nicht die digitalisierung an sich und auch nicht die vernetzung, die unsere selbstbestimmung beeinträchtigen und die menschen tendenziell zu marionetten degradieren. diese entwicklung ist das ergebnis einer wirtschaftlichen verwertungslogik in bezug auf unsere daten, die zu einer ungeheuren wissens-und machtkonzentration bei wenigen riesigen unternehmen geführt hat. das heutige internet ist -anders als in seinen anfängen -das gegenteil einer dezentralen informationsinfrastruktur und es steht zunehmend unter kontrolle weniger hochpotenter player. daten sind in ungeheurem umfang außerhalb der kontrolle der einzelnen nutzerinnen und nutzer konzentriert. die nicht ganz ernst gemeinte definition, dass sich cloudservices vor allem darin von klassischen it-strukturen unterscheiden, dass die daten nicht mehr auf dem eigenen, sondern auf einem fremden computer verarbeitet würden, beschreibt zutreffend einen wichtigen aspekt der heutigen informationstechnischen realität, wobei die kontrolle immer stärker bei wenigen, international tätigen unternehmen liegt. allerdings werden auch solche daten im interesse der datenoligopolisten verwendet, die auf dezentralen geräten gespeichert sind. die betreiber von anwendungsplattformen bestimmen weitgehend, welche apps zur nutzung freigegeben werden und welche informationen sie an wen ausspielen können. deutlich wurde deren einfluss jüngst bei der festlegung der standards für die nachverfolgung von kontakten im zusammenhang mit der bekämpfung der covid- pandemie. apple und google -und nicht die weltgesundheitsorganisation -legten die weltweit für das contact tracing gültigen standards fest. mag sein, dass das von den konzernen favorisierte "dezentrale" verfahren gegenüber dem konkurrierenden "zentralen" modell datenschutzrechtliche vorteile ausfweist. problematisch war aber gleichwohl, dass die entsprechenden weichenstellungen nicht durch demokratisch oder völkerrechtlich legitimierte akteure, sondern durch allein über die marktmacht der konzerne vorgenommen wurden. noch bedeutsamer sind die bemühungen der internetkonzerne, auch die kontrolle über die basisinfrastrukturen des netzes zu übernehmen, etwa indem sie -und nicht die telekommunikationsunternehmen -die nutzer mit dem netz verbinden. hinzuweisen ist hier etwa auf das projekt "loon" der google-mutter-alphabet [ ] zur versorgung abgelegener gegenden mit internet oder das vorhaben von amazon, eine globale internetversorgung über mehr als satelliten zu gewährleisten ("projekt kuiper") [ ] . auch facebook ist dabei, wenn neue märkte erschlossen werden: das unternehmen setzt dabei auf internetdrohnen, die auch in der wüste und im dschungel einen internetzugang ermöglichen sollen [ ] . auch weniger sichtbar, -aber nicht weniger wirksam -erweitern die konzerne ihre kontrolle über die internetinfrastrukturen. so ist der alphabet im begriff, sich über den service "google public dns" [ ] noch tiefer in der infrastruktur des internets zu verankern. letztlich versuchen global tätige unternehmen, ihre jeweiligen Ökotope zu synonymen des internets zu entwickeln, eines proprietäre "internets", in der letztlich nur eine instanz die vollständige kontrolle ausübt und damit über den umgang mit informationen im wirtschaftlichen eigeninteresse entscheidet. das in den er-jahren in seinen grundzügen entwickelte moderne datenschutzrecht sollte negativen folgen der digitalisierung entgegenwirken. ausgehend von dem bereits im . jahrhundert geforderten privatsphärenschutz (privacy) [ ] übertragen. allerdings behielt sich die us-regierung beim Übergang der verwaltung auf icann vor, den vertrag über die administration der zentralen rootzone (iana-vertrag) neu auszuschreiben [ ] . angesichts der fortbestehenden us-amerikanischen dominanz bei der festlegung der internetstandards verfolgen verschiedene staaten -allen voran china -eine neuordnung der internet-governance. die zentralen entscheidungen über die internetstandards sollen bei der internationalen fernmeldeunion (itu) gebündelt werden. unter dem stichwort "new ip" sollen zudem staatliche durchgriffsmöglichkeiten auf nationale internetsegmente gestärkt werden. in der letzten dekade ist ohnehin eine reihe virtueller "Ökosysteme" entstanden, die zwar nicht vollständig autonom agieren, bei denen die kommunikation und sonstige elektronische aktivitäten zensiert und überwacht werden. die kontrolle der interaktion von teilnehmern an einem virtuellen Ökosystem untereinander und mit akteuren in der außenwelt unterliegt der kontrolle durch die eigentümer bzw. in staatlich kontrollierten teilnetzen durch staatliche bürokratien. es ist zu befürchten, dass sich diese tendenz weiter verfestigt. der schon in den er-jahren feststellbare staatliche Überwachungsdruck auf das internet verstärkte sich insbesondere nach den islamistisch motivierten blutigen terror für geheimdienste und andere sicherheitsbehörden wurde das internet seither zu einem zentralen operationsfeld. sie nutzten und nutzen technische Überwachungsmöglichkeiten -vielfach über ihre gesetzlichen befugnisse hinaus. die machthaber autoritärer regimes -etwa des iran oder chinas -unterschätzten zunächst die möglichkeiten zur zivilgesellschaftlichen selbstorganisation, welche die neuen, internetbasierten kommunikationstechniken für ihre bevölkerung boten. insbesondere oppositionelle nutzten die chancen der unzensierten und weitgehend nicht überwachten internetkommunikation. weder der "arabische frühling" (vgl. [ ] ), der zum sturz mehrerer autokratischer herrscher führte, noch das zeitgleiche aufbegehren der iranischen jugend [ ] wären ohne internet, facebook und twitter denkbar gewesen. andererseits erkannten die regierenden -sofern sie nicht aus dem amt gejagt wurden -bald die Überwachungs-, zensur-und kontrollmöglichkeiten, die das netz bietet und sie zögerten nicht, davon gebrauch zu machen. deshalb ist kaum damit zu rechnen, dass das internet in absehbarer zeit wieder zu einem subversiven medium wird, das es früher bisweilen war. so hat die chinesischen regierung im rahmen des "golden shield projekts" das eigene internet durch eine virtuelle mauer weitgehend nach außen abschottet. ausgangspunkt waren proteste anlässlich des . jahrestags des massakers auf dem platz des himmlischen friedens im juni , bei denen sich protestler über das internet vernetzt hatten. heute ist das chinesische internet durch eine "great chinese firewall" abgeschottet, die alle unliebsamen inhalte herausfiltert. auch die sozialen netzwerke und messengerdienste werden vollständig überwacht. kritische beiträge werden zensiert, wie etwa kürzlich unliebsame berichte über den umgang der regierung mit der corona-epidemie anfang [ ] . andere staaten -russland, indien und der iran, um nur die wichtigsten zu nennen -sind dabei, diesem beispiel zu folgen. auch dort wird versucht, vollständige kontrolle über die elektronische kommunikation zu erlangen, allerdings bisher noch nicht mit so durchschlagendem erfolg wie china. auch in demokratien -etwa in frankreich, deutschland und den vereinigten staaten -wurden gesetzliche bestimmungen erlassen, die betreiber von internetdiensten zur ermöglichung staatlicher Überwachung verpflichten, etwa zur ausleitung der kommunikation mithilfe von sicherheitsbehörden kontrollierten Überwachungsschnittstellen. hinzu kommt ein immer rigideres, zumeist durch nationales recht bestimmtes regelwerk, das dienstanbieter zur inhaltskontrolle verpflichtet (zur durchsetzung des copyrights, zur unterbindung unzulässiger informationsweitergabe und zur verhinderung von "hasskommentaren"). die technisch und politisch motivierten eingriffe in die internetkommunikation haben dazu beigetragen, dass sich die aufteilung der verantwortlichkeiten im internet verschoben hat. früher hatten die betreiber der infrastruktur für die verfügbarkeit zu sorgen, während die anderen akteure für die integrität, vertraulichkeit und den datenschutz verantwortlich waren. heute bestimmen wenige globale konzerne zunehmend die funktionsweise des internets. google, facebook, microsoft, amazon und apple -um nur die wichtigsten und "westlichen" internetunternehmen zu nennenbetreiben sowohl infrastrukturen und bieten zugleich zahlreiche eigene dienste an. sie bestimmen als plattformbetreiber darüber, welche anderen unternehmen nach welchen konditionen mitspielen können. standards werden bewusst so definiert, dass sie die kommunikation über systemgrenzen hinweg erschweren oder unmöglich machen. auch wenn die entsprechenden mechanismen bei den privatwirtschaftlich betriebenen subnetzen/Ökosystemen weniger strikt sind als etwa bei staatlich vollständig kontrollierten netzsegmenten, ist es für deren nutzer vielfach schwierig, mit außenstehenden zu kommunizieren. dies gilt etwa für messangerdienste, die jeweils mit eigenen ("proprietären") standards funktionieren. während ein telefonnutzer ohne weiteres mit den kunden eines anderen anbieters telefonieren oder per sms nachrichten austauschen kann, ist der informationsaustausch bei whatsapp auf die dort registrierten teilnehmer begrenzt. auch bei vielen anderen diensten gibt es entsprechende beschränkungen. einschließungseffekte ("lock-in") sind die folge: nutzer haben -auch wenn es andere vergleichbare angebote gibt -keine praktikable alternative, weil die anderen familienangehörigen, kollegen, usw. ebenfalls mithilfe dieses dienstes kommunizieren oder diese plattform nutzen. bei der gestaltung der informationsgesellschaft der zukunft geht es um die frage, wie die menschen ihre selbstbestimmung zurückgewinnen können, die im Überwachungskapitalismus immer weiter eingeschränkt wird. allerdings ist die vorstellung völlig irreal, moderne gesellschaften könnten computer und internet einfach abschalten und zur analogen informationsverarbeitung zurückkehren. die frage nach der zukunft des internets und -allgemeiner -nach unserem zukünftigen umgang mit informationstechnik ist höchst politisch, denn es geht dabei um die fundamentale herausforderung, ob es den demokratischen rechtsstaaten gelingt, ihre zentralen ethischen und rechtlichen prinzipien zu verteidigen und mit leben zu füllen, welche die menschheit in ihrem jahrtausende langen zivilisatorischen prozess unter vielen schmerzen entwickelt hat. an erster stelle steht dabei die aufgabe, die grund-und menschenrechte im digitalen zeitalter zu gewährleisten und zu stärken. datenschutz ist dabei ein zwar wichtiger, aber eben nur ein aspekt [ , s. ]. er ist eine notwendige, aber keine hinreichende bedingung, wenn es um die digitale selbstbestimmung geht. zu entscheiden ist nicht allein über den umgang mit personenbezogenen daten, sondern darüber hinaus auch über "die selbstbestimmte wirtschaftliche verwertung der eigenen datenbestände sowie den selbstbestimmten umgang mit nicht-personenbezogenen daten, die etwa durch den wirkbetrieb eigener geräte generiert werden", wie die deutsche datenethikkommission einfordert [ ] . zum einen sind deshalb weitere rechtsgebiete gefordert, etwa das kartell-und wettbewerbsrecht, transparenzanforderungen, wie sie etwa durch informationsfreiheits-und transparenzgesetze formuliert werden und die verbesserung der im zivilrecht vorgesehenen gewährleistungs-und haftungsansprüche für die hersteller von hardund software. zum anderen geht es um die technische gestaltung von digitalen produkten, dienstleistungen und infrastrukturen, also um ansätze, die gewährleisten, dass die ethischen und rechtlichen anforderungen tief in der technik verankert werden, wie es ansatzweise schon im datenschutzrecht unter den stichworten "privacy by design" und "datenminimierung" geschehen ist, etwa in der eu-datenschutz-grundverordnung . die dsgvo kann kein endpunkt der entwicklung sein. im hinblick auf ihre weiterentwicklung stärker in den blick genommen werden müssen systeme, die für den einzelnen menschen oder für die gesellschaft wichtige entscheidungen selbst treffen oder vorbereiten. von besonderer bedeutung ist dabei die zusammenführung und auswertung von daten zum zwecke der bewertung (profilbildung) und der einsatz algorithmischer entscheidungssysteme, etwa im zusammenhang unter verwendung "künstlicher intelligenz" (ki) (vgl. [ ] ). dringend erforderlich ist darüber hinaus die e-privacy-verordnung (vgl. . ). bei aller bedeutung des europäischen rechts dürfen wir nicht aus dem blick verlieren, dass europa -schon allein aufgrund der digitalen technologieführerschaft der usa und chinas -das notwendige umsteuern nicht im alleingang bewältigen kann. verfehlte industriepolitische weichenstellungen und die unterschätzung der bedeutung des technologischen fortschritts durch das management der meisten großen europäischen unternehmen haben dazu beigetragen, dass die weitaus meisten global bedeutsamen digitalunternehmen in den usa und in asien angesiedelt sind. zur hoffnung anlass gibt aber, dass inzwischen auch in vielen außereuropäischen staaten eine datenschutzregulierung erfolgt, die sich an der dsgvo orientiert, wie etwa der am . januar in kraft getretene california consumer privacy act (ccpa). auch das am . februar zwischen der eu und japan in kraft getretene freihandelsabkommen (eu-japan economic partnership agreement) garantiert in dem so entstandenen gemeinsamen wirtschaftsraum ein der dsgvo entsprechendes datenschutzniveau. wir müssen verstehen, dass die herausforderungen der digitalisierung eine ähnliche dimension haben wie die herausforderungen der Ökologie, wobei durchaus berührungspunkte zwischen beiden bereichen bestehen, etwa hinsichtlich des ungeheuren energieverbrauchs (und der damit verbundenen co -emissionen) der riesigen datenzentren, die für die bereitstellung globaler cloud-und streamingdiens-te eingesetzt werden. auch weil die ökologische umorientierung nur unter einsatz digitaler techniken erfolgversprechend ist, liegt eine enge verknüpfung beider bereiche nahe. vielleicht hilft ja ein neuer gesellschaftsvertrag: ein "green new deal", erweitert um die digitale selbstbestimmung! Überwachtes netz. newthinking die globale Überwachung. droemer, münchen Überwachung total das zeitalter des Überwachungskapitalismus zugegriffen: jetzt auch amazon: mit mehr als satelliten internet für fast alle menschen per facebook-drohne: internet aus der luft the right to privacy Übereinkommen zum schutz des menschen bei der automatisierten verarbeitung personenbezogener daten v. . the cnil's restricted committee imposes a financial penalty of million euros against google llc digitaler frühling die neda-revolte. süddeutsche zeitung online v coronavirus, bloß nichts durchsickern lassen evaluation der datenschutz -grundverordnung -vorschläge zur weiterentwicklung des datenschutzrechts v key: cord- -ygmkul authors: khrennikov, andrei title: social laser model for the bandwagon effect: generation of coherent information waves date: - - journal: entropy (basel) doi: . /e sha: doc_id: cord_uid: ygmkul during recent years our society has often been exposed to coherent information waves of high amplitudes. these are waves of huge social energy. often they are of destructive character, a kind of information tsunami. however, they can also carry positive improvements in human society, as waves of decision-making matching rational recommendations of societal institutes. the main distinguishing features of these waves are their high amplitude, coherence (homogeneous character of social actions generated by them), and short time needed for their generation and relaxation. such waves can be treated as large-scale exhibitions of the bandwagon effect. we show that this socio-psychic phenomenon can be modeled based on the recently developed social laser theory. this theory can be used to model stimulated amplification of coherent social actions. “actions” are treated very generally, from mass protests to votes and other collective decisions, such as, e.g., acceptance (often unconscious) of some societal recommendations. in this paper, we concentrate on the theory of laser resonators, physical vs. social. for the latter, we analyze in detail the functioning of internet-based echo chambers. their main purpose is increasing of the power of the quantum information field as well as its coherence. of course, the bandwagon effect is well known and well studied in social psychology. however, social laser theory gives the possibility to model it by using general formalism of quantum field theory. the paper contains the minimum of mathematics and it can be read by researchers working in psychological, cognitive, social, and political sciences; it might also be interesting for experts in information theory and artificial intelligence. during recent years, the grounds of the modern world have been shocked by coherent information waves of very high amplitude. the basic distinguishing property of such waves is that they carry huge amounts of social energy. thus, they are not just the waves widely distributing some special information content throughout human society. instead, their information content is very restricted. typically, the content carried by a wave is reduced to one (or a few) labels, or "colors": one wave is "green", another is "yellow". at the same time, information waves carry very big emotional charge, a lot of social energy. therefore, they can have strong destructive as well as constructive impact on human society. in this paper, we present a model of the generation of very powerful and coherent information waves; a model based on the recently developed theory of social laser [ ] [ ] [ ] [ ] [ ] [ ] . we stress that social laser theory is part of the extended project on applications of formalism of quantum theory outside of physics, quantum-like modeling (see, e.g., monographs [ ] [ ] [ ] [ ] [ ] and some selection of papers [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] ). this terminology was invented by the author to distinguish this modeling from attempts to reduce human consciousness, cognition, and consequently behavior to genuine quantum physical processes in the brain (see, e.g., penrose [ ] or hameroff [ ] ). we do not criticize • indistinguishability of people. the human gain medium, population exposed to the information radiation, should be composed of social atoms, "creatures without tribe": the role of national, cultural, religious, and even gender differences should be reduced as much as possible. content ignorance. social atoms should process information communications without deep analyzing of their contents; they extract only the basic labels ("colors") encoding the communications. of course, humans are still humans, not social atoms; thus, in contrast to quantum physics, it is impossible to create human gain mediums composed of completely indistinguishable creatures. people still have names, gender, nationality, but such their characteristics are ignored in the regime of social lasing. one of the basic components of lasers, both physical and social, is a resonator [ ] . it plays the double role: • amplification of the beam (of physical vs. information) radiation; • improving coherence of this beam. social laser resonators play a crucial role in generation of coherent information waves of high amplitude. they are established via internet-based echo chambers associated with social networks, blogs, and youtube channels. their functioning is based on the feedback process of posting and commenting, the process that exponentially amplifies the information waves that are initially induced by mass media. echo chambers improve the coherence of the information flow through the statistical elimination of communications that do not match the main stream. this statistical elimination is a consequence of the bosonic nature of the quantum information field (sections . and . ) . although this quantum process of coherence generation dominates in echo chambers, we should not ignore other technicalities increasing coherence (sections . and . ), such as censorship of moderators and the dynamical evaluation system of search engines of, e.g., google, youtube, or yandex. the latter system elevates approachability of posts, comments, and videos depending on the history of their reading (seeing) and reactions to them say in the form of new comments. this is a good place to recall that the quantum-like hilbert space formalism if widely used for the modeling of information processing by internet search engines, and, in particular, for information retrieval [ ] [ ] [ ] [ ] [ ] [ ] . we compare functioning of optical and information mirrors (section . ). the latter represents the feedback process in internet systems such as, e.g., youtube. in contrast to the optical mirror, the information mirror not only reflects excitations of the quantum information field, but also multiplies them. thus, this is a kind of reflector-multiplier (section . ). as the result of this multiplication effect, social resonators are more effective than physical ones. however, as in physics, resonator efficiency depends on a variety of parameters. one of such parameters is the coefficient of reflection-multiplication (section . ). we analyze the multilayer structure of an information mirror and dependence of this coefficient on the layer (section . ). the main output of this paper is presented in section describing the quantum-like mechanism of the generation of big waves of coherent information excitations. we start the paper with compact recollection of the basics of social laser theory distilled from technical details and mathematical formulas. we present the basic notions of this theory such as social energy (section . )) and social atom, human gain medium (section . ), information field (section . ), the energy levels structure of social atoms (section . ), and spontaneous and stimulated emission of information excitations (section . ). finally, we conclude the introduction by the schematic presentation of the functioning of social laser theory (section ). the role of information overload in approaching indistinguishability of information communications, up to their basic labels, quasi-colors, is discussed in section . . this is a good place to mention studies on coupling indistinguishability and contextuality [ ] . finally, we point to coupling of the social laser project with foundations of quantum theory (appendix b). the basic component of a physical laser is a gain medium, an ensemble of atoms. energy is pumped into this medium aimed to approach the state of population inversion, i.e., the state where more than % of atoms are excited [ ] . then, a coherent bunch of photons is injected into the gain medium and this bunch stimulates the cascade process of emission of the coherent photon beam. if the power of pumping is very high, i.e., it is higher than the so-called lasing threshold, all energy of pumping is transferred into the output beam of coherent radiation. to make this beam essentially stronger, the laser is equipped by an additional component, the laser resonator (typically in the form of an optical cavity). the laser resonator also improves the coherence of the output beam, by eliminating from the beam photons that were generated via spontaneous emission in the gain medium [ ] . typically, in physics, coherence is formulated in physical waves terms, as electromagnetic waves going in phase with the same direction of propagation and frequency. for us it is convenient to reformulate this notion by excluding any reference to waves in the physical space, since we want to move to the information space. instead of the wave picture we can use the photon picture, so a propagating wave is represented as a cloud of energy quanta. (this is the fock representation in quantum field theory.) coherence means that they have the same energy (frequency) and the direction of propagation-photon's wave vector. we remark that a photon also has additional characteristics such as polarization, the quantum version of the ordinary polarization of light. for convenience of further considerations, let us call all characteristics of a photon additional to its energy quasi-color. we recall that the usual light's color is determined by photon energy (frequency). therefore, a photon has its color and quasi-color. the notion of social energy is the main novel component of our quantum-like modeling. to justify the use of a social analog of the physical energy, we use the quantum-mechanical interpretation of energy, not as an internal feature of a system, but as an observable quantity. thus, like in the case of an electron, we cannot assign to a human the concrete value of the social energy. there are mental states in the superposition of a few different values of the social energy. however, by designing proper measurement procedures we can measure human energy; see [ , ] for details. social energy is a special form of the psychic energy. we recall that at the end of th/beginning of th century psychology was strongly influenced by physics, classical statistical physics and thermodynamics (in works of james and freud), later by quantum physics (in works of jung). in particular, the leading psychologists of that time have actively operated with the notion of psychic energy [ ] [ ] [ ] [ ] . later psychologists essentially lost interest in the construction of general theories and, in particular, operating with the notion of the social energy. recently, the notion of social energy attracted a lot interest in economics and finance, multi-agent modeling, evolution theory and industrial dynamics [ ] [ ] [ ] . of course, these novel as well as old (freud-jung) studies support our model. however, we emphasize that the application of the quantum (copenhagen) methodology simplifies and clarifies essentially the issue of the social energy. we treat it operationally as an observable on a system, a human being. in contrast to, say, freud, we are not interested in psychic and neurophysiological processes of generation of psychic energy (see appendix a for a brief discussion). the basic component of social laser is a gain medium, an ensemble of people. as already mentioned, to initiate lasing, such a gain medium should consist of indistinguishable people, i.e., without tribe, without cultural, national, religious, and ideally sex differences. such beings are called social atoms. (it is not clear whether they still can be called humans). of course, people still have aforementioned characteristics, in some contexts they remember that they are men or women, or even christian, or swedish. we discuss contexts in which people behave as indistinguishable, as social atoms. creation of such behavioral contexts is the first step towards initiation of social lasing. we recall that in quantum physics the electromagnetic field is treated as a carrier of interactions. in the quantum framework, interaction cannot be represented as it was done classically, by force-functions. quantum interaction is of the information nature. in quantum information theory, excitations of the quantum electromagnetic field, photons, are carriers of information. at the same time, each excitation also carries a quantum of energy. this quantum picture is very useful for general modeling of information fields generated by mass media and the internet. communications emitted by newspapers, journals, tv, social networks, and blogs are modeled as excitations of a quantum information field, as quanta of information and social energy. as we know, the quantum description is operational; this is only the mathematical symbolism used for prediction of probabilities. even the quantum electromagnetic field cannot be imagined as a "real wave" propagating in spacetime. (in the formalism, this is a distribution, generalized function, with operator values. hence, this is a very abstract mathematical structure. it is useful for accounting for the numbers of energy quanta and description of the processes of their emission and absorption.) on one hand, this impossibility of visualization is a disadvantage of the quantum description compared to the classical one (we remark that the visualization of the classical electromagnetic field is also not as straightforward as might be imagined. the electromagnetic waves were invented as the waves propagating in the special media, the aether, similarly to acoustic wave propagating in air. later, einstein removed aether from physics. the picture of a vibrating medium became inapplicable. therefore, electromagnetic waves are vibrations of a vacuum. this is not so natural picture for the visualization of this process.). on the other hand, this is a great advantage, since it provides the possibility for generalizations having no connection with physical spacetime. thus, we model the information field as a quantum field with communications (generated, e.g., by mass media) as quanta carrying social energy and some additional characteristics related to communication content. as was already emphasized, quantum description is applicable to fields with indistinguishable excitations, where indistinguishability is considered to observable characteristics. in addition, "observable" means those characteristics that people assign to communications. these are labels of communications, say "terrorism", "war in syria", "coronavirus" and so on. such labels we shall call quasi-colors of information excitations, these are analogs of photon wave vector and polarization. thus, each communication is endowed with a quasi-color. it also carries a quantum of energy; its value we consider as communication color. thus, allegorically we can speak about red, blue, or violet information. content ignorance (up to communication quasi-color and color) is the crucial feature of the applicability of the quantum formalism. why do social atoms compress contents of communications to quasi-colors? the most important is information overload. the information flows generated by mass media and the internet are so powerful that people are not able to analyze communication content deeply, they just scan its quasi-color and absorb a quantum of the social energy carried by this communication. they simply do not have computational and time resources for such an analysis. it is also crucial that people lose their identity, so they become social atoms. for a social atom, there are no reasons, say cultural or religious, to analyze news; he is fine with just absorption of labels (quasi-color) and social energy (color) assigned to them. consider for simplicity social atoms with just two energy levels, excited and relaxed, e and e . the difference between these levels, is the basic parameter of a social atom, its color. a social atom reacts only to a communication carrying energy e c matching his color: if a communication carries too high-energy charge, e c larger than e a ("a social atom is yellow, but a communication is blue"), then an atom would not be able to absorb it. say a communication carrying social energy e c is a call for an uprising against the government. in addition, an atom is a bank clerk in moscow, who has liberal views and hates the regime, but the energy of his excited state is too small to react to this call. if e c is less than e a ("an atom is blue, but a communication is yellow"), then an atom would not be excited by this communication. the communication would be simply ignored. as well as a physical atom, a social atom cannot collect social energy continuously from communications carrying small portions of energy (compared to e a = e − e ), it either absorbs communication (if the colors of an atom and communication match each other) or it does not pay attention to it. in the same way, a social atom cannot "eat" just a portion of energy carried by too highly charged communication. in physics textbooks, the condition of absorption of energy quantum by atom is written as the precise equality: however, precise equalities are only mathematical idealizations of the real situation. the photon-absorption condition ( ) is satisfied only approximately: the spectral line broadening is always present. the difference between the energies of atom levels is the mean value (average) of the gaussian distribution, a bell centered at this point of the energy axis. the dispersion of the gaussian distribution depends on an ensemble of atoms. ensembles with small dispersion are better as gain mediums for lasing, but deviations from exact law ( ) are possible. it is natural to assume gaussian distribution realization of exact laws even for social systems; in particular, absorption of of excitations of the quantum information field by social atoms. thus, deviations from ( ) are possible. however, a good human gain medium should be energetic homogeneous. therefore, the corresponding gaussian distribution should have very small dispersion. shock news, say a catastrophe, war, killed people, epidemic, terror attack, is very good for energy pumping to a social gain medium. the modern west is characterized by the high degree of excitation, the energy e of the excited level is sufficiently high-otherwise one would not be able to survive: life in the megalopolis, long distances, high intensity of the working day, and so on. on the other hand, the energy e of the relaxation level is very low-one who is living on state support, say, in sweden, has practically zero excitement, often his state is depressive. hence, e a = e − e is high and a social atom would absorb only communications carrying very high energy: as in aforementioned shock news or say in tv shows, people should cry loudly, express highly emotional psychic states. since e a is high (blue), people would not pay attention to plain news (say red colored). even scientific news attracts attention only if it is very energetic, carries big emotional charge (blue or, even better, violet). however, shock news is very good for energy pumping not only because it carries a high charge of social energy, but also because it is very good at peeling communications from content. labels (quasi-colors) such as "coronavirus is a bio-weapon" leads to immediate absorption of communications, and social atoms react immediately to the instinctive feeling of danger. in our quantum-like model (similarly to physical atoms), social atoms can both absorb and emit quanta of the social energy. as in physics, there are two types of emission-spontaneous and stimulated. the spontaneous emission happens without external interaction, a social atom spontaneously emits a quantum of social energy, in the form of some social action. such spontaneous actions are not coherent, different atoms do different things, quasi-colors of social energy quanta emitted spontaneously can be totally different. such emissions generate a social noise in a human media, noise that is unwanted in social lasing. in particular, spontaneous emission noise disturbs functioning of internet echo chambers. on the other hand, the emission of quanta of social energy can be stimulated by excitations of the information field. in the very simplified picture, it looks like this. an excited social atom by interacting with an information excitation emits (with some probability) quantum of social energy. the most important feature of this process is that the quasi-color of the emitted quantum coincides with the quasi-color of stimulating communication. this is the root of the coherence in output beam of lasers, both social and physical. (the colors also coincide; see section . ). in reality, the process of stimulated emission is more complicated. it is important that the information field (similarly to the quantum electromagnetic field) satisfies bose-einstein statistics. this is a thermodynamic consequence [ ] of indistinguishability of excitations: two excitations with the same social energy and quasi-color are indistinguishable. as was shown in [ ] , by using the gibbs' approach based on consideration of virtual ensembles of indistinguishable systems (or any origin) we obtain the standard quantum classification of possible statistics, bose-einstein, fermi-dirac, and parastatistics. indistinguishability is up to energy (for the fixed quasi-color). hence, by taking into account that the number of communications carrying the same charge of social energy can be arbitrary, we derive the bose-einstein statistics for the quantum information field (see [ ] for derivation's details). interaction of atomic-like structures with bosonic fields are characterized by the following property: probability of stimulated emission from an atom increases very quickly with increasing of the power of a bosonic field. an excited social atom reacts rather weakly to the presence of a few information excitations. however, if they are many, then it cannot stay indifferent. in fact, this is just a socio-physical expression of the well-known bandwagon effect in humans' behavior [ ] . in contrast to psychology, we can provide the mathematical field-theoretical model for such an effect. we consider the fixed energy (frequency) mode of the quantum electromagnetic field. for fixed quasi-color mode α, n-photon state |n, α , can be represented in the form of the action of the photon creation operator a α corresponding to this mode on the vacuum state | : this representation gives the possibility to find that the transition probability amplitude from the state |n, α to the state |n + , α equals to (n + ). on the other hand, it is well known that the reverse process of absorption characterized by the transition probability amplitude from the state |n, α to the state |(n − ), α equals to √ n . generally, for a quantum bosonic field increasing the number of its quanta leads to increasing the probability of generation of one more quantum in the same state. this constitutes one of the basic quantum advantages of laser-stimulated emission showing that the emission of a coherent photon is more probable than the absorption. since, as shown in [ ] , indistinguishability, up to energy (color) and quasi-color, of information excitations leads to the bose-einstein statistics, we can use the quantum operational calculus for bosonic fields even for the quantum information field and formalize in this way the bandwagon effect in psychology [ ] . this is the good place to recall that in our considerations the notion "social action" is treated very widely, from a purely information action, as posting a communication at facebook or commenting one of already posted communications, to a real physical action, as participating in a demonstration against putin or trump, or supporting government's policy on "self-isolation". the previous works on social laser [ ] [ ] [ ] [ ] emphasized external representation of social actions, say in the well-known color revolutions. in this paper, we are more interested in their representation in information spaces, e.g., spaces of social networks. however, we are even more interested in internal representation of some social actions as decision makings. in addition, a decision can have different forms, not only "to do"-decisions, but also "not to do"-decisions. the decisions of the latter type also consume energy and social atoms transit from the excited state to the relaxed one. it is also important to point to the unconscious character of many (or may be majority) of our decisions. for example, people can support (or not support) societal policies totally unconsciously. to make such decisions, they consume social energy. mass media and internet pump social energy into a gain medium composed of social atoms to approach the population inversion-to transfer most atoms into excited states. then a bunch of communications of the same quasi-color and energy (color) matching with the resonant energy of social atoms is injected in the gain medium. in the simplified picture, each communication stimulates a social atom to emit a quantum of social energy with the same quasi-color as its stimulator. resulting two excitations stimulate two social atoms to emit two quanta, the latter two quanta generate four and so on, after say steps there are , approximately one million of information excitations of the same (quasi-)color. in reality, the process is probabilistic: an atom reacts to stimulating information excitation only with some probability. the later increases rapidly with increasing of the density of the quantum information field. now, we discuss the basic counterparts of social lasing in more detail: • each information communication carries a quantum of social energy. the corresponding mathematical model is of the quantum field type, the information field. quanta of social energy are its excitations. each social atom is characterized by the social energy spectrum; in the simplest case of two levels, this is the difference between the energies of the excitation and relaxation states, besides of social energy, the excitations of the information field are characterized by other labels, quasi-colors. coherence corresponds to social color sharpness; ideal social laser emits a single mode of quasi-color, denoted say by the symbol α. humans in the excited state interacting with α-colored excitations of the information field also emit α-colored excitations. the amount of the social energy carried by communications stimulating lasing should match with resonance energy e a of social atoms in the human gain medium. to approach the population inversion, the social energy is pumped into the gain medium. this energy pumping is generated by the mass media and the internet sources. the gain medium should be homogeneous with respect to the social energy spectrum. in the ideal case, all social atoms in the gain medium should have the same spectrum, e a . however, in reality, it is impossible to create such a human gain medium. as in physics, the spectral line broadening must be taken into account. for example, a gain medium consisting of humans in the excited state and stimulated by the anti-corruption colored information field would "radiate" a wave of anti-corruption protests. the same gain medium stimulated by an information field carrying another social color would generate the wave of actions corresponding this last color. the general theory of resonators for social lasers is presented in [ ] . here we shall consider in more detail special, but at the same very important type of social resonators, namely internet-based echo chambers. we recall that an echo chamber is a system in that some ideas and behavioral patterns are amplified and sharped through their feedback propagation inside this system. in parallel to such amplification, communications carrying (as quasi-color) ideas and behavioral patterns different from those determined by the concrete echo chamber are suppressed. in our terms, an echo chamber is a device for transmission and reflection of excitations of the quantum information field. its main purpose is amplification of this field and increasing its coherence via distilling from "social noise". the latter function will be discussed later in more detail. the echo chamber is also characterized by the resonance social energy e a of its social atoms. for simplicity, it is assumed that all social atoms have the same resonance energy e a . (in reality, resonance energy of social atoms is a gaussian random variable with mean value e a .) we underline that in this paper an echo chamber is considered to be a component of the social laser, its resonator. compared to physics we can say that this is an analog of an optical cavity of the physical laser, not optical cavity by itself. the coherent output of an echo chamber, the quasi-color of this output, is determined not only by the internal characteristics of the echo chamber, but also by the quasi-color of stimulating emission. let us consider functioning of some internet-based echo chamber; for example, one that is based on some social group in facebook (or its russian version "vkontakte") and composed of social atoms. the degree of their indistinguishability can vary depending on the concrete echo chamber. say, names are still present in facebook, but they have some meaning only for the restricted circle of friends; in instagram or snapchat, even names disappear and social atoms operate just with nicknames. by a social group we understand some sub-network of say facebook, for example, social group "quantum physics". the main feature of a social group is that all posts and comments are visible for all members of this social group. thus, if one from the group puts a post, then it would be visible for all members of this social group, and they would be able to put their own comments or posts related to my initiation post. this is simplification of the general structure of posting in facebook, with constraints that are set by clustering into "friends" and "followers". we assume that the ensemble of social atoms of this echo chamber approached population inversion, so most of them are already excited. a bunch of communications of the same quasi-color α and carrying quanta of social energy e c = e a is injected in the echo chamber. excited social atoms interact with the stimulating communications and emit (with some probability) information excitations of the same quasi-color as the injected stimulators. these emitted quanta of social energy are represented in the form of new posts in echo chamber's social group. each post plays the role of a mirror, it reflects the information excitation that has generated this post. however, the analogy with the optics is not straightforward. in classical optics, each light ray is reflected by a mirror again as one ray. in quantum optics, each photon reflected by a mirror is again just one photon. an ideal mirror reflects all photons (the real one absorbs some of them). in contrast, "the mirror of an echo chamber", the information mirror, is a multiplier. a physical analog of such a multiplier mirror would work in the following way. each light ray is reflected as a bunch of rays or in the quantum picture (matching better the situation), each photon by interacting with such a mirror generates a bunch of photons. of course, the usual physical mirror cannot reflect more photons than the number of incoming ones, due to the energy conservation law. hence, the discussed device is hypothetical. this is a good place to remark that as mentioned, a photon should not be imagined as a metal ball reflecting from mirror's surface. a photon interacts with the macro-system, the mirror, and the latter emits a new photon that is identical to the incoming one, up to the direction of spatial propagation. it seems to be possible to create a kind of a mirror with the complex internal structure (composed of special materials) such that it would generate emission of a bunch of photons. of course, such a multiplier mirror cannot function without the energy supply. the internet-based system of posting news and communications works as a multiplier mirror. each posted news or communication emits a bunch of "information rays" directed to all possible receivers-the social atoms of echo chamber's social group. in the quantum model, each post works as an information analog of photon's emitter. it emits quanta of social energy; the power of the information field increases. consequently, excited social atoms emit their own posts and comments with higher probability. we repeat that new posts have the same quasi-color as the initiating information excitations that were injected in the echo chamber. it is also important to remind that the process of stimulated emission is probabilistic. members of the social group would react to newly posted message only with some probability. in addition, resulting from the bosonic nature of the quantum information field, this probability increases rapidly with increasing of field's power. by reaction we understood emission of a new message, say a comment. if a social atom simply reads a posted communication, but does not emit its own information excitation, then we do not consider such reading as a reaction. for the moment, we consider only the process of stimulated emission. later we shall consider absorption as well. in the latter, reaction means transition from the ground state to the excited state; so, not simply reading. (in principle, a relaxed atom can read a post or a comment without absorbing a quantum of social energy sufficient for approaching the state of excitement.) the crucial difference from physics is an apparent violation of the energy conservation law (see appendix a for the discussion on this question). each post in a social group works as a social energy multiplier. thus, information excitations in the echo chamber generated by posted communications not only increase the probability of emission of new information excitations by excited atoms, but they also perform the function of additional energy pumping into the gain medium (social group). relaxed social atoms can absorb social energy not only from externally pumped messages from mass media, tv and other social networks, but even from their own echo chamber. then they also emit new posts and so on. the main distinguishing feature of the quantum information field is its bosonic nature. we now emphasize the impact of the bosonic structure to coherence of the information field inside of an echo chamber. as was already noted (section . ), the interaction of a social atom with the surrounding bosonic field depends crucially on the power of this field, the probability of emission of energy quantum by an excited social atom increases very quickly with increasing of field's power. now, we stress again that a social atom (as well as a physical atom) distinguishes the modes of the field corresponding to different quasi-colors. the probability of emission of a quantum of the fixed quasi-color α depends on the power of the field's mode colored by α. thus, if the power of the α-mode essentially higher than the power of the mode colored by β, then with very high probability social atoms would emit α-colored energy quanta (in the form of posts, comments, and videos). social atoms would ignore the β-colored energy quanta, the probability of emission of such quantum (and hence the increase of the power of the β-mode) is practically zero. if a social atom emits a communication, colored by β, then this information excitation would not attract attention of social atoms who are busy with communications colored by α. as was already emphasized, the crucial role is played by indistinguishability, up to the quasi-colors, of the excitations of the information field. social atoms should process information in the regime of label scanning, without analyzing its content. as was discussed, the easiest way to establish the indistinguishability regime of information processing is to generate an information overload in the gain medium composed of social atoms. of course, the loss of individuality by social atoms is also very important, people "without tribe" are better accommodated to perceive information in the label-scanning regime. in this regime, one would never absorb the main information of the β-labeled communication, say statistical data. in this section, we considered the quantum-like nature of coherence of the information waves generated in echo chambers. this indistinguishability of information excitations, the label-scanning regime. the information overload and the loss of individuality by social atoms are the main socio-psychological factors leading to this regime. in following sections . and . , we consider supplementary factors increasing information field's coherence. now, we connect a social resonator, e.g., in the form of an internet-based echo chamber, to the social laser device described in section . as the result of the feedback processing of information in the echo chamber, the power and coherence of the information field increases enormously. one of the ways to consume the huge energy of this information field is to realize it in the form of physical social actions, mass protests, e.g., demonstrations or even a wave of violence. this is the main mechanism of color revolutions and other social tsunamis [ ] [ ] [ ] [ ] . however, in this paper we are more interested in the purely information consumption of the social energy of the coherent information field prepared in an echo chamber, namely for internal decision-making. decision-making on a question that important for society is also a social action; in particular, it consumes social energy. now, suppose that say a government needs some coherent and rational (from its viewpoint) decision on some question. it can use a powerful social laser. this is a good place to remark that an ensemble of echo chambers can be used coherently with stimulation by the same quasi-color α corresponding to the desired decision. by emitting the information excitation, a social atom confirms his-her support of the α-decision. such social action is realized in the mental space of social atoms, but, of course, it has consequences even for associated actions in the physical space. if the wave in the information space generated by a powerful social laser can approach the steady state, then social atoms live in the regime of the repeated confirmation of the internal α-decision: an atom emits and relaxes, then he/she again absorbs another α-excitation and moves to the state of excitement and so on. in this situation of surrounding by the information field of huge power concentrated on the same α-mode, the colors of the energy pumping and stimulated emission coincide. such repeating of the same α-decision is similar to concentration on the idea-fix and can lead to the state of psychosis and panic (see freud [ ] ). as in physical lasing, the above ideal scheme is complicated by a few factors related to losses of social energy in the echo chamber. as is known, not all photons are reflected by mirrors of the optical cavity, a part of them is absorbed by the mirrors. the coefficient of reflection plays the fundamental role. the same problem arises in social lasing. an essential part of posts is absorbed by the information mirror of the echo chamber: for some posts, the probability that they would be read by members of the social group is practically zero. additional (essential) loss of social energy is resulted from getting rid of communications carrying quasi-colors different from the quasi-color α of the bunch of the communications initiating the feedback dynamics in the echo chamber. such communications are generated by spontaneous emission of atoms in the social group. the real model is even more complex. the information mirror is not homogeneous, "areas of its surface" differ by the degree of readability and reaction. the areas can be either rigidly incorporated in the structure of the social group or be formed in the process of its functioning. for example, "quantum physics" group has a few layers that are rigidly incorporated in its structure. one of them is "foundations and interpretations". this sublayer of the information mirror "quantum physics" has rather low visibility, due to a variety of reasons. once, i posted in "quantum physics" a discussion on quantum information and quantum nonlocality. in addition, i discovered that the social group moderators control rigidly the layer structure. the message that my post should be immediately moved to this very special area of the information mirror, "foundations and interpretations", approached me in a few minutes. it looks that even in such a politically neutral social group moderators work in the online regime. as an example of functionally created information layers, we can point to ones which are coupled to the names of some members of the social group, say "area" related to the posts of a nobel prize laureate has a high degree of readability and reaction. however, of course, one need not be such a big name to approach a high level of readability and reaction. for example, even in science the strategy of active following to the main stream speculations can have a very good effect. top bloggers and youtubers create areas of the information mirror with high coefficients of reflection-multiplication (see below ( )) through collecting subscriptions to their blogs and youtube channels. it is clear that the probability of readability and reaction to a post depends heavily on the area of its location in the information space of a social group or generally facebook, youtube, or instagram. the reflection-multiplication coefficient of the information mirror varies essentially. consider first the physical mirror and photons reflected by it. from the very beginning, it is convenient to consider an inhomogeneous mirror with the reflection coefficient depending on mirror's layers. suppose that k-photons are emitted to area x and n of them were reflected, i.e., (k − n) were absorbed. then the probability of reflection by this area p(x) ≈ n/k, for large k. now, for the information mirror, consider a sequence of posts, j = , , ..., k, that were put in its area x. let n j denotes the number of group's members who reacts to post j. each n j varies between and n, where n is the total number of group's members. then coefficient of reflection-multiplication p(x) ≈ ( k ∑ j n j )/kn, for large k, n. if practically all posts generate reactions of practically all members of the group, then n j ≈ n and p(x) ≈ . we have already discussed in detail the multilayer structure of the information mirror of an echo chamber. this is one of the basic information structures giving the possibility to generate inside it the information field of the very high degree of coherence: a very big wave of information excitations of the same quasi-color, the quasi-color of stimulating communications. it is sufficient to stimulate atoms with the potential of posting in the areas of the information surface with the high coefficients of reflection-multiplication. these areas would generate a huge information wave directed to the rest of the social group. spontaneously emitted communications would be directed to areas with the low coefficients of reflection-multiplication. how is this process directed by the internet engines? it is described by the model of the dynamical evaluation of the readability history of a post. we shall turn to this model in section . . although the dynamical evaluation plays the crucial role in generating the coherent information waves, one has not to ignore the impact of straightforward filtering. we again use the analogy with physics. in the process of lasing, the dynamical feedback process in the cavity excludes the excitation of the electromagnetic field propagating in the wrong directions. in this way, laser generates the sharply directed beam of light. however, one may want some additional specification for excitations in the light beam. for example, one wants that all photons would be of the same polarization. it can be easily done by putting the additional filter, the polarization filter that eliminates from the beam all photons with "wrong polarization". of course, the use of an additional filter would weaker the power of the output beam. the latter is the price for coherence increasing. in social lasing, the role of such polarization filters is played by say google, facebook, instagram, or yandex control filtering, e.g., with respect to the political correctness constraints. besides numerous moderators, this filtering system uses the keywords search engines as well as the rigid system of "self-control". in the latter, users report on "wrongly colored posts and comments" of each other; the reports are directed both to the provider and to social groups-to attract the attention to such posts and comments. the dynamical evaluation system used, e.g., by youtube, increases post's visibility based on its reading history, more readings imply higher visibility (at least theoretically). however, the multilayer structure of the information mirror of youtube should also be taken into account. the main internet platforms assign high visibility to biggest actors of the mass media, say bbc, euronews, rt, that started to use actively these platforms. then, and this is may be even more important, these internet platforms assigns high visibility to the most popular topics, say presently the coronavirus epidemic, videos, posts, and comments carrying this quasi-color are elevated automatically in the information mirrors of google, youtube, or yandex. of course, the real evaluation system of the main internet actors is more complicated and the aforementioned dynamical evaluation system is only one of its components, may be very important. we would never get the answer to the question so widely discussed in communities of bloggers and youtubers: how are the claims on unfair policy of internet platforms justified? by unfair policy they understand assigning additional readings and likes to some internet communications or withdraw some of them from other communications. (i can only appeal to my own rather unusual experience from the science field. once, i was a guest editor of a special issue (so a collection of papers about some topic). in particular, my own paper was published in the same issue. this is the open-access journal of v top ranking, a part of nature publishing group. presently, all open-access journals qualify papers by the number of downloads and readings. (therefore, this is a kind of youtubing of science.) my paper was rather highly estimated in these numbers. however, suddenly i got email from the editors that since i put so much efforts to prepare this issue, i shall get as a gift an additional downloads. of course, i was surprised, but i did not act in any way and really received this virtual gift... after this event, i am very suspicious of numbers of downloads and readings that i can see in the various internet systems. if such unfair behavior is possible even in science, then one can suspect that it is really not unusual.) starting with presentation of the basics of social lasing, we concentrated on functioning of one of the most important kinds of social resonators, namely internet-based echo chambers. we analyzed similarities and dissimilarities of optical and information mirrors. the main distinguishing feature of the latter is its ability not only reflect excitations of the quantum information field, but also multiply them in number. the coefficient of reflection-multiplication is the basic characteristic of the information mirror. we point to the layer structure of the information mirror of an echo chamber; the coefficient of reflection-multiplication varies depending on mirror's layer. we emphasized the bosonic nature of the quantum information field. this is a straightforward thermodynamic consequence [ ] of indistinguishability of information excitations, up their quasi-colors. being bosonic, the information field increases tremendously the speed and coherence of stimulated emission of information excitations by excited social atoms. social atoms, "creatures without tribe", form the gain medium of a social laser. in contrast to quantum physics, we cannot treat real humans as totally indistinguishable. this is a good place to remind once again that our social lasing as well as generally decision-making modeling is quantum-like. quantum features are satisfied only approximately. this point is often missed in the presentation of "quantum models" for cognition and decision-making. in sections . and . , we discuss some technicalities related to functioning of internet-based social groups and generally google and youtube. this discussion plays only a supplementary role for this paper. it would be fruitful to continue it and especially to discuss exploring of the quantum-like features of users and information supplied to them by the internet (cf., for example, with studies on quantum-like modeling of information retrieval [ ] [ ] [ ] [ ] [ ] [ ] ). in appendix a, we discussed very briefly interrelation between the psychic energy and the physical energy of cells' metabolism. it is very important to continue this study in cooperation with psychologists and neurophysiologists. the main output of this paper is description of the mechanism of generation of big waves of coherent information carrying huge social energy, a kind of information tsunamis. we especially emphasize listing of the basic conditions on human gain media and the information field generated by mass media and amplified in echo chambers leading to successful generation of such waves. the author recognizes very well that this study is still one of the first steps toward well elaborated theory. motion) interpretation of quantum mechanics [ ] , whereas quite different from other interpretations, such as the many-world interpretation [ , ] and the wise (wave function is the system entity) interpretation [ ] . it should be also pointed out that in recent years, some physics-based social or network models have been studied [ , ] . some new intersection of quantum and operational research are emerging, such as quantum machine learning [ , ] . towards information lasers social laser: action amplification by stimulated emission of social energy social laser model: from color revolutions to brexit and election of donald trump on interpretational questions for quantum-like modeling of social lasing concept of information laser: from quantum theory to behavioural dynamics phase transitions, collective emotions and decision-making problem in heterogeneous social systems information dynamics in cognitive, psychological, social, and anomalous phenomena; series: fundamental theories of physics ubiquitous quantum structure: from psychology to finances quantum models of cognition and decision quantum concepts in the social classical and quantum mechanics on information spaces with applications to cognitive, psychological, social and anomalous phenomena. found. phys on quantum-like probabilistic structure of mental information pilot-wave theory and financial option pricing quantum dynamics of human decision making a quantum probability explanation for violations of 'rational' decision theory a quantum theoretical explanation for probability judgment errors decision theory with prospect interference and entanglement mathematical structure of quantum decision theory quantum-like model of behavioral response computation using neural oscillators a quantum probability explanation in fock space for borderline contradictions an operator-like description of love affairs can quantum probability provide a new direction for cognitive modeling? context effects produced by question orders reveal quantum nature of human judgments are quantum-mechanical-like models possible, or necessary, outside quantum physics? the visualizable, the representable and the inconceivable: realist and non-realist mathematical models in physics and beyond the role of information in a two-traders market conjunction and negation of natural concepts: a quantum-theoretic modeling a generalized probability framework to model economic agents' decisions under uncertainty quantum dynamics applied to cognition: a consideration of available options new fundamental evidence of non-classical structure in the combination of natural concepts the emperor's new mind quantum coherence in microtubules. a neural basis for emergent consciousness? brain and physics of many-body problems dissipation and memory capacity in the quantum brain model my double unveiled: the dissipative quantum model of brain enough! electoral fraud, collective action problems, and post-communist coloured revolutions why it's kicking off everywhere: the new global revolutions the color revolutions mistrust we trust: can democracy survive when we don't trust our leaders? the new digital age: reshaping the future of people why did people vote for brexit? deep-seated grievances lie behind this vote why did people vote for donald trump? voters explain; the guardian the science of structure: synergetics synergetics: an introduction-nonequilibrium phase transitions and self-organisation in physics laser light dynamics lasers and synergetics. a colloquium on coherence and self-organization in nature subjective probability: a judgment of representativeness prospect theory: an analysis of decision under risk critique des postulats et axiomes de l' cole amricaine risk, ambiguity, and the savage axioms ambiguity, and the dark-dependence axioms the geometry of information retrieval what is quantum information retrieval? how quantum theory is developing the field of information retrieval? supporting polyrepresentation in a quantum-inspired geometrical retrieval framework introduction to information retrieval and quantum mechanics deriving a quantum information retrieval basis the principles of psychology the standard edition of complete psychological works of sigmund freud on the nature of the psyche atom and archetype: the pauli/jung letters - competition between collective and individual dynamics from the financial crisis to sustainability? potsdam, european climate forum. available online: www.europeanclimate-forum.net/index.php?id=ecfreports interaction ritual chains oxford dictionary of psychology interpreting quantum mechanics in terms of random discontinuous motion of particles relative state" formulation of quantum mechanics many-worlds interpretation of quantum mechanics general quantum interference principle and duality computer modeling viral diffusion using quantum computational network simulation mechanics-an endless frontier quantum machine learning recent advances in quantum machine learning. quant. eng. , , e . [crossref] c by the author funding: this research received no external funding. the authors declare no conflict of interest. above we wrote about an "apparent violation" of the law of conservation of the social energy. we briefly discuss this point. the social energy is the special form of the psychic energy. hence, by discussing the conservation law we cannot restrict consideration solely to the social energy. the detailed analysis of transformation of different forms of the psychic energy and its origin in neurophysiological processes and finally the physical energy generated by cells' metabolism was presented by freud [ ] . we do not plan to discuss here the freud's hydrodynamical model for psychic energy transformations. we want to elevate the crucial difference of the energy transfer from the information field to social atoms from the energy transfer from the electromagnetic field to physical atoms. in physics, energy is assigned to photons carriers of information, an atom by absorbing a photon receives its energy. in our social model, an excitation of the information field just carries the social energy label e c . a social atom absorbs this label and generate the corresponding portion of energy by itself, by transforming its psychical energy into the social energy. in addition, the former is generated by neurophysiological activity in the brain and the nervous system from the physical metabolic energy. thus, by taking into account the psychic energy, we understand that even for cognitive systems the law of energy conservation is not violated. we remark that development of the social laser model has also some relevance to the interpretations of quantum mechanics. in this model, the quantum nature is apparent, because the smallest unit in a society is a human person. this is quite like the rdm (random discontinuous key: cord- -kfjg i authors: jagodzinski, jan title: inflexions of deleuze|guattari: for a new ontology of media, when west-east meet date: - - journal: pedagogical explorations in a posthuman age doi: . / - - - - _ sha: doc_id: cord_uid: kfjg i deleuze|guattari’s philosophical oeuvre has become increasingly known in east asia as scholars recognize their importance for virtually all disciplines. in this chapter, i try to show why representational thought, which they overturn, continues to retain its force over us. by introducing concepts, which they developed, such as simulacrum, powers of the false, the event, and affect, i make the attempt to show how non-representational thought can be understood within visual art and media. lastly, when west-east meet, there is an attempt on my part to show the relationship between taoist understanding of chaos and deleuze|guattari’s use of the transcendental plane to grasp the genesis of creativity proper. the philosophical importance and impact of deleuze|guattari are experiencing a strong interest in asia. in , tamkang university in taipei taiwan, under the directorship of hanping chiu held the first international asia conference on deleuze. korean scholars were among its participants. in , japan's toyonaka university, in osaka, hosted the second international deleuze|guattari conference with theme of "islands." in , deleuze and asia (bogue et al. ) published papers from the taipei conference. the deleuze-asian connection continues. before this sudden uptake, there has been a trickle of essays that show the links between deleuze|guattari and zen-tao (maliavin ; vodka ; o'sullivan a; zhang ) . it is to the credit of korean society of education through art (kosea) to continue to recognize the importance of this line of thinking for the twenty-first century by hosting the th kosea international conference on october , , under the title: nomadism, art, and art education. this chapter addresses insights a deleuze|guattari understanding can bring to art and its education especially for the twenty-first century, a time of chapter inflexions of deleuze|guattari: for a new ontology of media, when west-east meet continued rapid technological change, capitalist globalization, and the planetary disaster of 'climate change.' the future of education depends on how we, as artists and teachers, address these questions with our children and students who face such difficult issues. youth around the globe expect a response despite the lack of initiative by many of the key global leaders, most notably donald trump. greta thunberg has been elevated to the post as the leading youth activist and spokesperson. she has managed to spearhead the fridays for future (fff) movement where students in many major cities around the globe skip school in their attempt to rally together and send a message to their legislators. the philosophy of deleuze|guattari forms a bridge between taoism that is pervasive in the east and a revival of non-representational thought in the west by what generally has been called materialist transcendental philosophies. the influences between them are obviously rhizomatic in the deleuze|guattarian sense. "if i had been born in china, i would have been a calligrapher, not a painter," picasso once famously said, hinting that there has always been an artistic exchange between east and west. the complexity and interrelationships that exist between taoism and deleuze|guattari philosophy require scholarship to explain the many difficult concepts that, at times, seem tenuous; nevertheless, they open up interesting questions. this is remarkably illustrated by amir vodka ( ) who makes the claim that the cinema of kung fu can be related to deleuze|guattari's body without organs via tao, which forms its own kind of bwo as "the shape that has no shape, the image that is without substance" (lao-tzu : ) . when deleuze|guattari ask, "is the tao masochist?" (tp, : ) , the answer is that pain, torture, and restraint are part of the kung fu master's 'way' to absolve his ego and provide a way to reconfigure his/her bwo. "the masochist uses suffering as a way of constituting a body without organs and bringing forth a plane of consistency of desire" (tp, ) . while vodka refers to the famous films of bruce lee, marvel's tv series iron fist is equally instructive, as the narrative follows danny rand as he searches for his identity, quelling his anger to focus his chi (material force of free energy) into the part-object as weapon: his fist. the intensive body without organs is tao, while iron fist becomes a channeled 'bloc of sensations.' vodka argues that 'becoming animal' is quite prevalent in many kung fu movies. it is taken up in chapter with the cinema of kim ki-duk, while 'becoming insect' is sequentially illustrated in chapter . 'becoming imperceptible' is equally at play in kung fu cinema when new potentials emerge from the virtual, and reality then changes through such events. vodka draws on the matrix, and such digitalized effects like 'bullet time' to show how for neo, the hero who 'wakes' up to a new reality is able to discern all durations that coexist on a shared plane of consistency-durations that are at once too slow and too fast, enabling him to dodge bullets as he is too fast for them (and they are too slow). this is 'suspended time,' the time of aion-the living present "precisely where the imperceptible is seen and heard" (tp, ) . in brief, digital cinema in this case is able to visualize the imperceptible as pure movement, an abstract plane understood as any-plane-what-soever that traces movements. there are only relations of its singularities. we can point to the artwork of morgan o'hara to illustrate what is being inferred. her 'scribble-art' presents the performative line drawings of a generic human, which are imperceptible movements that emerge through her translations of her lebenswelt. what is mistakenly taken to be the most extreme example of abstract art is in fact the most concrete. her drawings capture only movement, drawn at speeds with things she 'vibrates' with. this is not her 'translating' life experiences; rather, it is a collapse of subject and object, a direct expression. françois laruelle (see galloway : - ) says as much and illustrates this same phenomenon through the pencil-music drawings of august von briesen. while these drawings give graphical forms to musical sounds via lines, dashes, points, variations in length, line weight, width, orientation, curvature, and so on, a clear case of synaesthesia (see chapter in this book), laruelle takes this a step further. von briesen's syncretism is an "automatic registration" of the real itself. imperceptibility in this sense is a difficult concept. if it is analogous to an outside real, then it also has affinities with lacan's objet a; something from the outside, something imperceptible and non-categorizable makes us think. this part-object (like the iron 'fist') would be an "automatic registration" of chi much like von briesen's pencil drawings, who generated as many as , of these gestures a year! 'becoming imperceptible' is performative. it calls forth the experience of time, which deleuze, following the stoics called aion (as opposed to chronos) and, in taoist thought, referred to as 'pure' time, beginning-less, yet non-eternal. the tao is understood as a concept that can simultaneously hold opposite values. time is immanent, yet never changes. it has no beginning, yet it is not eternal; most strange of all: it is everything, yet it does not even exist. aion sustains this same idea of infinite time, the 'gaps' between past and the future. in quantum physics, the smallest imaginable unit of time is plank time ( . × − sec.) and plank distance ( − cm). at that point, time and distance become meaningless. in mathematics, continuous and discontinuous numbers remain complementary, the paradox repeating energy as both wave and particle. there is no unifying mathesis universalis. as deleuze maintains, self and other, subject and object, outside and inside collapse into an empty present, "which is subdivided ad infinitum into something that has just happened and something that is going to happen, always flying in both directions at once" (deleuze : performance artists place themselves into such a time, and by doing so, they become imperceptible to themselves as they move in this virtual gap, or virtual real of chronological time. they emerge changed after the performance. this is precisely where 'true' creativity occurs, but it is also the moment of greatest risk where life = death becomes a 'cut' in the fabric of the actual. the deleuzian 'w(hole),' as an open-becoming, has affinities with the taoist zero; it is the open, the impossible 'set of sets,' without exteriority. neo in the matrix should be regarded as a singularity, a 'zeroness' of the bwo, an open whole, perhaps best understood by their formula: pluralism = monism. here, we can push deleuze and guattari even further by suggesting a quantum twist to their claim by maintaining that this 'monism' is also an infinite "multiverse," as a majority of physicists claim. as a true infinite, the multiverse (as tao) is both everything and nothing, or 'happening' itself of infinite creation: pure potential. our universe is just one of many, which supports a taoist notion of the permanence as well as the impermanence of time. even the notion of rhizome's nonlinearity, non-hierarchy, and multi-dimensionality is not a simple task to comprehend, but describes the relationships and principles of tao, and the sensibility to chaos through the creativity of 'non-action,' as wu wei (non-doing, or doing nothing, to be in the flow, to be imperceptible). intentionality becomes transformed into intuition. intuition holds no rigid and set plan as much as a direction of speed and slowness. yet, the paradox is that this requires a repetition of practice, and in that practice, to discover always anew. the artistic process is intuitive rather than strictly methodological. overcoming dualisms, stressing creativity as affirmation, and overcoming transcendentalist thought are shared by both philosophies. as deleuze and parnet ( ) puts it: it would not be enough to oppose the east and the west here, the plane of immanence which comes from the east and the plane of transcendent organization which was always the disease of the west; for example, eastern poetry or drawing, the martial arts, which so often proceed by pure haecceities and grow from the 'middle.' the west itself is crisscrossed by this immense plane of immanence or of consistence, which carries off forms and strips them of their indications of speed, which dissolve subjects and extract their haecceities, nothing left but longitudes and latitudes. ( , added emphasis) the korean scholar, hyeyoung maeng ( ) has provided many parallels between deleuze's process ontology and the practice of korean bunche painting. she shows that many of their concepts can be transposed and illustrated through this painting process. perhaps more impressive are maeng's ( ) many attempts to provide the linkages to tao in her impressive doctoral thesis. there she notes, in a thousand plateaus there are many passages that confirm such rich parallelism. deleuze|guattari describe the tao as "a field of immanence in which desire lacks nothing and therefore cannot be linked to any external or transcendent criterion" ( ). where yin and yang are processes of stratification and destratification, while the state of tao as the plane of consistency can be thought of as "the totality of all bwo's, a pure multiplicity of immanence" ( ). there are many more parallels, of course. both taoism and deleuze|guattari are process philosophies: they are about change and becoming, and not about being and representation. it is very difficult to articulate the significance of deleuze|guattari for art and its education in any simple and easy way. there is no 'method' for art, only emergent processes that are not always successful; there is no 'research' in art as is so often wanted, only 'research creation.' the second stanza of chapter of tao te ching states: therefore the sage goes about doing nothing, teaching no talking. the ten thousand things rise and fall without cease, creating, yet not possessing, working, yet not taking credit, work is done, then forgotten. therefore it lasts forever. research creation in art speaks to 'the way' in tao. as lieh-tzu (graham ) says: "will the way end? at bottom it has had no beginning. will there never be more of it? at bottom it does not exist?" ( - ). james elkins ( ) once wrote a book maintaining that art cannot be taught. artistic research creation presents this paradox of c reation by artists who never 'arrive,' are never finished, and are constantly 'on their way,' searching for 'the way,' which does not exist, working with contingencies, surprises, accidents so as to enter a zone of imperceptibility: becoming-impersonal, becoming-indiscernible. the tao also seems to address the plane of immanence (or plane of consistency). as deleuze and guattari ( ) put it in what is philosophy: "the plane is clearly not a program, design, end or means: it is a plane of immanence that constitutes the absolute ground of philosophy, its earth or deterritorialization, the foundation on which it creates its concepts" ( ). we read in the book of lieh-tzu: "there was primal simplicity, there was primal commencement, there were primal beginnings, and there was a primal material. the primal simplicity preceded the appearance of the breath. the primal commencement was the beginning of the breath. the primal beginnings were the breath beginning to assume shape. the primal material was the breath when it began to assume substance. breath, shape and substance were complete, but things were not yet separated from each other; hence the name 'confusion.' 'confusion' means that the myriad things were confounded and not yet separated from each other" ( - , graham's translation). confusion and chaos, what félix guattari called 'chaosmosis,' appear to be the same concept, while the virtual provides for the quantum notion of superposition: the potential configurations of an assemblage-as an arrangement of particles or fields-that actualize as singularities within specific ecologies. thinking this way avoids considering only a 'big bang' theory of the universe-that is, a plank epoch or plank era that starts the arrow of time (e.g., . billion years). rather, it suggests a multiverse as a timeless state of indeterminacy, closer to a hartle-hawking universe, which claims it an emergent singularity. it is precisely singularity or an event, as deleuze and guattari stress, that generates form. perhaps the emergence comes from black holes, for it is in these w(holes) where chaos as 'confusion' reigns supreme; it is precisely here where light is bent and gravitational force becomes the stumbling block to complete the search for an all-encompassing unification of quantum mechanics and general relativity as a 'theory of everything' that would combine the four known forces together: gravity, electromagnetism, the weak, and the strong interaction. it is the quantum nature of gravity that remains enigmatic and allusive in this quest (snaches ) . it is not difficult to see how deleuze|guattari envision a cosmology and the interplay of the cosmological forces of chaos as inorganic energy (unleashed energy, zoë), a pure bwo, the tao of zero, laruelle's one (a radical real), the 'given as given,' and the material force of chi (ki) of tao, and so on, where 'mad' particles abound. for those who have tried to read deleuze|guattari will readily testify how difficult this task can be. one doesn't comprehend their writing style or philosophy easily. yet there has been an explosion in various areas concerning their work in virtually every scholarly area: including literary studies, cinema, education, urban planning, literary studies, architecturethe list goes on. perhaps the most significant shift by scholars who have embraced their work has been the realization that representational thought dominates most disciplines globally, and this needs to be urgently and constantly questioned. artistic research in particular needs a complete rethinking here where thought and image are in tension with one another (see especially vellodi ). in the field of art education, this is particularly true when it comes to visual cultural studies a sub-discipline that emerged out of cultural studies. (i discuss this development in chapter .) here, the usual approach of representational thinking is to understand difference always in relation to sameness. this is a very seductive way of thinking as it appears that social equality is being pursued, and that justice is being achieved as the question of human rights is meant to put everyone on an equal footing. from a deleuzian standpoint, representational thinking simply reinstates inequality by masking the situation. representational thought places the question of the other, either too far or too close. what do i mean by this? by 'too close,' difference is masked under a humanist universalism: we are (after all) all the same despite our differences. but, despite these differences, we somehow 'know' or try to 'know' and understand the other. we are 'like' them, and we try to communicate and get to know them despite the gulf between us, because, in the end, we are all 'human beings.' such a stance generates fantasies of compassion by maintaining distance, for it is we, after all who are reaching out to the other and trying to grasp their identity. it is not them who are reaching out to us. for a trained anthropologist, this seemed to be a commonsensical approach. after all, the intent is to get an inside ethnographic view to close the distance. ideally, this would lead to becoming the other, but when this happens, if the collapse is complete, then 'differences' are no longer discernible. "going native" now becomes ideologically questionable as does the question of 'cultural appropriation.' artists such as paul gauguin, emily carr, and 'haida' artist bill reid have all been criticized for getting 'too close' so that the self-other gap is reinstated and identity preserved. by being 'too far' from the other moves in the opposite direction. this means "we" are "unlike them," or unlike the other, so we must remain silent in relation to what is alien; otherwise, we do away with their difference, or interfere with it. this leads to fetishization and exoticization of the other because it is this very difference that supports our fantasies. this is how difference always forms a dualism with sameness. representational thinking, through this either-or logic always, generates a hierarchy in relation to the structure of sameness and difference. the other's difference is always 'used' to position the relationship in unequal ways on the grounds that equality is being pursued. theorists like balibar ( ) called this post-racism, neo-racism, and even cultural racism. let me offer an illustrative example. there is a rather ingenious advertisement by benetton. it shows three hearts, which are similar but slightly different. they are images of pig hearts but they seem human enough as photographed by the italian designer oliviero toscani in , who spearheaded its 'the united colors of benetton' campaign. benetton has since cut ties with toscani in over his remarks concerning the collapse of the morandi bridge in genoa. (i encourage the reader to look up this image on the internet as copyright laws prevent me from including it in this chapter.) despite their minor differences, the three images fall under the categorical signifier 'heart.' each heart, however, has another signifier that 'represents' a race in an essentialized way in terms of identity formation: white, black, and yellow are the colors that signify three different races. the implication being that all three races, beneath their skin, are equal or the same. at first glance, this is a very seductive advertisement, for it appears that benetton is certainly making a claim as to the inequality of the 'surface' color and comparing it with the 'depth' of bodily organs where we are all the 'same.' we all have 'heart.' we are all 'human' says the advertisement despite the color of our skin and despite that these are animal hearts. so, how can this possibly be 'post-racist' advertisement? this seems counter-intuitive. differences (as particularities) are contrasted to the generalization of sameness: we are all of the species homo sapiens (or 'pigs' in this case!). difference in-itself, as singularities, is what is negated, and it is precisely this oversight that deleuze|guattari attempt to overturn. they base their philosophy on pure difference alone. when negation happens (as discrimination for instance), the idealization of homo sapiens at a physical level of idealization is then 'smuggled' in. this has been the primary problem with the signifier 'human': who belongs and is given privilege and access to this category? like the hearts in this advertisement, it appears to be an all-encompassing inclusive category, yet we know historically who is justified by this signifier is selective and exclusive: the most obvious examples are women, indigenous peoples, and african americans. they quickly come to mind, but the entire post-colonial landscape is shaped by this exclusionary term, human, that appears inclusionary. the advertisement appeals to an impossible category of equality called 'human,' which is endlessly discriminated against due to the logic of the multiple binaries it generates within the sameness-difference logic of representation based on the purity of visible color: red, white, yellow. such color boundaries become troubled very quickly when shades, tints, and tones come into play; their intensities of affect begin to blur boundaries between colors, introducing confusions of 'distancing' and 'passing' which representational identities present politically and ethically. deleuze, in difference and repetition ( ) , succinctly listed four postulates of such logic: ( ) identity, ( ) opposition of predicates, ( ) analogy when it comes to judgment, and ( ) resemblance in perception. when it comes to benetton's unhate campaign (again, i encourage the reader to find these images on the internet), all the president of the united states (barack obama at the time) and the former general secretary of the communist party of china (hu jintao) have to do is simply kiss each other. collapse distance and then world harmony will begin. all will be forgiven. this is a very hegelian move: two opposites come together (thesis and anti-thesis) and an aufhebung will take place, elevating the contact to a higher plane of resolution. simple, isn't it? love overcomes all, in this case homoerotic love. but just how races (or nations, or ethnicities, or women, or men and so on) are the same, overcoming their dichotomous binds, remains a representational illusion. this is a transcendental ideal that can never be reached. the physical heart, just like the 'kiss,' is itself an ambiguous sign that says we are all biologically the same when we know that there are psychocultural differences that cannot be easily dismissed. traumas that prove too deep to be flippantly done away with. the judas 'kiss' of betrayal is repressed. north korea and south korea present this difficult paradox of difference and sameness. there is a half-truth to the unhate campaign, which again makes it so seductive. bodies do affect each other. "we do not know what a body can do," a phrase that is so often repeated in the secondary deleuzian literature. this was deleuze's great insight when he embraced spinozian philosophy. unquestionably, great diplomacy can overcome what are impossible differences, even when they are in the end thwarted, as happened when bill clinton brokered a peace agreement (the oslo accords) in between israel's prime minister yitzhak rabin and the leader of the palestine liberation organization (plo) yasser arafat. two years after, in , rabin was assassinated at the kings of israel square in tel aviv for his moderate liberal views by a radicalized right-wing rabbi extremist. east asian cultures tend to define the self in relation to a group or collective whereas within western cultures there is greater emphasis on the self to be viewed as unique and independent of the social group. in these benetton advertisements, the assumption is made that ( ) races are categorical, there is no overlap, and ( ) they resemble one another on the physiological level. we all have 'heart,' and therefore ( ) we are all equivalent under the general law that we are 'human beings' capable of great compassion and care. true. especially when it comes 'to our own.' this is the commonsense view that is continually reproduced in the current image of thought, especially in multicultural societies where there is a deep need to 'tolerate' differences under an assumed sameness of equality by human rights legislation. "diversity," as deleuze ( ) argued, "is not difference. diversity is given, but difference is that by which the given is given, that by which the given is given as diverse" ( ). diversity is rather the negotiation of distances and toleration for the other within representational logic. for deleuze, thought as reproduction, the common image of thought simply affirms identity; it is thought that produces the new images that are not in circulation, which stimulate and create the new. benetton manages to stabilize differences along colored lines via a universal ideal, or generalization that, on its surface, seems perfectly fine, but that means erasing all 'differences' as singularities so that we end up with really nothing at all in relation to justice, rights, equal treatments, peace rather than war, and so on. identity politics continues to sustain its affective force. difference in representational thought is always in relation to a conceived identity; it is a judged analogy (the other is too close) or an imagined opposition (the other is too far). for deleuze, thinking in terms of generalities and particulars operates within the limits of representation, which is a "transcendental illusion." "'i think' is the most general principle of representation" (deleuze : ) . this means that the human subject 'grounds' knowledge in relation to a metaphysical notion of an outside reality, as if a thinking subject is able to transcend and represent the meaning of an object (e.g., culture) in accordance with an external reality. the authority of the subject is always in play to establish truth claims. deleuze|guattari are against any forms of identity as the "dogmatic image of thought." all representational thinking presupposes a subject-object gap, a 'correlationism' (meillassoux ) . our consciousness produces images from or of things that our perception grounds in representation. this is the kantian axiom where the transcendental field is in the image of the empirical field. "i think" is added so that a transcendental ego is shaped; there is a "transcendental unity of apperception," and thus a distinct anthropocentrism established to all judgments. deleuze, on the other hand, maintains that this transcendental field remains differential, impersonal, and pre-individual. the condition for 'real experience' is virtual, which has no identity. correlationism is already the product of actualization of this prior differential field. against representational thought of any kind, deleuze|guattari develop a philosophy of becoming where time is always a factor (as discussed above). like the tao, which says, "the movement of tao is to return; the way of tao is to yield" (verse , tao te ching) (lao-tzu and takuan sōhō ) . in anti-oedipus ( ) , deleuze|guattari draw on the figure of the schizo as exemplifying such an ego-loss, a yielding: "[the schzo who practices a psychic minorization] produces himself as a free man, irresponsible, solitary, and joyous, finally able to say and do something simple in his own name, without asking permission; a desire lacking nothing, a flux that overcomes barriers and codes, a name that no longer designates any ego whatever" ( ). this suggests the creativity of singularity, of becoming imperceptible. the tao of zen seems to follow this: "[n]othing remains the same for two consecutive ksanas (the shortest imaginable periods of time)" (hanh : ) . nothing has a fixed representational identity, what persists is the paradox of difference itself. the most sophisticated forms of representation are the performative and constructivist approaches to subjectivity as exposed by michel foucault's discourse theory. representation is considered in terms of what it does and not in terms of truth or accuracy; the question is how both the subject and the object are mutually structured or constructed by a discourse. the discursive image implies an ideal subject position or an ideal ego in lacanian terms, and asks the spectators to emulate or take part in this image through their own ego ideal. here the performative notion of the image is usually shaped cleverly by and through company image or logo branding. a brilliant example comes from the country of peru that re-branded its national identity in the globalized world. in what is a case of reverse post-colonialism, a busload of well-known peruvians-actors, dancers, and chefs-traveled to peru, nebraska in the united states to sell the peruvian culture to a population of . nebraskans danced, ate peruvian food, and learnt a few words of spanish. this delightful 'invasion' can be viewed on youtube https://www.youtube.com/ watch?v=tor lhymocu. it is a marvelous example of performative and representative notion of the image: what it can 'do,' that is, the power of its affect to influence and shape desire. what an image's effects are for global capitalism is what deleuze|guattari worry about as deleuze ( ) briefly mapped out in a short article on a 'societies of control' that draws on the capacity of third-generation cybernetics to program algorithms to run smart machines (as discussed throughout this book). when it comes to performative imagery, the subject is never certain. one has to be lured into comply with an ideal ego that is being offered. the inhabitants of the small village of peru, nebraska have to identify with something that is ideally, essentially, and nationalistically peruvian in latin america to insure recognition of the peruvian logo that categorizes their difference (as diversity). what is 'different' as abject is meant to turn into its opposite, as something that becomes acceptable, even enjoyable, and thereby increasing the comfort level of being with the other. its performative affect is thereby achieved. this is the repetition of difference in the postmodern sense. the so-called interactive social media attempt to continually adjust what customers and spectators are saying, voting, tweeting, posting, and so on, so that the discursive formation of the ideal ego can be maintained by the company, state, or those in charge. management becomes a huge industry while prosumers do their bidding inadvertently. in a modulated society of control, says deleuze ( ) and guattari, consumers become produces and vice versa in a never-ending look to maintain some sort of structure so markets are stable within certain degrees of freedom. as judith butler ( ) has developed this, the performative (or active) notion resides in the possibility of always contesting the ideal ego image: the so-called norm or ideal is always changing historically. the problem is that such resistance or protest is always co-opted inside this machine; the negative resistance fuels change through conflict, if only to reverse trends in a never-ending question to have equity and equality that can never be achieved, a metaphysical ideal. there are/were 'accelerationists' like nick land in the s (warwick university) who thought that speeding up the cyber-technology invented by global capitalist system would lead eventually to its collapse via a 'cyberpunk phuturism' complete with a 'zombie apocalypse.' although this wasn't likely to be the case, one wonders today about such a scenario when sars (covid- ) has been let lose? land's 'dark enlightenment manifesto' ( ) continues to be thought provoking if not an earth-shattering view of a dystopian capitalist future. simon o'sullivan ( b), with strong roots to deleuze|guattari, has attempted to provide a strong and worrisome assessment of accelerationist thought as has benjamin noys ( ) . but, on to more humous note. an intentionally funny and ironic example of this never-ending quest for equality in representational terms comes from the american film culture. the film, the lone ranger ( , d. gore verbinski) (screen image . ), based on a well-known television series in the s (screen image . ), was meant to intervene in the myth of the wild, wild west in order to subvert the historically inequitable indian|white man relationship. try as the film might, the exotic first nations 'warrior,' tonto (played by johnny depp), set up as the true brains behind the lone ranger (played by armie hammer), was just not convincing enough. the film was a flop. identity politics around gender, sex, ethnicity, race, and so on is continually in flux in representational thought. the 'hollywood' solution to this is to work out a politically correct neutrality by having 'one of each' so to speak, for equal screen image . lone ranger ( ) representation. hence, an equal number of men and women, or an equal number of races represented in films. companies like disney have become particularly sensitive to such political correctness. again, this is precisely what deleuze|guattari in tp attempt to dispel and query. they introduce the term 'minoritarian' rather than minority to worry and confuse the numerology that majority-minority dualisms represent. minoritarian is always an ethico-political positioning that requires action when challenging power and discrimination to deterritorialize itself from the majority, and hence not necessarily bound by number. it marks the potentiality for change through a new becoming or singularity. most of art and its education are structured by this "transcendental illusion." the power of deleuze|guattari's shift is to a position beyond representation, toward anti-representationalism that strives to think beyond all subjectivism, to free thought of the illusion of transcendence. in this sense, deleuze|guattari are closer to eastern thought than western, which remains caught by representation. deleuze|guattari drew on what was a "minor philosophy" in the west-spinoza, whitehead, nietzsche, hume-to develop non-representation processes of creative becoming. of course, it must be understood that an encounter with art that creates thought is not an experience that is decided in advance; its advent cannot be planned. deleuze|guattari reserve the term becoming imperceptible (as developed above) where what is experienced cannot be recognized; a limit is reached when common sense becomes unhinged. for deleuze|guattari, this is an indication that an event of art, an encounter has indeed taken place. a thought without image has been engendered, which gives a hint of what 'imperceptibility' is about. one of the main concepts deleuze develops against representational thought of any kind is that of the simulacrum. the most common notion of a simulacrum is a copy of a copy, like photorealism where the painting is a copy of a photograph, which itself is a copy or representation of the world 'out there' that acts as the originating model. whereas a copy is made to stand in for its model, simulacra as repetitions of the model introduce a difference that turns against the 'original' in order to open up a new space and a new world. it affirms its own difference and unfolds a multiplicity of new problems and solutions. it is what deleuze refers to as the powers of the false. simulacra are non-representative because they are nothing but appearances-just 'qualities' of things as part-objects. there is nothing 'behind' the image. the will to art (kunstwollen) consists of extracting differences from repetition by reversing copies into simulacra. art neither represents nor imitates because it simply repeats. everyday life is characterized by repetition, as a return to the same, through habit, primarily through standardized production of commodities and the proliferation of information. art, from a deleuzian standpoint, is not opposed to such a mechanical, stereotypical, and habitual repetition, but it embraces this standardized production of the commodity to expose its limits and extract what is differential and virtual within it. the task of art is to open a "line of flight" (ligne de fuite) that passes from the virtual to the actual by interrupting repetition with a difference. gradually, repetition is transformed from the repetition of the return of the same to creation based in difference. the contemporary artist, matthew ritchie's repetitions that involve his process approach to drawing, is an appropriate example (screen images . and . were taken from the pbs art: featured segment on ritchie called 'structures'). ritchie works with a computer program that enables him to play with fractal spaces. he can enlarge drawings, manipulate them, and further turn them into sculptures. each time the installation travels to a new gallery, a difference 'structures' has been introduced to the point where the exhibition itself has morphed into something different. richie's 'play' with drawing enables a de-anthropocentrism to take place as human scale is no longer a viable measure. the assemblage consists of the computer program, richie, as one of many assistants who rework the installation each time, and the new gallery space he is assigned. completing each renewed assemblage are viewers who are thrown into a rhizomatic web of lines that never seem end. virtual in the deleuze|guattari lexicon does not mean 'virtual reality'-the 'virtual' as seen using special vr headsets or d cinema glasses. rather, it refers to a realm of the invisible where memory and different orders and combinations of time are in play (past-future, present-past, present-future). these enfoldments of time are worked out in deleuze's two cinema books (the movement-image and the time-image). like tao, or the way, time is an eternal intuitive process, not chronological. the act of genesis does not have some set method. it is profoundly enigmatic. for instance, no one knows how or from where the 'blueprint' for the 'origin' of any species takes place. there are only speculations like those of the 'heretic' rupert sheldrake's ( ) 'morphic resonance' theory where a shared species memory is posited, perhaps located in the nowhere time-space of aion, or tao, being the dark matter and energy itself. where do the morphogenic field instructions for cell growth come from to form an embryo? how does such self-organization come about? epigenetic processes have put a damper to the jubilation on any claims to genetic determination. either way, it is the virtual that is of key concern. deleuze rejects the notion of the 'possible,' which remains attached to the logic of representation. it is an empirical event that will be actualized sometime in the future. it is based on prediction and calculation. possible events are posited, then, depending on circumstances, one of them will be realized, but this a 'hedged' future. in contrast, deleuze also maintains that the virtual is fully 'real.' the virtual "is not opposed to the real, it possesses a full reality by itself" ( : ). for genesis to occur, the 'real' multiplicity within the virtual must be actualized, which is to say, the potentiality of co-existent differences becomes 'real.' there is no predictability in this, only radicalized contingency as to how this virtual transcendental field is integrated or resolved. this is a true 'event' where the future has not been predicted; that is, evolution proper for deleuze where the virtual is creatively actualized, whereas the possible is simply 'realized' (deleuze and parnet : - ) . this is simply to say that the possible 'real' is already preordained either through limitation or resemblance of a preconceived image as to what is about to happen. following this logic, deleuze ( ) can say in his book on bergson that the 'virtual' is not opposed to 'real' but opposed to 'actual,' whereas the 'real' is opposed to the 'possible.' repeating his proust formula, he says that such states of virtuality are "real without being actual, and ideal without being abstract" ( ). the creative event belongs to the real as an actualization of something new. the artist draws on an event that has moved him or her. this event is a disruption of chronological time; it belongs to cosmic time. the artist draws from cosmological forces-the rhythms of 'a life' or zoë. these terms are used interchangeably as in deleuze's last monograph called pure immanence: essays on a life ( ). zoë is unbound creative energy as inorganic life (it has no organs). this is like the tao of the 'uncarved block,' which is capable of infinite characterizations, all possible ordered worlds, what deleuze calls univocity. the tao te ching, the main text of taoism uses the term 'pu' as this 'uncarved block': all nature is as its most powerful when considered in its original form-as energy (or zoë), which would be quite consistent again with pluralism = monism, monism being equivocal to pu = uncarved block as the virtual of potentiality for multiple actualization, not unlike the concept of a 'perfect quantum vacuum.' simulacra another example of simulacra is in order as developed by gregory minissale ( ) . everyone is familiar with duchamp's infamous fountain, his ready-made that put to question art 'outside' the sacred halls of the gallery or museum. unfortunately, his fountain was destroyed, but he also miniaturized it hundreds of times, times to be exact in his boite-en-valise ( ) ( ) ( ) ( ) ( ) ( ) ( ) series. the box in a suitcase began as a -box edition. it then went through a further editions of , until the final series was completed in . this sequence of duchamp's art process is a precursor to the developments deleuze discusses in difference and repetition in . duchamp raised concerns over the copy or facsimile in an age of reproduction, which has only intensified in our digitalized age. the miniatures of duchamp's work that are reproduced in the suitcase are meticulous replicas of his oeuvre. it took , reproductions and the development of a new laborious process of replication to 'copy' his work over the span of years. this was to be the 'end game' (like the chess that he loved so much) of the idea that a gallery and a museum were the custodians of unique original works of art. duchamp offers a strange autobiography where his past works are preserved, yet an impossible equivalence is maintained between the original and the copy, where a and a′ are 'different' only because, a′ (as a perfect clone) is displaced only by space and time. it is the idea or the problematic that duchamp wishes to 'preserve' by embracing the copy or multiple, which now becomes an uncanny simulacrum, like the difficulty of telling a recording from someone actually singing. duchamp, in effect, was articulating what deleuze was to do by overturning the platonic distinction between original, copy, and simulacra wherein the idea of an essentialized ideal reality-a superior realityjustified the judgment of inferiority. identity prevails. copies as icons most closely resembled the eternal idea, whereas simulacra in which no semblance was found were the most inferior. deleuze transformed this platonic idea as an intensive multiplicity. "overthrowing platonism refers to allowing simulacra to assert their rights over icons or copies. the motive of platonism is to distinguish essence from appearance, the intelligible from the sensible, the idea from the image, the original from the copy, the model from the simulacrum" (deleuze : ). deleuze's simulacrum is not a false copy, nor is it a lack of similarity; rather, it is an image without a resemblance to a model. simulacra have internalized differentiation, or difference-in-itself as justified by an 'eternal return,' wherein difference in kind, not degree, occurs. intensity changes and is actualized differently. the platonic idea now becomes a virtual multiplicity, a chaosmosis of differences. it is qualitative multiplicity. it cannot be divided up without changing its nature as its intensity would then change. it is continuous, non-numerical, subject to change whenever it is divided. but, there is also a quantitative or external multiplicity: discontinuous, measurable, and calculable. the idea as a virtual multiplicity forms various problematics, which are actualized through a process of differenciation where numerous differences in kind are created by its movement. differenciation refers to the processual virtual content of an idea. in contrast, deleuze reserves the word differentiation for quantitative multiplicities of representation. ideas, the power and potential of thinking as such, unfold a never-ending creative problematic. this problematic is only realized 'after the fact': looking back as much as forward. in tao, the way forms a visible trace through sensation. the virtuality of the problem is revealed, so to speak, by the conditions that were in place. the idea of the anthropocene, for example, is of great concern. it forms the current problematic. while the above discussion is complex, it helps to grasp duchamp's problematic; his idea of asking about art its own status, and it brings us back to repetition with a difference. i draw on some remarks by alain badiou ( ) on duchamp here to make my points. part of duchamp's anti-retinal and anti-romantic problematic was the idea of the "infra thin," "to pass into the infra thin interval that separated two identical things." this is the foundation for his use of reduplication, copies, and multiples that constituted his reputation. he would sign a copy, or a miniature or a multiplication of one of his works by the famous inscription "certified true copy," even when done by an assistant. the infra thin is a point of minimal discontinuity from the same to the other same. as badiou says, "the new productive and reproductive thought must pass by this point." this point, then, belongs to the tao of the immanent plane, the in-between, the middle-of-things, the rhizomatic line of flight, the infinite regress between and of complex numbers, like the square root of - . it is the same concept as paul klee's graupunkt, the grey-point of chaos. the graupunkt is cosmological and virtual. it is the point before color establishes itself, the tao of color, like the pre-plank universe. duchamp's idea is not 'embodied' in the artwork, or by his oeuvre. it is there 'on the surface,' like the way is 'on the surface,' but at the infra thin "point that separates that being-there from itself." it is but a touch, a trace. simulacra and repetition take on yet another twist via the writings of jacques baudrillard ( ) . his claim was that we had entered an age of hyperreality, a world of simulacra-a copy of a copy, or a hyper-copy where artificial rubber trees have stronger scents in disneyworld than they have in nature. however, this is not the simulacra of deleuze. it is the affect or the 'force' of the artwork that begins to matter. deleuze in was heading in the same direction as duchamp, as were fluxes artists who continued this line of thought. andré malraux's "museum without walls" presents a similar idea: in the twentieth century, images become uprooted from their geographical contexts and can be organized purely along stylistic grounds. the entire history of images becomes a virtual past that then allows both cross-cultural and transcultural (that is global) comparisons among artists to take place. deleuze picks up on this as well through the writings of henry bergson. images are now put on a virtual plane, and in a similar way, walter benjamin's "optical unconscious" drew on the idea that images have been released from site specificity, to "any place whatsoever." baudrillard's simulacra become a world of screen images; his "precession of simulacra" meant the blurring of true|false, real|imaginary. simulation begins when opposites collapse. this brings to fore a whole new problematic. duchamp's fountain can no longer be held up as an original. when we look at the variety of t-shirts that are found advertised on the internet (screen image . ), these are obvious tacky imitations reproduced commercially, we immediately wish to dismiss them as just cheap copies-mere simulacra following the platonic aesthetic. yet, the "t" shirt raises the same status of art that duchamp's fountain did. underneath it says, "the great attack on the institution of art." the t-shirt is not found in an art gallery or institution, while duchamp's mass-produced urinal is, and yet both the urinal and the t-shirt are mass produced, mundane objects, one referencing the status of the other. the t-shirt raises the same issue about the status of art as duchamp's fountain. it raises the questions about identity (jokingly it can be an invitation to urinate on it) and status of fashion discourse in relation to screen image . duchamp t-shirts, screen image body and self-image. the t-shirt is a simulacrum; it takes its idea from duchamp's fountain and sets it in motion to the realm of fashion, uncomfortably destabilizing the world of ordinary objects and the world of art. a "fountain dress" (available on the internet), designed and copyrighted by philip colbert, offers another example of the duchampian idea. the concave urinal, a container for oncoming urine, is presented as being visually the opposite of a full pregnant woman's belly when placed on this model and worn like a dress. the sexual joke of the model having a penis is also evident as the drawing of a pipe that drains the urinal is placed directly below the model's waist emerging from her pubic area. what is this dress performing? what are its affects when worn in public, or even on a model runway? this can lead to many misogynistic and homophobic examples of duchamp's fountain, all repetitively exploring the idea in a variety of representations. there are a number of examples to cite here. meike van schijndel has designed a "kiss urinal," essentially a woman's open mouth with full-bodied crimson-colored lips that forms the urinal itself, with two white teeth showing on the top lip. its implications are pornographic, alluding to 'water games.' there is also a 'jaw urinal.' imagine the dark open mouth of a shark with long razor teeth. i gather that male urination in this case is both out of fear and against fear itself. there are outright misogynistic urinals as well: quite literally a urinal is placed between a model of a woman's outspread legs, as if the urinal was an exaggerated vulva (see image , https://www.oddee.com/item_ .aspx). the 'holy mary' urinal is a half bust of the mary figure; the male urinates into the cavity of her chest (image , same website). there is also a homophobic urinal that has a contorted male figure with testicles and erect penis, his hands seem to be spreading his anus, while his feet are vertically in the air (https://imgur.com/gallery/qxtjota). i encourage the reader to find these urinal designs on the internet or to go to the websites i have listed. so here we have a series of simulacra that are ethically worrisome as they perpetuate a male|female binary. these urinal variations become a never-ending loop, a form of representation where no difference is allowed in, no duchampian 'infra thin interval' at play. the public toilet becomes a marginal space, personal and yet shared, to spread hate and division among gender and sexes. the tao of yin and yang seems perverted rather than complicated. each play of fountain, as simulacrum, introduces a difference, which parodies the idea of art in an "age of mechanical reproduction," to echo walter benjamin's wellknown and studied essay. there are many, many more variations of the urinal. when the urinal becomes a self-repeating loop, an eternal binary that shuttles between male and female, simply an object where a female is to receive a mouthful of urine or her vaginal filled by male urine, then difference in the deleuze|guattrian sense is not possible as the dualism plays into the same|difference of representational thought. there is no 'difference' that makes a difference in any of these examples: again there is no thought-provoking interval that might lead us into the cosmology of sex, the yin-yang of tao, or difference-in-itself that duchamp's idea problematizes. as deleuze puts it, an idea is "neither one nor multiple, but a multiplicity constituted of differential elements" (deleuze : ) . in contrast, korean director kim ki-duk's film dream ( ) problematizes yin and yang forces (see chapter for this exploration). however, sherrie levine's fountain (buddha) ( ) might be an exception here as well. copyright prevents me from showing you the image, but i encourage you to look it up on the internet. the sculpture is clearly a parody on the shape of duchamp's urinal as a 'significant form.' the urinal looks like a 'sitting' buddha. she seems to be riffing on duchamp's buddhist leanings when thinking outside the box, at the same time it is made of shinning brass-as if gold, suggesting that duchamp has also profited from his transgression making it a 'golden commodity.' the paradox is that duchamp introduces a taoist idea, yet it remains caught by the simulacra of capitalism. of course, there is a lot of nonsense spoken about this piece, which adds to the nonsense that i am providing-the best being that levine is alluding to brancusi's bronze sculptures who duchamp mentioned when referring to his own urinal. it seems the blatant allusion to buddha drops out in such formalist analysis that dwells on artistic genealogies. the simulacrum dispels any form of representational truth or essence or category of an object or thing, and works with what deleuze calls the "powers of the false," where the only truth is time itself; that is change, the "eternal return of difference" as theorized by nietzsche, where and when the new emerges. lacan made the same differentiation between wiederholung and wiederkehr, only the latter brought a 'true' change. what is significant for art and its education is that a deleuzian approach is always concerned with becoming, but a becoming-other. how might it be possible to break the categorization of an object through art? this is a huge challenge. one way to break with representation is to place the object in such an arrangement or frame or modification so that the object or thing takes on a 'supplementary dimension'; sides of existence emerge that were never seen before or thought of before, like mathew barney's strange cremaster cycle of films, for example, that explore the inner workings of the male cremaster muscle that raises or lowers the testes in response to temperature. the object is made strange, abject, or made to tell a story never imagined before. the object is made to do what it was never imaged to do, or show its secret life-that is, its other life to produce a seeing that is beyond the object's everyday existence as we know it. in other words, other qualities of the object, what deleuze calls its virtual multiple dimensions, are exposed. he called this a shift from the object to the objectile, from the subject to the superject, and from essence to event (savat ). all objects are "part-objects" in this sense, so the usual representational understanding of 'object' is mistaken. an object is always in flux, never stable, always in process and elusive. when such an object becomes strange, lacan referred to it as objet a. it is alluring, strange, coming from the outside, magical in its own way. part-objects are always formless and functionless, composed of particles that increase and decrease at various speeds and intensities. they are cosmic in this sense, like those allusive elementary quark particles that are mathematically posited, but are yet to be 'discovered.' such an understanding of 'object' follows the taoist saying: "though formless and intangible-it gives rise to form; though vague and illusive-it gives rise to shapes; though dark and obscure-it is the spirit, the essence, the life-breath of all things" (lao-tzu, verse ) (lao-tzu and takuan sōhō ) . in a deleuzian saying (coming from spinoza): "we never fully know what a body can do," or what it is, or how other bodies affect it, or how it affects other bodies when placed in different assemblages. let me show you two examples of this as related to visual culture. we all know that bottled plastic water damages the environment. the coca cola company is the worst offender but does everything in its power to claim it is concerned about water for the earth. it is difficult for people to stop drinking bottled water, especially in the west, which already has good quality water. how could art intervene in such a problematic, a term deleuze|guattari use, for there are no easy solutions? the first example is directed at commodity culture. duchamp's repetition and infra thin point are used effectively in a creative way. seattle-based artist chris jordan in his exhibition running the numbers i and ii has tried to intervene in this problematic by exposing the waste in our throwaway capitalist economies so that the numbers of thrown away commodities become sublime and incomprehensible for us to grasp. we are unable to imagine the number of thrown away goods that happens every second of every day. the numbers become non-representable. his illustrations transport the viewer to this place of incomprehensibility; they become cosmic equations of the tao. i encourage the reader to view jordan's website to view the many examples that startle you as to how wasteful capitalist commodity societies are. in one such image, , phones are shown that are thrown away each day in the united states. through repetition and miniaturization, these objects confuse the anthropocentrism of scale that keeps the frame of art together. jussi parikka ( ) called this state of affairs the anthrobscene rather than the anthropocene. the rare minerals needed to keep the technological machinery in production are likely to run out. the second example is an artistic intervention into this ecological problematic: advertisements that are part of the filter for good campaign. i encourage the reader to go on the internet and look at the images for this campaign. you will see two portraits of a man and a woman with black tar running out of their mouths, onto the white t-shirts they are wearing. these advertisements encourage you to drink filtered water and try to precisely intervene in our throwaway society of bottled water by showing a side of plastic we don't realize. it is an 'assemblage' that recognizes the agency of plastic as a petroleum product, which is made of tar, and tar is shown in these images as an abject object that is not to be swallowed. yet, it appears that we have done so inadvertently, and it is now oozing out of our mouths on its own. plastic becomes transmogrified (mutated) so that as an object its affects are felt at the deep level of our nerve-body, what deleuze|guattari called a body without organs (bwo). these images make us hesitate and (potentially) stop drinking water out of a plastic bottle. plastic becomes more 'nonhuman' in the way it interacts with us; at the same time, its agential force is recognized as it has now become visible and materialized. it becomes an agent in its own right as it acts on us. this is what deleuze|guattari call an assemblage, when the human and the nonhuman come together in a form that is entirely unexpected. heterogeneous elements combine in this case to present a repulsive affect. they encourage a political strategy where striated space, that is, space that is already occupied by the coca cola corporation is intervened by a 'smooth space,' an opening that introduces difference into an already occupied and full molar assemblage. one can say that this intervention is an n- that opens up a new space and time. again, this is a taoist move when we read "the most yielding thing in the world will overcome the most rigid; the most empty thing … will overcome the most full" (lao-tzu verse , verse ) (lao-tzu and takuan sōhō ) . there is always a way to the outside to force thought. plastic as a petroleum product becomes an objet a in lacan's terms. the subjectivity of the artist, for deleuze|guattari, is not one that elevates the ego; rather, it is a subjectivity of what they call nomadic becoming-to rid the "self" in such a way as to become 'imperceptible'-that is, to become someone who falls outside the usual categorizations, and who creates possible worlds through their art, provoking sensibilities that enable encounters of another 'compossible world.' duchamp was such a nomad. deleuze calls on jorge louis borge's story "the garden of the forking paths," where the chinese philosopher, ts'ui pên, presents the tao of the myriad of bifurcating paths that can be followed. of course, as a virtual multiplicity, they are all simultaneously available, like all colors are available at the graupunkt of klee's color theory. incompossibilities and dissonances that the virtual real presents have to be 'walked' by the nomad to find the way. there is always an attempt made to be attuned with the world in deleuze|guattari terms, and this is very taoist attitude. to become 'imperceptible' as an artist means constantly being attuned to difference in the world, what deleuze|guattari call 'signs' that affect you when the world looks at you, rather than you looking at it. something from the outside forces thought. some artists, as we know, use their own bodies as a technology of self-knowledge to become nomadic and imperceptible. i think here of the famous performance artist orlan in this regard. for example, in her 'self-hybridization series' (internet image search: 'orlan, self-hybridization series' will generate an array of these images. copyright prevents me from inserting these examples into this chapter), orlan tries to confuse any notion of racial, national, and ethnic hybridity by playfully generating possible visions of beauty that remain outside of western consciousness. she brings in identity elements of beauty from other cultures like the long-ringed necks or flat small noses. these hybridizations are compossible worlds. in this way, orlan goes against what deleuze|guattari call the "faciality machine" as developed in tp, the way we are always produced and coded as signifying subjects into various categories so that a meaningful and stable world is maintained, like the benetton advertisement i spoke of earlier. beyond the face lies altogether a different inhumanity and different modes of organization that form strange new becomings. orlan explores this inhumanity. she becomes a "probe-head" in deleuze|guattari's terms (tp, - , o'sullivan ; mills ) . such a probe-head is cosmic as well; it belongs to the tao of the face, or in françois laurelle's terms to the 'generic human. ' artists, according to deleuze|guattari, try to work with blocs of percepts and affects emerging from the transcendental cosmic plane so as to impact and shake up our commonsense perceptions. they do so, so that the duality of subject and object can be overcome via creative becoming; the same principle as yin |yang where the yin as passive nonbeing is equated to the deleuzian understanding of the virtual and yang, as the active being is equated with the actual. these two forces are also analogous to the qualitative and quantitative multiplicities. let me explore this through a well-known korean contemporary artist, myoung ho lee, working in seoul who explores the tao of trees. she isolates trees in such a way that we see 'tree' again as a singularity. (internet image search: 'myoung ho lee tree' will generate an array of these images. copyright prevents me from inserting these examples in this chapter.) a singular 'tree,' by being extrapolated from its environment and then photographed, raises questions of scale and interconnectedness with nature. they take on a singularity in their personalities, a materialist vitalism of spirituality that deleuze argues is part of the plane of immanence as the membrane of the cosmos. each photographed tree is like a cosmic dogon egg as deleuze|guattari discuss in tp ( ). each tree is a bwo. it is both permanent and impermanent at once, under constant transformation, which itself is a permanent process. 'treeing' presents this very paradox. the tree's seed contains both its existence and non-existence, like the embryo discussed earlier; like the dogon egg. in the taoist sense, each tree has its own te, or virtue, its own intrinsic excellence. the notion of 'site' specificity of lee's trees disappears and it becomes (in the deleuzian sense) "any-site-whatsoever," raising questions about the empirical world as such. each thing has its own internal and immanent order. the idea of tree is not transcendentalist and not platonic; rather, the idea of tree becomes a singularity: its own order. yet, its seed bears its existence and non-existence at once; it is both nothing and everything bringing us back to the tao of tree. it is a series, again repetition, that preoccupies lee's photography (refer to internet image search). the series, another deleuzian concept, requires that each tree, like the duchampian urinal series that constitutes his 'strange' autobiography, introduces a difference. one cannot help but think of the japanese art of bonzai. as lao-tzu wrote in the tao te ching: "seeing your own smallness is insight" (verse ) (lao-tzu and takuan sōhō ) . bonzai-ing the natural environment through serialization amplifies it by repetition to make a new imaginary field. such an installation shifts the corporeal scale of the viewer to see the order of tree differently, perhaps rhizomatically. bonzai = the tree of life as it is tended from generation to the next generation in potentially an unending succession (or serialization). in much different way, mark dion's neukom vivarium is also a tree, a dying western hemlock tree. (internet image search: 'mark dion, neukom vivarium' will generate an array of images of this installation. copyright prevents me from inserting these examples in this chapter.) it is perhaps the western response to the care of a bonzai tree. begun in and completed in , housed in the seattle art museum, it has received wide attention and shows the irony of how to maintain a 'dying' western hemlock tree by putting it on artificial life support systems within the confines of an art gallery. once again, a smooth space is created within a gallery setting, and a pedagogical element is also a part of this installation to explore its newly developing ecosystems. the tree becomes das ding [the thing] through its isolation. the rhizomatic connection of the assemblage of creatures, microbes, and artificial greenhouse apparatus that keeps it alive once more vivifies the singularity and interconnectedness of things themselves, which lee's tree so vividly shows. we can also point to jean-claude didler's installation, trapped inside ( ) as yet another tree put on life support on the grounds of the united nations environment programme located in gigiri, nairobi. this time the warburgia ugandensis (african greenheart) tree was specifically selected because of its endangered status and its spirituality, holding a special place given its use in traditional medicine. there is a youtube video of myoung ho lee's tree exhibition where a gallerygoer made a one-minute video. you can watch it at: www. youtube.com/watch?v= lso_ukndy. i use it to illustrate my next deleuzian point. as you will see, what is startling is the scale of these images, but worse is that the spectator is rushing from photograph to photograph, merely recording these tree-images. it raises the question as to when art becomes an object of encounter. pedagogically and artistically, the encounter has become more and more of a concern in consumer culture; consumers have become participants in order to become producers of their own desire; in other words, good corporate marketing solicits what their consumers want, and then sells these wants back to them to maintain their brand and market share. desire in the capitalist order is shaped by lack as lacan constantly reminded his audiences. when i brought my volvo a few years back, i became part of the volvo family that sold me a particular lifestyle primarily based on the idea that my volvo was one of the safest cars around. should i be in an accident, i would most likely survive. my volvo is sold as a fetish object, a desirable commodity, which is precisely what capitalism does. the encounter i had to buy my volvo was shaped by my desire for safety, especially in canadian winters. this is not the encounter deleuze|guattari are interested in, and it is not the type of participation displayed on this video. many gallerygoers simply take images of artworks with their cell phones never to look at them again. an encounter with art depends very much on the participantspectator; something has to happen, something has to "look back at you." it's an encounter only when the object; or the world of things deterritorializes you so to speak. lee's trees have to "speak" to you as signs. such a moment can only be 'sensed.' your time has to be thrown "out of joint" as deleuze says: there has to be a desubjectification taking place; you as a subject have to be dissolved as you become in tune with that object in such a way that your common sense or frame of reference is somewhat shaken or even shocked. if the object is simply recognized, like in the above video, either through your imagination, or perception, then your experience becomes limited. you fall again into representation. my volvo was such an object. i had seen many televised commercials as to its safety record and its performance in winter storms. with objects of recognition, we are reconfirmed and reaffirmed into everyday life as we understand it. there is no thought in the deleuzian terms, only cognition supported by a feeling of security. an art object of encounter has to rupture this self-confirming mechanism. it performs a cut where a different kind of subjectivity might begin. this is again a taoist principle, as a deleuzian encounter means you try to dwell in its center as a form of emptying the 'self,' which is a becoming-itself of tao. deleuze calls this 'becoming-imperceptible.' in contrast, my volvo supported by ego; it fed the image of a future accident or being struck in the snow. i became my own prosumer. deleuze calls such an encounter an event. as one's world is ruptured another is opened up. such a rupture need not be huge. it can be small, but then develop into something huge. just like chaos theory that says a butterfly flapping its wings may cause a storm somewhere else on the earth, this is precisely what deleuze|guattari mean by their notion of event. only when an event-encounter is responded to can you say the participant-spectator has been affected, changed at the unconscious molecular level. the notion of affect is very important here, and much has been written about it-as the so-called 'affective turn' in the west (clough ) . deleuze|guattari are all about the virtual intensity of the affect, the way the nerves are affected most at the unconscious and intrinsic body levels. this is where thought without an image can take place. affect is something that happens to us below the level of language and the image. often, we can call affect as life force of zoë, taoist for chi, before it is captured into emotion, what has been called bios, life energy that has been categorized or actualized. the event is able to rupture what deleuze|guattari call striated space and open up a smooth space, a deterritorialization of space and time so that thought can take place. to generate affect requires an event. let me introduce you to a spectacular example via the photographic work of spencer tunick (internet image search: 'spencer tunick, aletsch glacier, switzerland' will generate this image. copyright prevents me from inserting this example in this chapter). spencer tunick's (along with greenpeace switzerland) installation of a 'living sculpture' (hundreds of naked men and women) was situated in the aletsch glacier, switzerland, on august , . this mass of bodies, all huddled together, symbolized the vulnerability of glaciers under climate change. the site of the installation created a smooth space in the public domain of the national park, a domain that is rapidly changing so that attention can be brought to the state of global warming. an element of duration is also evident. the disappearing speed of the glacier and the time it took to set up such a photo formed a basis of comparison. they illustrate the speed and intensity of the ecological processes that were being explored. we could say that this was akin to duchamp's infra thin interval. by photographing a mass of nude bodies in various environments, spencer tunick's photographs attend to a non-representational sense of equality; for example, volunteers stripped in front of sydney's famed opera house during australia's gay and lesbian alliance parade. gay men and women lay naked with straight neighbors to deliver this message. (turnick's photos of this event, 'sea, earth, change' can be found on the internet). deleuze|guattri's aesthetic is constantly about becoming; hence, the notion of commodity is always being questioned. it is what art can 'do' that matters; what's its force does is the educational question. what is its political and performative ethic? it is but one attempt to avoid the commodity structure of capitalism, and the forces of its capture. such art cannot be 'hung' on a wall, sold, traded, and so forth. it tries its hardest not to be a commodity by 'disappearing,' becoming anonymous and imperceptible once its "forcework" has been done (ziarek ) . relations affirmatively change here as wellthe relation with the inhuman glacier in tunick's case, and the relations between the huddling mass, cold, and shivering naked bodies who are affected. one can imagine the exchanges that took place among this 'living sculptural mass' in relation to themselves and the emergent 'will' that this project manifested. the stark exposure of nude bodies raises many questions regarding human vulnerability. it is an event that can then be counter-actualized in deleuze's terms, revisited for its transvaluational affects, to engage with its monumentalism as it addresses climate change, and our species vulnerability in the face of it. lastly, i want to end by describing another event where the commodity form disappears and an encounter can take place. army of melting men by brazilian artist néle azevedo is a repeated installation performed in brazil, france, japan, italy, and germany. (for images, please do an internet search: 'néle azevedo's army of melting men.' copyright prevents me from including images in this chapter.) each repetition is counter-actualized and introduces a difference, not unlike the expansive drawing installations of mathew ritchie mentioned earlier. azevedo's installation addresses global warming and presents the precariousness of existence under climate change. - cast mold ice figurines, generically male and female, approximately inches high, are placed on site usually on the steps of some well-known state building of legislative authority (but not necessarily) by a participating public. like the melting of the arctic ice in greenland and antarctica (sea levels will rise over a meter by ), these statuettes begin to 'disappear' as they melt-as quickly as minutes. they are vulnerable and frail just like turnick's sculptural mass of nude bodies. during this duration, the melting 'sculpturines' undergo subtle differences of form before 'becoming extinct.' their inactivity as they melt away (not unlike turnick's stilled nude bodies without gestured movement) speaks directly to the inactivity of humankind toward climate change. the sculptural minimalism and autonomy addresses 'every[man]' who cannot escape, regardless of class, wealth, and power, the impeding apocalypse. i would like to end this chapter by reiterating that there are many touchstones between taoism and deleuze|guattari's philosophy. creativity is always an invention to bring new life into the world for a "people to come" as deleuze says. art education must take seriously the ecological crisis we face today and both of these process philosophies can give us a meeting of east and west that is needed. some remarks concerning duchamp. the symptom is there a 'neo-racism deleuze and asia gender trouble: feminism and the subversion of identity the affective turn: political economy, biomedia and bodies. theory plato and the simulacrum (r. krauss, trans.). october logic of sense societies of control difference and repetition pure immanence: essays on a life (a. boyman anti-oedipus: capitalism and schizophrenia thousand plateaus: capitalism and schizophrenia what is philosophy? why art can't be taught: a handbook for art students laruelle: against the digital the book of lieh-tzu: a classic of tao zen keys tao te ching: zen teachings on the taoist classic (t. clear, trans.) dark enlightenment manifesto documentation art and korean bunche painting: an investigation of deleuze use of deleuze's process ontology self-forgetting and its history: a taoist perspective on postmodern phenomenology after finitude: an essay on the necessity of contingency (r. brassier, trans.) talking probe-heads: escaping faciality in contemporary aesthetics duchamp's fountain and deleuze's repetition and difference. drain magazine the philosophy of the daodejing malign velocities: accelerationism and capitalism pragmatics for the production of subjectivity: time for probe-heads a life between the finite and infinite: remarks on deleuze the missing subject of accelerationism a geology of media and a new materialism: in conversation with annika richterich dis)connected: deleuze's superject and the internet a new science of life: the hypothesis of formative causation new quantum structure of space-time thought beyond research: a deleuzian critique of artistic research the tao of bwo: deleuzian becomings in kung fu cinema deleuze and zen: an interological adventure the force of art usa) pbs: key: cord- -rm c vu authors: odusanya, kayode; adetutu, morakinyo title: exploring the determinants of internet usage in nigeria: a micro-spatial approach date: - - journal: responsible design, implementation and use of information and communication technology doi: . / - - - - _ sha: doc_id: cord_uid: rm c vu the dearth of information communication technology (ict) infrastructure in the sub-saharan africa region underscores the argument that the spread of broadband infrastructure can foster internet adoption in the region. consequently, the aim of this paper is to present results on the determinants of internet adoption in a sub-saharan african country. drawing on a dataset of households in nigeria, this study presents findings on the demographic, socio-economic and infrastructure factors that predict internet usage in nigeria. the novelty of our analysis stems from a unique dataset constructed by matching geo-referenced information from an inventory of network equipment to a nationally representative street-level survey of over , nigerians, by far one of the largest technology adoption surveys in sub-saharan africa to date within the information systems literature. the results are discussed and concluding remarks highlighting next steps are made. internet access is perhaps one of the most significant indicators of human and socioeconomic development. it fosters productivity and innovation (avgerou ; paunov and rollo ) , social interactions (liang and guo ) and reduces communication and search costs (beard et al. ). yet, there is a digital divide in the level of internet access among developing countries, especially those in sub-saharan africa (ssa) compared to the rest of the world. within ssa, the lack of internet access is a recognized barrier to the adoption of information communication technologies (icts) (afolayan et al. ) . this argument seems to be supported by regional broadband statistics, as shown in fig. where ssa is portrayed to have the lowest levels of internet penetration and wireless broadband infrastructure per capita, relative to other regions of the world. while previous studies have shed light on the regional variation in internet adoption by focusing on the determinants of internet penetration in the context of the "digital divide" (e.g. oyelaran-oyeyinka and lal ; chinn and fairlie ) , these studies often employ infrastructure measures/proxies (such as fixed telephone lines per sq. km, main telephone lines per capita, etc.,) thereby overlooking the peculiar nature of broadband infrastructure in the ssa region. more specifically, we note that internet use across the ssa region is undertaken mainly via wireless broadband access, rather than fixed-line broadband. thus, in this paper, we draw on a unique dataset that combines geo-referenced (longitude and latitude) information on g and g wireless network equipment to examine factors that predict of internet usage in nigeriaa sub-saharan country. consequently, this study offers two potential contributions to the literature. first, unlike previous studies, we demonstrate a measure of broadband infrastructure is micro-spatial in nature, i.e., it is based on the density or concentration of g and g equipment around each household's dwelling over specified spatial domains. to achieve this, we match street-level information from the technology survey to the global positioning system (gps) coordinates of g and g cell towers. this microspatial approach renders more nuanced and invaluable insights on whether/how spatial proximity or access to wireless internet connection shapes adoption decisions at the local level. afterall, without connectivity, it is practically impossible to adopt/use broadband services. furthermore, this spatial approach also embodies the reality that signal quality, a key determinant of actual usage behaviour, is shaped by the physical proximity of users to broadband connections (neto et al. ; destefano et al. ) . hence, in addition to treating the spatial diffusion of wireless network infrastructure as an appropriate measure of broadband infrastructure, it is also a quality-weighted indicator that enriches our analysis. second, our focus on nigeria provides a plausible and timely case study of the effect of broadband infrastructure diffusion on internet usage in ssa and the broader developing country context. given that it accounts for the largest proportion ( %) of the entire ssa region's . billion population (world development indicators, ), we would argue that nigeria is the most representative country of the ssa region. furthermore, itu estimates indicate that more than % of the youth population in developed countries use the internet compared to % in less developed regions of the world. considering that the proportion of nigeria's population below years is projected to reach % by , broadband penetration is likely to have a significant role in shaping nigeria's participation in an increasingly digitalized future economy. finally, despite growing to become africa's largest economy and one of the major emerging economies in the world, nigeria epitomizes the co-existing low levels of internet penetration and ict infrastructure deficit (the economist ). the remainder of this paper is organized as follows. in sect. , we undertake a critical review of related literature on internet adoption and ict infrastructure. this is followed by the model specification and the methods section. in sect. , we discuss the results and conclude in sect. highlighting next steps and expected contributions of our study to the information systems literature and policy. the literature on internet adoption and penetration is large. consequently, due to space constraints, the goal of this review is not to present an extensive discussion of existing literature. rather, we highlight a gap in the use of proxy measures in accounting for technology adoption. a more comprehensive review can be found in cardona et al. ( ) . a dominant strand in the literature pertains to cross-country studies aimed at evaluating the determinants of internet usage and penetration in the context of the digital divide in developing countries. one of such studies is dasgupta et al. ( ) who investigated the determinants of internet intensity (internet subscriptions per telephone mainline) for a cross country sample of oecd and developed countries. similarly, chinn and fairlie ( ) provide an analysis of internet penetration using a larger sample of based on panel data for countries over the - period. they control for a range of macro-level determinants on telecoms prices, per capita income, education, age structure, urbanization, regulatory environment, etc. although this study also controls for ict infrastructure, their reliance on telephone density may not be applicable to the african context. other internet diffusion/penetration studies use large cross-country samples from developing regions of the world (e.g. chinn and fairlie ) . a few studies however focus on internet penetration for more specific sub-groups/regions of the world such as oecd (lin and wu ) , the americas (galperin and ruzzier ), apec (liu and san ) , africa (oyelaran-oyeyinka and lal ) and asia (feng ) . greenstein and spiller ( ) investigate the impact of telecommunication infrastructure (measured by the amount of fiber-optic cables employed by local exchange telephone companies) on economic growth in the u.s. roller and waverman ( ) investigate the linkages between broadband investment and economic growth across oecd countries and developing countries during to . more recent studies have focused specifically on the impact of broadband infrastructure. for instance, czernich et al. ( ) investigated the effect of broadband infrastructure on the economic growth for a panel of oecd countries over the period [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] . however, they model this relationship by analyzing how broadband infrastructure (proxied by broadband penetration) shifts the growth parameter of technological progress within a macroeconomic production function setting. tranos ( ) explored the causal effect of broadband infrastructure (internet backbone capacity) on the economic development across european city regions over the period - with ict infrastructure found to stimulate economic development. three important observation can be gleaned from the above literature review. first, the cross-country studies tend to focus on the digital divide, attempting to explain internet adoption and penetration based on changes in economic, social, demographic and regulatory/institutional factors. even when these studies attempt to account for the role of telecoms infrastructure, they do so by using variables such as personal computers per people, main telephone lines per people, customers' equipment (e.g. telephone set, facsimile machine), etc. second, studies based on microdata tend to focus on developed country contexts, while also relying on infrastructure proxies that are analogous to the cross-country studies. moreover, the developing country studies are not free from this problem too. third, even studies that focus mainly on the effect of ict infrastructure on economic measures and internet adoption also employ similar ict infrastructure measures such as investments in telephone/broadband cables. in comparison to the cited studies, we take a different approach by employing a true measure of internet infrastructure that is based on the prevalence of wireless network equipment (i.e. towers and radios). we then explore the effect of wireless network access on internet adoption at the individual level, based on physical proximity to broadband connection. research has shown that variations in technology adoption are shaped by heterogeneity in geographical network access, such that the physical proximity to broadband connection or infrastructure can be expected to shape adoption decisions at the local level. for instance, it is well established that urban areas benefit from higher concentration of ict infrastructure, which may enable social learning and adoption of ict technologies (liu and san ) . hence, unless this type of analysis is undertaken, these disparities in network access may be inadvertently explained away as differences across individual income or demographic characteristics. this paper is also related to a small and evolving body of studies that exploit spatial data towards analyzing the diffusion of telecommunications technologies. our review shows that research conducted by buys et al. ( ) and hodler and raschky ( ) come close to the line of inquiry pursued in this study. however, they differ significantly from this study in two crucial ways. firstly, both studies conduct country-level spatial analysis while we adopt a micro-level spatial approach. secondly, while the former investigates the determinants of mobile operators' spatial location of network sites across sub-saharan africa, the latter explores the role of ethnic politics in shaping the spatial diffusion of mobile phone infrastructure in africa. as far as we know, the closest relative to this study is destefano et al. ( ) where the authors investigated how the arrival of adsl broadband technology influenced the it-productivity gap among uk firms. comparatively, this study is therefore the first to explore the individual-level influence of broadband infrastructure on internet adoption using a micro-spatial approach, especially in a developing country context. this study uses a probit model to examine the determinants of internet usage in nigeria. probit models are used when the dependent variable is dichotomous. we employ a set of explanatory variables in our model, namely: demographic factors include age, gender, marital status and religious affiliation; socio-economic factors such as expenditure, education, whether they are employed or not and a location variable indicating whether respondents live in rural or urban areas. finally, we also include a micro-spatial infrastructure representing the number of internet infrastructure within a radius of where they live (i.e. towers) and the average tariff per megabyte of data used (tariff). both towers and tariffs represent our infrastructure variables included in our model. thus, our model is represented by the equation: internet use = f (age, gender, income and education, marital status, religious affiliation, towers, tariffs, urban). we expect monthly expenditures (our proxy for income) to affect broadband adoption/usage positively. the probability of this adoption decision is also likely to rise for more educated individuals. however, the need for possessing an internet subscription, however, may fall for older and unemployed respondents. in terms of age, one could argue that, whereas younger individuals may have lower income, they tend to demonstrate a greater degree of technological affinity (hübler and hartje ) . we add a gender variable as an additional characteristic, given that the preferences and decisions of men are more dominant than the preferences of their spouse(s) in patriarchal societies (bulte et al. ) . similar considerations can be extended to the marital status of an individual on the adoption of internet technologies. finally, technology adoption decisions are often shaped by religious reasons, as some religious beliefs may restrict the adoption of conventional technological products (fungáčová, et al. ). hence, we control for the religious beliefs of each sampled individual. to examine the effect of broadband infrastructure on internet adoption, used a measure that captures individual-level access to wireless broadband network. hence, we employ a microspatial variable that captures the prevalence of wireless network infrastructure at the individual level. we calculate this variable using information from our two data sources in four steps. first, inspired by hodler and raschky ( ), we extract and map the g and g cell tower locations from opencellid (see fig. b in the appendix section). secondly, we extract and map the street-level dwelling locations from the survey data. thirdly, to use both data for our purpose, we link them by overlaying the towers map with the dwelling location map. finally, we compute a micro-spatial infrastructure variable as the total number of cell towers within -km radius of everyone. this -km specification is premised on the fact that network coverage in ssa is mainly based on base stations that can provide service up to a - km radius (aker and mbiti , p. ) . we achieve this using the stata 'spmap' command. to investigate the effect of wireless connectivity on internet usage, we rely on a unique nationally representative market survey of nigeria carried out by africa's largest mobile operator, mtn during april-july . the mtn survey, which covers localities (i.e., villages or towns), is by far and away one of the largest and most comprehensive technology surveys in sub-saharan africa to date. the data was collected via the use of paper questionnaire distributed to respondents in all states of the country. figure a in the appendix section plots the centroid gps coordinates of surveyed areas at the municipality level. the wide geographical spread of the survey areas confirms the nationally representative nature of the survey. a total number of , observations were obtained for the final analysis. our second data resource is the opencellid database which contains information on the micro-spatial independent variable (towers). the database contains raw information on the geo-location (longitude and latitude) of around seven million unique cell sites across the world (hodler and raschky ) . one key benefit of the opencellid database is the possibility to identify the technology (radio) type for each telecommunications tower (i.e. gsm, umts, lte, etc.). this allowed us to identify the two wireless network classes: "umts" (third-generation technology, g) and the more advanced "lte" (fourthgeneration technology, g) types. specifically, we identified a total of unique tower locations from the opencellid data consisting of g and g sites. these sites are geo-coded at the gps (longitude and latitude) -level (see fig. b in the appendix). our main dependent variable is represented by a broadband usage variable for the use of broadband services. to construct this indicator variable, we convert survey responses on broadband subscription using the question: "which of the following telecommunication services do you use nowadays?". we then calculate the dependent indicator variable as a dummy that takes the value " " if "data service (accessing internet)" was selected in response to the question. otherwise, a dummy value of " " was assigned to the observation. table provides the summary statistics of the variables employed in this study. the internet adoption rate within our dataset is %, falling firmly within the same ballpark as the . % and . % broadband penetration rates reported by business monitor international (bmi ) and the itu, respectively. the summary statistics in table also indicate that there are on average wireless network towers within km of each respondent's street. however, the standard deviation of towers within the km radius suggests a reasonable spread or dispersion of the cell tower variable. table presents the marginal effects from baseline probit estimations. in column , we start by measuring the network infrastructure effects on broadband adoption: probability of adopting internet services, without any control variables or locality effects. due to the cross-sectional nature of our data, we interpret these results as associations. it is clear from the results in columns ( ) that a strong positive correlation exists between the concentration of broadband infrastructure around each individual and internet adoption. this coefficient is significant at the % level. from column to , we add the control variables one by one, but the infrastructure coefficient remains statistically significant at the % level, albeit the magnitude of the coefficient drops. in column , we include locality effects to draw inference only from the variation in individual adoption decisions. besides employing controls and locality effects, we use heteroskedasticity-robust standard errors clustered at the locality level that allow the data to be independent across localities by restricting the error terms to be correlated for individuals within the same areas on account of omitted regional characteristics. ÃÃÃ, ÃÃ, and à denote significance at the %, %, and % level, respectively. as seen in column , the infrastructure coefficient retains its statistical significance but drops further. specifically, infrastructure coefficient of . suggests that a unit increase in the number of cell towers within km of a respondent increases the probability of internet adoption by . %, which corresponds to an increase in the likelihood of adoption from around % to around . % in an average respondent. also, in the full specification in column , the coefficients on the control variables are consistent with intuition and they are all statistically significant at the %-level. for instance, older and unemployed respondents are less likely to adopt broadband services, whereas higher income earners, more educated individuals and urban dwellers are more likely to adopt the internet. for religion, being a christian increases the probability of adoption. the positive coefficient on the male gender variable is consistent with the patriarchal nature of the nigerian society, which indicates that the men are more likely to adopt internet services, perhaps reflecting the stronger socioeconomic power of the male gender. interestingly, in terms of the magnitude of the coefficients, we find age and gender to have the greatest effect on broadband adoption, with both coefficients indicating % and % positive impact on the probability of adoption, respectively. the age coefficient lends weight to our opening arguments on the implications of broadband adoption for the participation of the large projected youth population of nigeria in an increasingly digitalized global economy in the future. for all the specifications in table , the results also show that individuals in areas with a higher concentration of network infrastructure are more likely to adopt and use broadband services. this paper presents first-stage results showing factors that influence internet adoption in nigeria. using geo-referenced information of an inventory of broadband network infrastructure, we employ a more appropriate micro-spatial measure of internet infrastructure based on g/ g network equipment at the local level alongside a range of explanatory variables to explain internet adoption in nigeria. in general, the results obtained are economically important, and they can help explain the adoption patterns of broadband services particularly when network infrastructure effects are likely to play an important role in driving internet penetration. for instance, in many regions across developing countries, network coverage is usually the first modern technology of any kind (aker and mbiti ) . hence, we would argue that the failure to control for this network infrastructure effect in the study of broadband adoption across developing countries could well result in significant omitted variable bias. furthermore, the network infrastructure effect may also explain the nuances embodied in the varied adoption of broadband services across different regions, given that the quality of service (qos) and user experience may well depend on the diffusion and reliability of the underlying network infrastructure. although the first-stage results presented in this paper revealed preliminary drivers of internet adoption, they also provide interesting avenues for further study which we aim to explore in subsequent analysis of the dataset. in the first place, the significant urban variable shows that there are likely to be regional differences with regards to the factors that influence internet adoption in nigeria. given that nigeria has one of the largest rural population in africa (world bank ), we expect that further analysis testing the relationships in this study across rural and urban dwellers, will contribute significantly to the policy debate on bridging the urban-rural internet divide in nigeria. further, having microeconomic information that is representative of the national population of the type used in this research, permits clear visualization of the digital divide as an additional form of inequality that can hinder access to other internet-reliant technologies. for instance, the telecommunications sector in nigeria is currently undergoing several policy changes, one of which is the granting of mobile money licenses to mobile network operators. therefore, it is earmarked that further analysis with our data will explore diffusion constraints for other technologies covered in the survey data and how usage patterns might vary in different parts of the country. we hope that the results of these avenues of research will provide new insights that contribute to both policy and research. − . *** − . *** − . *** − . *** − . *** − . *** − . *** − . *** *** − . *** − . *** − . *** − . *** − . *** information technology usage in smes in a developing economy mobile phones and economic development in africa information systems in developing countries: a critical research review internet use and job search business monitor international (bmi): nigeria telecommunications report gender training and female empowerment: experimental evidence from vietnam determinants of a digital divide in sub-saharan africa: a spatial econometric analysis of cell phone coverage ict and productivity: conclusions from the empirical literature the determinants of the global digital divide: a cross-country analysis of computer and internet penetration ict use in the developing world: an analysis of differences in computer and internet penetration broadband infrastructure and economic growth policy reform, economic growth and the digital divide broadband infrastructure, ict use and firm performance: evidence for uk firms determinants of internet diffusion: a focus on china trust in banks price elasticity of demand for broadband: evidence from latin america and the caribbean modern telecommunications infrastructure and economic activity: an empirical investigation ethnic politics and the diffusion of mobile technology in africa are smartphones smart for economic development? social interaction, internet access and stock market participation-an empirical study in china identifying the determinants of broadband adoption by diffusion stage in oecd countries social learning and digital divides: a case study of internet technology diffusion fostering pro-competitive regional connectivity in sub-saharan africa. global ict department internet diffusion in sub-saharan africa: a cross-country analysis has the internet fostered inclusive innovation in the developing world? world dev telecommunications infrastructure and economic development: a simultaneous approach the causal effect of the internet infrastructure on the economic development of european city regions rural population -nigeria acknowledgements. the authors also gratefully acknowledge the support of africa's leading mobile operator mtn, for providing the market survey and operator data employed in this study. special thanks to the staff of the business intelligence and research departments. we also like to thank participants at various seminars and workshops for their helpful comments. the usual disclaimer applies. opencellid data) key: cord- -rkuzcng authors: chang, i-chiu; li, yi-chang; hung, won-fu; hwang, hisn-ginn title: an empirical study on the impact of quality antecedents on tax payers' acceptance of internet tax-filing systems date: - - journal: gov inf q doi: . /j.giq. . . sha: doc_id: cord_uid: rkuzcng this study used the technology acceptance model (tam) to examine taxpayers' acceptance of the internet tax-filing system. based on data collected from experienced taxpayers in taiwan, the acceptance and the impact of quality antecedents on taxpayers' perceived usefulness (pu) and perceived ease of use (peou) of the system were assessed and evaluated. the results indicated that the model of internet tax-filing system was accepted with a reasonable goodness-of-fit. three important findings include the following items. first, tam proves to be a valid model to explain the taxpayers' acceptance of the internet tax-filers' system. meanwhile, pu has created more impact than peou on taxpayers' intention to use the system. second, pu is positively influenced by such factors as information system quality (isq), information quality (iq), as well as perceived credibility (pc). third, iq has a positive impact on peou. based on the research findings, implications and limitations are then discussed for future possible research. in taiwan, there are three major tax-filing methods: manual filing, two-dimension ( d) bar code filing, and internet filing. since , the taiwan government has moved aggressively to promote internet tax filings under the e-government initiative, the goals of which are a paperless environment, an efficient process, and the public's convenience in contacting government agencies. the tax authority in taiwan, the national tax administration (nta), promoted the system by providing incentives, such as faster tax refunds and online credit card payment, which results in an increasing number of internet tax filers (see table ). with regard to the aforementioned three methods, public satisfaction is greatest with the d bar code filing, and the manual filing method is the least satisfactory. despite all the efforts of nta's promotion and the low satisfaction, traditional manual filing remains the most widespread method. though d bar code is classified as an electronic tax-filing method, it still requires taxpayers to print their return in paper form and mail it to the government agent. therefore, this study focuses on the internet tax-filing method, the only method that would eventually achieve one of the e-government's goals to be paperless, and examines the critical factors that influence the acceptance of the internet tax-filing system. the technology acceptance model (tam) is among the most influential and discussed theories in explaining and predicting the individual's acceptance of information technology. several past studies have examined the relationship of perceived ease of use (peou), perceived usefulness (pu), attitudes, intention, and the usage of information technologies. lee et al. summarized the information systems examined by tam in articles published by leading is journals and conferences from to into four different classes: communication system ( %), general purpose system ( %), office system ( %), and specialized business system ( %). among this research, little was done in the e-government area. however, applying tam to the e-government system can lead to a better understanding not only of public acceptance of the internet tax-filing system but also of the possible differences between the e-government and other types of internet applications. since the internet tax-filing system is still new to most of the public, the earlier adopters may share the same personal characteristics. as a result, this paper focuses only on the quality factors that influence the acceptance of internet tax-filing software. information system quality (isq) and information quality (iq) were often used in the evaluation of system usability. - many researchers have validated the relationships between these two quality source: national tax authority, taiwan. factors and system use. [ ] [ ] [ ] [ ] [ ] [ ] [ ] additionally, perceived credibility (pc) was indicated as an influential factor affecting the electronic tax-filing system. altogether, the authors proposed a tam-like research framework with the isq, iq, and pc of the internet tax-filing system that affect the taxpayers' pu and peou, and additionally the attitude and the intention to utilize the system. the internet tax-filing system provides online tax service to taxpayers and is one kind of government to customer (g c) electronic service. furthermore, g c e-service is one part of the e-government domain. consequently, the definitions and characteristics of e-government and the internet tax filing are reviewed in this section. in addition, the related research about tam and delone and mclean's taxonomy, which provides the basis for this research framework, is also reviewed in this section. e-government can be defined as the use of information technology in general, and the utilization of the internet in particular, to provide citizens and organizations with more convenient access to government's information and services. on the other hand, government can deliver public services to citizens and organizations electronically and even facilitate them to these aforementioned services through it. , from a technical perspective, e-government can be seen as a new technology used by government to help simplify and automate transactions among governments and constituents, businesses, or other governments. from an economics perspective, e-government defines a new market and a new type of government-a powerful channel to distribute public services interactively. as to the context of the e-government, an integral part of the binformation societyq and bdigital economyq visions were formed. under the e-government concept, governments usually find themselves confronted with a broad range of political themes arising from the need to re-establish their vision and role and re-structure their services around the concept of bcitizen as customer.q just like business to customer (b c) in e-commerce, a well-established g c e-government can provide citizens with all the information they need on the web. citizens can ask questions and receive corresponding answers, pay taxes and bills, receive payments, view documents, and access other services that are available hours a day through the internet. in the information systems domain, a variety of users' behavior effects have been theoretically discussed and empirically demonstrated in the previous research. however, few studies about g c e-services are done. because of the challenges from the inner and outer environmental changes, the government needs to keep up with the latest information technology and provide citizens with more convenient access to government information and services. west analyzed government web sites from different nations and presented an updated global e-government report in three consecutive years. the criterions used in the study to compare these aforementioned sites include factors such as publication, database, audio or video clip, restriction, foreign language, privacy, security, credit card payment, number of online services, etc. twenty-one percent of the visited web sites actually provide online services. west's definition of online services, however, included only those services that were fully executable online. in other words, if a citizen has to print out a form and mail/take it to a government agency to execute the service, it is not counted as an online service. in a updated report, singapore was ranked at the top of countries. the rest of the top five nations that scored well in terms of e-government service included the united states, canada, australia, and taiwan. this report confirmed the effort made by the taiwanese government in rendering its electronic services. however, the high services score does not guarantee popularity. the usage of online internet tax filing in taiwan is still not satisfactory at this stage. as discussed earlier, there are three major tax-filing methods in taiwan: manual filing, d bar code filing, and the internet tax filing. the later two are usually considered electronic tax filing. for manual filing, taxpayers fill out a standard printed form, usually by hand. complex calculations are performed using either pencil and paper or calculator. the tax agencies use either manual data entry or image processing to input taxpayers' data into their computers. this traditional filing method consumes large amounts of time and effort of both taxpayers and the tax agencies. the d bar code filing was used around the same time as the internet filing. these two electronic filing methods provide the taxpayer an opportunity to use software to run tax return preparation with a cd-rom or through an internet download. with tax-filing software, the data entry errors and calculation errors can be automatically detected, and the best tax return option is provided for the taxpayer. using d bar code filing, the taxpayers need to print out two to three pages of a paper form with a d bar code, which encodes all data (nearly bytes) within the form. the tax agencies receive the paper form by mail and then scan the d bar code to input taxpayers' data into their computer system for further operation. while in internet filing, the taxpayers file directly online, and data are transferred into the agencies' computer system automatically. hwang adapted the haines and petit instrument to conduct an experiment and survey to evaluate the above three tax-filing methods. the results show that the user satisfaction is greater in d filing than in internet and manual filing. the survey indicates that the major users of electronic filing are males who have a college education, who have more than three years of computer experience, and one to three years of internet experience. wang added a new construct, bperceived credibility,q and critical individual difference variables to extend the tam study in the context of electronic tax-filing systems. based on a sample of users, obtained through telephone interviews, wang's study results show strong support of the extended tam. the tam (see fig. ) adapts the theory of reasoned act (tra) model to explore the it acceptance. tam and tra state that an actual behavior is determined by the intention to perform that behavior. intention itself is determined by the individual's attitude toward the behavior. tam indicates both pu and peou as key independent variables that determine or influence potential users' attitudes (att) toward it intention of use (bi). davis called for future research to consider the role of additional external variables that will influence pu and peou. despite the great success of this widely applied theoretical model, mixed results of these external variables were found in the is field. lee et al. summarized from articles published by leading is journals and conferences from to and classified the external variables of tam into the following four categories. they include ( ) rogers' five innovation characteristics, which include relative advantage, compatibility, complexity, observability, and trialability; ( ) individual differences such as self-efficacy, personal innovativeness, subjective norms, computer anxiety, computer attitude, and prior experience; ( ) system quality, including output or information quality, end user support, and result demonstrability; and ( ) others such as voluntariness, perceived enjoyment, and management support, etc. this research explained that the antecedents of pu and peou are important in the tam research area. delone and mclean (d&m) introduced the information systems success model based on a review of more than published papers. this model organized a rich, but somewhat confusing, body of research and provided a comprehensive view of is success. since , nearly articles in leading journals have referred to or adopted the d&m is success model. among these success variables, the conceptualization of the buseq construct was especially confusing in the original d&m model. the is use was defined as the consumption of is output which is consistent with a process model of is success. seddon identifies three distinct models intermingled in the d&m model, each reflecting a different interpretation of is use. from a process model of is success, buseq depicts the sequence of events relating to an is. a second meaning is a variable that proxies for the benefits from buse.q a third meaning is the dependent variable in a variance model of future is use in which the role of is use is used to describe behavior. to avoid the confusion associated with d&m, this study used d&m-like quality antecedents to summarize the external variables of the tam acceptance model for the internet tax-filing system. during the last several years, one of the most important issues in business is bquality.q regardless of the industry (manufacturing, healthcare, education, or government), professionals of all kinds are wrestling with the issue of how to improve their quality in order to gain competitive advantage. however, bqualityq means different things to different people under different contexts. therefore, researchers should develop an appropriate scale contextually when referring to the quality. taylor defined quality as ba user criterion that has to do with excellence or in some cases truthfulness in labeling.q from the customer view, quality was considered to meet a customer's expectations of the product or service being delivered. , d&m embraced the most frequently cited quality factors of the information system in the is academic research into their updated is success model: the isq, iq, and service quality. isq is associated with the issue of whether the technical components (including hardware, software, help screens, and user manuals) of delivered is provide the quality of information and service required by stakeholders. isq is often measured by multi-items such as useful functionality, accessibility, flexibility, integration among sub-systems, response time, reliability, accuracy of data processing, ease-of-use, and ease-of-learning. , , , the iq represents the users' perception of the output quality generated by an information system and includes such issues as the relevance, timeliness, and accuracy. , service quality was introduced in the is originally to measure the quality of services provided by it departments in organizations. researchers have applied and tested the servqual measurement instrument from marketing , into the d&m original model. some debates challenged the servqual metric, identifying bproblems with the reliability, discriminant validity, convergent validity, and predictive validity of the measure.q service quality was often measured by tangibles, reliability, responsiveness, assurance, and empathy. d&m suggested that each of the three quality dimensions (iq, isq, and service quality) has different weights depending upon the level of analysis. they also suggested, bto measure the success of a single system, information quality or system quality may be the most important quality component. for measuring the overall success of the is department, as opposed to individual systems, service quality may become a more important variable.q credibility can be simply defined as believability. credibility was seen as a perceived quality of a computer system. , in the internet era, security and privacy issues have been the primary concerns to internet users. a credible web site needs to safeguard personal information from unauthorized access or disclosure, accidental loss, and alteration or destruction. studies indicated that the users' perceptions of credibility regarding security and privacy were found to influence the user's intention to use webbased it application. , , wang further explained that bperceived fears of divulging personal information and users' feelings of insecurity provide unique challenges to planners to find ways in which to develop users' perceived credibility of electronic taxfiling systems.q therefore, perceived credibility is considered as another important quality factor in the internet tax-filing system. lucas concluded that unused systems are failures. therefore, it is reasonably assumed that heavily used systems are successful. different constructs influence the usage of various systems differently. lewis predicted system usability in three dimensions: system usefulness, interface quality, and information quality. d&m used iq, isq, and service quality, while tam used pu, peou, attitude, and intention to forecast the usage. however, only a few empirical studies , , were done to find causality among these constructs. fig. depicts the research model of this study. the development of our research model is based on two points of view. the first point of view, from tam, is that an it usage is determined by the beliefs that a user holds about the pu and peou of the system. meanwhile, external variables of tam affect usage behavior only through their impact on beliefs of pu and peou. , the second viewpoint is the quality antecedents of usage from d&m and wang. the authors then extracted part of the davis model that has been validated by many studies: pu and peou impact on att, and the att impacts on bi. the external variables that impact pu and peou generated significant and insignificant mixed results. as a result, the external variables are summarized using d&m-like quality delone and mclean included the service quality aspect of e-commerce; however, the internet tax-filing system is not a commercial product. since its first launching in until now, most users of the internet tax-filing system are assumed to be early adopters who have greater capability to deal with abstractions and rationality, and are assumed to have a more favorable attitude toward changes than later adopters. to this end, they may emphasize service quality less. meanwhile, service quality is also involved in some controversial issues. due to the aforementioned issues, service quality was not included in this study. in the internet era, research that focuses on trust dimensions has received a lot of attention. trust is an expectation that who one chooses to trust will not behave opportunistically by taking advantage of the situation. however, g c e-service is different from b c e-commerce. government will behave in a dependable, ethical, and socially appropriate manner. therefore, instead of the trust construct, wang's perceived credibility regarding security and privacy that influenced the user's intention of using electronic tax filing is included in this study. the quality antecedents in our model mimic that of the d&m structure, but separate variables related to pu, peou, and pc to fit in the context of the internet tax-filing system. consistent with these aforementioned two points, the authors propose that the constructs hypothesized (e.g., iq, isq, and pc) affect the use of internet tax filing indirectly through their effect on pu and peou. the definition of iq, isq, and pc in this study will be explained in the next section. considering that the filing of individual income taxes is done annually, the actual behavior of continued usage is excluded in this study. the dependent variable is the construct of bbehavior intention,q which means respondents' intention to use internet filing in the next year. altogether, ten hypotheses are formulated below: h . user attitude on using the internet tax-filing system positively affects the users' intention to file their tax on internet next year. h . the pu of the internet tax-filing system positively affects the users' intention to file their tax on internet next year. h . the pu of the internet tax-filing system positively affects the users' attitude on using the internet tax-filing system. h . the peou of the internet tax-filing system positively affects the users' attitude on using the internet tax-filing system. h . the peou of the internet tax-filing system positively affects the pu of using the internet tax-filing system. h . the isq of the internet tax-filing system positively affects the pu of using the internet tax-filing system. h . the iq of the internet tax-filing system positively affects the pu of using the internet taxfiling system. h . the pc of the internet tax-filing system positively affects the pu of using the internet tax-filing system. h . the isq of the internet tax-filing system positively affects the peou of using the internet tax-filing system. h . the iq of the internet tax-filing system positively affects the peou of using the internet tax-filing system. a taxpayer will prepare and file his/her tax on the internet only if the filing method assists him/her in completing tax filing effectively. therefore, this study assumes that pu of internet tax filing will positively affect the attitude and intention of using the system. in addition, the authors assume that the easier it is to learn and use internet tax filing, the more willing taxpayers will be to file their tax on the internet. as suggested by davis et al. and venkatesh and davis, peou is a direct determinant of pu. all else being equal, the less effort spent on learning the internet tax-filing system, the more it is perceived to be useful. higher isq, such as faster response, error reduction, tax calculation, what-if analysis, easy navigation of the internet tax-filing system, and ease of locating needed information, is assumed to enhance taxpayers' performance and therefore reinforces the perceived benefits of the system. before computer-based tax-filing software, most taxpayers paid, and some still pay, professionals to prepare their taxes each year in order to get higher output or information quality (e.g., accurate and complete tax information, suitable format of prepared tax form, understandable presentation of tax-filing steps). therefore, higher iq is assumed to enhance taxpayers' performance on tax-filing output and therefore reinforce their perceived usefulness of the system. perceived credibility may reduce risk perceptions (e.g., personal information disclosure) in users' minds and consequently enhance taxpayers' confidence in internet filing. both is and isq factors may reduce the user's effort spent in it usage and are assumed to positively impact user's peou of the system. multiple items were used for measuring the research variables using a seven-point likert scale. the selected items in the instrument for each construct were mainly adapted from prior studies to ensure the content validity. a draft questionnaire was translated into chinese by a bilingual research associate. the translation accuracy was then refined and verified by two mis professors. for content and preface validity, a focus group comprised of five experienced internet tax-filing users was held to review the questionnaire and derive appropriate quality attributes that the internet tax-filing system should possess. the modified version was then examined by two senior is managers in e-government for wording improvement. table summarized the operational definition of constructs and sources of the questionnaire items. the final instrument is included in the appendix section. behavior intention is defined as the user's likelihood to use the internet tax-filing system in the next tax season and was measured by three items adapted from taylor and todd. attitude is defined in terms of individual preferences and interests via feelings/evaluations regarding use of the internet tax-filing system, and it was measured by three items adapted from davis. pu was defined as the degree to which a person believes that using the internet tax-filing system would enhance/improve his/her job performance. according to the taxpayer focus group, the main benefits of internet filing are greater error reduction in preparing tax return, better time saving on tax calculation and form preparation, lower communication costs, and faster refund than conventional filing tax method. consequently, these items were used to measure the pu of internet tax filing. peou was defined as the degree to which a user expects the use of the internet tax-filing system to be free of effort and was measured by four items adapted from davis' study. isq is judged globally by the degree to which the technical components (including software, help screens, and user manuals) of internet tax filing provide the quality information and service required by users. isq is often measured by attributes such as useful functionality, accessibility, flexibility, integration among sub-systems, response time, reliability, accuracy of data processing, ease-of-use, and ease-of-learning. , , , after screening out the overlap of pu and peou, and modifying according to input from the focus group, isq, in this study, was measured by four attributes: usefulness of functions, process reliability, response time, and ease of navigation. iq is judged globally by the degree to which users are provided with quality information with regard to their needs. iq is often measured by attributes such as accuracy, precision, understandability, readability, clarity, suitable format, completeness, relevancy, timeliness, and freedom from bias. , , after screening out the overlap of pu and peou, and modifying according to input from the focus group, iq, in this study, was measured by four attributes: information reliability, relevance, adequacy, and understandability. finally, perceived credibility was defined as the extent of users' confidence in the internet taxfiling system's ability to protect the user's personal information and security. these aforementioned two items were adapted from wang. according to hwang's survey results, the dominant users of electronic tax filing in taiwan were men with a college education, and with years of computer-related experiences. to this end, appropriate sampling was employed in this research. approximately questionnaires were distributed to faculty and part-time undergraduate students of four universities who had previously filed their income tax using internet filing. among subjects, respondents completed the questionnaires, a . % response rate. detailed demographic data are provided in table . the composite reliability measure gives a truer indication of reliability than the traditional measure of alpha coefficient as it takes into account the possibility that the indicators may have different factor loadings and error variances. , the value of composite reliabilities higher than the threshold level of . was deemed to provide satisfactory reliability. in this research, the composite reliability for each construct in the measurement model was above . (see table ). further, another measure of reliability is the average variance extracted. this measure reflects the overall amount of variance in the indicators accounted for by the latent construct. higher variance extracted value occurs when the indicators are truly representative of the latent construct. the average variance extracted measure is a complementary measure to the construct reliability value. guidelines suggest that the average extracted variance value should exceed a . level for a construct, which means more than one-half of the variances observed. in table , average variances extracted for all constructs were above . . as a result, it is concluded that all the constructs used in this study were highly reliable. the constructs employed were further examined using convergent and discriminant validity analysis to validate the prediction of taxpayers' acceptance of the internet tax-filing system. convergent validity is assessed by testing whether the factor loading that relate each indicator to the construct of interest are all significant, , i.e., t values greater than . ( p b . ). in this study, all t values are between . and . , which indicates good convergent validity. discriminant validity is the degree to which measures of different concepts are distinct. to test discriminant validity, fornell and larcker suggested that the squared correlations between two different measures in any two constructs should be statistically lower than the variance shared by the measures of a construct. all shared variances between any two different constructs were, in fact, less than the amount of variance extracted by one of the two constructs (see table ). therefore, the constructs of the proposed research model exhibit adequate discriminant validity. table contains the lisrelcalculated correlations among the constructs. the lisrel analysis of the path model developed in this study shows reasonable fit for the structural model (see fig. , v /df = . , gfi = . , agfi = . , rmsr = . , nfi = . , nnfi = . ). , the explanatory power of the model for individual construct was examined using the resulting r for pu, peou, att, and bi are %, %, %, respectively. two hypotheses, h (beta = . , t = . ) and h (beta = . , t = . ), were not supported by data. in other words, the influence of isq on peou was not significant, and neither was the influence of pu on bi. however, other hypotheses, h (beta = . , t = . ), h (beta = . , t = . ), h (beta = . , t = . ), h (beta = . , t = . ), h (beta = . , t = . ), h (beta = . , t = . ), h (beta = . , t = . ), and h (beta = . , t = . ), were significantly supported by the data. the results of hypotheses testing are summarized in table . the multicollinearity among independent variables may reduce any single independent variable's predictive power. as a rule of thumb, a tolerance value less than . indicates a multicollinearity problem. in this study, multicollinearity was not a serious concern in our proposed model, since all relevant checks returned a tolerance value above . for all independent variables (see appendix a). this study confirmed most of tam's conclusions in previous research of applying other it (h , h , h , h ) . in other words, the results from this proposed study imply that it is suitable to apply tam in the g c area. in this study, the authors found that the pu has no direct impact on bi but has significant influences on att, which consequently impacts on bi of using the system. that is, the effect of pu on bi was mediated through att. this finding is consistent with the study from lederer et al. on www acceptance. dishaw and strong; hu, et al.; karahanna, et al.; taylor and todd h puybi not supported davis; davis et al.; hu et al.; lederer, et al. (ns); taylor and todd; venkatesh and davis not supported (ns): p value is not significant. both pu and peou have significantly positive impacts on att of using the system. the effect of pu (b = . ) is twice the significance of peou (b = . ) on att. this finding was consistent with prior research results , , , , but was contrary to electronic tax-filing system acceptance. in wang's research, peou is more important than pu on taxpayers' acceptance of electronic tax filing. the possible explanations include the following: first, the effect of peou on it usage often decreases with user familiarity with the it. only experienced internet tax-filing taxpayers were selected in this study, and most of them were already familiar with the operation of the internet taxfiling method. therefore, the effect of peou on internet filing acceptance is not as important as pu. second, the respondents' education level in this study was, on average, higher than that in wang's study. in general, higher education assures more computer and internet experience. therefore, peou may not be an important concern for these respondents. the iq was found to have a positive impact on pu and peou in the proposed study, which is consistent with previous studies. , the isq was found to have an association with pu instead of peou. this result is contrary to the findings of lucas and spitler and lin and lu. lucas and spitler found that users' peou declined significantly as isq of workstations did. lin and lu found that the response time of a web site greatly affects the user's belief of pu and peou, and affects on peou more than on pu. the possible explanation may be that experienced computer/internet users have less patience for low isq. a slow response system is considered to be useless and not worth the wait compared with a difficult-to-use system. lastly, the empirical result shows that pc has created a certain impact on pu. this finding is also consistent with shih and wang. the taxpayers were concerned with security and privacy issues during uploading tax return data and payment phases. this empirical study has five limitations. first, because of personal confidentiality concerns, general taxpayers were unwilling to respond to the questionnaire. therefore, a convenient sampling method, according to hwang's study, rather than a random sampling method was employed in this study. because highly educated taxpayers who were experienced with the internet tax-filing system were selected as the survey sample, the external validity of the research results may be limited. second, the overall goodness-of-fit of the proposed structural model (gfi = . ) is slightly lower than the commonly cited threshold: gfi n . . though most published papers in leading mis journals seldom show excellent fit values in all the indices, a gfi value of . or larger is acceptable for such an exploratory study. therefore, the proposed model may present a reasonable goodness-of-fit as an exploratory study. although is service quality was the subject of controversial debate, - , - d&m included service quality in the e-commerce success model. however, this study did not include service quality because of the sample subjects and because the attention was focused on iq, isq, and pc of the internet tax-filing system of g c. to promote internet filing with all taxpayers, service quality may play an important role. therefore, future research should extend to include service quality as a quality antecedent of tam. finally, instead of pc positively impacting peou, ong, et al. and wang found that peou positively influences users' pc in interaction with is. the authors also tested the revised lisrel model with this linkage. however, gfi of the revised model decreased dramatically from . to . , which indicated that the presence of the relationship between peou and pc in the proposed model could not significantly explain the obtained data results. the inadequate results may be caused by the difference in the sampling characteristic or the different information system applied. future research could also be extended to examine the linkage between peou and pc in the e-government service context. the implications of this study are twofold: the first is to retain the current users; the second is to create new users. since the current experienced users weigh pu heavily, the nta should propose some promotional plans that capitalize on the benefits of using the internet tax-filing system in order to retain/keep the current users. the pu has a three times stronger impact than peou (pu: . , peou: . ) on attitude toward the system, which then influences the intention to use the system next year and afterwards. to increase the pu of using the internet tax-filing system, the tax agency needs to provide a better isq, iq, and pc. for example, faster response time of the internet tax-filing system will lead users to perceive higher usefulness of the system. under budget constraints, the priority of the improvement should start from isq, then iq, and finally pc, respectively, since each has an impact on pu as isq: . ; iq: . ; and pc: . . besides, the empirical results also show that the quality factors (e.g., iq, isq,) of the internet tax-filing system have positive impacts on taxpayers' peou of using the system. to create new users, the nta needs to provide more training courses and simplify the interface design to make the system easier to use. for costs consideration, taiwanese government often outsourced most of its is systems to slim sown the human resource overhead. the e-government projects are outsourced to professional computer system's vendors. without ongoing quality monitoring, e-government service may deviate, or depart somewhat, from citizens' requirements and expectations. the authors suggest that explicit statements of quality concern should be added in the future maintenance contract with e-government system providers. moreover, the nta should be responsible for the following jobs: ( ) to define standards and prepare reliable quality measurements of isq, iq, and pc. the instrument of this study may provide a good reference, ( ) to apply different evaluation methods of the internet tax-filing system to detect ongoing quality, and ( ) to define a corrective action plan when quality levels deviate from the standards. though the e-government services in taiwan are ranked in the top five worldwide, this may not guarantee public acceptance. to fulfill the concept of citizen as customer, there is still room for improvement. in taiwan, there are about . million households using the internet, and the registered online users number about . million. these are the major potential internet tax-filing users. in addition, the number of internet tax-filers doubled over that of (see table ). a major reason was the severe acute respiratory syndrome outbreak in taiwan during the tax-filing season. therefore, applying tam to the internet tax-filing system can lead to an understanding of public acceptance of the system and can shed some light on research of the g c e-government system in the future. behavior intention (bi) bi . i intent to file my tax on internet in the next tax return season. bi . i intent to use internet tax-filing system in my preparation of tax return in the next tax return season. bi . to the extent possible, i would try to file my tax on internet in the next tax return season. attitude (att) att . filing tax return on the internet is a good idea. att . filing tax return on internet will be a pleasant experience. att . i like the idea of internet tax-filing. perceived usefulness (pu) pu . using the internet tax-filing enables me to accomplish my tax filing more quickly. pu . using the internet tax-filing enables me to accomplish my tax filing in lower communication cost. pu . using the internet tax-filing enables me to reduce the error in my tax filing process. pu . using the internet tax-filing enables me to get refund from tax agency more quickly. perceived ease of use (peou) peou . learning to use the internet tax-filing software is easy for me. peou . i can use the internet tax-filing software in a manner that allows me to get the appropriate information i want. peou . my interaction with the internet tax-filing software is clear and understandable. peou . in general, i find the internet tax-filing software easy-to-use. information system quality (isq) usefulness of functions, reliability, response time, and ease of navigation. isq . the internet filing system provides the useful functions as i need when i prepare and file my tax return. isq . when i prepare and file my tax return, the operation of internet filing system is reliable. isq . when i prepare and file my tax return, the response of internet filing system is quick. isq . when i prepare and file my tax return, i can navigate the system to finish my tax filing easily. information quality (iq) iq . the internet filing system provides the reliable information when i prepare and file my tax return. iq . the internet filing system provides the adequate information when i prepare and file my tax return. iq . the internet filing system provides the information which is easy to understand. iq . the internet filing system provides the relevant information as i need. perceived credibility(pc) pc . using the internet filing system would not divulge my personal information. pc . i find the internet filing system secure in preparing and filing tax returns. information systems success: the quest for the dependent variable the delone and mclean model of information systems success: a ten-year update are individual differences germane to the acceptance of new information technologies? decision sciences the contribution of information technology to consumer welfare exploring the impact of online service quality on portal site usage the three faces of it value: theory and evidence towards an understanding of the behavioral intention to use a web site technology use and performance: a field study of broker workstations. decision sciences a respecification and extension of delone and mclean's model of is success the adoption of electronic tax filing systems: an empirical study e-government-a vision for new zealanders electronic commerce: a managerial perspective racing to e-government: using the internet for citizen service delivery benchmarking the egovernment revolution meeting the challenge of internet commerce: key issues and concerns a vision of the future global e-government conditions for successful human resource information systems user acceptance of information technology: system characteristics, user perceptions and behavior impacts diffusion of innovations value-added processes in information systems tools and methods for the improvement of quality handbook of quality tools: the japanese approach development of a tool for measuring and analyzing computer user satisfaction dss component design through field experimentation: an application to emergency management service quality: a measure of information systems effectiveness perceived service quality and user satisfaction with the information services function cautions on the use of the servqual measure to assess the quality of information systems services the elements of computer credibility trust and tam in online shopping: an integrated model perceived security and world wide web purchase intention factors affecting engineers' acceptance of asynchronous e-learning systems in high-tech companies implementation: the key to successful information systems a partial test and development of the delone and mclean model of is success ga association for information systems a motivational view of information systems success: a reinterpretation of delone and mclean's model. paper presented at the australian conference on information systems an empirical study on predicting user acceptance of e-shopping on the web attitude, intention and behavior: an introduction to theory and research the psychological origins of perceived usefulness and ease-of-use user acceptance of computer technology: a comparison of two theoretical models a model of the antecedents of perceived ease of use: development and test assessing it usage: the role of prior experience perceived usefulness, perceived ease of use, and user acceptance of information technology assessing information system value measuring effectiveness of computer based information systems in the financial services sector assessing the validity of is success models: an empirical test and theoretical analysis an approach for confirmatory measurement and structural equation modeling of organizational properties multivariate data analysis assessing construct validity in organizational research validating instruments in mis research evaluating structural equation models with unobservable variables and measurement error alternative ways of assessing model fit in publishing in the testing structural equation models the technology acceptance model and the world wide web an empirical assessment of a modified technology acceptance model information technology adoption across time: a cross-sectional comparison of pre-adoption and post-adoption beliefs user acceptance of information technology: toward a unified view validation in is research: a state-ofthe-art assessment measuring information system service quality: servqual from the other side pragmatic perspectives on the measurement of information systems service quality measuring information systems service quality: concerns on the use of the servqual questionnaire measuring information systems service quality: lesson from two longitudinal case studies extending the technology acceptance model with task-technology-fit constructs examining the technology acceptance model using physician acceptance of telemedicine technology a theoretical extension of the technology acceptance model: four longitudinal field studies cognitive fit: theory-based analyses of the graphs versus tables literature the effect of information presentation on decision making: a costbenefit analysis key: cord- -lj x qdd authors: niner, holly j.; johri, shaili; meyer, judith; wassermann, sophia n. title: the pandemic push: can covid- reinvent conferences to models rooted in sustainability, equitability and inclusion? date: - - journal: socio ecol pract res doi: . /s - - -y sha: doc_id: cord_uid: lj x qdd the covid- pandemic necessitates a change in conference formats for . this shift offers a unique opportunity to address long-standing inequities in access and issues of sustainability associated with traditional conference formats, through testing online platforms. however, moving online is not a panacea for all of these concerns, particularly those arising from uneven distribution of access to the internet and other technology. with conferences and events being forced to move online, this is a critical juncture to examine how online formats can be used to best effect and to reduce the inequities of in-person meetings. in this article, we highlight that a thoughtful and equitable move to online formats could vastly strengthen the global socio-ecological research community and foster cohesive and effective collaborations, with ecology and society being the ultimate beneficiaries. international conferences are valued as important for the development of both researchers and knowledge (fraser et al. , p. ; timperley et al. , pp. - ). yet, the traditional conference model that brings delegates together in a single 'destination' demands costly international travel and often high registration fees. given the associated carbon emissions and inequities in access, there is a particularly strong moral onus for those engaged in the field of socio-ecology to develop conference models or practices that do not contribute to the very problems that the discipline seeks to address. the forcing hand of covid- has opened an opportunity to trial online formats and to reinvent conferences as a core institution of research and practice. a global rise in community goodwill and flexibility in response to the challenges of 'lockdown' (morgan ) provides an opportunity to address some of the long-term ethical quandaries that relate to both sustainability and accessibility (ford et al. ; arend and bruijns ; timperley et al. ) posed by traditional conference formats. equal participation at conferences is restricted by access to financial support for travel and registration fees. the costs of visas, flights, registration and accommodation are an insurmountable obstacle for many involved in socioecological practice (xiang ) and research. funding for attendance is frequently limited for those in practitioner positions outside of the academy and those based in lowincome nations (fullick ; waruru ) . in our experience organising several non-profit international socio-ecological conferences, raising finance to support attendance in constrained funding landscapes remains challenging, despite the ethical and moral imperatives to diversify participation. beyond access, societies have moved to address issues of inequality faced by participants. these efforts include codes of conducts to manage in-conference interaction (favaro et al. ) , mentoring (timperley et al. , p. ) , childcare provision, employing mediators to manage disputes and incentivising leadership roles and participation of historically and structurally marginalised groups including women, people of colour, those with physical and mental health disabilities, and lgbtq + (e.g. imcc ). additionally, travelling internationally to a conference is not sustainable. some conferences offset their participants' travel emissions (holden et al. (holden et al. , p. . however, the compensation of carbon emissions is controversial and, according to the carbon management hierarchy, should only be used as a last resort after exhausting all other options for mitigation (hyams and fawcett , p. ). online conference formats remove the need for travel and reduce the costs of attendance, but they do not preclude inequality in access and participation. new challenges are posed by a move online, such as replicating the much valued spontaneous and informal opportunities of traditional in-person conferences, where non-verbal cues are more easily detected (fish et al. , p. ; erickson , p. ) . furthermore, the different norms of interaction in an online setting may exacerbate inequalities in participation. for instance, online communication is often associated with a degradation in politeness (hardaker , p. ) . cultural insensitivity and impoliteness are known causes of lower levels of minority faculty representation in the academy (louque and thompson , p. and p. ) . accordingly, online communication may make it harder for inexperienced or minority community members to establish themselves in a global network of colleagues. this may have long-term impacts on the diversity and innovation potential of the socio-ecological research community. another immediate issue is access to technology and infrastructure for online participation. minority participants likely experience the digital divide disproportionately. for example, only % of urban households in india have internet access, dropping to . % for rural households (government of india p. ). further, primary users of the internet in india are male ( %) with only % of women having access to mobile internet, the primary mode of digital connectivity (gsma , p. ). this underlying disparity in digital access is partially neutralised by reliable internet access for those engaged at a subset of governmental, private and higher education institutes. however, with the current shelter-in-place restrictions, institutional access is restricted, forcing users to rely on in-home or mobile internet, which is frequently unavailable, with only . % of urban households having access to a computer (government of india , p. ). access is likely even worse for conservation practitioners based in rural settings globally and for students 'sent home' from universities. in africa in march , only . % of the total population could access the internet, compared to the rest of the world at . %, with smartphone access at % in south africa, % in kenya and extremely low at % in tanzania (ngware ) . increasing global access to the internet is central to achieving the un sustainable development goals, and a shift to online conferences supports this aim. however, increasing access will increase the carbon emissions associated with the internet, which are estimated to exceed those of the aviation industry (boston consulting group , pp. - ; malmodin and lundén , pp. - ) . evidently, 'going online' does not completely neutralise the carbon emissions of a conference (taylor ). as such, accounting for the carbon footprint of conferences remains relevant for online formats, but could legitimately meet the demands of the carbon management hierarchy. the covid- pandemic offers an opportunity to understand and demonstrate how online platforms can address issues of equity in access and sustainability. the elimination of travel costs clearly reduces the barriers to participation as long as conference registration fees associated with inperson events are similarly reduced. however, engagement online must be fostered to allow online formats to confer the same value as traditional formats. additionally, the associated technological requirements could risk widening the inequities for participants who are already the most disadvantaged in the socio-ecological community (martin ; un news ) . a key 'entry cost' to participation remains in the issue of technology access and internet infrastructure. while many conference participants may have adequate access to internet and technology, to address issues of diversity, equity and inclusion, online platform selection should consider associated requirements for high bandwidth, high-performing devices and training in these technologies. in the absence of uniform digital access, conferences can make use of a variety of more widely available technologies, such as dissemination of conference content in recorded, live and audio-only formats accessible via telephone, radio (ngware ) , podcasts and other popular smartphone applications such as whatsapp (bouhnik and deshen , pp. - ; taru ) . meeting format can address inequalities in participation, but success will depend on community goodwill to actively engage in the programme. formats can be adapted to support the cultivation of 'social presence' through careful consideration of scheduling around time zones, and a combination of live and recorded presentations and interactive events, such as live question and answer sessions or break out groups (tu and mcisaac ) . mentorship and clear communication of expectations of engagement as set out by a code of conduct could also assist in supporting aims of equitable inclusion and in providing space for all voices to be heard. the optimal model for online conferences may differ more drastically from the in-person format. rather than large, immersive conferences held over a set time period, it could include perennial platforms (e.g. professional societies or established communication hubs) integrating a range of technologies that host 'special issues' or informal networking events. these 'smaller' yet more frequent events may ensure active and sustained participation across geographic and disciplinary sectors, ensuring a true diverse and inclusive conference model. the potential to address long-standing inequities in the socio-ecological community through online conferences is a bright spot in the post-covid- landscape. for now, online formats tend to follow the traditions of in-person conferences, being focussed over a defined period and based around thematically grouped presentations or posters. the current pandemic has led to a shift in forms of communication (taru ; wen ) , and as people adapt to forging and maintaining relationships online rather than in person, innovation of the conference model to avoid the risks of continuing or exacerbating issues of inclusion and access online will be key. the reinvention of conferences required by covid- shows us that there are viable options for professional and knowledge development that do not sit at odds with ambitions for an equitable and sustainable future. however, this reinvention must carefully consider the requirements for equitable access and will depend on ongoing and enthusiastic engagement of audiences. as the covid- response challenges the need for and moral standing of the traditional conference, we have been given an opportunity to experiment and begin to explore what works best for all sectors of society. funding open access funding provided by projekt deal. open access this article is licensed under a creative commons attribution . international license, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the creative commons licence, and indicate if changes were made. the images or other third party material in this article are included in the article's creative commons licence, unless indicated otherwise in a credit line to the material. if material is not included in the article's creative commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. to view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/ . /. disparity in conference registration cost for delegates from low-and middle-income backgrounds gesi smarter : the role of ict in driving a sustainable future whatsapp goes to school: mobile instant messaging between teachers and students synchronous interaction among hundreds: an evaluation of a conference in an avatar-based virtual environment your science conference should have a code of conduct video as a technology for informal communication gender inequity in speaking opportunities at the american geophysical union fall meeting the value of virtual conferencing for ecology and conservation it's time to rethink academic conference funding, university affairs key indicators of household social consumption on education in india: nss th round trolling in asynchronous computer-mediated communication: from user discussions to academic definitions academic conferences urgently need environmental policies the ethics of carbon offsetting % registration discount for female symposium leaders at imcc , scb marine exposing the "culture of arrogance" in the academy: a blueprint for increasing black faculty satisfaction in higher education the energy and carbon footprint of the global ict and e & m sectors - . sustain (switzerland) where are the women in ecology as coronavirus shrinks our world, resurgent community spirit offers hope, the guardian startling disparities in digital learning emerge as covid- spreads: un education agency delivering education online: coronavirus underscores what's missing in africa, the conversation digital tragedy: doing online teaching in zimbabwe during the pandemic why going digital in pandemic times might not be as green as you think he moana pukepuke: navigating gender and ethnic inequality in early career academics' conference attendance the relationship of social presence and interaction in online classes the relationship of social presence and interaction african and asian researchers are hampered by visa problems s-how-lockd own-helps -those -who-fear-the-phone xiang w-n ( ) ecopracticology: the study of socio-ecological practice. socio-ecological practice research dr. holly niner is a research and knowledge exchange fellow working on the one ocean hub at the university of plymouth. through her work, she seeks to support the sustainable, equitable and just use of marine and coastal resources and space. she is a co-chair of imcc .dr. shaili johri (ph.d. genetics) is pursuing postdoctoral work at stanford university with a research focus on conservation genomics of marine wildlife. in her capacity as the diversity officer of the society for conservation biology marine section, shaili engages in addressing issues of diversity, equity and inclusion affecting conservation research/practice. after a -year postdoc position at geomar helmholtz centre for ocean research kiel, she transitioned into science management in and is now employed as scientific project manager at the center for ocean and society in kiel, germany.sophia wassermann is in the last few months of her ph.d. in marine science at the national university of ireland, galway. her research focuses on applying theoretical concepts and quantitative methods to practical issues in fisheries and ecosystem sustainability. she is a co-chair of imcc . key: cord- - ug xxd authors: porras, cynthia; fajardo, jenny; rosete, alejandro; pelta, david a. title: planning wi-fi access points activation in havana city: a proposal and preliminary results date: - - journal: information processing and management of uncertainty in knowledge-based systems doi: . / - - - - _ sha: doc_id: cord_uid: ug xxd the availability of wi-fi connection points or hotspots in places such as parks, transport stations, libraries, and so on is one of the key aspects to allow people the usage of internet resources (to study, work or meet). this is even more important in central america and caribbean countries where the deployment of huge cost infrastructure (like optical fiber) to provide internet access at home is not envisaged neither in the short or mid term. and this is clearly the case in havana, cuba. this contribution presents the problem of planning the wi-fi access points activation, where each point can have different signal power levels and availability along the time. due to power consumption constraints, it is impossible to have all the points activated simultaneously with maximum signal strength. the problem is modelled as a dynamic maximal covering location one with facility types and time dependant availability. a metaheuristic approach is used to solve the problem by using an algorithm portfolio and examples on how solutions can be analyzed (beyond the coverage provided) are shown. in a recent report from the international telecommunication union (itu) (united nations specialized agency for information and communication technologies) [ ] , one can read: ... internet use continues to grow globally, with . billion people now using the internet, or . % of the global population. however, an estimated . billion people remain offline, with the majority of the unconnected living in the least developed countries where an average of just two out of every ten people are online. in all regions of the world, households are more likely to have internet access at home than to have a computer because internet access is also possible through other devices. while % of the world's population lives within reach of a mobile broadband (or internet) service, just over % actually uses the internet. according to the world bank' databank (an analysis and visualisation tool that contains collections of time series data on a variety of topics) cuba has the following figures: although internet usage and mobile cellular subscriptions are steadily increasing, they are still quite low. also in the rest of central america and caribbean countries the deployment of huge cost infrastructure (like optical fiber) to provide internet access at home is not envisaged neither in the short or mid term. so deploying wireless connection is one of the steps needed to enlarge internet usage among the people. these wi-fi access points (waps), provide signal to the users if they are within its radius of signal. that means, the location of the wap is very important for the satisfactory access to internet. with this approach, the questions of where to locate the waps and how to plan their activation/deactivation can be treated as special cases of location problems. a classic location problem is the maximal covering location problem (mclp) proposed by church and revelle in [ ] . the mclp aims to locate a limited number of facilities in order to maximize the coverage over a set of demand nodes. the term "facility" should be understood in a wide sense, ranging from warehouses, bus stops or, as in our case, waps. several examples of the application of mclp for placing wireless devices may be found in the literature. in [ ] , one the first contributions in telecommunications scenarios was proposed. a greedy-add method was used over the mclp and the set covering location problem (sclp) to design a network of cellular mobile communication system. in [ ] the mclp was used to locate wireless routers, taking into account a central tower with a main signal. the routers share its signal, therefore they must be located at a certain distance from each other to maintain the signal strength. in [ ] a variant of mclp was proposed to locate routers and gateways. the model has capacity constraint to guarantee the correct flow on the connections. in [ ] a multi-objective mclp was proposed to locate the nodes in a wireless network with bandwidth capacity constraints. the objectives were: to maximize the coverage, to minimize the bandwidth usage, and to minimize the cost of the network and the interference signal. the nsga-ii algorithm was used to obtain the pareto-front of the problem. in [ ] an extension of mclp was used to locate a set of access points in a wireless network in order to maximize the total offloaded traffic in an area. a greedy adding algorithm with substitution was used to solve the problem. in [ ] a fuzzy mclp was proposed to locate wi-fi antennas in havana city. the fuzzy constraint describe uncertain information about the availability of antennas due to meteorological events near the coast. a parametric approach was used to solve the fuzzy problem. in [ ] a variant of dmclp to locate flying base stations was proposed. these bases move according to the change in population density over time. the bases have user connectivity capacity constraints. let's now consider the following situation. havana city is exploring ways to provide internet access to their citizens. there is information about the population density and how they may move. during a day the number of users requiring an internet connection can change. for example, in the early morning there are less users than the afternoon. besides, there is a set of deployed waps that provide a wireless signal, which provide coverage up to a maximal distance, named the coverage radius. a wap has different signal levels or strength (e.g. strong, medium and low), that affects the coverage radius and just one level can be activated. due to energy saving policies, it is impossible to activate all of the available waps with a strong signal level (which would be the obvious solution to the problem). for example, during the early morning all signal levels may be available, since it is when the overall power consumption in the city is lower. however, at night, only the lowest signal level can be used, since it is where the highest energy consumption is perceived. the previous situation can be modelled using the location problem presented in [ ] : the dynamic maximal covering location problem with facilities types and time dependant availability (dmclp-ft). in this contribution we apply the dmclp-ft to manage the waps activation in havana, cuba. the paper is organized as follows: sect. describes the dmclp-ft. section presents the dmclp-ft to activate a set of waps in havana city, cuba and comments the principals results and analysis. finally, we present the conclusions of our contribution. several features should be considered when solving the problem but we highlight two of them: ) waps may not have available all the different signal levels and ) the demand at a geographic area may change over the time. for example, the number of users can decreased or increased depending on the schedule, in the morning there more users on-line than the early morning. these features can be modelled using a generalization of dmclp presented in [ ] : the dynamic maximal covering location problem with facility types and time dependent availability (dmclp-ft). the mathematical formulation of the dmclp-ft, applied to the waps activation, is the follows: the objective function is: subject to: t∈t j∈j the objective function ( ) is aimed at maximizing the coverage of the nodes (sets of users). constraint ( ) shows that a node can be covered if an activated wap with level k in period t belong to set n itk . constraint ( ) shows that the number of waps with level k to activate must be equal to p k . constraint ( ) shows that the total of waps with level k that will be activated should be less than the total of available waps. constraint ( ) shows that the availability of wap with level k must be greater than the number of waps with level k that will be activated in all time periods. constraint ( ) shows that a wap must have at least one available signal level in some period t. constraint ( ) shows that a wap j can be activated with only one level k. finally, constraint ( ) shows that a wap can be activated with an available level k. in this section we describe the problem of planning the waps activation for a whole day in havana city, cuba. the waps can adjust their signal levels (signal power), thus affecting the covered area. the stronger the signal level is, the higher the covered area (the higher the number of users served). due to energy saving policies in cuba, it is not possible to have the already deployed waps working simultaneously with a strong signal. thus, in short, the problem is to determine which wap (and when) should be activated and with which signal level. there are several factors that may influence the availability of a given signal level in a schedule. firstly, the consumption of electrical energy. at night, the energy consumption is higher and therefore, it is not possible to use strong signal levels, since it consumes more energy. secondly, geographical features during working hours (morning and afternoon). during these schedules of the day, activities are carried out that need waps with a certain signal level. for example, cultural or economic activities, where the signal levels suitable for a wap are known. finally, at early morning, the energy consumption is low, so there are no restrictions on the signal levels. as the user may notice, within a day the availability of the signal levels per access point would change, depending on the access point location. in this case study we aim at planning the activation just for one day divided in four stages. part of the data available for the problem is taken from [ ] . we include t = time periods that represent times of a day: t = is early morning, t = is morning, t = is afternoon and t = is night. there are i = areas or user nodes and for each node i and time period t there is a demand estimation or number of users to be covered a it . at period t = , the values a i are taken from the data set used in [ ] . then, the demand a it for t = , , was randomly generated using values between [min(a i ) − ; max(a i ) + ]. the total demand/users to be covered for each time period are { , , , } for to t = (early morning), t = (morning), t = (afternoon) and t = (night) respectively. there are j = waps deployed, and k = signal levels for each wap j are considered. the availability of the signal levels is as follows: at t = every waps can be activated with any signal level. at t = and t = , % of j have available the low power level k = , % with medium power level k = and % with strong power level k = . please note that a wap may have available more than one type of signal during a time period. finally, at t = the waps can only be activated with a low power signal level k = . for each signal level k, the corresponding coverage radius s k is calculated as follows. firstly, it is computed the maxd = max(d ij ). then, for each signal level, the coverage radii are defined as: s = maxd × . , s = maxd × . and s = maxd × . . in a day, we can activate p = waps with low signal, p = with medium signal and p = with a strong signal. as we assume that imprecision can be present in the problem' information, we propose to solve it by means of an approximate method. instead of choosing just a single strategy (like an isolated metaheuristic), we focus in an algorithm portfolio (ap) presented in [ ] . the ap is composed by different metaheuristics, that are used in a cooperative way to obtain good solutions for optimization problems. here, the portfolio is composed by: a hill climbing, an evolutionary strategy and a basic genetic algorithm, both with a population size = and a simulated annealing with and initial temperature t initial = , a cooling rate α = . , a number of iterations at each temperature iter = and a final temperature t f inal = . [ ] . a solution of the dmclp-ft can be represented as an integers' list where the value means that the wap is off, k means that the wap is activated with level k and − means that the wap has not any signal level in that period. as we are not aiming at performing comparisons among several metaheuristics method, we omit more details about the implementation used. the ap method was run times. figure displays the best solution obtained. the overall coverage was ( . %). the activated waps have a circle that represents the radius corresponding to the signal level used. it is interesting to observe the number of users covered at each time period. at the center of fig. a pie chart appears showing the number of waps activated per time period. the black/grey/white series indicates the level k = , , respectively. at period t = there are waps activated: wap with level k = and with level k = . at the period t = , waps were activated: wap with level k = , with k = and with k = . at period t = , waps are activated: waps with level k = , with level k = and with level k = were activated. finally, at period t = , a higher number of waps is used because they can only use a low level signal strength. figure also shows the geographical distribution of the waps, together with their coverage radii (associated with the circle radii). in the solution, we can see that the waps were activated in areas where there is high concentration of users to be connected. it can be observed that at period t = , just a few waps were activated. this period corresponds with the early morning, where the demand is lower. at t = , waps with medium and strong signals are used. at t = , the covered area is quite similar but with a different configuration for the waps. by the end of the day (t = ), the covered demand is reduced due to the use of waps with low level signals. as we made runs of the portfolio, we analyse a set of different solutions with a similar value of demand covered. figure shows a graphic where we can see the behaviour in the number of activated waps during a day. we denoted each level using the previous scheme. for visualization purposes, just solutions (out of ) achieving different coverage values are selected. the solutions are sorted by the coverage value (solution is the one showed in fig. ). for each solution we indicate the number of users covered. the end of the periods are marked by lines and for each one we show the number of users covered. for each series we can see the number of waps that were activated. in general, we can see that the variability among the solutions is low. the graph provides insight on how the coverage is attained. the use of waps in the period t = was not prioritized. the algorithm makes use of waps with level k = in the last period t = , and not in the previous periods. an example is the best solution ( ), which provides new and varied combinations that resulted in a better planning. the activated waps with level k = were the most used in the period t = . it is interesting to note that solution uses the same distribution of levels as the best solution. when we explore the location of the waps we observed that both solutions differ in activated waps. the selection used by the best solution allowed to obtain a greater number of users covered. we can also note that solution has the same number of waps activated in the period t = as the solution . however, solution has more users covered in that period than the solution . we could obtain an even better solution than solution , if we keep the distribution of waps activated in the periods t = , t = and t = of solution , and replace the distribution of t = obtained in solution . in general, in the period t = , we can see few difference between the solutions, in terms of the number of users covered. you can also note that the solution was the only one that used a wap with level k = at t = , but it does not yield to better results. in the best solutions, the use of low power level k = is almost neglected in the first periods. from solutions to , we can see that the main difference was in the period t = , where the quality of the distribution of the waps with signal level k = made the difference. in this contribution we propose a model and an algorithm to deal with the planning of waps activation in havana. a dynamic maximal covering location problem with facility types and time dependant availability were used to model the situation. as we consider the data available as imprecise, we discard solving the problem with an exact solver. instead, we applied an algorithm portfolio composed by classic metaheuristics that allows us to obtain a set of good solutions. we explore several ways to analyse them, ranging from simple coverage values, to the type of signal levels used in the waps, and to the waps geographical distribution. several venues are opened now. one is regarding how the quality of a solution should be assessed. in our opinion, a sort of max-min approach can be useful: to maximize the minimum level of coverage provided at any time of the day. second is to consider some activation/deactivation costs for the waps. both aspects are under study. a genetic algorithm based decision support system for the multi-objective node placement problem in next wireless generation network rethinking offloading wifi access point deployment from user perspective the maximal covering location problem uav base station location optimization for next generation wireless networks: overview and future research directions modeling and solution of maximal covering problem considering gradual coverage with variable radius over multi-periods an algorithm portfolio for the dynamic maximal covering location problem placing wi-fi hotspots in havana with locations availability based on fuzzy constraints international telecommunication union maximal covering with network survivability requirements in wireless mesh networks dynamic maximal covering location problem with facility types and time dependent availability maximizing wireless mesh network coverage demand-based radio network planning of cellular mobile communication systems acknowledgments. d. pelta acknowledges support of project tin - -p (spanish ministry of economy and competitiveness, includes feder funds from the european union). j. fajardo, acknowledges support received funding from the european union horizon research and innovation programme under grant agreement no. . key: cord- -pwmr m o authors: gupta, deepti; bhatt, smriti; gupta, maanak; tosun, ali saman title: future smart connected communities to fight covid- outbreak date: - - journal: nan doi: nan sha: doc_id: cord_uid: pwmr m o internet of things (iot) has grown rapidly in the last decade and continue to develop in terms of dimension and complexity offering wide range of devices to support diverse set of applications. with ubiquitous internet, connected sensors and actuators, networking and communication technology, and artificial intelligence (ai), smart cyber-physical systems (cps) provide services rendering assistance to humans in their daily lives. however, the recent outbreak of covid- (also known as coronavirus) pandemic has exposed and highlighted the limitations of current technological deployments to curtail this disease. iot and smart connected technologies together with data-driven applications can play a crucial role not only in prevention, continuous monitoring, and mitigation of the disease, but also enable prompt enforcement of guidelines, rules and government orders to contain such future outbreaks. in this paper, we envision an iot-enabled ecosystem for intelligent monitoring, pro-active prevention and control, and mitigation of covid- . we propose different architectures, applications and technology systems for various smart infrastructures including e-health, smart home, smart supply chain management, smart locality, and smart city, to develop future connected communities to manage and mitigate similar outbreaks. furthermore, we present research challenges together with future directions to enable and develop these smart communities and infrastructures to fight and prepare against such outbreaks. covid- is an infectious disease caused by a newly discovered coronavirus (sars-cov- ) and is rapidly spreading around the world. according to world health organization (who ), covid- has already affected countries and territories around the world and continue to spread rapidly across other regions. the highly contagious coronavirus outbreak was declared a "pandemic" on march , . who reports that the number of positive cases has dramatically surged, with nearly millions reported cases and , fatalities as of april , . the fatalities are assuredly the most tragic cost of this disease. in order to control the spread the pandemic, lockdowns, quarantines and stay home orders have been issued by several nations across the globe, which have crippled national and world economy with critical consequences to workers, employers and investors. in addition, the industries, businesses and travel restrictions restrain the supply of goods and services, and the economic disruptions will continue to have a long term impact on global supply chains and economy. in the united states, unemployment rates spiked to . % -its highest level since the great depression, in addition to fear of stronger second wave of the disease looming during winter. currently, with no cure or vaccine for this disease, the first line of defense to fight against this pandemic is preventative measures and mitigation strategies. as suggested by the who, the u.s centers for disease control and prevention (cdc ) and several other federal organizations suggest personal protective measure (ppe), social distancing, environmental surface cleaning, self isolation, travel restrictions, local and national lockdowns, quarantine, limits on large gatherings, restrictions on opening businesses, and school closures, as some of the preventive measures that are needed to limit the spread of the disease. however, these guidelines impose restrictions which hinder the way of normal life for humans. it has become a huge challenge to swiftly implement and enforce such measures on a large scale across cities, nations, and around the world. we believe that to effectively enforce and monitor the preventive controls and mitigation strategies for covid- , iot together with its key enabling technologies including cloud computing and artificial intelligence (ai) and data-driven applications can play an important role. there are several existing examples of use of technology to control the spread of covid- and manage the large gatherings of already infected patients and possibly infected cases. the u.s. cdc has introduced a self-checker application enabled with cloud platform, which helps a patient to make decision to find appropriate healthcare service through questionnaires. however, most people do not have any symptoms who are also known as the silent spreaders/asymptomatic carriers. therefore, conducting over the last few years, there has been a huge surge in the number of iot devices and different types of smart sensors have been introduced. with new technological advancements, this trend is expected to continue and grow in the future. iot market is currently valued at $ billion per year and is expected to reach $ billion by . another recent article predicted more than billion devices will be internet-connected by . in today's connected world, not having network capability in a device limits the market potential for that device. as a result, there are large number and various types of network connected iot devices providing convenience and ease of life to humans. smart devices have the potential to be a major breakthrough in efforts to control and fight against the current pandemic situation. iot is an emerging field of research, however, the ubiquitous availability of smart technologies, as well as increased risks of infectious disease spread through the globalization and interconnection of the world necessitates its use for predicting, preventing, and mitigating diseases like covid- . iot includes large number of novel consumer devices including hdmi sticks, ip cameras, smartwatches, connected light bulbs, smart thermostats, health and fitness trackers, smart locks, connected sprinkler systems, garage connectivity kits, window and door sensors, smart light switch, home security systems, smart ovens, smart baby monitors, and blood pressure monitors. however, mostly these iot devices are used in a distributed manner based on user and their requirements. iot technology including smart sensors, actuators, and devices and data driven applications can enable smart connected com- https://www.forbes.com/sites/louiscolumbus/ / / /iot-market-predicted-to-double-by- -reaching- b/# f f https://www.cisco.com/c/dam/en/us/products/collateral/se/internet-of-things/at-a-glance-c - .pdf munities to strengthen the health and economical postures of the nations to fight against the current covid- situation and other future pandemics efficiently. in this paper, our main goal is to present a holistic vision of iot-enabled smart communities utilizing various iot devices, applications, and relevant technologies (e.g., ai, machine learning (ml), etc.). here, we propose a vision of smart connected ecosystem, as shown in figure , with real-world scenarios in various applications domains with a focus on detecting, preventing and mitigating covid- outbreak. the major contributions of this paper are outlined below. • we outline some of covid- symptoms, preventive measures, mitigation strategies, and current problems, challenges and present an overview of adaptable multi-layered iot architecture and depict interactions between layers focusing on smart communities for covid- requirements. • we design a smart connected ecosystem by developing multiple conceptual iot application frameworks including e-health, smart home, smart supply chain management, smart locality, and smart city. we introduce use-cases and applications scenarios for early covid- detection, prevention, and mitigation. • we identify and highlight challenges to implement this smart ecosystem including security and privacy, performance efficiency, interoperability and iot federation, implementation challenges, policy and guidelines, machine learning and big data analytic. finally, we discuss the interdisciplinary research directions to enable and empower future smart connected communities. the remainder of this paper is organized as follows. section discusses the essential characteristics to diagnose, prevent and mitigate covid- disease. section presents the multi-layered architecture for iot, whereas section discusses smart connected ecosystem scenarios in various iot application domains. section highlights open research challenges and future directions. finally, section draws conclusion to this research paper. essential characteristics to diagnose, prevent and mitigate coronavirus is transmitted mainly by the infected person's saliva and nasal drips which spread during coughing and sneezing and infects anybody in close contact. another source of infection is contaminated surfaces in surrounding and high risk areas, such as door handles, railings, elevators, and public restrooms. covid- is a highly contagious virus with the incubation period stretched from days to weeks after exposure. symptoms of covid- range from mild symptoms including fever, coughing and shortness of coping with anxiety disorder, depression issues, and mental health problems. shortness of breath or cannot breathe deeply enough to fill your lungs with air, chills. avoid face-to-face meetings, practice social distance from other people outside of the home. monitor symptoms regularly, wear a cloth covering or n- mask over nose and mouth. knowledge gaps to understand virus transmission, no specific antiviral treatment, and no vaccine available. loss of the sense of smell is most likely to occur by the third day of infection and some patients also have experience a loss of the sense of taste. cover mouth and nose with a cloth or wear mask when around others, wear gloves and discard them properly. manufacturers use of all cleaning and disinfection products, follow the workplace protocol and provide ppe to their employee. lack of testing and essential resources such as ventilators, masks, beds, and health staffs, cancel elective surgery. diarrhea and nausea a few days prior to fever, cdc says a sudden confusion or an inability to wake up and be alert may be a serious sign. cover cough or sneeze with a tissue, then throw the tissue in the trash. hospital task force such as increase the number of testing, available the ppe for their staff members, and increase the incentive care. food and drug administration (fda ) provided emergency use authorization for hydroxychloroquine medicine to treat the people who are suffering from this virus in hospitals. later, on april , fda warned against use of hydroxychloroquine to treat this disease outside of the hospital setting or a clinical trial due to risk of heart rhythm problems. people can protect themselves by following some protective measures and help to slow the spread using mitigation strategies. table provides a comprehensive overview of the symptoms, preventive measures, mitigation strategies and some challenges fighting covid- disease. one of the easiest preventive measure is to wash your hands frequently and thoroughly with soap and water for at least seconds or use hand sanitizer or an alcohol-based hand rub when soap and water are not available. people should keep social distancing (six feet distance) from others especially from people who are coughing or sneezing. it is suggested to wear mask and gloves in outdoor locations, and to avoid touching the face and surfaces such as the button at a traffic light, a keypad to add a tip for the restaurant take-out order, elevator buttons, etc. many surfaces are touched by hands accidentally and virus can be potentially picked up and then transmitted to other surfaces and locations. once the hands are contaminated, the virus can be transferred through eyes, nose or mouth, thus, it enters human body. respiratory hygiene is another protective measure. there is a need to avoid cough or sneeze into the hands, and to cover mouth and nose with a tissue or bent elbow during cough or sneeze and throw away the tissue immediately. groceries and packets can be contaminated from coronavirus, and it is recommended to wash grocery items carefully and wipe packets using disinfectant spray. local public health administrations regularly issue health guidelines, which people should follow. the who, governments, and healthcare workers are all urging people to stay home if they can. on top of basic illness prevention, experts said that the best (and only real) defense against disease is a strong immune system. in addition to the physical health, taking care of mental health is also necessary. high stress levels can take a toll on human's immune system, which is the opposite of what people want in this situation. in addition, mitigation strategies are a set of actions applied by the people and communities (hospitals, grocery stores, and cities) to help slow the spread of respiratory virus infections. these mechanisms can be scaled up or down depending on the evolving local situation. at individual level, if a person is infected with coronavirus, then he/she should self-isolate and follow the guidelines of quarantine provided by hospital. the hospitals must support healthcare workforce, increase testing and intensive care capacity, and availability of personal protective equipment. city governments can appoint task https://www.fda.gov/media/ /download https://www.fda.gov/drugs/drug-safety-and-availability/fda-cautions-against-use-hydroxychloroquine-or-chloroquine-covid- -outside-hospital-setting-or force, open shelters for homeless people, and maintain availability of resources to implement preventive and mitigation strategies for this disease. while each community is unique, appropriate mitigation strategies vary based on the level of community transmission, characteristics of the community and their populations, and the local capacity to implement strategies. nonetheless, it is crucial to understand the characteristics of this novel virus and spread awareness and up-to-date information across communities through appropriate technology. consequently, it is essential to address the challenges with significant research and implementation of strategies as shown in the table. multi-layered architecture in this section, we explain an integrated multi-layer iot architecture which can fundamentally change the infrastructure and underlying technologies for smart communities including hospitals, grocery retail stores, transportation, and city etc. as shown in figure . our proposed architecture extends and adapts existing iot and cps architectures [ ] [ ] [ ] [ ] [ ] [ ] , and focuses on the need of swift enforcement of policies, laws and public guidelines, in order to curtail the widespread of such disease. the architecture integrates a hybrid cloud and edge computing nodes together with iot and smart sensor devices, to enable real-time and data-driven services and applications needed in covid- pandemics. overall, the architecture consists of six layers: object layer, edge layer, virtual object layer, cloud layer, network communication, and application layer. the object layer is a rich set of iot devices including sensors, actuators, embedded devices, road side infrastructures, vehicles, etc. these physical objects are spread across and implemented in smart communities like hospitals, retail-stores, homes, parking lots. the edge layer provides local real-time computation and analysis needed for smart resource constrained physical objects. this layer incorporates edge gateways and cloudlets [ ] which can enable local computation at this layer overcoming limited bandwidth and latency requirements, and also impacts the usability of the iot applications. this multi layer architecture has integrated the concept of virtual objects (vos) [ ] , which are the digital delineation of physical iot devices. vos show the current state of corresponding physical objects in the digital space when they are connected, and can also store a future state for these devices when they are offline. cloud layer provides various services like remote storage, computation, big data analysis, and data-driven ai applications etc. for huge amount of information generated by billions of iot devices connected to the cloud. we defined a computation layer which comprises of edge layer, virtual object layer, and cloud layer. computation, data analytic and processing services are performed in this layer. network communication layer run among different layers to establish the interaction. it is responsible for connecting physical sensors, smart devices, edge compute nodes or cloudlets, and cloud services with different technologies, and is also used for transmitting and processing sensor data. application layer delivers specific services to end users through different iot applications. in the multi-layered smart communities architecture, this application integrates mobile phones, edge computing, cloud computing, ai based analytic, and data-driven services. this architecture can incorporate the iot application frameworks within different domains as discussed in the section , and different use case scenarios can be mapped and implemented using relevant technologies associated with each layer of the architecture. in this section, we discuss various iot use case scenarios to monitor, diagnose, detect, and mitigate it is expected that the global internet of medical things (iomt) market will grow to a . billions in year . as of , there are . million medical devices in use that are connected and monitor body parameters of the users to inform data-driven applications in making real-time healthcare decisions. improving the efficiency and quality of healthcare services in hospitals have been an important and critical challenge during the covid- pandemic time. in e-health use cases, we will discuss three important scenarios which can help reduce the risk and spread of coronavirus infection. e-health set up comprises of a smart hospital, a remote patient monitoring, and a smart testing booth as shown in figure . these use-cases involve connected smart sensors, connected devices, robots, patients, hospital practitioners, workers etc. together with iot applications, edge devices and cloud services to offer data-driven services. other scenarios including smart pharmacy, smart ambulance and smart parking in hospital's parking area, are also briefly discussed. such scenarios can be extended with the current proposed architectures described in section . various components of smart hospital concept has been studied in the literature [ ] [ ] [ ] [ ] [ ] . however, it is still a challenge to track covid- patient's record and keep track of essential resources in hospital during such pandemic crisis. the hospitals are overflowing with patients, and running out of hospital beds, ppe and other essential resources needed for treatment and prevention. to overcome these problems, we propose a smart hospital use case here, whihc extends the existing infrastructure to enable coordinated actions for coronavirus patients. within a smart hospital, rfid sensors can be used to track inventory items like masks, face shields, gauzes, disposable patient examination papers, boxes of gloves, and plastic bottles and vials. these rfid tags also could be an ideal way to keep track of large equipment as well, such as smart beds, ventilators within a smart hospital. towel, sheets, and blankets must be washed and disinfected regularly, and such items also can be tracked through rfid laundry tags. in wuhan, wuchang field hospital provided wearable smart bracelets and rings embedded with multiple sensors to each patient in the hospital, where these iot devices are synced with the cloud ai platform so that patient's vital signs, body temperature, heart rate and blood oxygen levels, can be monitored regularly by hospital practitioners. in addition, all hospital workers and staff members also wear these smart bracelets and rings to notice any early symptoms of coronavirus infection. iot devices generate tremendous amount of data and this data can be collected using edge servers deployed in the hospital facilities. these data sets can be used for training with federated deep learning [ ] technique to enhance the intelligence of data-assisted applications which can be used to predict coronavirus infections for hospital practitioners, staff members and also provide insights on coronavirus characteristics and infection trends for the future. hospital practitioners can check the patient's data through remote applications which will also help in reducing the number of visits to the patient's room to measure her/his vital parameters. in this way, hospital practitioners can not only collect more data in less time with minimal in-person contact but can also reduce the risk for cross-infection from the patients. this technique can significantly help reduce the workload and increase the efficiency of hospital practitioners. moreover, smart hospitals can utilize smart beds, which sense the presence of a covid- patient and automatically adjust the bed to a good angle if the patient is short of breathe to provide proper support without the need for a nurse to intervene. a singapore based medical device company invented a smart ventilator, which allows inpatient monitoring process through remote access via an online portal. these ventilators measure the amount of oxygen automatically, or monitor the rate of delivery to the patient, as high pressure to force in more oxygen can damages the lungs of the patient also. these smart ventilators can communicate through patient's smart bracelets with embedded sensors and can respond according to patient's body parameters. besides, there are other iot devices, such as disinfected robots which can autonomously disinfect a patient room regularly and after the patient is discharged, or a specific hospital area post contamination as needed. a study if a patient is in the car. the pharmacist communicates with patient through an application and provides prescribed medicine to them. before attending the other person, the pharmacist must take some time to sanitize the smart pickup box. in the above scenarios, data and information collected from smart devices are sent to edge gateways, services, or cloud. due to high security and privacy concerns in health domain, it is important to understand that these edge gateways and cloud-iot platforms will be owned only by authorized entities, such as hospitals or other highly trusted entities through some private cloud. we elaborate these challenges in detail in section . various aspects of smart home for health have been investigated in the literature [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] . it is nearly impossible to keep everything in the real-world virus-free. however, individuals must exercise due pre- the novel coronavirus has people boarded up inside their homes due to lockdowns and stay home orders. however, people can still go outside for essential services. it is critical to take precautions during outside time and must be aware of possibly infected items and exposure to the virus that they may accidentally bring home. it is paramount to keep our smart homes disinfected and sanitized using different iot during the pandemic, there is a requirement to improvise supply chain management to adapt automatic business processes and also improve the inventory with delivery of essential items. in this subsection, we will discuss two scenarios: smart inventory, and smart retail stores, which illustrate how iot devices and technologies can enable efficient supply-chain and help in slowing the covid- spread using various prevention and detection mechanisms. the complete scenario of smart supply chain management is shown in figure . various aspects of smart inventory systems have been investigated in the literature [ ] [ ] [ ] [ ] [ ] [ ] [ ] . most of them involve iot devices where staff use handheld readers to scan the bar code of goods, and then write the storage information to the rfid tags to complete the inventory. smart inventory systems show how iot technology can be leveraged globally to plan and respond under the current pandemic situation. inventories are facing an unprecedented challenge in coping with the fallout from covid- . however, a smart inventory system can provide a safe and secure environment to the workers using iot technologies. within the inventories, drones can be used to track all the employees to check their temperature using thermal sensors, and also measure their social distancing practices. inventory manager can also provide smart wearable devices connected to the centralized cloud to each employee to monitor and track them. if an employee or his/her family member is infected with coronavirus, inventory manager can get notification through data-driven iot applications. in addition, disinfectant spray can be attached to the shelves and can start to spray when associated sensor senses the sound of sneezing. california had the earliest stay-at-home order issued on march . in california, there has been an early rise in truck activity since the week of march . autonomous delivery robots can also help in smart inventory and help to reduce cross infection. however, truck activity in california has fallen . % from early february. iot sensors like thermal sensors, gps, motion sensors can be attached with delivery trucks to maintain the temperature for perishable items and to track the location. this data can be stored on inventory cloud, and can help predict demand and supply for next month. these tools will become the foundation on which supply chain managers gain insight into their markets and erratic supply and demand trends. the rfid antenna scans the number of units on the sales floor and alerts a store manager in case it's low. iot allows store managers to automate product orders, is capable of notifying when a certain product needs to be re-ordered, gathers data regarding the popularity of a certain item, and prevents employee theft. the retail industry is seeing a rapid transformation, with the iot solutions taking the center stage in the sector. smart grocery store have been widely investigated in the literature [ , [ ] [ ] [ ] [ ] [ ] . iot along with ai and ml technologies can help slow the spread of infection by enforcing prevention and detection mechanisms through connected sensor devices in a smart grocery store. due to the stay home order, people are panicking and stocking up grocery items. they need to stand in queues for hours outside the store to buy groceries. by employing iot sensors around the store and wearable iot devices, a store manager can get a better understanding to slow the spread in the store. chinese tech firm kuang-chi technologies has developed a smart helmet that is used to identify and target those people who are at high-risk for virus transmission in the retail store. the customer will wear a rfid tag based smart bracelet at the store. at the entrance each customer's smart bracelet will be scanned by a scanner, which will show https://talkbusiness.net/ / /groups-share-data-quantifying-covid- -impacts-on-trucking-industry/ https://www.yicaiglobal.com/news/chinese-tech-firm-debuts-five-meter-fever-finding-smart-helmet his/her body parameters. if the records shows any symptoms of covid- , an alert can be sent to the store manager and individual can be restricted to enter in the store. similarly, there can be thermal cameras and microphone sensors installed at the store which can detect the people who are coughing in store during shopping and take pictures. these areas will be disinfected and identified individuals may be reported for testing based on their other symptoms and will be categorized as risky customers. this information can be maintained in store for short-period of time to assist in identifying these individuals during their future visits to the store. from a customer's perspective, the user can enable alerts on his smartphone regarding his grocery list, can see the map of the store and crowded aisles and plan accordingly to maintain social distance while shopping. the customer can visit desired aisles and get items from the smart shelves, which will allow to pick some limited number of items as per family size, and then put items in the smart cart. smart shelves will have three common elements -an rfid tag, an rfid reader, and an antenna. data collected by smart shelves during the day will be analyzed and shopper buying trends, patterns, shopper traffic, etc. will be shared with a store manager to provide customer-related insights to efficiently manage the store inventory and restocking goods. most retail stores now allow only ten people at minutes shopping interval slot to avoid large gatherings inside the store. social distancing can be measured and enforced by autonomous retail worker (robot) together with smart cameras, and smart microphone sensors and speakers to alert the customers, using ai technologies. if two customers come in same aisle and violate the physical distancing norm, an autonomous retail worker will go there to warn them or a loudspeaker attached with smart camera will announce to keep maintain social distancing. autonomous retail worker can roam around the store and can take note of misplaced items, or products running out of stock (smart shelves also keep track of items and can send alert for restocking as needed). at&t with xenex and brain corporation has already launched iot robots to help grocery stores in keeping them clean, killing germs and maintaining well stocked shelves more efficiently. the uvd robot also uses ultraviolet light to zap infection viruses and sanitize surfaces. in smart pickup, sensors and other ai based techniques are used to determine whether order is ready to pickup and a person is here to pickup his order. for instance, a parking space or driveway at the store might allow covid- patient or elderly people firstly to avoid waiting time. smart pickup can automatically allow most vulnerable and infected people first and enforce these rules to inform the vehicles. a restaurant takeout service can follow the same protocol for smart pickup. intelligent transportation system (its) [ ] can support to deliver resources to essential services and delivery https://spectrum.ieee.org/news-from-around-ieee/the-institute/ieee-member-news/thermal-cameras-are-beingoutfitted-to-detect-fever-and-conduct-contact-tracing-for-covid https://www.fiercewireless.com/iot/at-t- g-lte-connects-iot-robots-to-kill-germs-keep-shelves-stocked robot can enhance contact-less delivery, which reduce the spread of the virus. gupta et al [ ] have also elaborated how its and smart city infrastructures can be used to enable and enforce social distancing community measures in covid- outbreak. smart localities have been widely investigated in the literature [ ] [ ] [ ] [ ] [ ] [ ] [ ] . such localities are a collection of various interdependent human and physical systems, where iot represents the sensing and actuating infrastructure to estimate the state of human and physical systems and assist in adapting/changing these systems. here, we discuss two scenarios, which can help humans avoid coronavirus infections and adapt to the 'new normal' in covid- situation living in the smart locality. these two scenarios including other relevant scenarios are shown in figure . every individual who lives in a smart neighborhood will receive notifications regarding allotted time for outside activities, such as riding a bike, a walk on the trail, etc. in order to maintain the social distancing while being outside in the locality common areas. in a smart locality, motion sensors and cameras will sense and count the number of people and send the data to the locality cloud- as shown in figure . cloud smart analytics service can analyze the locality data and send notification to people regarding https://cloud.google.com/solutions/smart-analytics/ number of positives cases and categorize risk zones with different colors (e.g., red -high risk, yellowmedium risk, and green -low risk) in smart neighborhood. when a person will go for a walk in a smart neighborhood, he/she will receive an alert if any infected person/pet are around and also alert them to avoid high risk (red) zones in smart neighborhood. there will be disinfected sprinkler installed that can spray on the possibly infected areas, such as pedestrian path, common areas, etc. when sensor will sense the presence of infected person in the area through notifications from the locality cloud- . the u.s. cdc we have discussed in previous scenarios. figure shows that the body parameters of the elderly people can be taken through attached body bracelet and sent to edge devices and gateways. iot devices and applications connected through locality cloud- can share information of covid- patients to smart hospitals. the nursing home staff can monitor the patient's body parameters regularly, and will also track other elderly people in the smart nursing home. in a smart gym, multiple sensors, devices, autonomous devices can be connected through gateway, and gym manager/coach can access the information of each member at different access levels and can communicates to other locality through locality cloud- . a gym member will receive a notification regarding to come to the gym, it must required to maintain the % occupancy at the gym and time interval to sanitize all gym equipment and surface. to enable multi-cloud secure data and information sharing and communications between locality clouds, there is need to be decentralized trust framework in place using advanced technologies like blockchain and trusted distributed computing. smart and connected city infrastructures have been investigated in the literature [ , [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] . daegu has setup a novel system using large amount of data gathered from various sensors and devices, such as surveillance camera footage and credit card transactions of confirmed coronavirus patients to recreate their movements. the newcastle university urban observatory developed a way of tracking of pedestrian, car parks, traffic movement to understand how social distancing is being followed in tyne and wear. however, other major cities need to prepare themselves for coronavirus future outbreak waves. countries have used cell phone data to track citizens' movements during the pandemic showing the geographic data on hot spots and risk zones where people are more likely to get infection. smart and connected vehicles have been extensively investigated in the literature [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] . in order to keep patients and healthcare providers safe, drive-thru coronavirus testing sites have been popping up in the city. an autonomous testing vehicle can be used for covid- testing in urban and rural areas. the smart testing vehicle can include infrared body temperature, oxygen level sensors, smart test kit, camera, microphone, and local edge services. it can help reduce exposure of old-age people with preexisting conditions. in these vehicles, a person can enter from one side of the glass-walled area in the car. in-build sensors can record person's body temperature and oxygen level and can store on city cloud- . it can also provide a test kit to individuals who can test themselves and return it through the car window. in rural areas, autonomous testing vehicles are largely applicable and can help in testing people, as well as inform and make people aware of the covid- pandemic, symptoms and preventive measures. to flatten the curve of confirmed cases, smart city can provide mass quarantine for coronavirus patients, who have mild symptoms but with higher risk of cross-infection to others. a smart hall or large stadium or facilities can be setup for quarantine with installed sensors, smart devices, robots, and connected to city cloud- (as shown in figure ). disinfectant robot is an autonomous robot that can sterilize floors in these large areas as discussed in other scenarios. the large-scale disinfectant robot can also be used to clean the roads of the city. autonomous and self driving vehicles can be used for delivering the post, which will also help reducing the human contact and cutting down the number of covid- cases. smart city will also provide immunity-based rfid tags to those people, who recover from the disease and allow the tag holders to return to work with extra-precautions. in the future, once covid- vaccines are available, the individual with vaccination can get similar immunity-based rfid tags to prove their immunity. the development of the proposed smart connected ecosystem requires to address several challenges and needs inter-disciplinary research from an integrated perspective involving different domains and stakeholders. in this section, we will discuss these challenges in detail with examples from each of the proposed scenarios, as illustrated in figure . one of the major challenges in the deployment of the smart infrastructure is the security and privacy concerns pertaining to iot and cps users, smart devices, data, and applications in different application domains like healthcare, smart home, supply chain management, transportation, and smart city. in health care industry, it is still a challenge to secure connected medical devices and ensure user privacy. in e-health scenario, for instance, a user visits smart testing booth for covid- testing, and his/her data is transmitted and stored on smart hospital private cloud. hospitals then share this data with state healthcare staff or city government for tracking and monitoring the user activities. to secure the identity of user and ensure privacy, differential privacy [ ] and data masking techniques [ ] , such as pseudonymize [ ] and anonymize [ ] , can be used. however, there are limitations intrinsic to these solutions. in pseudonymize technique, data can be traced back into its original state with high risk of compromising user privacy, whereas it becomes impossible to return data into its original state in anonymize. it is critical to ensure user privacy while deploying iot and data-driven applications for their wide-adoption in preventing, monitoring, and mitigating covid- . secure authentication mechanisms including access control and communication control models are necessary for cloud-enabled iot platforms to defend against unauthorized access and securing data, both at rest and in motion. several iot access control models have been developed in the literature [ ] , with cloud-assisted iot access control models for aws [ ] , google [ ] , and azure [ ] . traditional access control models are not adequate in addressing dynamic and evolving access control requirements in iot. attribute-based access control (abac) [ , ] , offers a flexible and dynamic access control model, which fits more into distributed iot environments, such as smart home [ ] , connected vehicles [ , ] , and wearable iot devices [ , ] . in addition to access control, communications in terms of data flow between various components in cloud-enabled iot platform need to be secured from unauthorized data access and modifications. thus, attribute-based communication control (abcc) [ ] a pertaining risk to these and other ai assisted system and applications is adversarial machine learning [ ] using which adversaries compromise user data and privacy. in order to protect the data sets, differential privacy [ ] can be applied to add noise. cloud based medical data storage and the upfront challenges have been extensively addressed in the literature [ , ] . study [ ] conducted semistructured interviews with fifteen people living in smart homes to learn about how they use their smart homes, and to understand their security and privacy concerns, expectations, and actions. in future research, there are requirements to conduct interviews of practitioners to understand the security and privacy concerns while developing the smart hospital, and need to apply similar approach involving community residents, infrastructure manufacturers and stakeholders to develop other components of the smart connected ecosystem. privacy preserving deep learning approaches such as collaborative deep learning or federated deep learning also need to be explored to train and deploy local models at the edge devices. within a connected ecosystem, users are constantly interacting with numerous smart devices and applications. one of main challenges in such an environment with billions of smart devices is performance https://spectrum.ieee.org/telecom/security/tracking-covid -with-the-iot-may-put-your-privacy-at-risk efficiency and quality of service (qos iot is an emerging technology that is being adopted by several nations across the world. one reason for the lack of constitutions and policies may be because the iot differs from other network technologies and there is a lack of specific iot standards. the research on constitution and policy, including engagement on public policy development debates, and iot standards is necessary to successfully integrate privacy, accuracy, property and accessibility solutions in the smart communities. to develop effective constitutional policies and standards, collaboration across governmental and nongovernmental organizations and industry partners as cloud providers, iot manufacturers, would be beneficial. enabling a smart community requires thousands of low-power and low cost embedded devices together with large scale data analytics and applications. there are several implementation challenges involved in developing such large scale smart infrastructure. fault tolerance and resilience are the challenges for reliable delivery of sensor data from smart devices to distributed cloud service. various failures can occur including face recognition, community infrastructure management, and http://smartcities.gov.in/content/innerpage/guidelines.php https://www.transportation.gov/smartcity https://www.govtech.com/opinion/if-only-one-us-city-wins-the-smart-city-race-the-whole-nation-loses.html emergency response in smart infrastructure. geographically correlated resilient overlay networks (geocron) [ ] is developed to capture the localized nature of community iot deployments in the context of small failures. research in [ ] proposed a new fault-tolerant routing technique for hierarchical sensor networks. another challenge for constant running and managing these iot devices is costs related to energy , communication, computation, infrastructure, and operation. there is generally a tradeoff between benefit and cost for iot applications [ ] , however in the scenario of covid- pandemic, expected benefits (saving lives, economic growth) should outweigh the operational and deployment costs. another challenge, for instance, various iot devices communicate to each other or server to build a prediction model. if the local model is not able to predict accurately due to data duplication or other reasons, there will be no point to build such a model. this study [ ] proposed a game theoretical analysis to allocate more storage capacity in a cost-effective manner, which achieves to maximize the benefits. for the future directions, game theoretical approach can be used to analyze the smart infrastructures in terms of cost-benefit analysis. furthermore, interdisciplinary research collaboration is inevitable to implement a smart connected ecosystem. there are several areas of research and engineering aspects, as machine-to-machine technology, artificial intelligence, deep and machine learning, predictive analysis, security and privacy, and others need to be merged and collaborative research approach is necessary in implementing, deploying, and managing a smart connected ecosystem. iot generate tremendous amount of data collected by physical devices, and this raw data is converted into valuable knowledge using ai and machine learning technologies. the " v" (value, velocity, volume, variety, variability, and veracity) big data challenges for iot applications are discussed in [ , ] . the volume of data from iot devices overwhelms storage capacities. there is not only storage issue, but the data needs to be organized properly so that it can be retrieved and processed in a timely manner. data duplication is a data storage issue when an organization has multiple copies of the same data source. for example, a user has multiple wearable smart bracelets (smart hospital bracelet, smart grocery bracelet, and rfid antibody tag bracelet), these wearable devices will collect similar kind of data from a user which can create an issue of data duplication. machine learning (ml) based applications require a large amount of valuable data for correct prediction, however, complicated and insufficient data can be an issue to the accuracy of the learning and predictive models. in addition, ml approaches need further research and development to deal with such heterogeneous and constantly evolving sensory data inputs. for instance, a locality-based covid- patient detection model based on early symptoms learns with the collaboration with smart nursing home data sets and smart child care data sets. the prediction model can be biased towards elderly people if the number of patients in smart nursing home are more than smart child care. to overcome this problem, both models can learn at their edge networks using collaborative deep learning [ ] . research on these open challenges will help early development and deployment of future smart communities. in this paper, we propose future smart connected community scenarios, which are blueprints to develop smart and intelligent infrastructures against covid- and stop similar pandemic situations in the future. the autonomous operation with low human intervention in smart communities enable safe environment and enforce preventive measures for controlling the spread of infection in communities. data-driven and ai assisted applications facilitate increased testing, monitoring and tracking of covid- patients, and help to enforce social distance measure, predict possible infections based on symptoms and human activities, optimize the delivery of essential services and resources in a swift and efficient manner. the paper discussed different use case scenarios to reflect smart applications and ecosystem. the plethora of iot devices enable huge data collection in different sectors including healthcare, home, supply chain management, transportation, environment, and city, which raises user concerns. in addition, the implementation of proposed smart connected scenarios face other challenges including legislation and policy, deployment cost, interoperability etc. which have also been discussed in the paper. we believe that our vision of smart communities will ignite interdisciplinary research and development of connected ecosystem to prepare our world for future such outbreaks. smart items, fog and cloud computing as enablers of servitization in healthcare enabling health monitoring as a service in the cloud opportunities and challenges of the internet of things for healthcare: systems engineering perspective security and privacy in smart farming: challenges and opportunities cloud-assisted industrial internet of things (iiot)-enabled framework for health monitoring an access control framework for cloud-enabled wearable internet of things the case for vm-based cloudlets in mobile computing the virtual object as a major element of the internet of things: a survey an iot-aware architecture for smart healthcare systems enhancing the quality of life through wearable technology an iot-aware architecture for smart healthcare systems fall detection -principles and methods flexible technologies and smart clothing for citizen medicine, home healthcare, and disease prevention privacy-preserving deep learning a health smart home system to report incidents for disabled people correlation between real and simulated data of the activity of the elderly person living independently in a health smart home study and implementation of a network point health smart home electrocardiographic model and simulator of the activity of the elderly person in a health smart home health smart home -towards an assistant tool for automatic assessment of the dependence of elders detecting health and behavior change by analyzing smart home sensor data patient status monitoring for smart home healthcare automated cognitive health assessment from smart home-based behavior data towards a distributed estimator in smart home environment access control model for google cloud iot a closer look into privacy and security of chromecast multimedia cloud communications investigating security and privacy of a cloud-based wireless ip camera: netcam a testbed for privacy and security of iot devices an experimental framework for investigating security and privacy of iot devices analysis of iot traffic using http proxy convergence of manet and wsn in iot urban scenarios a method to make accurate inventory of smart meters in multi-tags group-reading environment smart spare part inventory management system with sensor data updating an iot application for inventory management with a self-adaptive decision model an iot based inventory system for high value laboratory equipment iot applications on secure smart shopping system aeon: a smart medicine delivery and inventory system for cebu city government's long life medical assistance program study of smart warehouse management system based on the iot iot based grocery management system: smart refrigerator and smart cabinet short on time? context-aware shopping lists to the rescue: an experimental evaluation of a smart shopping cart the smart shopping basket based on iot applications sysmart indoor services: a system of smart and connected supermarkets enabling rfid in retail secure v v and v i communication in intelligent transportation using cloudlets enabling and enforcing social distancing measures using smart city and its infrastructures: a covid- use case smart community: an internet of things application internet of things and big data analytics for smart and connected communities smart cities, big data, and communities: reasoning from the viewpoint of attractors a community-based iot personalized wireless healthcare solution trial an integrated cloud-based smart home management system with community hierarchy modeling 'thriving communities' using a systems architecture to improve smart cities technology approaches research on key technology for data storage in smart community based on big data a use case in cybersecurity based in blockchain to deal with the security and privacy of citizens and smart cities cyberinfrastructures internet of things for smart cities understanding smart cities: an integrative framework an information framework for creating a smart city through internet of things foundations for smarter cities long-range communications in unlicensed bands: the rising stars in the iot and smart city scenarios smart health: a context-aware health paradigm within smart cities smarter cities and their innovation challenges everything you wanted to know about smart cities: the internet of things is the backbone uav-enabled intelligent transportation systems for the smart city: applications and challenges a collaborative mechanism for private data publication in smart cities scalable mobile sensing for smart cities: the musanet experience an introduction to multi-sensor data fusion smart cars on smart roads: problems of control the security and privacy of smart vehicles real-time object detection for "smart" vehicles predictive active steering control for autonomous vehicle systems collision avoidance and stabilization for autonomous vehicles in emergency scenarios learning driving styles for autonomous vehicles from demonstration dynamic groups and attribute-based access control for next-generation smart cars the algorithmic foundations of differential privacy on the security and privacy of internet of things architectures and systems privacy through pseudonymity in user-adaptive systems privacy protection: p-sensitive k-anonymity property anas abou elkalam, and abdellah ait ouahman. access control in the internet of things: big challenges and new opportunities access control model for aws internet of things parbac: priority-attribute-based rbac model for azure iot cloud guide to attribute based access control (abac) definition and considerations (draft). nist special publication a unified attribute-based access control model covering dac, mac and rbac authorizations in cloud-based internet of things: current trends and use cases authorization framework for secure cloud assisted connected cars and vehicular internet of things poster: iot sentinel-an abac approach against cyber-warfare in organizations abac-cc: attribute-based access control and communication control for internet of things iot passport: a blockchain-based trust framework for collaborative internet-of-things adversarial attacks on medical machine learning development of private cloud storage for medical image research data extensive medical data storage with prominent symmetric algorithms on clouda protected framework end user security and privacy concerns with smart homes mifaas: a mobile-iot-federation-asa-service model for dynamic cooperation of iot cloud providers internet of things-new security and privacy challenges a framework for internet of things-enabled smart government: a case of iot cybersecurity policies and use cases in us federal government the united kingdom's emerging internet of things (iot) policy landscape. tanczer, lm, brass, i the united kingdom's emerging internet of things (iot) policy landscape the internet of things (iot) and its impact on individual privacy: an australian perspective resilient overlays for iotbased community infrastructure communications enabling reliable and resilient iot based smart city applications cost-benefit analysis at runtime for self-adaptive systems applied to an internet of things application cost-benefit analysis game for efficient storage allocation in cloud-centric internet of things systems: a game theoretic perspective internet of things: vision, future directions and opportunities deep learning for iot big data and streaming analytics: a survey learner's dilemma: iot devices training strategies in collaborative deep learning key: cord- -e yfo authors: rainwater-lovett, kaitlin; rodriguez-barraquer, isabel; moss, william j. title: viral epidemiology: tracking viruses with smartphones and social media date: - - journal: viral pathogenesis doi: . /b - - - - . - sha: doc_id: cord_uid: e yfo the science of epidemiology has been developed over the last years, using traditional methods to describe the distribution of diseases by person, place, and time. however, in the last several decades, a new set of technologies has become available, based on the methods of computer sciences, systems biology, and the extraordinary powers of the internet. technological and analytical advances can enhance traditional epidemiological methods to study the emergence, epidemiology, and transmission dynamics of viruses and associated diseases. social media are increasingly used to detect the emergence and geographic spread of viral disease outbreaks. large-scale population movement can be estimated using satellite imagery and mobile phone use, and fine-scale population movement can be tracked using global positioning system loggers, allowing estimation of transmission pathways and contact patterns at different spatial scales. advances in genomic sequencing and bioinformatics permit more accurate determination of viral evolution and the construction of transmission networks, also at different spatial and temporal scales. phylodynamics links evolutionary and epidemiological processes to better understand viral transmission patterns. more complex and realistic mathematical models of virus transmission within human and animal populations, including detailed agent-based models, are increasingly used to predict transmission patterns and the impact of control interventions such as vaccination and quarantine. in this chapter, we will briefly review traditional epidemiological methods and then describe the new technologies with some examples of their application. the science of epidemiology has been developed over the last years, using traditional methods to describe the distribution of diseases by person, place, and time. however, in the last several decades, a new set of technologies has become available, based on the methods of computer sciences, systems biology, and the extraordinary powers of the internet. technological and analytical advances can enhance traditional epidemiological methods to study the emergence, epidemiology, and transmission dynamics of viruses and associated diseases. social media are increasingly used to detect the emergence and geographic spread of viral disease outbreaks. large-scale population movement can be estimated using satellite imagery and mobile phone use, and fine-scale population movement can be tracked using global positioning system (gps) loggers, allowing estimation of transmission pathways and contact patterns at different spatial scales. advances in genomic sequencing and bioinformatics permit more accurate determination of viral evolution and the construction of transmission networks, also at different spatial and temporal scales. phylodynamics links evolutionary and epidemiological processes to better understand viral transmission patterns. more complex and realistic mathematical models of virus transmission within human and animal populations, including detailed agent-based models, are increasingly used to predict transmission patterns and the impact of control interventions such as vaccination and quarantine. in this chapter, we will briefly review traditional epidemiological methods and then describe the new technologies with some examples of their application. insight into the epidemiology of viral infections long preceded the recognition and characterization of viruses as communicable agents of disease in humans and animals, extending at least as far back as the treatise of abu becr (rhazes) on measles and smallpox in the tenth century. successful efforts to alter the epidemiology of viral infections can be traced to the practice of variolation, the deliberate inoculation of infectious material from persons with smallpox (see chapter on the history of viral pathogenesis). documented use of variolation dates to the fifteenth century in china. edward jenner greatly improved the practice of variolation in using the less-virulent cowpox virus, establishing the field of vaccinology. an early example of rigorous epidemiological study prior to the discovery of viruses was the work of the danish physician peter panum who investigated an outbreak of measles on the faroe islands in . through careful documentation of clinical cases and contact histories, panum provided evidence of the contagious nature of measles, accurate measurement of the incubation period, and demonstration of the long-term protective immunity conferred by measles. the discovery of viruses as "filterable agents" in the late-nineteenth and early twentieth centuries greatly enhanced the study of viral epidemiology, allowing the characterization of infected individuals, risk factors for infection and disease, and transmission pathways. traditional epidemiological methods measure the distribution of viral infections, diseases, and associated risk factors in populations in terms of person, place, and time using standard measures of disease frequency, study designs, and approaches to causal inference. populations are often defined in terms of target and study populations, and individuals within study populations in terms of exposure and outcome status. the purpose of much traditional epidemiological research is to quantify the strength of association between exposures and outcomes by comparing characteristics of groups of individuals. exposures or risk factors include demographic, social, genetic, and environmental factors, and outcomes include infection or disease. in viral epidemiology, infection status is determined using diagnostic methods to detect viral proteins or nucleic acids, and serologic assays to measure immunologic markers of exposure to viral antigens. infection status can be defined as acute, chronic, or latent. standard measures of disease frequency include incidence, the number of new cases per period of observation (e.g., person-years), and prevalence, the number of all cases in a defined population and time period. prevalence is a function of both incidence and duration of infection and can increase despite declining incidence, as observed with the introduction of antiretroviral therapy for human immunodeficiency virus (hiv) infection in the united states. although the number of new cases of hiv infection declined, the prevalence of hiv infection increased as treated individuals survived longer. commonly used study designs include, l cross-sectional studies in which individuals are sampled or surveyed for exposure and disease status within a narrow time frame, l cohort studies in which exposed and unexposed individuals are observed over time for the onset of specified outcomes, l case-control studies in which those with and without the outcome (infection or disease) are compared on exposure status, and l clinical trials in which individuals are randomized to an exposure such as a vaccine or drug and observed for the onset of specified outcomes appropriate study design, rigorous adherence to study protocols, and statistical methods are used to address threats to causal inference (i.e., whether observed associations between exposure and outcome are causal), such as bias and confounding. much can be learned about the epidemiology of viral infections using such traditional methods and many examples could be cited to establish the importance of these approaches, including demonstration of the mode of transmission of viruses by mosquitoes (e.g., yellow fever and west nile viruses), the causal relationship between maternal viral infection and fetal abnormalities (e.g., rubella virus and cytomegalovirus), and the role of viruses in the etiology of cancer (e.g., epstein-barr and human papilloma viruses). the epidemiology of communicable infectious diseases is distinguishable from the epidemiology of noncommunicable diseases in that the former must account for "dependent happenings." this term was introduced by ronald ross to capture the fact that infectious agents are transmitted between individuals or from a common source. traditional epidemiological and statistical methods often assume disease events in a population are independent of one another. in infectious disease epidemiology, individuals are defined in terms of susceptible, exposed, infectious, and recovered or immune. key characteristics of viral infections that determine the frequency and timing of transmission, and thus the epidemiology, include the mode of transmission (e.g., respiratory, gastrointestinal, sexual, bloodborne, and vector-borne), whether infection is transient or persistent, and whether immunity is short or long lasting. temporal changes in the transmission dynamics of viral infection can be displayed with epidemic curves, by plotting the number or incidence of new infections over time to demonstrate outbreaks, seasonality, and the response to interventions. key metrics in infectious disease epidemiology that capture the dependent nature of communicable diseases include: ( ) the latent period, the average time from infection to the onset of infectiousness; ( ) the infectious period, the average duration of infectiousness; ( ) the generation time, the average period between infection in one individual and transmission to another; and ( ) the basic reproductive number (r ), the average number of new infections initiated by a single infectious individual in a completely susceptible population over the course of that individual's infectious period. if r is larger than one, the number of infected individuals and hence the size of the outbreak will increase. if r is smaller than one, each infectious individual infects on average less than one other individual and the number of infected individuals will decrease and the outbreak ceases. the reproductive number (r) is a function not only of characteristics of the viral pathogen (e.g., mode of transmission), but also the social contact network within which it is transmitted and changes over time in response to a decreasing number of susceptible individuals and control interventions. an important concept related to the interdependence of transmission events is herd immunity, the protection of susceptible individuals against infection in populations with a high proportion of immune individuals because of the low probability of an infectious individual coming in contact with a susceptible individual. the concepts and methods of infectious disease epidemiology provide the tools to understand changes in temporal and spatial patterns of viral infections and the impact of interventions. traditional epidemiological methods provide powerful analytical approaches to measure associations between exposures (risk factors) and outcomes (infection or disease). recent technological advances enhance these methods and permit novel approaches to investigate the emergence, epidemiology, and transmission dynamics of viruses and associated diseases. expanded access to the internet and social media has revolutionized outbreak detection and viral disease surveillance by providing novel sources of data in real time (chunara, ) . traditional epidemiologic surveillance systems rely on standardized case definitions, with individual cases typically classified as suspected, probable, or confirmed based on the level of evidence. confirmed cases require laboratory evidence of viral infection. surveillance systems are either active or passive. active surveillance involves the purposeful search for cases within populations whereas passive surveillance relies on routine reporting of cases, typically by health care workers, health care facilities, and laboratories. data acquired through active surveillance are often of higher quality because of better adherence to standardized case definitions and completeness of case ascertainment but are more expensive and resource intensive. however, both active and passive surveillance are prone to delays in data reporting. the major advantage of using the internet and social media to monitor disease activity is that the signal can be detected without the lag associated with traditional surveillance systems. influenza is the most common viral infection for which the internet and social media have been used for disease surveillance because of its high incidence, wide geographic distribution, discrete seasonality, short symptomatic period, and relatively specific set of signs and symptoms. however, the internet and social media have several limitations compared to traditional active and passive surveillance systems and complement rather than replace these methods. these limitations include lack of specificity in the "diagnosis," and waxing and waning interest and attention in social media independent of disease frequency. in , the internet company google developed a webbased tool called google flu trends, for early detection of influenza outbreaks. google flu trends is based on the fact that millions of people use the google search engine each day to obtain health-related information (ginsberg, ). logs of user key words for pathogens, diseases, symptoms, and treatments, as well as information on user location contained in computer internet protocol (ip) addresses, allow temporal and spatial analyses of trends in search terms ( figure ). early results suggested that google flu trends detected regional outbreaks of influenza - days before conventional surveillance by the centers for disease control and prevention (carneiro, ). however, accurate prediction was not as reliable as initially thought, and google estimates did not closely match measured activity during the - influenza season. google now reevaluates estimates using data from traditional surveillance systems (specifically those of the centers for disease control and prevention) to refine model and parameter estimates. these refinements more accurately capture the start of the influenza season, the time of peak influenza virus transmission, and the severity of the influenza season. a similar approach, called google dengue trends, is used to track dengue virus infections by aggregating historical logs of anonymous online google search queries associated with dengue, using the methods developed for google flu trends. early observations suggest google queries are correlated with national-level dengue surveillance data, and this novel data source may have the potential to provide information faster than traditional surveillance systems ( figure ). other internet sources are being explored to enhance viral surveillance. wikipedia is a free, online encyclopedia written collaboratively by users and is one of the most commonly used internet resources since it was started in . as with google searches, the use of disease-specific queries to wikipedia are expected to correlate with disease activity. the number of times specific influenza-related wikipedia sites were accessed provided accurate estimates of influenza-like illnesses in the united states weeks earlier than standard surveillance systems and performed better than google flu trends (mciver, ) . similarly, social media data are being evaluated for surveillance purposes. twitter is a free social networking service that enables users to exchange text-based messages of up to characters known as tweets. as with google flu trends, the number of tweets related to influenza activity is correlated with the number of symptomatic individuals. several published studies reported correlations between twitter activity and reported influenza-like illnesses (chew, ; signorini, ; figure ). limitations to using social media, such as twitter, to monitor disease activity are illustrated by the ebola virus outbreak in west africa in early . despite the fact that ebola had not yet occurred in the united states, posts to twitter on ebola rose dramatically, likely in response to intense media coverage and fear. clearly, such tweets could not be interpreted to indicate ebola disease activity in the united states. studies reporting misleading associations, or the lack of correlation between social media and disease activity, are rarely published, providing a cautionary note. while initial efforts using data from the internet for viral disease surveillance offer promising results, concerns have been raised regarding the utility and robustness of these approaches (lazer, ) . integration into existing surveillance frameworks will be necessary to maximize the utility of these data streams. the internet allows rapid processing and communication of health-related information, including the aggregation and display of surveillance data for viral infections. traditional surveillance networks can be linked through the internet to allow rapid integration and dissemination of information. information on viral disease outbreaks available through internet postings of health care agencies such as the world health organization (who) and centers for disease control and prevention (cdc), as well as press reports and blogs, can provide data that are more current than traditional surveillance systems. information from these online sources can be made available to a large, global audience. several of the most commonly used surveillance sites report animal as well as human diseases (see sidebar and figure ). mapping spatial patterns of disease and relationships with environmental variables preceded the development of modern epidemiology. the classic example is john snow's hand-drawn map of london cholera cases of . however, routine mapping of health data only became commonplace in the s after desktop geographic information systems became widely available. combined with satellite imagery and remotely sensed environmental and ecological data, spatial mapping of viral infections is a powerful tool for surveillance and epidemiological research. spatial epidemiology is typically used to identify and monitor areas of differential risk. an early example was a large outbreak of st. louis encephalitis virus infection in houston, texas in . spatial analysis showed that the outbreak was concentrated in the city center, with lower incidence at the outskirts. further investigation revealed that the city center was associated with the lowest economic strata, unscreened windows, lack of air-conditioning and pools of standing water, factors facilitating virus transmission. investigation into the spatiotemporal dynamics of viral diseases at smaller spatial scales has become promed, the program for monitoring emerging diseases, is an internet-based reporting system established in that compiles information on outbreaks of infectious diseases affecting humans, animals, and food plants. promed relies on official announcements, media reports, and local observers, including the network of subscribers. a team of experts screen, review, and investigate reports before posting and often provide commentary. reports are distributed by email to direct subscribers and posted on the promed-mail web site. promed-mail currently reaches over , subscribers in at least countries. started in by epidemiologists and software developers at boston children's hospital, healthmap monitors disease outbreaks and provides real-time surveillance of emerging public health threats, including viral infections (figure ) . healthmap organizes and displays data on disease outbreaks and surveillance using an automated process. data sources include online news aggregators, eyewitness reports, expertcurated discussions and validated official reports. google flu trends http://www.google.org/flutrends google dengue trends http://www.google.org/denguetrends healthmap http://healthmap.org promed http://www.promedmail.org possible with increasing availability of global positioning systems devices and geocoding algorithms. such studies have revealed spatial heterogeneity in the local transmission of some directly (e.g., hiv and influenza) and indirectly (e.g., dengue and chikungunya) transmitted viruses. for example, clustering analyses of the residential locations of people with dengue in bangkok over a -year period showed evidence of localized transmission at distances less than km (salje, ; figure ). analyses of data from a large population-based cohort of hiv-infected persons in rakai district, uganda revealed strong within-household clustering of prevalent and incident hiv cases as well clustering of prevalent cases up to m (grabowski, ) . beyond descriptive applications, mapping spatiotemporal patterns of viral infections can provide fundamental insights into transmission dynamics at different spatial scales. traveling waves from large cities to small towns were shown to drive the spatiotemporal dynamics of measles in england and wales (xia, ) . the incidence of dengue hemorrhagic fever across thailand manifested as a traveling wave emanating from bangkok and moving radially at a speed of km/month (cummings, ) . insight into the spatial epidemiology of viral infections and associations with environmental risk factors can be greatly enhanced when information on the spatial location of cases is combined with remotely sensed environmental data (rodgers, ) . the spatial coordinates of cases can be overlaid on satellite imagery to demonstrate relationships with environmental features-such as bodies of water-and formally analyzed using spatial statistical techniques. satellite sensors that detect reflected visible or infrared radiation provide additional information on temperature, rainfall, humidity, and vegetation among other variables, which are particularly important for the transmission dynamics of vector-borne viral infections. satellite data for epidemiologic analyses are provided by a number of sources such as: ( ) earth-observing satellites with high spatial resolution ( - m) but low repeat frequencies such as ikonos and landsat satellites; ( ) oceanographic and atmospheric satellites such as modis and aster with lower spatial resolution ( . - km) that provide images of the earth surface twice a day; and ( ) geostationary weather satellites such as geos with large spatial resolution ( - km). the statistical relationships between cases and environmental risk factors can be used to construct risk maps. risk maps display the similarity of environmental features in unsampled locations to environmental features in locations where the disease is measured to be present or absent. spatial analysis of the initial cases of west nile virus infection in new york city in identified a significant spatial cluster (brownstein, ) . using models incorporating measures of vegetation cover from satellite imagery, the risk of west nile virus could be estimated throughout the city. a more recent risk map for west nile virus in suffolk county, new york, was generated with data on vector habitat, landscape, virus activity, and socioeconomic variables derived from publicly available data sets (rochlin, ; figure ). population movement plays a crucial role in the spread of viral infections. in the past, quantifying the contribution of movement to viral transmission dynamics at different spatial scales was challenging, due to limited data. as an early example, the impact of restrictions of animal movement on transmission of foot-and-mouth disease in was estimated, using detailed contact-tracing data from farms in the united kingdom (shirley, ) . however, such detailed data are rarely available for patterns of human movement. studies have attempted to model the impact of long-range human movement on the spread of viral diseases using measures such as distance between cities, commuting rates, and data on air travel. this approach has been used to explain regional and interregional spread of influenza viruses. data on air traffic volume, distance between areas, and population sizes have been invoked to describe and predict local and regional spread of chikungunya virus in the americas (tatem, ) . new technologies have greatly enhanced the capacity to study the impact of human movement on transmission dynamics of infectious diseases. data from mobile phones and gps loggers can be used to characterize individual movement patterns and the time spent in different locations (figure ) . individual movement patterns can be overlaid on risk maps to quantify movement to and from areas of high (sources) and low risk (sinks) as well as to estimate potential contact patterns. gps data loggers generated . million gps data points to track the fine-scale mobility patterns of residents from two neighborhoods in iquitos, peru, to better understand the epidemiology of viral infections (vazquez-prokopec, ) . most movement occurred within km of an individual's home. however, potential contacts between individuals were irregular and temporally unstructured, with fewer than half of the tracked participants having a regular, predictable routine. the investigators explored the potential impact of these temporally unstructured daily routines and contact patterns on the simulated spread of influenza virus. the projected outbreak size was % larger as a consequence of these unstructured contact patterns, in comparison to scenarios modeling temporally structured contacts. in addition to identifying individual and environmental characteristics associated with temporal and spatial patterns of viral infections, transmission networks are critical drivers of the dynamics of viral infections. analysis of transmission networks defines the host contact structure within which directly transmitted viral infections spread. network theory and analysis are complex subjects with a long history in mathematics and sociology, but have recently been adapted by infectious disease epidemiologists. the epidemiologic study of social networks is facilitated by unique study designs, including snowball sampling or respondent-driven sampling, in which study participants are asked to recruit additional participants among their social contacts. differing sexual contact patterns serve as an example of the importance of contact networks to the understanding of viral epidemiology. concurrent sexual partnerships amplify the spread of hiv compared with serial monogamy. this could partially explain the dramatic differences in the prevalence of hiv in different countries. social networks were shown to affect transmission of the h n influenza virus, and were responsible for cyclical patterns of transmission between schools, communities, and households. technological advances in quantifying contact patterns, with wearable sensors and the use of viral genetic signatures, have greatly enhanced the ability to understand complex transmission networks. self-reported contact histories and contact tracing are the traditional epidemiological methods to define transmission networks. contact tracing has a long history in public health, particularly in the control of sexually transmitted diseases and tuberculosis, and is critical to the control of outbreaks of viral infections such as the middle east respiratory syndrome coronavirus (mers-cov) and ebola virus. to better understand the nature of human contact patterns, sensor nodes or motes have been used to characterize the frequency and duration of contacts between individuals in settings such as schools and health-care facilities. these technologies offer opportunities to validate and complement data collected using questionnaires and contact diaries. as an example, investigators used wireless sensor network technology to obtain data on social contacts within m for high school students in the united states, enabling construction of the social network within which a respiratory pathogen could be transmitted (salathe, ) . the data revealed a high-density network with typical small-world properties, in which a small number of steps link any two individuals. computer simulations of the spread of an influenza-like virus on the weighted contact graph were in good agreement with absentee data collected during the influenza season. analysis of targeted immunization strategies suggested that contact network data can be employed to design targeted vaccination strategies that are significantly more effective than random vaccination. advances in nucleic acid sequencing and bioinformatics have led to major advances in viral epidemiology. population (sanger) sequencing has been the standard method for dna sequencing but is increasingly replaced by deep sequencing in which variants within a viral swarm are distinguished. sequencing allows for the detection of single nucleotide polymorphisms (snps) and nucleotide insertions or deletions ("indels"), analysis of synonymous and nonsynonymous mutations, and phylogenetic analysis (see chapter on virus evolution). sequencing techniques can be applied to both viral and host genomes. snps may be associated with changes in viral pathogenesis, virulence, or drug resistance. molecular techniques applied to pathogens also have been fundamental to the study of the animal origins of many viral infections including hiv and mers. phylogeographic approaches were used to trace the origins of the hiv pandemic to spillover events in central africa (sharp, ) . more recently, sequence data were used to track the animal reservoirs of mers-cov associated with the outbreaks (haagmans, ) , and to compare the ebola virus strain circulating in the west africa outbreak to strains from prior outbreaks (gire, ) . epidemiologic studies that probe host genomes can be either candidate gene studies or genome-wide association studies. the goal of these studies is to link specific changes with an increased risk of infection or disease. as an example, a small subset of individuals who failed to acquire hiv infection despite exposure, prompted studies to determine how these individuals differed from those who acquired infection. a -base-pair deletion in the human ccr gene, now referred to as ccr -delta , accounted for the resistance of these subjects. individuals who are ccr -delta homozygotes are protected against hiv infection by ccr tropic hiv strains, while heterozygotes have decreased disease severity. infectious disease epidemiologists are increasingly linking evolutionary, immunologic, and epidemiological processes, a field referred to as phylodynamics voltz, ) . because of the high mutation rates of viral pathogens, particularly rna viruses, evolutionary and epidemiological processes take place on a similar timescale (see chapter on virus evolution). according to this framework, phylodynamic processes that determine the degree of viral diversity are a function of host immune selective pressures and epidemiological patterns of transmission ( figure ). intrahost phylodynamic processes begin with molecular characteristics of the virus as well as the host's permissiveness and response to infection. for example, a single amino acid substitution in epstein-barr virus was shown to disrupt antigen presentation by specific human leukocyte antigen polymorphisms (liu, ) . this resulted in decreased t-cell receptor recognition and successful viral immune escape. the virus must also induce an "optimal" host immune response to maximize transmission to new hosts. if the virus induces a strong, proinflammatory immune response not balanced by the appropriate anti-inflammatory responses, the host may succumb to the overabundance of inflammation and cannot propagate viral transmission. alternatively, a virus that fails to stimulate an immune response may also replicate uninhibited, overwhelming, and killing the host prior to transmission. selective pressures maximize replication while sustaining transmission between hosts. interhost dynamics are affected by several factors including evolutionary pressures, timescales of infection, viral latent periods, and host population structures. typically, only a small number of virions are transmitted between hosts, creating a genetic bottleneck that limits viral diversity. a virus that mutates to cause highly pathogenic disease but is not transmitted cannot propagate its pathogenicity. cross-immunity between viral strains also precludes the replication of particular viral lineages. influenza vaccine strains require annual changes due to new circulating influenza strains that have escaped immune pressures through high mutation rates and gene re-assortment. the strong selection pressure of cross-immunity is reflected in the short branch lengths in a phylogenetic tree of influenza viruses isolated from infected individuals. thus, the selection of influenza strains for future vaccines is partly determined by cross-immunity to prior circulating strains, because influenza viral strains that circulated in the past may elicit immune protection against currently circulating strains. at the population level, phylodynamic methods have been used to estimate r for hiv and hepatitis c virus, for which reporting and surveillance data are often incomplete (volz, ) . phylodynamic and phylogeographic models also have been useful in reconstructing the spatial spread of viruses to reveal hidden patterns of transmission. for example, epidemiological and molecular studies of influenza virus transmission were compared at different spatial scales to highlight the similarities and differences between these data sources (viboud, ) . the findings were broadly consistent with large-scale studies of interregional or inter-hemispheric spread in temperate regions with multiple viral introductions resulting in epidemics followed by interepidemic periods driven by seasonal bottlenecks. however, at smaller spatial scalessuch as a country or community-epidemiological studies revealed spatially structured diffusion patterns that were not identified in molecular studies. phylogenetic analyses of gag and env genes were used to assess the spatial dynamics of hiv transmission in rural rakai district, uganda, using data from a cohort of , individuals residing in communities (grabowski, ) . of the phylogenetic clusters identified, almost half comprised two individuals sharing a household. among phylodynamics links evolutionary, immunologic, and epidemiologic processes to explain viral diversity, as shown here for equine influenza virus. for viral evolution, these processes take place on a similar timescale. within host mutations ( ) result from an interplay between optimization of viral shedding, immunologic selective pressures and host pathogenicity. transmission bottlenecks and host heterogeneity ( ) further determine the population genetic structure of the virus, which in turn influences and is determined by the epidemic dynamics. larger scale spatial dynamics at local, regional, and global levels ( ) the remaining clusters, almost three-quarters involved individuals living in different communities, suggesting transmission chains frequently extend beyond local communities in rural uganda. the timescale of infection is also important for viral diversity and transmission dynamics. some viruses are capable of initiating an acute infection that is cleared within days, while other viral infections are chronic and persist for a lifetime. the duration of infection impacts how quickly a virus must be transmitted and has implications for the infectious period and the potential to be transmitted to new hosts. viruses with long latent periods create interhost phylogenetic trees with longer branch lengths. the long duration between infection and transmission permits accumulation of viral changes through many rounds of viral replication before transmission to the next host. examples include hepatitis b virus, hepatitis c virus, and human immunodeficiency virus. availability of computational resources allows widespread use and development of classic approaches to the mathematical modeling of viral transmission dynamics, such as compartmental, metapopulation and network models, to address epidemiologic questions (see chapter on mathematical methods). these models have been used extensively in the study of viral dynamics and to explore the potential impact of control interventions. new sources of high-resolution spatial, temporal, and genetic data create opportunities for models that integrate these data with traditional epidemiological data. such analyses improve estimates of key transmission parameters and understanding of the mechanisms driving virus spread. agent-based models (also known as individual-based models) can now be run using desktop computers, and offer advantages over more traditional mathematical models. because each unit in a population is modeled explicitly in space and time and assigned specific attributes, agent-based models can reproduce the heterogeneity and complexity observed in the real world. more traditional compartmental, differential equation models often require simplifying assumptions that limit applicability. agent-based models have been used to study the spread of viruses in populations as well as the evolution of viruses within and across populations. while agent-based models are intuitive and easy to formulate, these models are often difficult to construct due to the large number of parameters necessary to describe the behavior and interaction between individual units. commercial frameworks that offer large computational power and intuitive user interphases have also become increasingly available. the global epidemic and mobility model (gleam) on the gleamviz platform (www.gleamviz.com), for example, contains extensive data on populations and human mobility, and allows stochastic simulation of the global spread of infectious diseases using user-defined transmission models. our understanding of the epidemiology of viral infections is being revolutionized by the integration of traditional epidemiological information with novel sources of data. l data streams from the internet are promising sources to enhance traditional surveillance but have yet to be fully validated. l molecular data on viral genomic sequences provide unprecedented opportunities to characterize viral transmission pathways. l phylodynamic and phylogeographic models have been used to estimate r , and characterize the spatial spread of viruses. l network analysis reveals hidden patterns of transmission between population subgroups that are not easy to capture with traditional epidemiological methods. l novel analytical and computational resources are playing a key role in integrating information from multiple large data banks. these more comprehensive methods improve our ability to estimate the impact of infection control measures. the combination of traditional and evolving methodologies is closing the gap between epidemiological studies and viral pathogenesis. these developments have laid the foundation for exciting future research that will complement other approaches to the pathogenesis of viral diseases. with these evolving technologies in mind, it is timely to ask: is the world able to control viral diseases more effectively? it is a mixed score card. on the one hand, smallpox has been eradicated and we are on the verge of elimination of wild polioviruses. furthermore, deaths of children under the age of years (which are mainly due to viral and other infectious diseases) have decreased by almost % in the last few decades. on the other hand, the aids pandemic continues to rage in low-income countries, with only a slight reduction in the annual incidence of new infections. the united states has not done any better in reducing hiv incidence which has been unchanged for at least years. the - ebola pandemic in west africa reflects the limited capacity for dealing with new and emerging viral diseases on a global basis. in conclusion, epidemiological science continues to advance with evolving new technologies, but their application to public health remains a future challenge and opportunity. new technologies for reporting real-time emergent infections google trends: a web-based tool for real-time surveillance of disease outbreaks unifying the epidemiological and evolutionary dynamics of pathogens studying the global distribution of infectious diseases using gis and rs spatial analysis of west nile virus: rapid risk assessment of an introduced vector-borne zoonosis pandemics in the age of twitter: content analysis of tweets during the h n outbreak travelling waves in the occurrence of dengue haemorrhagic fever in thailand detecting influenza epidemics using search engine query data rakai health sciences program. the role of viral introductions in sustaining community-based hiv epidemics in rural uganda: evidence from spatial clustering, phylogenetics, and egocentric transmission models middle east respiratory syndrome coronavirus in dromedary camels: an outbreak investigation the parable of google flu: traps in big data analysis wikipedia usage estimates prevalence of influenza-like illness in the united states in near real-time a molecular basis for the interplay between t cells, viral mutants, and human leukocyte antigen micropolymorphism assessing and maximizing the acceptability of global positioning system device use for studying the role of human movement in dengue virus transmission in iquitos predictive mapping of human risk for west nile virus (wnv) based on environmental and socioeconomic factors a high-resolution human contact network for infectious disease transmission revealing the microscale spatial signature of dengue transmission and immunity in an urban population the evolution of hiv- and the origin of where diseases and networks collide: lessons to be learnt from a study of the foot-and-mouth disease epidemic the use of twitter to track levels of disease activity and public concern in the u.s. during the influenza a h n pandemic air travel and vectorborne disease movement usefulness of commercially available gps data-loggers for tracking human movement and exposure to dengue virus using gps technology to quantify human mobility, dynamic contacts and infectious disease dynamics in a resource-poor urban environment contrasting the epidemiological and evolutionary dynamics of influenza spatial transmission viral phylodynamics measles metapopulation dynamics: a gravity model for epidemiological coupling and dynamics key: cord- -fcinsz z authors: kumar, krishna; kumar, narendra; shah, rachna title: role of iot to avoid spreading of covid- date: - - journal: international journal of intelligent networks doi: . /j.ijin. . . sha: doc_id: cord_uid: fcinsz z abstract covid- has become pandemic, spreading all over the world. scientists and engineers are working day and night to develop a vaccine, to evolve more testing facilities, and to enhance monitoring systems. mobile and web-based applications, based on questionnaires, have already been developed to monitor the health of individuals. internet of things (iot) can be used to avoid the spreading of covid- . internet of things is an interconnection of physical devices and the internet. devices are not only sensel and record, but can also monitor and respond. in this paper, we have reviewed the literature available on covid- , monitoring techniques, and suggested an iot based architecture, which can be used to minimize the spreading of covid- . an increase in the use of mobile technology and smart devices in the healthcare sector results in a significant impact on the world. potential development of new smart and powerful devices for monitoring of individuals' health, health experts are taking advantage of these technologies, thus a substantial improvement in healthcare in clinical settings and out of them. iot allows integrating physical devices capable of connecting to the internet and provides real-time health status of the patients to doctors. chronic diseases such as diabetes, heart, blood pressor are remarkable in the world economic and social level problems. it can also provide a platform that allows public health agencies to access the data for monitoring covid- pandemic. fig. shows the present trends of new cases of the top ten counties. new cases in the usa are increasing rapidly than in other countries. china has controlled the covid- , hence the rate of increase in new cases in china is the lowest. coronavirus is transmitted to humans, birds, camels, pigs, rats, bats, and cats. bat origin coronavirus hku was responsible for acute diarrhea syndrome in pigs in . in november , a novel beta coronavirus resulted in nearly human infections and deaths in countries. in , saudi arabia identified the middle east respiratory syndrome (mers) coronavirus (mers-cov), which was the seventh member of the coronavirus family. covid- is closely connected to other bat-origin beta coronaviruses. zhu et al. [ ] covid- 's first case was identified, and a cluster of unknown patients with betacoronavirus pneumonia linked to the seafood wholesale market in wuhan, china. a novel cov ( -ncov) was detected in hospitalized patients in wuhan, china, from december until january . evidence for this virus's discovery includes the identification of whole-, direct pcr, and bronchoalveolar fluid culture in three patients. phylogenetic studies indicated that -ncov falls into the genus betacoronavirus, which provides for coronaviruses found in humans, bats, and other wildlife (sars-, sars-cov bat, and others). guan et al. [ ] analyzed the clinical characteristics of coronavirus and extracted data from hospitals in provinces, autonomous regions, and municipalities in china from laboratory-confirmed covid- patients through january , . the patients 'median age was years; the patients were . % female. the primary composite outcome occurred in patients ( . %), including . % who were admitted to icu, . % who were subjected to intrusive mechanical ventilation, and . % who died. joseph et al. [ ] analyzed social media data based on three approaches: content, descriptive, and network analysis. findings show that it can be used to extract the information of individuals likings and dislikings. misra et al. [ ] presented a review of iot and depicted critical challenges in the same field. the article offers an outline of the iot concept and its related technologies, application, and future scope of research of the area. gil et al. [ ] reviewed the current iot technologies, approaches, and models. g omeza et al. [ ] developed an architecture based on an ontology for monitoring the health and workout and provides recommendations to patients with chronic diseases. the model developed for the system proved to be efficient when making inferences related to the context. li et al. [ ] developed an iot based system (ncapp) to diagnose covid- earlier. according to existing data, questionnaires, and check results, the diagnosis is automatically generated as confirmed, suspected, or suspicious. zaheer et al. [ ] highlighted the need for standardization of protocols for smart city communication. noah et al. [ ] in the medical and healthcare field and highlighted the potential. mohammed et al. [ ] developed a smart helmet-mounted with thermal imaging systems for identifying the infected among the crowd. it also equipped the facial recognition system. fig. shows the workflow of the smart helmet. scanning of the crowd using infrared camera and if the high temperature of any person detected, then it will capture the face using an optical camera. it also provides the location of the infected person through gps. a coronavirus is a large group of a virus, they consist of genetic material and envelope with protein spikes, and these spikes are known as a crown (shown in fig. ). there are different types of corona virus-like respiratory, gastrointestinal, etc. the respiratory disease ranges from common cold and pneumonia, and most of the cases, people generally have mild disease. these include the sars-cov coronavirus. however, coronavirus sarsc-cov was first identified in china in the year . mers-cov coronavirus was identified in saudi arabia in . n-cov first identified in china in ; this type of virus comes from animals and sometimes transmitted from animal to human, called spillover. the number of symptoms ranges from mild to severe fever, cough, and shortness of breath. the infection can be a diagnosis by a test called pcr (polymerase chain reaction). this test identified by the genetic fingerprint. at present, no specific medication, supportive care, and vaccines are available. we can only prevent the transmission of the virus. humans should avoid close contact with those who are sick, use alcohol-based hand sanitizer. the human should avoid optical camera face recognition of patient ip camera for virtual conference/meeting/news broadcasting unnecessary contact with animals. assure that animal products appropriately cooked before consuming. iot is an application-specific, low power, effective, and easy to use a solution of any real-time problems. sensors are the input providers from the physical world, which transferred over a network, and actuators allow things to act or react according to the input received from sensors. a list of iot sensors and their application areas are shown in table . a proposed iot architecture to avoid the spreading of covid- is shown in fig. . data communication is through a gateway device that will further transferred to the cloud gateway. in the big data warehouse, filtering of data, i.e., meaning full data, is extracted. a big data warehouse only contains structured data. machine learning is used to create models of the system based on requirements and received data. data analysis can be used for visualization of results, performance comparison. ir sensors can be used in public toilets for the automatic operation of doors and water supply. infrared thermometers can be used to check the body temperature to identify the infected among crowd and face recognition by using optical camera at the entry points of gates of airports, railway stations, bus stand, malls, etc. similarly, sensors, as proposed in the architecture, can be installed to monitor the body temperature, automatic door operation, water supply control at public places and toilets, online conference to avoid direct contact with the physical world and humans interact. ai and deep learning can help to understand healthcare trends, model risk associations, and predict outcomes. for small applications or for individuals, the configuration of one temperature sensor, one nodemcu, or arduino board with sensors and the internet can be used. a mobile app can be developed using app inventor, which is an open-source platform provided by mit. thingspeak, which is an open-source web service api of matlab to store and retrieve data from things using the http and mqtt protocol over the internet or via a local area network, can be used. figs. and show the sequence of actuator activation and reading data from the sensor. identifying an infected person in-crowd is very difficult. isolating people from the infected is the only solution to avoid the spreading of this virus. the use of iot with smart sensors to measure and record the body temperature of individuals will help to identify the infected. it will also help to maintain social distance. iot based health care systems connected through cloud computing and using data analysis to make an effective decision based on real-time data can be used. johns hopkins university's center for systems science and engineering website a novel coronavirus from patients with pneumonia in china review of discussions on internet of things ( iot ): insights from review on internet of things ( iot ): making the world smart internet of things : a review of surveys based on context aware patient monitoring system based on internet of things, procedia -procedia comput sci clinical ehealth chinese experts ' consensus on the internet of things-aided diagnosis and treatment of coronavirus disease (covid- ) on the coronavirus ( covid- ) outbreak and the smart city network : universal data sharing standards coupled with artificial intelligence (ai) to benefit urban health monitoring and management covid- ) epidemics, the newest and biggest global health threats: what lessons have we learned? an internet of things-based smart homes and healthcare monitoring and management system : review novel covid- detection and diagnosis system using iot based key: cord- -zjwlmzj authors: liu, shinan; schmitt, paul; bronzino, francesco; feamster, nick title: characterizing service provider response to the covid- pandemic in the united states date: - - journal: nan doi: nan sha: doc_id: cord_uid: zjwlmzj the covid- pandemic has resulted in dramatic changes to the daily habits of billions of people. users increasingly have to rely on home broadband internet access for work, education, and other activities. these changes have resulted in corresponding changes to internet traffic patterns. this paper aims to characterize the effects of these changes with respect to internet service providers in the united states. we study three questions: ( )how did traffic demands change in the united states as a result of the covid- pandemic?; ( )what effects have these changes had on internet performance?; ( )how did service providers respond to these changes? we study these questions using data from a diverse collection of sources. our analysis of interconnection data for two large isps in the united states shows a - % increase in peak traffic rates in the first quarter of . in particular, we observe traffic downstream peak volumes for a major isp increase of - % while upstream peaks increased by more than %. further, we observe significant variation in performance across isps in conjunction with the traffic volume shifts, with evident latency increases after stay-at-home orders were issued, followed by a stabilization of traffic after april. finally, we observe that in response to changes in usage, isps have aggressively augmented capacity at interconnects, at more than twice the rate of normal capacity augmentation. similarly, video conferencing applications have increased their network footprint, more than doubling their advertised ip address space. the covid- pandemic has resulted in dramatic shifts in the behavioral patterns of billions of people. these shifts have resulted in corresponding changes in how people use the internet. notably, people are increasingly reliant on home broadband internet access for work, education, and other activities. the changes in usage patterns have resulted in corresponding changes in network traffic demands observed by internet service providers. many reports have noted some of the effects of these changes from service provider networks [ , ] , application providers [ , ] , and internet exchange points [ ] . generally, previous findings and conventional wisdom suggest that while overall traffic demands increased, the internet responded well in response to these changing demands. previous work has shed light on the nature of the resulting changes in traffic patterns. in europe, internet exchange points saw a - % increase in overall traffic volumes [ ] , in some cases resulting in peaks in round trip latency in some countries (e.g., italy) that were approximately % higher than normal [ ] . because users were less mobile, downlink traffic volume decreased by up to % for cellular networks in the uk [ ] . while some of the characteristics of shifting traffic demands are known, and certain aspects of the internet's resilience in the face of the traffic shifts are undoubtedly a result of robust design of the network and protocols, some aspects of the internet's resilience are a direct result of providers' swift responses to these changing traffic patterns. this paper explores these traffic effects from a longitudinal perspective-exploring traffic characteristics during the first half of to previous years-and also explores how service providers responded to the changes in traffic patterns. service providers and regulatory agencies implemented various responses to the traffic shifts resulting from covid- . at&t and comcast have made public announcements about capacity increases in response to increases in network load [ , ] . the federal communications commission (fcc) also announced the "keep americans connected" initiative to grant providers (such as at&t, sprint, t-mobile, u.s. cellular, verizon, and others) additional spectrum to support increased broadband usage [ ] . web conferencing applications zoom and webex were also granted temporary relief from regulatory actions [ ] . these public documents provide some perspectives on responses, but to date, there are few independent reports and studies of provider responses. this paper provides an initial view into how some providers responded in the united states. we study the effects of the shifts in internet traffic resulting from the covid- pandemic response on internet infrastructure. we study three questions: • how did traffic patterns change as a result of covid- ? traffic volumes and network utilization are changing as a reaction to changes in user behaviors. it is critical to measure the exact alterations in a long time span. • what were the resulting effects on performance? considering an expected surge around the dates when states issued stayat-home orders or declared states of emergency, we seek to observe possible changes in the latency and throughput of network traffic across locations. further, different isps also have different capacity and provisioning strategies, which provides us a finer granularity based on these differences. • how did isps and service providers respond? finally, to deal with the usage boosts and performance degradations during the covid- response, operations and reactions of isps and service providers were taken which may explain the changes in network performance. the answer to this question informs us of the networks robustness and their effective disaster provisioning strategies. these questions have become increasingly critical during the covid- pandemic, as large fractions of the population have come to depend on reliable internet access that performs well for a variety of applications, from video conferencing to remote learning and healthcare. to answer these questions, we study a diverse collection of datasets about network traffic load, through granular measurements, proprietary data sharing agreements, and user experiences, as well as extensive baseline data spanning over two years. summary of findings. first, we study the traffic pattern changes in the united states ( § ) and find that, similar to the changes previously explored for european networks, our analysis reveals a - % increase in peak traffic volumes. in the comcast network in particular, we find that downstream peak traffic volume increased - %, while upstream peak traffic volume increases by more than %. certain interconnect peers exhibit significant changes in the magnitude of traffic during the lockdown. second, we observe a temporary, statistically significant increase in latency lasting approximately two months ( § ). we observe a temporary increase of about % in average latency around the time that stay-at-home orders were issued. typical latency values returned to normal a few months after these orders were put in place. we also find heterogeneity between different isps. finally, we explore how service providers responded to this increase in traffic demands by adding capacity ( § ). isps aggressively added capacity at interconnects, more than x the usual rates. on a similar note, application service providers (e.g., video conferencing apps) increased the advertised ip address space by . - x to cope with the corresponding - x increase in traffic demand. the pandemic response has modified people's habits, causing them to rely heavily on the internet for remote work, e-learning, video streaming, etc. in this section, we present some previous efforts in measuring the effects of covid- and past disaster responses on networks and applications. network measurements during covid- . previous work has largely focused on aggregate traffic statistics surrounding the initial covid- lockdowns. traffic surged about % in europe for broadband networks [ ] . in the united states, a blog post [ ] reveals that the national downstream peak traffic has recently stabilized, but in the early weeks of the pandemic, it showed a growth of . %. for wireless networks in the us, volume increases of up to . % for voice and . % for data by the top four providers were shown in an industry report [ ] . mobile networks in the uk reported roughly % drops in downlink data traffic volume [ ] . industry operators have self-reported on their network responses largely through blog posts [ , , , ] . for traffic performance changes, different patterns appear in different regions. facebook shows that less-developed regions exhibited larger performance degradations through their analysis of edge networks [ ] . network latencies were approximately % higher during the lockdown in italy [ ] . according to an ncta report, networks in the united states saw less congestion [ ] . due to decreased user mobility, cellular network patterns have shifted [ ] : the authors found a decrease in the average user throughput as well as decreased handoffs. feldmann et al. [ ] observed that the fixed-line internet infrastructure was able to sustain the - % increase in traffic that happened rapidly during a short window of one week. our work differs from and builds on these previous studies in several ways: first, this study extends over a longer time frame, and it also uses longitudinal data to compare traffic patterns during the past six months to traffic patterns in previous years. due to the nascent and evolving nature of covid- and corresponding isp responses, previous studies have been limited to relatively short time frames, and have mainly focused on europe. second, this work explores the isp response to the shifting demands and traffic patterns; to our knowledge, this work is the first to begin to explore isp and service provider responses. application measurements during covid- . previous work has also studied application usage and performance, such as increases in web conferencing traffic, vpn, gaming, and messaging [ ] . favale et al. studied ingress and egress traffic from the perspective of a university network and found that the internet proved capable of coping with the sudden spike in demand in italy [ ] . another paper used network traffic to determine campus occupancy at the effect of covid- related policies on three campus populations across singapore and the united states [ ] . the cybercrime market was also statistically modeled during the covid- era to characterize its economic and social changes [ ] . network measurements of other disasters. while covid- responses are ongoing and evolving, making measurement efforts incomplete, network responses under other disastrous events can be informative. in , the japan earthquake of magnitude . caused circuit failures and subsequent repairs within a major isp. nationwide, traffic fell by roughly % immediately after the earthquake. however, surprisingly little disruption was observed from outside [ ] . in , hurricane sandy hit the eastern seaboard of the united states and caused regional outages and variances over the network [ ] . for human-caused disasters such as the september th attacks, routing, and protocol data were analyzed to demonstrate the resilience of the internet under stress. their findings showed that although unexpected blackouts did happen, they only had a local effect [ ] . oppressive regimes have also caused internet outages, such as a complete internet shutdown due to censorship actions during the egypt and libya revolts [ ] , where packet drops and bgp route withdrawals were triggered intentionally. although there have been several preliminary measurements of the effects of the covid- response, none have holistically studied traffic data, performance analysis, routing data, and isp capacity information together, as we do in this paper. it is crucial to collect and correlate such information to better understand the nature of both traffic demands, the effects of these changes on performance, and the corresponding responses. this paper does so, illuminating the collaborative view of responses of service providers in the united states. we leverage multiple network traffic datasets to facilitate our study: traffic demands and interconnect capacity: internet connection measurement project. we leverage a dataset that includes network interconnection statistics for links between anonymized access isps and their neighboring partner networks in the united states. the dataset consists of ipfix-based statistics over five-minute intervals, including: timestamp, region (as access isps may connect to a partner network in multiple geographic regions), anonymized partner network, access isp, ingress bytes, egress bytes, and link capacity. the dataset contains roughly % of links from all participating isps. all of the links represented in the dataset are private (i.e., they do not involve public ixp switch fabrics). each five-minute interval includes the sum of the utilization of traffic flows that were active during that interval. we also calculate secondary statistics from the dataset, including: timestamp for the peak ingress and egress hour for each day on each link in terms of usage, ingress/egress peak hour bytes, and daily th and th percentile usage. performance data: federal communications commission measuring broadband america (mba). we analyze the fcc's ongoing nationwide performance measurement of broadband service in the united states [ ] . the raw data is collected from a collection of distributed measurement devices (named whiteboxes) placed in volunteer's homes and operated by samknows. measurements are conducted on an hourly basis. the dataset includes raw measurements of several performance metrics, such as timestamp, unit id, target server, round trip time, traffic volume, etc. each whitebox also includes information pertaining to its isp, technology, and state where it is located. we also define dates related to the status of the pandemic response (e.g., stay-at-home orders, state of emergency declaration, etc.). based on these, we can compute more statistics for specified groups (e.g., break into isps): average and standard deviation among whiteboxes, daily th and th percentile latency/throughput. to keep the network capacity consistent and to record eventual changes solely based on utilization factors, we pre-process the mba dataset with several filters. first, we filter the non-continuous data within the dates of interest (dec. st, to june, th , and the previous year) to capture successive shifts. then, we eliminate the whiteboxes which do not aggregate a statistically significant amount of data, such as some states, isps, and technologies with limited data (e.g., satellite). finally, we choose the measurements from whiteboxes to the top most targeted servers across the united states to represent the overall us performance. these servers are sparsely located in major cities of the us and they have the most whiteboxes (over for each isp) connecting with them. ip prefix advertisements: routeviews. to gain insight into changes in ip address space, we parse internet-wide bgp information globally from several locations and backbones via routeviews. raw ribs (routing information bases) files were obtained from route-views [ ] data on a weekly basis. the average of each tuesday is computed to represent that week. the ribs are then parsed to obtain ipv prefix-to-autonomous system (as) relationships, including mappings of ip prefix, prefix length, paths of as numbers. in section . , we compute the total advertised ipv spaces for as numbers associated with two popular video conferencing applications: zoom and cisco webex [ ] . because most previous studies [ , , ] focus on europe, we begin our explorations by validating whether similar traffic changes are observed in the united states. we consider peak hour link utilization from the interconnect measurement project as a measure of traffic demand. we pre-process the interconnect dataset and remove anomalous data points that are caused by failures in the measurement system. in particular, we do not analyze dates that are greater than two standard deviations outside of a -day rolling mean for each link. figure shows both the absolute utilization and the utilization normalized against the link capacity for two anonymized isps. for each isp, we plot the value corresponding to the th percentile link utilization for a given day. we observe from figure a that isp a saw a dramatic increase in raw utilization at roughly the same time as the initial covid- lockdowns (early march ), with values tapering off slightly over the summer of . isp b, on the other hand, saw a smaller raw increase in utilization for its th percentile links. to better understand whether isp b's smaller increase is a byproduct of different operating behaviors, we explore possible trends in the normalized data (figure b) . here we see that both isps experienced significant increases in utilization in march and april . we also investigated how traffic patterns changed between isp a and each of its peers, in both the upstream and downstream directions. for this analysis, we focused on the dates around the utilization peaks shown in figure . we compared the peak hour download and upload rates on all of isp a's interconnects on ( ) january , , and ( ) april , ( figure ). in general, we see that traffic patterns to peers do not vary greatly between the two dates. we do see, however, that traffic volumes to (and from) some peers change significantly-some by several orders of magnitude. the identities of the peers are anonymous in the dataset, but some patterns are nonetheless clear: for example, some peers show an increase of upstream utilization by two or three orders of magnitude. such drastic changes may be attributable to users working from home and connecting to services that would cause more traffic to traverse the peer link in the upstream direction. we confirmed these results with the operators at isp a and report that they observed that streaming video traffic decreased from to % of the total traffic, but video conferencing increased from % to % as a percentage of overall traffic. the surge in interconnect utilization poses a challenge for service providers, as high utilization of interconnects can potentially introduce high delays for interactive traffic, packet loss, or both. these effects can ultimately be observed through changes in latency (and, potentially, short-term throughput). to examine whether we can observe these effects, we look into the latency and throughput reported by the measuring broadband america (mba) dataset [ ] . we explore these effects over the course of several years to understand whether (and how) performance anomalies that we observe during covid- lockdown differ significantly from performance anomalies observed during other time periods. to better understand how performance changed as a result of changing traffic patterns during the covid- lockdown in the united states, we explored how latency evolved over the course of . to establish a basis for comparison, we show the time period from late through mid- . the appendix also contains a similar analysis for the - time period. we compute the average latency per-whitebox per-day, and subsequently explore distributions across whiteboxes for each isp. (as discussed in section , we consider only whiteboxes in fixed-line isps for which there are an adequate number of whiteboxes and samples.) we use march th , the average declaration of emergency date [ ] , to mark the beginning of the covid- pandmic phase (red shaded for figures). note that this is also the launch date of call of duty warzone peer peak download rate on jan figure shows a seven-day moving average of average round-trip latencies between all whiteboxes in this study. we observe an increase in average round-trip latency by as much as %, this increase in mean latency is significant, corresponding to x standard deviation among all whiteboxes. at the end of april, latencies return to early levels. it is worth noting that, although this increase in average latency is both sizable and significant, similar deviations and increases in latency have been observed before (see the appendix for comparable data from - ). thus, although some performance effects are visible during the covid- lockdown, the event and its effect on network performance are not significantly different from other performance aberrations. part of the reason for this, we believe, may be the providers' rapid response to adding capacity during the first quarter of , which we explore in more detail in section . in addition to the overall changes in performance, we also explored per-isp latency and throughput effects before and during the covid- lockdown period. figures and show these effects, showing (respectively) the th and th percentiles of average round-trip latency across the whiteboxes. these results show that, overall th percentile latency across most isps remained stable; th percentile latency, on the other hand, did show some deviations from normal levels during lockdown for certain isps. notably, however, in many cases the same isps experienced deviations in latency during other periods of time, as well (e.g., during the december holidays). high latencies can sometimes be reflected in achieved throughput, given the inverse relationship between tcp throughput and roundtrip latency. to explore whether latency aberrations ultimately result in throughput effects, as well as how those effects manifest at different times of day, we explored the distribution of latencies before covid- emergency declarations (ed), after the ed but before the stay-at-home order (so). our hypothesis was that we might see higher latencies (and lower throughputs) during "peak hours" of the day from broadband access networks, with the peak hours effectively expanded to the weekday working hours, in accordance with previous descriptions of these effects [ ] . we explored these metrics for a baseline period predating covid- , the time between state declaration of emergency and stay at home ordered [ ] , after stay-at-home declarations were ordered, and two months after stay-at-home ordered. because these dates differed across states, we used known dates for each state [ ] and matched the corresponding dates for each state against the known location of the whiteboxes. figure shows the distribution throughput and latency distributions across all whiteboxes for four time intervals, plotted in four-hour intervals. from figure a , it is clear that the quantiles, median, and maximum latencies all exhibit effects that correlate with these time periods. the period between ed and so corresponds to abrupt routing changes, and the latency data thus reflects a corresponding degradation during this time interval, perhaps at least partially due to the fact that providers cannot immediately respond after the initial emergency declaration (we discuss the timeframes during which capacity was added to the networks in section ). as the transition continues, so appears to be a point in time where latency stabilizes. figure b shows that distributions of throughput measurements are more robust, although the upper end of the distribution is clearly affected, with maximum achieved throughputs lower. the median and minimum have negligible changes during time periods in late april suggesting (and corresponding to) aggressive capacity augmentation, which we discuss in more detail in section . how did service providers respond? in this section, we study how service providers responded to the changes in traffic demands. we focus on the capacity changes during lockdown by inspecting two data sources: ( ) to understand how isps responded by adding capacity to interconnects, we study the interconnect capacity of two large isps in the united states; and ( ) to understand how video service providers expanded their network footprints in response to increasing demand, we analyze ipv address space from two major video conference providers-webex and zoom-and find that both providers substantially increased advertised ip address space. we begin by exploring how isps responded to changing traffic demands by adding network capacity at interconnect links. to do so, we use the interconnect measurement project dataset. we calculate the total interconnect capacity for each isp by summing the capacities for all of the links associated with the isp. to enable comparison between isps that may have more or less infrastructure overall, we normalize the capacity values for each using min-max normalization. we again filter out date values that are beyond two standard deviations from a rolling -day window mean. to show aggregate infrastructure changes over time, we take all of the data points in each fiscal quarter and perform a least-squares linear regression using scikit learn. this regression yields a slope for each quarter that illustrates the best-fit rate of capacity increases over that quarter. we scale the slope value to show what the increase would be if the pace was maintained for days (i.e., a slope of would result in a doubling of capacity over the course of a year). figure shows the resulting capacity plots. the overall trend shows how these two isps in the united states aggressively added capacity at interconnects-at more than twice the rate at which they were adding capacity over a comparable time period in the previous year. second, both isps significantly added capacity in the first quarter of -at a far greater rate than they were adding capacity in the first quarter of . recall from the usage patterns shown in figure , isp a tends to operate their links at nearly full capacity, in contrast to isp b, where aggregate utilization is well below %. both isps witnessed a jump in usage around the lockdown; the response of aggressively adding capacity appears to have mitigated possible adverse effects of high utilization rates. the increase in capacity was necessary to cope with the increased volume: although network performance and utilization ratios returned to pre-covid- levels, the absolute traffic volumes remain high. to cope with abrupt changes caused by covid- , application service providers also took action to expand their infrastructure. previous work has observed shifted traffic in communication applications (such as video conferencing apps, email, and messaging) after lockdown [ ] . it has been reported informally that many application providers expanded serving infrastructure, changed the routes of certain application traffic flows, and even altered the bitrates of services to cope with increased utilization. while not all of these purported responses are directly observable in public datasets; however, routeviews makes available global routing information, which can provide some hints about routes and infrastructure, and how various characteristics of the internet routing infrastructure change over time. this data can provide some indication of expanding infrastructure, such as the amount of ipv address space that a particular autonomous system (as) is advertising. in the case of video conference providers, where some of the services may be hosted on cloud service providers or where the video service is a part of a larger as that offers other services (e.g., google meet), such a metric is clearly imperfect, but it can offer some indication of response. to understand how service providers announced additional ipv address space, we parsed bgp routing tables from routeviews [ ] . for each route that originates from ases of certain application providers, we aggregate ip prefixes and translate the resulting prefixes into a single count of overall ipv address space. we focus on two popular video conferencing applications, zoom and webex, since they are two of the largest web conference providers in the united states-as also recognized by the fcc in their recent order for regulatory relief [ ] . we track the evolution of the advertised ip address space from the beginning of through october . figure demonstrates how each provider added advertised ipv address space from before the pandemic through october . after the beginning of the covid- pandemic, both zoom and webex rapidly begin to advertise additional ipv address space. table enumerates the absolute values of advertised ip address space: zoom and webex increased advertised ip address space by about x and . x respectively, roughly corresponding to the - x increase in video conferencing traffic. this paper has explored how traffic demands changed as a result of the abrupt daily patterns caused by the covid- lockdown, how these changing traffic patterns affected the performance of isps in the united states, both in aggregate and for specific isps, and how service providers responded to these shifts in demand. we observed a - % increase in peak traffic rates for two major isps in the us corresponding with significant increases in latency in early weeks of lockdown, followed by a return to pre-lockdown levels, corresponding with aggressive capacity augmentation at isp interconnects and the addition of ipv address space from video conferencing providers. although this paper presented the first known study of interconnect utilization and service provider responses to changes in patterns resulting from the covid- pandemic, this study still offers a somewhat limited viewpoint into these effects and characteristics. future work could potentially confirm or extend these findings by exploring these trends for other isps, over the continued lockdown period, and for other service providers. this section provides a basis for performance comparison in section . following the same analysis, we choose the exact same time period in the previous year (i.e., late to mid- ) in the united states. we compute the average latency per-whitebox per-day, and subsequently explore distributions across whiteboxes for each isp. longitudinal evolution of aggregate, average round-trip latency. figure shows the aggregate average latency per-whitebox per-day. the previous year has an overall latency of about ms lower than . we observe that the latency keeps stable until the end of april, where a deviation of about ms is shown. the rate of increase is of about %, echoing similar effects around lockdown. longitudinal evolution of per-isp latencies. we further break the aggregate results into the granularity of isps. we report both th and th percentile latencies here. note that in the th percentile plot, we show the groups differently, mainly because of major differences of latency for mediacom and at&t. from figure , we find that the majority of isps performed stably, while mediacom has a large variance in the average rtt. they both have a tail that contributes to what we observed in . figure is grouped the same as figure , which shows that for certain isps, they experience similar deviations during similar periods of different years. we put a supplementary figure referred to in section in this appendix. it shows the distributional changes in latency and throughput on a -hour basis. detailed explanations are in the main text. : changes in latency and throughput before and after the lockdown. ed means "emergency is declared" so means "stay-at-home ordered". a- covid- : our response how the internet reacted to covid- -a perspective from facebook's edge network impact of the covid- pandemic on the internet latency: a large-scale study the japan earthquake: the impact on traffic and routing observed by a local isp covid- network update the wireless industry responds to covid- : network performance analysis of country-wide internet outages caused by censorship campus traffic and e-learning during covid- pandemic keep americans connected measuring broadband raw data releases the lockdown effect: implications of the covid- pandemic on internet traffic wikimedia foundation. accessed october, . u.s. state and local government responses to the covid- pandemic keeping our network infrastructure strong amid covid- a preliminary analysis of network outages during hurricane sandy a characterization of the covid- pandemic impact on a mobile network operator traffic parts of a whole: effect of covid- on us internet traffic covid- : how cable's internet networks are performing: metrics, trends & observations network traffic insights in the time of covid- keeping the internet up and running in times of crisis the internet under crisis conditions: learning from university of oregon route views project turning up the dial: the evolution of a cybercrime market through set-up, stable, and covid- eras prashant shenoy, and rajesh balan. . analyzing the impact of covid- control policies on campus occupancy and mobility via passive wifi sensing key: cord- -l zyubwx authors: cao, hung title: what is the next innovation after the internet of things? date: - - journal: nan doi: nan sha: doc_id: cord_uid: l zyubwx the world had witnessed several generations of the internet. starting with the fixed internet, then the mobile internet, scientists now focus on many types of research related to the"thing"internet (or internet of things). the question is"what is the next internet generation after the thing internet?"this paper envisions about the tactile internet which could be the next internet generation in the near future. the paper will introduce what is the tactile internet, why it could be the next future internet, as well as the impact and its application in the future society. furthermore, some challenges and the requirements are presented to guide further research in this near future field. today's world had experienced several generation of the internet. if the first generation of fixed internet [ ] which had virtually connected the infinite network of computer offered a chance for people collaboration and interaction between each other without regard for geographic location, the second generation of the internet which is the mobile internet [ , ] provided more flexibility and convenience to the user by combining telecommunication with the internet. so people can connect anywhere, anytime with their mobile devices and accelerating the rapid increase in the number of internet users recently. this is also paving a new path to the next generation of the internet where every things and objects can be ubiquitously connected to the internet to create the internet of things (iot) [ ] . one of the question is "what is the future internet after the era of iot?" research scientists now start to discuss about the next vision in the future which is the tactile internet [ , ] . the tactile internet is expected to create a potential market and a plethora of new business opportunities as well as applications that could reshape our life and economy. first coined in [ ] , tactile internet is considered as the communication method over the internet with typical characteristic such as ultra-low latency in combination with high availability, reliability and security in order to mimic the same as human tactile reaction sense on the internet environment. therefore, the data lifecycle must be made within ms from the sensors to the actuators otherwise 'cyber-sickness' may occur [ ] when users feel disoriented in an experience similar to the motion sickness sometimes suffered in the air or on the road. developing new architectures and enabling extremely low-latency endto-end communications to render the tactile internet vision realistic is one of the main motivation. there are some review papers for the tactile internet can be found in [ , , ] . maier et al. ( ) [ ] elaborates some commonalities, differences between the tactile internet, the internet of things and g vision, and survey some recent progress and enabling technologies proposed for the tactile internet. some of the most strictly design challenges as well as specific solutions to enable the tactile internet revolution can be found in [ ] . the purpose of this paper is to introduce about the tactile internet, its impact to our society in the near future as well as its challenges, infrastructure requirement to apply this new technology in real life. to this end, the paper is organized as follows. section outlined some typical foreseen and unforeseen effects of the tactile internet on the society, economic and cultures. next, plurality of application fields are illustrated. section gives some challenges and requirement toward the tactile internet. finally, in section , conclusions and future work are drawn. the potential impact of the tactile internet is expected to bring a new dimension and method to human-tomachine, human-to-human interaction in a plurality of different society aspects including healthcare, education, energy, smart city, and culture.  education: the immediate reaction times in the tactile internet allows haptic overlay of the teacher and learner. as a result, novel learning experiences are driven to a new level of possibilities by the improvements in the training of specific fine-motor skills.  healthcare: by reuniting the medical experts and doctors via the tactile internet during remote diagnosis and treatment, the better healthcare quality could be happen. in addition, experienced surgeons' tactile sense could be combined with the high spatial precision of robot-assisted operations to cure patients in isolated environments.  energy: the tactile internet could be one of the foundation technologies for smart grid to improve energy efficiency and stability in electricity networks. the powerful computing capability and super agility of the tactile internet infrastructure enables dynamic activation and deactivation of local power generation and consumption in the decentralized electrical energy generation and distribution networks. this will potentially result in minimize the generation of unusable reactive power.  smart city: the tactile internet can offer some solutions for the management and operation in the smart city. it can create of a personal spatial safety zone, or 'bubble', to assist people to interact with nearby objects. the traffic flow in the smart city can be optimized thanks to cooperative traffic methods in the tactile internet. guided autonomous driving will enable for a continuous traffic flow in which safety and energy efficiency can be significantly improved in the smart city environment.  culture: by providing new methods in human-tohuman, human-to-machine interaction, the tactile internet can create new behaviors, habits, perceptions not only in many aspect of culture, society, but also in the economy. the tactile internet will enhance the way of communication and lead to more realistic social interaction in various environments. in this section, some main examples are provided to show the ground-breaking potential of the tactile internet. the tactile internet will benefit virtual and augmented reality by providing the low-latency communication where several users are physically coupled via a vr or ar simulation to perform tasks together. for example, a group of artists in different country want to create a sculpture together. currently, they are cooperating by sending the picture, design, images over the internet to simulate a d prototype on the computer. they cannot directly design a real object in the real-life. but the tactile internet maybe the game-changing. with the robot engine, the haptic/tactile sensors, the artist can together create the specific sculpture in real-time interaction over the tactile internet while they are in different geographical locations. tele-presence, tele-diagnosis, tele-surgery and telerehabilitation combining with the robotic technology are just some of the many potential applications of the tactile internet in robotic, healthcare service. the world had witnessed many dangerous diseases such as 'ebola', severe acute respiratory syndrome (sars). people who have these kind of diseases need to be isolated. in the near future, tactile internet can be applied to assist the doctor and physician to diagnose, cure, surgery patients without directly contact to them. the doctor will be able to sit far from the patient's location and command the motion of a tele-robot, then receive not only audio-visual information but also critical haptic feedback. the tactile internet can bring road-traffic safety and efficiency thank to its quickly computation and response characteristic. with the connected cars, the traffic light may be disappeared. vehicles will detect a highly dynamic object such as a pedestrian, obstacle, another car by radar, lidar, sonar or video technology, and disseminate this information to neighboring vehicles within ms latency. so vehicles could slide through intersections without crashing. also, the tactile internet allow a pedestrian enable a personal bubble with his/her cell phone, which makes sure that no car hits the person in vicinity when crossing roads. the improvement in learning experiences for students and teachers can be gained by applying tactile internet. typically, this new technology can be apply in the music instrument teaching class. teacher can stay far or close to the class but they can go over the tactile internet combining with virtual/augmented reality to teach the students hand-by-hand or to show them how to play the music instruments. the tactile internet can be applied in smart grid where the latency is strictly requirements. intelligent monitors can be achieved to optimize consumers' power supply based on information on the status of the power grid over the tactile internet. as a result, associated costs can be reduced. besides, this new technology can be applied in assembly-line at the industrial zone and other applications to manufacture the mass production. to achieve the envisioned tactile internet, there are some major problems and the architecture design requirements need to be solved. the biggest challenge is how to provide ultra-low end-to-end latency of ms along with the highest possible reliability for real-time response of the data lifecycle in this technology. figure . presents a typical latency budget of a data lifecycle from value sensing to decisions making on the tactile internet environment. besides, data security, the availability and dependability of systems that non-violate the low latency requirement due to encryption and authentication delays are the important criteria. these technical requirements of the tactile internet pose tremendous challenges for communications systems. the physical limit of the speed of light causing communication delay needs to be considered as well. for example, the maximum distance for a steering/control server to be placed from the point of tactile interaction by the users is km away (speed of light = km/ms) [ ] . paradigm shifts are necessary to overcome these obstacles. by distributed and decentralized service platform architecture to keep tactile applications local and close to the users, the combination of mobile edge-clouds, mini cloud and a multi-stage hierarchy of cloud platform can be deployed as figure . additionally, communications from sensors to actuators is associated with extremely demanding requirements on both hardware and software. intelligent data compression methods combining with "in-network processing" such as network coding, software defined network [ ] at every node between the source and the destination can speed up the process of decision making and reduce the delay. furthermore, new ideas and concepts to boost the requirement of stringent latency, reliability, and also capacity in both wired and wireless access network. last but not least, edge intelligence to facilitate predictive caching as well as interpolation/extrapolation of human actions is the solution to overcome ms-at-speed-of-lightlimit. this allow spatially decouple the active and reactive end(s) of the tactile internet since the tactile experience is virtually emulated on either ends. as a result, allows a much wider geographic separation between the tactile ends. it is necessary to develop novel artificial intelligence techniques based on predictive actuation for edge cloud architectures. this paper discuss the next vision after the era of iot which is the tactile internet. the big challenges of the tactile internet are extremely low-latency, reliability and high availability of the data life cycle. to achieve this, some infrastructure and technology requirements need to be performed such as decentralized architecture, in network processing, improving access networks, and applying intelligence at the edge. in the near future, the tactile internet is still in its fancy and can bring big impacts on education, healthcare, energy, smart city, culture and many other unforeseen of aspect of society. the internet revolution an empirical study of the use contexts and usability problems in mobile internet what do we know about mobile internet adopters? a cluster analysis internet of things in the g era: enablers, architecture, and business models the tactile internet: applications and challenges the tactile internet-iot, g and cloud on steroids. g radio tech the tactile internet the tactile internet: vision, recent progress, and open challenges realizing the tactile internet: haptic communications over next generation g cellular networks towards the tactile internet: decreasing communication latency with network coding and software defined networking key: cord- -gc hdqis authors: chen, peter john; stilinovic, milica title: new media and youth political engagement date: - - journal: jays doi: . /s - - - sha: doc_id: cord_uid: gc hdqis this article critically examines the role new media can play in the political engagement of young people in australia. moving away from “deficit” descriptions, which assert low levels of political engagement among young people, it argues two major points. first, that there is a well-established model of contemporary political mobilisation that employs both new media and large data analysis that can and have been effectively applied to young people in electoral and non-electoral contexts. second, that new media, and particularly social media, are not democratic by nature. their general use and adoption by young and older people do not necessarily cultivate democratic values. this is primarily due to the type of participation afforded in the emerging “surveillance economy”. the article argues that a focus on scale as drivers of influence, the underlying foundation of their affordances based on algorithms, and the centralised editorial control of these platforms make them highly participative, but unequal sites for political socialisation and practice. thus, recent examples of youth mobilisation, such as seen in recent climate justice movements, should be seen through the lens of cycles of contestation, rather than as technologically determined. at the turn of the century, considerable interest was focused on new internet-based technologies and their potential to stimulate democratic improvements around the world. attention was particularly given to their role in revitalising the public due to its diverse meanings, and its applications subject to endless contestation (spicer ) , democracy is an essentially contested concept. active definitions for democracy lie across a spectrum of performances and values, from the "minimalist" versions that captures simple measures such as voter registration and turnout, to "maximalist" definitions that include activities like associational membership, active information seeking and civic dialogue (dahl ) . the discussion of democracy in this article focuses on a maximalist value of equality over other values, such as maximal individual liberty. given the recent prevenance of "collective crises" like the climate and pandemic, prioritising maximal individual liberty contrasts the necessary collective nature of contemporary, complex societies (wildavsky ). prioritising values like maximal individual liberty can be unsustainable and/or create inequalities that structurally undermine participation by others through self-interest (wildavsky ). in contrast, a maximalist view of democracy equally emphasises the production of democratic culture and institutions that promote just outcomes that sustain democratic practice. this sees generalised civic culture as important in developing practices by citizens that are realised through or by institutions that permit democratic modes of expression and collective action. this is important as recent challenges to individual well-being have collective origins (climate, pandemic, economic inequality). as such, it is complementary to a study of youth participation in the political processes of evolved democracies, such as australia, and the internet-based technologies that afford them access. in recent times, youth participation in democratic processes has been subject to controversy. krinsky ( ) notes that it is unremarkable that young people are often the focus of media and moral panics. therefore, it comes as no surprise that this particular demographic has been implicated as the focal point for three politically focused "crises" within the twenty-first century. the first panic is that young people are the source of "democratic decline". this reductionist view is commonly associated with lower formal participation rates, particularly voting, but also membership in key institutions like political parties (milner ) . in the compulsory voting context of australia, print highlights the focus on young people as a state educational project to become "active and informed citizens" ( ) . which, in the context of today's technologically driven political environment, would garner access to political discourse, engagement, and the use of advancing technologies to communicate, coordinate and mobilise. the second and third crises pertain to the increasing levels of structural inequality and the inability of the post- s neoliberal economic model-with a focus on egoistic individualism, and the resultant social and political acceptance of enduring and reproducing inequality (nozick )-to ameliorate the causes of, resultant social conflict over, the environmental crisis. while these last two are empirical facts, the former is more contestable. to unpack this youth-focused concern, a good example is the often-cited lowy institute annual poll (kassam ) , which, at times, has shown a % gap between australians aged to years, and those over , in response to a question that asks if "democracy is preferable to any other kind of government". these types of findings are often reported in the media in rejectionist terms that overreads the data set and does not interrogate its context. for example, this type of finding has led to sensationalist claims in the media that "fewer than half of australian adults under the age of actually believe in democratic government" (hildebrand ) . this type of coverage commonly is predicated on a discourse that young people are expected to perform a high degree of nativity about the political world, which, when displayed by older people, is attributed to pragmatism and experience. at the core of this is an implicit message that the status quo must be observed as a normative good. thus, young people are at the intersection of multiple fast and slow-moving crises, real or phantasmagorical. yet, with higher levels of concern for issues of social and climate justice (sealey and mckenzie ) , it becomes critical for them to have the capacity to engage in political practices and advance these concerns and question the foundations of political practice that have created or contributed to these social problems. therefore, contestable claims about current and potential democratic capacity have to be explored, particularly in the context of claims about technologies that afford or impede on youth participation. emerging information and communication technologies provide new (or remediate) "affordances", or possibilities for human action. affordances are important because of the way they encourage, allow, discourage and prevent particular behaviours. these can be deliberately or accidentally designed into a technology, be visible, or concealed (livingstone and das ) . when thinking about the application of this concept to politics, this is frequently captured in the "cost" hypothesis: the internet reduces the costs of political participation and allows some "natural" human desire to be afforded in greater abundance (negroponte ) . the positive aspects of the "cost" hypothesis become evident when considering the affordances made possible through advancing technologies. from the late s onwards, the very nature of the internet-as a tool to communicate, aggregate and coordinate-has been associated with its democratising potential. therefore, at first instance, it appears logical to assume that contemporary youth, who have grown alongside these evolving technologies, would employ the internet as a communication tool to engage with political discourse, much in the same way that low-cost printing played an important role in youth politics of the s and s. more recently, with the advent of "platform" technologies (technologies that facilitate a range of applications, rather that provide a narrow set of specific functions) and social media channels, it is noted that young people increasingly use social media to engage in political discourse (yang and dehart ) . nonetheless, affordances not only have the ability to promote an engagement in political discourse, the design of certain technologies can also hinder participation. in retrospect, claims that assert the democratising potential of the internet have been predicated on loose understandings of the underlying character of the technology under discussion, that is, an exaggeration of its "true" network characteristics. the internet is not a "mesh" where each node has equal power relative to its peers, but a "powerscape" which virtually mirrors the hierarchical nature of power in the physical space where certain agendas, people and even locations are prioritised. equally, contestable are claims about the impacts of the technology such as deterritorialisation, or the notion that these technologies may separate the individual from the physical context as a primary definer of their social, economic and cultural needs (chen ). importantly, the cost claim, once so important to early arguments about the levelling effect of the internet (a view subject to very early empirical criticism; see small ) , can now be understood as generating compensatory costs: as data abundance increased, scarcity has shifted from the production of content to its consumption, and considerable time (cost) is now spent sorting, filtering and killfiling the vast amount of content generated and pushed at individuals, particularly through online automation (aka "bots"). additionally, free entry to the internet's public spheres is not cost-free for those marginalised subaltern populations-a term coined by postcolonial theorists to describe faction of society excluded from hierarchal structures of power-who also experience exclusion at a personal cost. virtual violence and harassment in online spaces have forcefully attempted to exclude these marginalised groups from the digital public sphere and are well-documented. in this context, youth within established democracies, despite having access to these virtual public spheres, form a part of not only the subaltern identity due to their cultural standing, but also the repression they experience from institutions such as the education system (spivak ) . as such, their participation, much like any other faction of subaltern society, is intensely contested (hartounian ; dhrodia ) . in thinking about social media from the perspective of democratic affordances, it is important to consider the political implications of its underlying technological and intuitional characteristics (howard and parks ) . that is, social media is largely only possible because of its reliance on large database systems that afford horizontal visibility within peer groups. thus, it is unsurprising that social media has been politically useful in the processes of political mobilisation. as evidenced in the work of groups like getup! (an australian-based independent movement for progressive participation in democratic processes), along with others, groups have successfully capitalised off of internet-based technologies to disseminate their message and motivate collective action (vromen ) . equally, xenos et al. ( ) have argued strongly for a positive relationship between young people's time spent on social media and political participation. based on a survey of young people ( - ) in the usa, uk and australia, and drawn from online panels, they argued that social media was positively related to increase political participation and produce a good regression analysis in support of this claim. the deterministic interpretation of this research can be contested, however. this analysis also strongly correlated reported levels of participation with respondents' sense of personal political efficacy. this leaves open the real possibility (as the authors identify) that their observations about technology use and political participation may be an expression of some other unmeasured causal agents, or that tool use is epiphenomenal to the connection between political interest and expression that would occur in any other socio-technical setting. significantly, reflecting our concern about the dominance of individualism, the same volume includes a longitudinal analysis of internet use that concludes that "…facilities on the internet often described as 'social' media offer environments which mainly draw young people's attention away from common concerns" (ekstrom et al. ). thus, the actual relationship remains open for investigation, and youth engagement in political participation on the internet is questionable, opening up the potential to explore how the use of social media and other internet-based technologies could mobilise youths into political engagement. recent attention has particularly been paid to youth mobilisations around climate issues, including the role of young people as leadership figures (i.e. greta thunberg) and peer mobilisation using new media (collin and mccormack ) . these observations are commonly placed into the now-familiar causal narrative of new media as inherently facilitative of collective action. however, until end-to-end case research is conducted, caution needs to be taken in ascribing causation. that is, participants may take a bus to participate in a demonstration. however, the bus itself has little to do with political action, much in the same way that social media might not necessarily be the driver for collective action. more specifically, to argue that social media was the driver behind climate youth protests remains a mostly correlative explanation when dealing with a population so ensconced in a mediated lifeworld, a reality in which all the immediate experiences of an individual are directly impacted and influenced by evolving media technologies. many of these mobilisation case examples are embedded in established social movement industries and, importantly, are not outside the scale of mobilisations seen in pre-internet youthled movements during the cold war. in similar "existential" issues of concern for young people (such as anti-conscription in the s or anti-nuclear movements of the s), mobilisation of youth movements was significant, preinternet. an alternative hypothesis is that we can see this as part of the routine, periodical "cycle of contention" of post-war youth mobilisations in which "good, decent, little people" with an apparent distrust towards establishment rally against the "corrupt and evil forces from above whose policies are responsible for their pain and suffering" (kazin ) . equally, we could argue that established collective action theory might be hierarchically higher than social media-specific theorising in explaining case examples, as it provides a better-substantiated explanation of a greater number of recurrent phenomena. further, the basic premises of the existence of a "deficit" have been challenged. collin ( ) , for example, argues that claims about youth disengagement are exaggerated. she points to volunteering and social movement participation rates as correctives to reliance on "formal" institutional measures of (dis)engagement. while longitudinal data on social media and volunteering in australia is scarce and unreliable (walsh and black ) , internationally, there is evidence that increased volunteering rates pre-date widespread internet adoption and may be associated with motivations like experience-gathering to enhance employability or college entry (jones ) . again, membership and volunteering may now be afforded via online channels, but this does not demonstrate a causal connection between the means and social practice. lastly, flexible definitions of participation serve as a correction of the institutionally oriented "democratic decline" literature by expanding what political participation looks like. they do so by recognising a shift towards informality in terms of participation in the public sphere ("everyday making"; bang and sørensen ) . this draws us to bennett's ( ) analysis of the implications of a social shift towards citizenship as "social movement citizenship" over dutiful/republican models that privilege participation within or through formal institutions. bennett's model emphasises a focus on concern for specific issues as a primary driver for the "hitand-run" participation of everyday making, combined with modes of participation that are more informal and expressive. this not only sits within a post-modern/post-war notion of justice as including recognition as well as other "rightscentric" motivators, but also recognises that the large participation rates in political organisations in the nineteenth and first part of the twentieth century might have less to do with their political functions, as much as their provision of social services, recreational opportunities and networking resources. these benefits are now seen outside of these explicitly political groups with postwar consumer culture, hyperpluralism and social diversification. while participation in the activities of formal political institutions is essential in liberal democracies, a decline in interest in more conventional models of government presents problems in realising political wins or accepting political compromises, the importance of linking these types of rights and recognition concerns with just structural outcomes (fraser ) . overall, social movement citizenship, or everyday making, presents challenges to an outcome-focused democratic analysis due to a tendency towards adhocracy, paradoxical disconnection and rapid demobilisation by political participants following their "hit-and-run" engagement. each is discussed in turn. movement politics tend towards fluid structures which more commonly produce flexible adhocracy. while these non-hierarchal power structures are an established advantage of movements, giving them the flexibility, dynamism and resistance to repression, reliance on adhocracy may not produce democratic socialisation. in the context of virtual collective spaces afforded by internet technologies, adhocracy, through decentralised electronic and online methods of collective action, tends to situate issues and problems in the context of "unique" or unusual issues that may require extra-normal methods to address. these types of organisational forms place politics into states of exception where the framing of the problem as exceptional encourages solutions based on the sovereign's ability to transcend the rule of law in the name of the "public good" (schmitt ) . therefore, they are less, not more, likely to consider democratic norms and suffer from low accountability and less-drawing potential. while these forms of governance can ameliorate the crises, the longer-term governance legitimated by invoking sovereign power is problematic under this model (see wallach's ( ) discussion of advocacy and the global financial crisis). while the lack of lesson-drawing limits the "developmental" value of participation in these forms of governance structure, this critique is emerging in response to state responses to the covid- crisis of . second, while rejecting arguments about social media as fundamentally "siloing" its users (the so-called filter bubbles argument), issue-based politics can disconnect participants from other issue groups and meta-narratives seen either as generally necessary for social functioning or as important canvases against which popular debate is framed. bouvier ( ) equates this to reduced personal ownership of claims made online, due to anonymity, and the collapse of the existence of a shared symbolic order (the "big other"). new media, in undermining the cultural dominance of mass political media, has played an essential part in this process. as a type of networked politics, horizontal visibility can be low. one of the difficulties of the study of younger people's political engagement lies in its comparative "invisibility" within social media that is not readily observable to wider publics. as schuster ( ) observed, this invisibility can create a "generational divide" within movements, with older activists unaware of the depth of engagement of younger activists. this reinforces findings that social networks may not create social capital as anticipated (valenzuela et al. ). indeed, there are concerns that high levels of social media consumption may be alienating (hunt et al. ) . finally, and related the factors of velocity and transience associated with "internet time", rapid mobilisation and "flocking" (where attention shifts towards the next exceptional space or incident that garners high visibility, leaving the previous platform or issue empty) can be associated with rapid demobilisation (jackson and chen ) . as uldam and vestergaard ( ) argue, there is a need to refocus on civic participation beyond movement-based and protest-focused analysis. image is not action, and considerable over-attention to visible movement action raises questions about the extent to which the transition from expressive politics to agenda building to policy design, implementation and monitoring occurs. this saw considerable interest from post-war social movement "pracademics" (academics engaged in instrumental and action research), asking questions about "realising wins" and the problem of follow-through post demobilisation (see, for example, the practical work of moyer et al. , see in the map model for social movements). to understand the relationship between social media and democratic practice, we need to determine what type of practice space social media affords. "practice space" is used deliberately here over the more popular "public sphere" due to the authors' view of the tendency to misapply this concept to the new media environment. more specifically, the attraction of online media theorists to habermasian deliberation may not have been the right choice because this particular democratic model emphasises early parts of the policy process over later aspects of it highlighted above. thus, rather than see the commercial social media platforms as public spheres (true sandboxes), we can see them as sites with non-trivial visible and invisible geographies of power that not only provide political affordances, but also shape social expectation of social media citizenship. the monetisation of online spaces, combined with the collapse of the conventional advertising driven internet economy, has increasingly shifted social media into a primary role in what is called "surveillance capitalism". surveillance capitalism produces value through the observation, quantification and commodification of individuals' online behaviours. this data becomes the core product of these services, providing a new market with the potential to capitalise off knowledge about users' preferences (zuboff ) . this has implications for participation in and through these systems due to the role that surveillance plays in creating self-censorship, and the way preference engines generate sameness in the information consumed by individuals. these tendencies-in stark contrast to the view of information in markets as facilitating fair exchange, or the free-speech ethos maintained by the entrepreneurs who run social media enterprises-are problematic for democratic participation. this is due to the reduced capacity for preference formation (attacking performative aspects of speech practice) and preference realisation (via the selective satisfaction of wants at low cost). from the preference formation perspective, the impact of surveillance is demonstrated in stoycheff's ( ) study of the effect of social media users' awareness of surveillance. this experiment found that-even in "strong" free-speech jurisdictionspriming users' awareness of the possibility of surveillance produced more conservative online behaviour and speech. this impact should be concerning to developmental democrats. dahl ( ) , for example, emphasis on the processes of preference formation as a critical aspect of developmental citizenship, something that continues life long, but is vital in the transition into civic life in youth. preference formation is both an individualised practice, in that it is a developmentally acquired skill that individuals exhibit different levels of capacity in, and a collective capacity under conditions of equality. specifically, discourse within groups develops the capacity of the group to undertake political discourse through observation and the presence of relevant information. from a preference realisation perspective, the existence of these so-called preference-knowing machines has implications for human agency. this observation is made because surveillance capitalism, unlike traditional top-down political paternalism exhibited in democratic and authoritarian societies, is fully compatible with high levels of perceived individual efficacy. where "search" once drove the core economy of the internet and provided efficacy through agency, "sharing" and automagical result systems replace agency with expectation. "sharing" allows users to expect that aspects of their "wants" are provided automatically as these systems offer solutions to personal wants and adjust the platform in line with anticipated user desires. thus, efficacy is obtained in these surveillance regimes. significantly, this is achieved not through the type of agency commonly associated with democratic participation, rather a negative agency, surrendering to the panoptic view in recognition of its capacity to service the individual within very narrow and uncontested spheres. this exchange has a psychic cost. hoffmann et al. ( ) identified "privacy cynicism" as the tendency for users to engage in "a cognitive coping mechanism, allowing users to rationalise taking advantage of online services despite serious privacy concerns." this type of preference servicing/channelling is not something we see in corporate media spaces. through the continued valorisation of corporate modes of production in the political sphere, these types of negative agency, passive efficacy systems are proliferating within institutions (public, private and non-profit), and in the new interest in behavioural and persuasive "nudge" economics evidenced in the public sector. these projects-wrapped in the discourse of "big data" analytics-seek to understand their stakeholder groups, be they citizens, clients or employees, but only to the extent to which that knowledge fulfils institutional objectives. this is most obvious in the authoritarian internet of china, where the hard power has most recently been combined with "soft" incentivising through the "social credit system" (ramadan ) but can also be seen, to a seemingly less intrusive extent, in developed democracies through the aforementioned corporatised nature of e-governance and behavioural policy units. the naturalisation of the technologies underpinning these management systems, be they in liberal democracies or authoritarian regimes, further erodes the capacity of users to express consent (with legitimacy implications important in democratic regimes) and participate in process design, eroding the capacity for the transference of democratic capacity into other areas of life. socialisation within these systems of expectations thus displaces developmental citizenship as citizens are increasingly embedded in these systems of affordance. it is essential to learn from the failures of the "electronic democracy" (e-democracy) movement as a development which, at the turn of the century, aimed to streamline bureaucratic processes and motivate higher levels of participative engagement between power elites and their publics. attempts to create and propagate participative platforms advocated at this time significantly failed through a combination of low utilisation and limited state interest in cultivating and connecting with them. even where public management has embraced notions of popular legitimacy, such as the focus on connecting administration to public legitimacy through the reclassification of public sector management as the creation of "public value", this has not led to an embrace of participative media by public managers (bolívar ) . where participative design is undertaken, understanding the constitutive nature of affordances is essential. as mentioned, design choices activate and allow behaviours. thorson ( ) , for example, has argued that there are important differences in the value of new media associated with how the technologies are employed, with "active" use (i.e. search over sociability) correlated with more challenging modes of citizenship, or seeking divergent opinion exposure. matei and britt ( ) provide a useful analysis of wikipedia as an example of a platform that uses advocacy, creates a social hierarchy in production, but sustains openness to new entrants, accountability, and has sustained itself in the face of attacks on its primary function of knowledge production. as indicated, however, this cannot be seen as an engineering problem, and the cultivation of interest and practices to employ, discover and create new democratic affordances is necessary. the contemporary problem is that most civics education is undertaken in bad faith. as bennett ( ) observed, most academic and educational representation of democratic practice bears little resemblance to its actual practice, precisely because this practice in most established democracies falls far short of the underlying idealistic motivations embedded in civics instruction. as "telling people to participate in bad institutions is mere propaganda" (levine ), there has been a tendency to focus on political spaces that are presented as tabula rasa, be this technological spaces over the last two decades, or an emphasis on social movement participation because ever-renewing social movements maintain the appearance of both democratic because they are inherently participative, and new because they tend to be continuously renamed. coleman ( ) has highlighted the importance of skilling for "autonomous ecitizenship" that capacity building and skilling overcome the problem of bad faith by allowing young people to develop their own aspirational spaces. this has been demonstrated in applied research in schools (see black ) . this model presumes, however, that these skills are produced in a context where their democratic application is explicit, and that these self-actualising young e-citizens would employ these skills in a frictionless environment that provides them with an even playing field. if this illusion was able to be sustained by coleman in , a decade later it cannot be. the corporate-dominated internet has proven libertarians wrong: atomised "heroic" individuals cannot find self-actualisation in the face of such overwhelming institutional power. individual liberalism is impossible in an age of such incredible institutional power. the "hidden affordances" of the sub-systems most internet users spend the majority of their time in online (as opposed to the delusion of the internet as a "sandbox") may be malleable to a literate few, but this presents the prospect of meritocracy heavily tilted by neoliberal access to education, not a sustainable democracy of equality. further, social media may provide greater affordances for reactionary and counter-democratic agents (the entrenched, corporations, the state) to limit movement and political agency (cammaerts ; uldam ) . this latter power is contestable. while online activists have attempted to moderate the policies of major internet platform operators, their capacity to act against organisations of this scale and transnational character has been most effective only where similar transnational capacity exists. thus, while there have been a range of statelevel investigations and proposed regulatory interventions in the way platforms operate (in australia primarily focused on ensuring representative democracy), the most significant developments have been seen at the level of the european union. this remains a moving target but provides an example of the need to revisit the politics of accessibility but focused on interventions within the walled gardens online. the internet is no longer a land without a history, and optimistic projections that new affordances have a deterministic correlation with democratisation need to be checked against the visible history of these claims, and the broader background context of societies that are more technologically saturated, and less democratic overall. while new media is undoubtedly at the heart of recent youth mobilisation politics, it is not clear if this is more than correlative or epiphenomenal. younger people are disproportionately institutionally situated in contexts obsessed by the "human capital" developmental school (becker ) in which citizenship rights are primarily held in abeyance, secondary to instrumental aims associated with participation in the private world's production and consumption. thus, recent youth mobilisations can be seen as remarkable contestations in the face of these systems, rather than a result of them to avoid falling into the twin traps of bad faith promotion of weak systems or magical thinking about new spaces and places for participation, systematic democratisation of spaces and places-both online and off-must be the focus of reform. participation without democratisation is the possible panacea that can consume the efforts of reformers. as the internet splits into three distinct jurisdictions-the surveillance capitalism space, authoritarian spaces and the regulated internet-new possibilities and natural experiments present themselves for investigation in the online space. to take advantage of these affordances civics education has demonstrated, it can produce positive improvements in rates of participation, but this will need to be met by increased opportunities for democratic agency in our classrooms, workplaces and the public sphere. the everyday maker: a new challenge to democratic governance human capital: a theoretical and empirical analysis, with special reference to education the wealth of networks: how social production transforms markets and freedom active citizenship and the 'making' of active citizens in australian schools policymakers' perceptions on the citizen participation and knowledge sharing in public sector delivery. in: sobaci mz (ed) social media and local governments what is a discourse approach to twitter, facebook, youtube and other social media: connecting with other academic fields building citizen-based electronic democracy efforts. paper presented to the internet and politics: the modernization of democracy through the electronic media conference doing it for themselves: management versus autonomy in youth e-citizenship young citizens and political participation in a digital society: addressing the democratic disconnect young people and politics the marketplace, e-government and e-democracy we tracked , abusive tweets sent to women mps-half were directed at diane abbott the networked young citizen: social media political participation and civic engagement the internet, democracy and democratization. democratization recognition without ethics? the electronic republic: reshaping democracy in america. viking why young voters are turning their backs on democracy privacy cynicism: a new approach to the privacy paradox social media and political change: capacity, constraint, and consequence no more fomo: limiting social media decreases loneliness and depression rapid mobilisation of demonstrators in march australia youth volunteering on the rise lowy institute poll : understanding australian attitudes to the world moral panics over contemporary children and youth. routledge, london levine p ( ) macarthur online discussions on civic engagement a companion to new media dynamics structural differentiation in social media adhocracy, entropy, and the " % effect the internet generation: engaged citizens or political dropouts doing democracy: the map model for organizing social movements new york nozick r ( ) anarchy, state, and utopia. blackwell, oxford print m ( ) a "good" citizen and the australian curriculum the gamification of trust: the case of china's 'social credit the virtual community: homesteading on the electronic frontier political theology: four chapters on the concept of sovereignty, g schwab (trans.) invisible feminists? social media and young women's political participation agenda for action: what young australians want from the election. youth action and australian research alliance for children and youth equal access, unequal success-major and minor canadian parties on the net what do we mean by democracy? reflections on an essentially contested concept and its relationship to politics and public administration can the subaltern speak? under surveillance: examining facebook's spiral of silence effects in the wake of nsa internet monitoring the networked young citizen: social media political participation and civic engagement civic engagement and social media: political participation beyond protest is there social capital in a social network site? facebook use and college students' life satisfaction, trust, and participation digital citizenship and political engagement: the challenge from online campaigning and advocacy organisations australian research alliance for children and youth wildavsky a ( ) culture and social theory the great equaliser? patterns of social media use and youth political engagement in three advanced democracies social media use and online political participation among college students during the us election the age of surveillance capitalism: the fight for a human future at the new frontier of power key: cord- -mvgqlufq authors: thorp, andrea w.; brown, lance title: accessibility of internet references in annals of emergency medicine: is it time to require archiving? date: - - journal: annals of emergency medicine doi: . /j.annemergmed. . . sha: doc_id: cord_uid: mvgqlufq study objective we seek to evaluate the accessibility of all internet references appearing in annals of emergency medicine from , , and . secondary objectives are to determine whether the number of internet references is increasing and to describe how internet references are inaccessible. methods we visually scanned all articles for references made in the printed version of annals of emergency medicine for , , and . we identified the internet references and grouped them into categories according to the results of entering the uniform resource locator (url) into the internet browser. results we identified , references published in annals of emergency medicine. the proportion of internet references increased from % of the total references in to . % in . internet references were not readily accessible for of internet references in ( %; % confidence interval [ci] % to %), of internet references in ( %; % ci % to %), and of internet references in ( %; % ci % to %). inaccessibility was most commonly manifested by urls that no longer link to active web sites ( of inaccessible internet references [ %]; % ci % to %) and urls that linked to generic home pages where the authors’ referenced material could not be found ( of inaccessible internet references [ %]; % ci % to %). conclusion in annals of emergency medicine, older internet references appear to be less accessible than newer references. internet reference archiving is one solution to preserving this information for future readers. the desire to preserve important writings for future generations has existed for centuries as people have designed ways to store, catalogue, and retrieve information. [ ] [ ] [ ] [ ] [ ] recently, the internet has become a transformative technologic advance for dispersing information to a widespread audience. because searching, organizing, and sharing information has never been easier, authors of medical and scientific articles have been incorporating internet references to support material in their publications. , approximately % of all references within recently published scientific and medical journals are internet references. , in our own experience retrieving internet references, we found that many of them were no longer active or no longer contained the material originally referenced in the article. using internet references within published articles becomes problematic for the readers of scientific and medical journals if the supporting material for factual statements disappears. this problem has been identified in high-impactfactor journals such as journal of the american medical association, science, and new england journal of medicine. [ ] [ ] [ ] our objective was to evaluate the accessibility of all internet references appearing in annals of emergency medicine for , , and . secondary objectives were to determine whether the number of internet references is increasing and to describe how internet references are inaccessible. all references published in the printed version of annals of emergency medicine from january through december december , december , and were included in our study. we excluded references within advertisements or supplemental publications. we grouped each reference into one of categories: medical/ scientific journals, books and book chapters, government documents, presentations and announcements, newsletters and newspapers, databases and software, pamphlets and package inserts, and internet references. the internet reference category comprised our main study group. throughout a -month period (august to september ), we attempted to access all internet references cited in annals of emergency medicine during our study years. in a post hoc analysis, we classified the internet references into categories (table ) . to better interpret the data, we then organized the categories in main groups. the first group consisted of internet references that were readily accessible and contained information that the author intended to cite. the second group consisted of internet references that failed to contain readily accessible information that the author intended to cite. we entered all the internet references into a spreadsheet and checked the spelling of each uniform resource locator (url) against the original published internet reference at the time of data entry and again if the internet reference was placed in the "no site found" category. the spreadsheet was linked to an internet browser (firefox, version . . . ; mozilla foundation, mountain view, ca), and double-clicking on each url directly from the spreadsheet accessed the urls. these spreadsheets can be found in appendix e , available online at http://www.annemergmed.com. we were concerned that the dynamic nature of the internet would give the "no site found" message or "temporarily unavailable" message erroneously. to avoid misclassification when these messages appeared, we retried each site at least hours after the failed attempt. we frequently encountered a situation in which the url would access a generic home page. the authors' intended information would be related to the content on the home page, but the specific topic cited by the author could not be identified by scrolling through the web site. for example, a url that links to the new york times web site will be active, but the information the authors intended to cite will no longer be on the home page. because the main purpose of the study was to evaluate the accessibility of referenced information on the internet, we thought that classifying these urls as active sites what is already known on this topic with increasing frequency, scientific articles cite internet materials in their reference section. the cited sites may not be stable. what question this study addressed how often are internet sites cited in annals of emergency medicine, and do they remain accessible? what this study adds to our knowledge less than one quarter of web sites cited in articles and less than half of web sites cited in articles were accessible in . unlike references to printed material, references to internet material are unstable and decay over time. in light of this evidence, journals should consider requiring that all internet references be made through webcite or other free, permanent archival services. was inappropriate. in an attempt to locate the authors' intended information, we used the site-specific search engine. phrases that represented the authors' intended information as published were then entered into the data spreadsheet. we used the copy and paste functions to transfer the text directly from the "intended information" column in the spreadsheet to the site-specific search bar on the web site. the first results produced by each search were reviewed, and the internet references were then categorized according to the results of these individual searches. if the authors' intended information was found, the internet reference was classified in the "reference found with the site search" category. if the information could not be found, the internet reference was placed in the "reference not found with the site search" category. the number of internet references in each category was totaled and % confidence intervals (cis) were calculated using the cii, wilson command (stata, version . ; statacorp, college station, tx). because our study contained information available to the public and did not involve human subject intervention, institutional review board review or oversight was not required. we identified , references published in annals of emergency medicine in medicine in , medicine in , and . references made to the medical/scientific journals and magazines composed % of the total references identified during the years. the internet reference category was the third largest group of references and contributed % of the total references ( table ) . the proportion of references that were internet references increased throughout our study years. fifty-one of , total references ( %; % ci . % to . %) in , of , references ( . %; % ci . % to . %) in , and of , ( . %; % ci . % to . %) in were internet references. when comparing the categories of internet references, we found that the number of internet references that no longer accessed the authors' intended information increased over time. forty of the internet references in ( %; % ci % to %), of the internet references in ( %; % ci % to %), and of the internet references in ( %; % ci % to %) no longer allowed access to the authors' intended information. to better describe how the internet references are inaccessible, we compared the categories of internet references that did not readily link to the authors' intended information. during the years, internet references ( % of inaccessible internet references; % ci % to %) could not locate a web site. one hundred fifteen ( % of inaccessible internet references; % ci % to %) of the inaccessible internet references linked to a generic and related web site, but the authors' intended information could not be identified (table ) . of the internet references that gave a "temporarily unavailable" message initially ( % of total internet references; % ci . % to . %), none of those internet references became readily accessible during our study period and, therefore, remained in this category. we did not have a computerized method for identifying references within annals of emergency medicine, so we manually reviewed each issue. it is conceivable that we overlooked or misclassified references. some references were in languages other than english, but none of these references had the format of an internet reference. we are not aware of a validated method for determining whether internet references contain the authors' intended information. difficulty in determining whether the the internet references published in annals of emergency medicine appear to become less readily accessible over time. five years after publication, % of the internet references no longer allowed the reader to readily access the authors' intended information as cited in the published reference section. previous publications have reported that . % of internet references become inaccessible after months, % after year, and % at years. even though previous publications have different definitions of accessibility, the gradual loss of internet information demonstrated in our study agrees with previous reports. of the internet references identified in our study, were not readily accessible. of these inaccessible internet references, % could not link to an active web site, and % would link to a generic but related web site that did not contain the authors' intended information. during our attempts at accessing these internet references, we often could not decide whether we were viewing the material the authors had viewed when writing the article. a central function of references is to provide support for statements put forward by an author. if a reader questions a fact, he or she can go to the supporting reference and review that information. traditionally, if a journal article is cited in a reference section, that journal can be accessed and the reader can be confident that the article he or she is reading was the exact article that the author intended to cite. the reader may then participate in a dialogue with the author by challenging his interpretation of the data presented in the reference. if this dialogue is held in a public forum such as in a letter to the editor, the readers may benefit from a deeper insight into the topic at hand. authors are held to a high degree of accountability for the factual statements they make and reference. if the supporting material in a reference disappears or changes over time, components of a meaningful dialogue and author accountability are damaged. internet references composed approximately % of the total references appearing in recently published scientific and medical journals. , our data demonstrate increased use of internet references in annals of emergency medicine. from to , internet references increased from % to more than % of the total references. because of the convenience of the internet, authors are recognizing the value of the internet as a resource. if the use of internet references continues to increase, the problems associated with inaccessible information will increase also. several solutions to the problem of inaccessible internet references have been proposed. published reports have suggested that the author or publisher keep a digital or hard copy of the information contained within an internet reference. , , , these solutions are not intuitively appealing because authors may become unavailable and publishers may go out of business. now that the technology exists to digitally archive information, a better solution available to authors and publishers is the archiving of internet references. the archiving systems that are available are internet archive (available at http://www.archive.org) and webcite (available at http:// www.webcitation.org). [ ] [ ] [ ] [ ] both of these systems can be used free by authors, readers, journals, and publishers. , in comparing internet archive to webcite, we feel that webcite is the better system. internet archive is limited in that it can store only static information, thus excluding video, representation of databases, and other dynamic sources. , in addition, internet archive acts as a crawler, so internet references that have password restrictions or restrictions to "crawlers" may not be accessible. webcite, on the other hand, takes a "snapshot" of the web site and immediately provides a "permalink" url that the author can publish. the permalink url will immediately take the reader to the snapshot of the web site. the webcite's use of the permalink would ensure that the reader was viewing the exact web site the author had viewed when writing the article. last, webcite has formed partnerships with libraries and preservation consortia that have legally agreed to support the service to ensure long-term availability to authors, readers, and publishers. there is no standardized requirement among scientific or medical journals' instruction to authors for archiving internet references. as of , % of high-impact journals do not provide instructions to authors on archiving digital information. given the increasing inaccessibility of internet references over time, we concur with previous recommendations that the time to require archiving of internet references has arrived. - , , supervising editor: david l. schriger, md, mph author contributions: lb conceived the project idea. awt and lb designed the project, participated in data collection, and analyzed the data. awt drafted the article, and lb contributed substantially to its revision. awt takes responsibility for the paper as a whole. funding and support: by annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article, that might create any potential conflict of interest. see the manuscript submission agreement in this issue for examples of specific conflicts covered by this statement. this project was solely supported by internal departmental funds. online is now reserved for acep members and print subscribers. online is now available for acep members and all print subscribers. to activate your individual online subscription, please visit annals of emergency medicine online by pointing your browser to http://www.annemergmed.com, follow the prompts to activate your online access, and follow the instructions. to activate your account, you will need your acep member number or your subscriber account number, which you can find on your mailing label. if you need further assistance to access the online journal, please contact periodicals services at - - . personal subscriptions to annals of emergency medicine online are for individual use only and may not be transferred. use of annals of emergency medicine online is subject to agreement to the terms and conditions as indicated online. information is also available at acep's home page at www.acep.org. manuscript history. wikipedia: the free encyclopedia saint catherine's monastery: mount sinai. wikipedia: the free encyclopedia wikipedia: the free encyclopedia key: cord- - ey kr b authors: kawchuk, greg; hartvigsen, jan; innes, stan; simpson, j. keith; gushaty, brian title: the use of internet analytics by a canadian provincial chiropractic regulator to monitor, evaluate and remediate misleading claims regarding specific health conditions, pregnancy, and covid- date: - - journal: chiropr man therap doi: . /s - - - sha: doc_id: cord_uid: ey kr b background: internet analytics are increasingly being integrated into public health regulation. one specific application is to monitor compliance of website and social media activity with respect to jurisdictional regulations. these data may then identify breaches of compliance and inform disciplinary actions. our study aimed to evaluate the novel use of internet analytics by a canadian chiropractic regulator to determine their registrants compliance with three regulations related to specific health conditions, pregnancy conditions and most recently, claims of improved immunity during the covid- crisis. methods: a customized internet search tool (market review tool, mrt) was used by the college of chiropractors of british columbia (ccbc), canada to audit registrants websites and social media activity. the audits extracted words whose use within specific contexts is not permitted under ccbc guidelines. the mrt was first used in october of to identify words related to specific health conditions. the mrt was again used in december for words related to pregnancy and most recently in march for words related to covid- . in these three mrt applications, potential cases of word misuse were evaluated by the regulator who then notified the practitioner to comply with existing regulations by a specific date. the mrt was then used on that date to determine compliance. those found to be non-compliant were referred to the regulator’s inquiry committee. we mapped this process and reported the outcomes with permission of the regulator. results: in september , inappropriate mentions of specific health conditions were detected from approximately registrants with failing to comply. the second scan for pregnancy related terms of approximately practitioners revealed inappropriate mentions. following notification, all cases were compliant within the specified timeframe. regarding covid- related words, inappropriate mentions of the word “immune” were detected from registrants with cases of non-compliance. conclusion: internet analytics are an effective way for regulators to monitor internet activity to protect the public from misleading statements. the processes described were effective at bringing about rapid practitioner compliance. given the increasing volume of internet activity by healthcare professionals, internet analytics are an important addition for health care regulators to protect the public they serve. (continued from previous page) conclusion: internet analytics are an effective way for regulators to monitor internet activity to protect the public from misleading statements. the processes described were effective at bringing about rapid practitioner compliance. given the increasing volume of internet activity by healthcare professionals, internet analytics are an important addition for health care regulators to protect the public they serve. keywords: chiropractic, internet analytics, regulation, misinformation background on march , , the world health organization (who) declared the covid- outbreak to be a pandemic [ ] . appropriately, the global health care community responded by emphasising how to combat the disease through frequent hand-washing, sanitizing, and social distancing [ ] . at the present time, who has stated that no vaccine has been identified for covid- and there are no known interventions effective in preventing, or treating, a covid- infection. unfortunately, some clinicians have used the covid- pandemic to promote misinformation whether intentionally or otherwise. for example, cases have been identified where chiropractors have used their internet presence to promote interventions they claim will boost immunity [ , ] . such claims are wholly unsubstantiated as outlined in a recent report from the world federation of chiropractic [ ] ; a point emphasized emphatically by multiple national associations globally [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] . without a doubt, the chiropractic profession recognizes this misinformation as potentially dangerous to the public and damaging to professional credibility [ ] [ ] [ ] . in most countries, health care professions like chiropractic are overseen by a regulatory body whose duties include registration of members, ensuring ongoing competency and public protection. in protecting the public in british columbia, canada, the registrar investigates, adjudicates and forwards any noncompliant activities to its inquiry committee. in identifying unsubstantiated claims and statements promoted by its registrants, regulatory bodies have traditionally relied on public surveillance to monitor and report potential missteps. whereas this traditional approach may have dealt with a relatively low volume of regulatory incursions in the past (e.g. evaluation of printed media only), the current volume at which a profession can market itself electronically requires a parallel improvement in regulation. this is especially the case if the volume of complaints remains low and therefore less easy for the public to identify. thus today, a passive regulatory approach of relying solely on public input may be insufficient given the rapidity that information is spread on the internet. despite electronic communications providing new challenges for regulators, this same technology also provides a potential solution. one such solution is technology that searches specific websites and social media accounts. this technology can be modified to proactively monitor activity then allow regulatory bodies to construct timely interventions [ , ] . while this approach is still evolving in many sectors, proactive monitoring has been used effectively within law enforcement to monitor and detect illegal activity on the internet for decades and continues to be recommended [ ] [ ] [ ] . a similar approach for monitoring healthcare professions could address many of the issues that arise with traditional, time-consuming methods of regulation that rely solely on public reporting [ ] . recently, the college of chiropractors of british columbia (ccbc) became the first chiropractic regulatory body that we are aware of to use internet analytics in their regulation of registrants. specifically, the ccbc commissioned software to identify internet content of their registrants that may contravene their established regulations to facilitate rapid remediation of unacceptable advertising. the ccbc's use of internet analytics also provides researchers with a unique opportunity to better describe the frequency and content of registrant internet activity and to evaluate the effectiveness of the regulatory process in remediating these claimssomething not yet reported in the literature. should active use of internet analytics be effective in professional regulation, other jurisdictions may consider adopting this approach to enhance public safety. in this paper, we report on internet analytics by the ccbc to monitor, evaluate and remediate potentially misleading health claims at three discrete time periods: when internet analytics were first employed by the ccbc in to identify inappropriate claims related to specific health conditions, then again in for specific claims related to pregnancy care and most recently in to identify inappropriate claims during the covid- pandemic (details below). customized software was commissioned by the ccbc to provide internet analytics of their registrants and was delivered to the college in (compliance verification tools, vancouver, british columbia, canada). this software, known as the marketing review tool (mrt), scans websites and social media activity of all chiropractors in british columbia. the software specifically searches for pre-defined words from a target list approved and maintained by the ccbc. using this target list, searches performed by the mrt are made twice per month for websites, daily for social media and on demand as required. following a search, the software returns cases where a word from the target list is identified. the ccbc then manually categorizes these cases into a) acceptable use of the target word (e.g. "our office is closed during the covid crisis") and b) unacceptable use (e.g. "chiropractic can boost immunity). for any case deemed unacceptable, the chiropractor(s) registered with the website or social media account is sent a notice to remediate the questionable material immediately (e.g. deletion of content). a mrt scan can then be performed multiple times to determine the rate at which compliance is achieved. when the stated deadline for mandatory compliance is reached, a follow up scan can then be performed and final compliance determined. any non-compliant activity is then dealt with by the ccbc who forwards these cases to its inquiry committee. the mrt was first used by the ccbc to review the internet activity of their registrants in relation to specific health conditions not permitted to be promoted or treated by registrants. these conditions were made known to registrants on october , with mandatory compliance to occur by november , . the regulation providing the targeted words states: as stated in section ( )(f) of the ccbc's professional conduct handbook ("pch"), chiropractors must not advertise health benefits of their services when there is no acceptable evidence that those benefits can be achieved. see appendix "n" to the handbook and the efficacy claims policy for additional information. the board is concerned registrants may be making claims in marketing or directly to patients that chiropractic care has beneficial effects on some diseases, disorders and conditions when there is no acceptable evidence for those claims. this policy identifies efficacy claims that are not supported by acceptable evidence, and therefore, must not be made. https://www.chirobc.com/efficacyclaims-policy/ due to the absence of acceptable evidence supporting such claims, registrants must not represent to patients or the public that chiropractic: (a) can be used to treat diseases, disorders or conditions such as: alzheimer's disease, cancer, diabetes, infections, infertility, or tourette's syndrome, or (b) has any beneficial effect on childhood diseases, disorders or conditions such as: adhd (or add), autism spectrum disorders including asperger syndrome, cerebral palsy, down syndrome, fetal alcohol syndrome, or developmental and speech disorders. a scan of registrant internet analytics was then performed by the mrt in october, before the compliance deadline. registrants associated with inappropriate messaging were notified before the compliance deadline so they could take corrective action. a subsequent scan was then performed on the compliance date and noncompliant cases were notified and forwarded to the ccbc inquiry committee. on december , , the ccbc released amendments to the professional conduct handbook and the efficacy claims policy regarding pregnancy. the regulation can be found here (https://www.chirobc.com/amendments-to-theprofessional-conduct-handbook-and-efficacy-claims-policywebster-technique-and-pregnancy-related-conditions/) and states: due to the absence of acceptable evidence supporting such claims, registrants must not represent to patients or the public that chiropractic: (a) has any beneficial effect on fetal development or position such as: breech/ breech turning or position and intrauterine/in utero constraint. (b) has any beneficial effect on labour or birth such as: easier or shorter labour, preventing the need for medical interventions and preventing premature or traumatic birth. (c) has any beneficial effect on hormone function or postpartum depression. after the adoption of this policy and subsequent notification of registrants, internet analytics based on selected pregnancy target words were generated from the mrt on december , prior to the compliance deadline of january , . registrants associated with inappropriate messaging were notified before the compliance deadline so they could take corrective action. a subsequent scan was then performed on the compliance date and noncompliant cases were notified and forwarded to the ccbc inquiry committee. on march , , target words were added to the mrt system in response to the covid pandemic with the expected date of compliance set to the same day. adding these words to the target list did not require new regulations be passed by the ccbc as claims related to infectious disease were already disallowed through existing regulations. the pch states: . the prevention and treatment of infectious disease is not within the scope of chiropractic practice. the ccbc also released an announcement to the public that claims promoting treatment or supplements to improve immunity were inappropriate. (https://www. chirobc.com/novel-coronavirus-covid- /) a surveillance scan was performed on march , for words such as covid, corona, and immune and derivative words. noncompliant cases were notified to take immediate corrective action. further scans were then performed and noncompliant cases were notified and forwarded to the ccbc inquiry committee by march , . in all three applications of the mrt, our research team was provided with anonymized, aggregated data from the ccbc beginning march , and ending march , . this anonymized data (provided by the ccbc with permission given for analysis) consisted solely of numerical totals and dates for all three mrt applications: estimates of the number of websites/social media accounts reviewed by mrt, subsequent cases of potentially inappropriate or non-compliant word use, estimates of the number of registrants and the number of non-compliant cases. approval for this project was provided by the university of alberta human research ethics board (pro ). approximately websites and social media pages of approximately ccbc registrants were reviewed with the mrt resulting in potentially inappropriate communications (table , fig. ) . a scan was then conducted on the compliance deadline that identified registrants as non-compliant. subsequent scans were then performed which showed that all non-compliant cases were resolved except for that remained outstanding and were later resolved through the inquiry committee. this mrt scan identified potential cases from approximated registrants (table , fig. ). further evaluation by the ccbc classified cases as inappropriate. internet analytics generated by mrt on the compliance date showed full compliance by all registrants. the mrt scan returned over potential cases from approximately registrants. specifically, there were potential cases for "corona" and/or "covid" and cases for "immune" (table , fig. ). after review by the ccbc to eliminate acceptable use of target words (e.g. wash your hands frequently during the covid crisis), cases were inappropriate in their use of the word "immune". no inappropriate cases were identified for use of "corona" or "covid". notices to remove inappropriate material were emailed in march . subsequent mrt scans showed that the majority of registrants responded promptly and removed questionable content. as of march , , there were sites yet to comply and the names of registrants associated with these sites were passed to the ccbc inquiry committee for further investigation. we describe the novel use of internet analytics to monitor the social media activity of the chiropractic profession in british columbia, canada. data presented demonstrate this technology can effectively monitor vast amounts of internet activity by registrants over a short period enabling the regulatory body to effectively bring about compliance. although other canadian professions are now in possession of this software, we are not aware of any prior report in the literature that describes the use of internet analytics to improve health care regulation in the chiropractic profession. traditionally, regulators have relied on publicreporting of inappropriate communications to provide input into their regulatory processes. although we did not receive data regarding the number of public complaints made to the ccbc during the time in which mrt has been in use, the ccbc communicated that there are occasions when a public concern is brought to their attention before the next scheduled mrt scan. these concerns are acted on immediately. still, it is doubtful that this level of public surveillance can identify the number of potential cases of inappropriate word use compared to mrt scanning. while public input remains important to the regulatory process, a modern regulatory body whose registrants are increasingly engaged with the public on the internet should not rely exclusively on public input to monitor their registrants when internet analytics are now available. while both the public and internet surveillance components of a modern regulatory approach are important, neither can be effective without significant human effort. on the public side, effort is needed to not only identify and interpret incursions, but also to ensure they are reported to the regulator promptly. the public must be made aware by the regulator of what is appropriate and inappropriate to make safe health care decisions. similarly, use of internet analytics by a regulator requires commitment to performing and analyzing scans regularly. either way this information is obtained, the regulator must use its existing processes to the fullest to address potential concerns. this is important for protecting the public and to act as a deterrent for future incursions given a rapidly evolving health care environment where competition for patients is becoming substantially more intense. this is especially true for the chiropractic profession whose integration into mainstream healthcare is evolving but whose record of advertising practices in some jurisdictions is unsatisfactory [ , ] . there would appear to be no substantial downside for regulators to adopt internet analytics in their daily operations to evaluate registrant compliance of regulatory directives [ ] although challenges may exist regarding cost, training, and/or time to process identified cases. current scanning techniques used with the mrt software do not detect target words placed over images (e.g. memes) or text/voice content within video sources. technologies to address these situations are being tested. the mrt software cannot presently scan all facebook content as only recent posts are available to the scanner; older posts require user scrolling to be visible. the mrt software is only effective as the target words used in its searches. the software cannot scan for implication or meaning. as the software allows the user to input new target words, the scan can be expanded as needed to include terms for existing or new regulations. while we have been informed verbally that mrt software has been ordered by health care professions other than chiropractic, we are not aware that the mrt software is being actively employed in these jurisdictions. widespread use of mrt software and communication between regulatory bodies as they gain experience in internet analytics may further improve its efficiency. importantly, mrt scanning also identifies cases from non-chiropractors who are associated with the registrant through a common clinic website or social media account (e.g. physiotherapy, massage, etc). cases from associated professions are not pursued by the ccbc nor are they passed forward to other regulatory agencies. finally, we cannot assume that all chiropractors knowingly post misinformation on the internet. although not presented here, the authors know of paid subscription services that provide chiropractors and their websites/ social media with "news feeds" that may provide questionable content. still, individual practitioners remain responsible for what is displayed in their name. this places the onus on these services to review their standards of business and become vigilant, evidence-based content providers. internet analytics are an effective way to monitor website and social media activity of registrants. use of internet analytics through mrt software is useful for regulatory bodies to keep pace with the high volume of internet activity produced by their registrants and may bring about rapid compliance to existing regulations. who director-general's opening remarks at the media briefing on covid- . in: who [internet coronavirus disease -advice for chiropractors a cross-sectional study of website claims related to diagnoses and treatment of non-musculoskeletal conditions. chiropractic manual therapies putting lives at risk: misinformation, chiropractic and the covid- pandemic the effect of spinal adjustment / manipulation on immunity and the immune system: a rapid review of relevant literature the canadian chiropractic association let's work together to protect and serve our patients, staff, families and communities false and misleading advertising on covid- urgent covid- statement vetenskapliga rådets kommentar på ica-dokumentet rcc research bulletins -spinal manipulation and the immune system afefc/ifec ajustement et augmentation de l'immunité coronavirus: the spread of misinformation statement in support of the scientists, public health professionals, and medical professionals of china combatting covid- surge of virus misinformation stumps facebook and twitter the application of internetbased sources for public health surveillance (infoveillance): systematic review toward real-time infoveillance of twitter health messages an inspector calls: the importance of proactive enforcement in the regulatory context keeping an eye on fraud: proactive and reactive options for statutory watchdogs the evidence-based policing matrix google trends in infodemiology and infoveillance: methodology framework at-risk advertising by australian chiropractors and physiotherapists complaints about chiropractors, osteopaths, and physiotherapists: a retrospective cohort study of health, performance, and conduct concerns health sector accreditation research: a systematic review publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations the authors would like to thank the ccbc for their data as well as their insight and professionalism. we would also like to thank compliance verification tools for providing us with information related to the functioning of the mrt. all authors (gk, jh, si, jks, bg) developed, wrote, edited and proofread this work. the author(s) read and approved the final manuscript. none applicable. all data generated or analysed during this study are included in this published article.ethics approval and consent to participate approval for this project was provided by the university of alberta human research ethics board (pro ). not applicable. fees for medical-legal expertise unrelated to this work from the canadian chiropractic protective association. jh reports that he holds multiple research grants from danish and international funding agencies and charities. he has received coverage of travel expenditures from multiple sources internationally in connection with speaking engagements. within the past year he has received speaking fees from parker seminars and novartis. he is a member of the world federation of chiropractic research council. si has no declarations. jks declares that he has no conflict of interest related to the subject matter or materials discussed in this manuscript. bg has no declarations. author details key: cord- -d s d ni authors: fareed, naleef; swoboda, christine m.; jonnalagadda, pallavi; huerta, timothy r. title: persistent digital divide in health-related internet use among cancer survivors: findings from the health information national trends survey, – date: - - journal: j cancer surviv doi: . /s - - - sha: doc_id: cord_uid: d s d ni purpose: prior research on the use of the internet among cancer survivors indicates a digital divide. the online landscape and patterns of information consumption, however, have notably changed over the past decade necessitating an updated examination of health-related internet use (hriu) among cancer survivors. methods: using survey data from , , , , , , and health information national trends survey (hints) iterations, the objectives of this study were to report prevalence, trends, and user profiles in hriu in terms of emailing doctors, buying medicine online, and support group participation. descriptive analyses and weighted multivariate logistic regression analyses were performed. results: cancer survivors who reported not using the internet were more likely to be older, belong to ethnic minorities, be less educated, and reside in rural areas as compared with those who reported using the internet. except for participation in online support groups, all other types of hriu increased in prevalence across the years. conclusions: consistent with the increased penetration of the internet and the altered online health communication landscape, we found increased prevalence of hriu among cancer survivors. however, the digital divide persists in terms of internet access. these findings can inform initiatives to bridge the gap among survivors of varying profiles in using the internet for their health needs. implications for cancer survivors: there is increased reliance on online platforms to obtain and communicate health-related information. the risk with this approach is potential oversight of ensuring equity in terms of internet access and technology literacy among survivors. electronic supplementary material: the online version of this article ( . /s - - - ) contains supplementary material, which is available to authorized users. the s have been marked by an altered online landscape. the internet has emerged as the primary resource for healthrelated information for an increasing percentage of the population [ ] . the internet is considered a conduit for extensive information, convenient access to specific types of information, and interactive communication for real-time feedback [ ] [ ] [ ] [ ] . thirty-five percent of us adults reported going online to diagnose a medical condition either for themselves or for someone else. of those diagnosing themselves online, % sought medical attention based on the information they found. furthermore, % of those diagnosing themselves online claimed that their diagnoses were confirmed by a medical professional [ ] . in a pew survey, % of us adults reported having read about or watched another person's health experience and % reported going online to find others who shared their health concern [ ] . the health information technology for economic and clinical health (hitech) act of [ ] , incentivized payments to eligible providers to meet meaningful use mandates heralding a new era in how healthcare is delivered in the usa [ ] . health information technology (hit) like electronic health records (ehrs) and technologies like mobile applications, social media, and video conferencing have been increasingly adopted since [ ] effectively transforming the online health communication landscape over the last decade [ ] . in , social media overtook traditional print media as the primary source of news [ ] . according to a pew research center survey, nine out of adults went online, % owned a smartphone, and % used social media [ ] . as connectedness becomes the standard expectation of individuals in the broader community, the implications of internet use and the associated disparities are critical to our understanding of cancer survivorship. the number of cancer survivors is projected to increase to million by [ ] . the internet as a source of information was second only to the healthcare provider for cancer survivors [ , ] . cancer survivors may use the internet for the following reasons: for information pertaining to disease management, complementary and alternative medicines, clinical trials and research, nutrition, personal rights, and medical language; for support in terms of advice on the cancer experience, emotional support, alleviation of isolation, and social interactions; and lastly, for practical reasons like communication with healthcare professionals, prescription refills, and scheduling. reasons precluding cancer survivors from using the internet cited in the literature include lack of computer skills, lack of access to computers, concerns over privacy, and quality of information [ ] . chou and colleagues have documented an increasing trend in health-related internet use (hriu) among cancer survivors, which they defined as emailing doctors, buying medicine online, online support group participation, and seeking cancer information from the internet as the first source [ ] . notably, they found that cancer survivors not using the internet were more likely to be less educated, belong to racial/ethnic minorities, live in non-metropolitan areas, and report worse general health suggesting a digital divide-defined as the gulf between people who have and do not have access to information technologies [ ] . since the report on hriu among cancer survivors by chou and colleagues, the online landscape has considerably shifted, where % of americans reported using the internet compared with the year , when nearly half the american adult population did not use the internet, per a report from the pew research center [ ] . in the general us population, the use of the internet as the first source for health information has risen across the years [ , ] . upon examining the individual types of hriu in cancer survivors, chou and colleagues found a significant trend across the years for all types of hriu except cancer information seeking. in their study on cancer survivors, while they did not examine cancer information seeking on the internet, jiang and liu reported a rising trend in internet health information seeking (ihis), defined as having looked for medical information on the internet for oneself or someone else in the past months [ ] . another recent study on online patient provider communication (oppc), which includes emailing doctors, among cancer survivors reported a rising trend from to [ ] . these reports suggest a shrinking, yet persistent, digital divide. long-term cancer survivors reported a high prevalence of unmet information needs pertaining to side effects and symptoms, test and treatment, health promotion, interpersonal and emotional, insurance, and sexual functioning and fertility [ ] . engagement in hriu was associated with psychological benefits for survivors and was positively associated with the patient-provider relationship [ ] [ ] [ ] . therefore, it is of vital importance to understand the factors associated with hriu to identify user profiles for developing tailored health promotion and communication strategies. the current landscape is characterized by use of personalized technology and the internet of things influencing the dissemination of health information, hit, and how patients and providers interact [ ] highlighting the importance of updated information on trends in internet usage in the cancer survivor population. therefore, the objectives of this analysis were to replicate and update the findings by chou and colleagues [ ] using a pooled, cross-sectional analysis based on data over multiple health information national trends survey (hints) iterations to examine prevalence, trends in hriu, and user profiles of cancer survivors across the years. to this end, we used a combined sample of survey data from hints corresponding to , , , , , , and . this study used data obtained through hints [ ] sponsored by the national cancer institute (nci). hints is fielded to a representative sample of us civilian, non-institutionalized adults over years of age. hints collects data about how respondents seek and use information about cancer, as well as cancer risk perception, cancer prevention behavior, and demographics. this study used data from hints ( ), hints ( ), hints ( ), hints , cycles ( ) and ( ), and hints , cycles ( ) and ( ). the hints and surveys collected data by random-digitdialing respondents to participate in a telephone interview, while hints in used two modes: random-digitdialing and also a mailed questionnaire. no differences were detected between survey administration modes based on chisquare comparison of key variables (i.e., age and gender), resulting in combining both telephone and mailed survey responses for hints . the fourth and fifth iterations of hints were administered by mail. the survey uses a stratified postal address frame to randomly sample residential addresses. hints is published with survey weights to allow the results to be more generalizable to the population [ ].the response rates for each year were hints ( %), hints ( %), hints ( % telephone mode, % mail mode), hints cycle ( %), hints cycle ( %), hints cycle ( %), and hints cycle ( %). to facilitate comparison with the findings identified by chou et al. [ ] , tables and have analyses by year. following chou's approach, analyses controlled for gender, age, race/ethnicity, education level, rural/urban status, insurance status, general health status, and time since cancer diagnosis. in contrast to chou et al., distress was removed as a covariate, as this was not collected after hints in . in addition, the first source of cancer information was not asked after and was removed as a dependent variable. the main population of interest was cancer survivors, consisting of those who answered "yes" to the question "have you ever been diagnosed as having cancer?" sociodemographic and clinical characteristics used as covariates in this analysis included gender (male/female), age ( - / - / +), race/ethnicity (hispanic/non-hispanic white/ non-hispanic black/other), education (high school graduate or less/some college/bachelor's degree or more), metropolitan status (metro area/non-metro area), general health (excellent, very good, or good/fair or poor), insurance status (insured/ uninsured), and time since diagnosis ( year or less/ - years/ - years/ + years). dependent variables used in this analysis include internet use, emailing doctors, buying medicine online, internet support groups, and seeking any cancer information. internet use behavior was quantified by asking "do you ever go on-line to access the internet or world wide web, or to send and receive e-mail?" with responses of "yes," or "no." if people responded "yes" to using the internet at all, the ways people used the internet were assessed using the question "in the last months, have you used the internet for any of the following reasons?". options included in this analysis were responses to the following: "used e-mail or the internet to communicate with a doctor or a doctor's office" (emailing doctors), "bought medicine or vitamins online" (buying medicine online), and "participated in an online forum or support group for people with a similar health or medical issue" (support group). finally, responses to the question "have you ever looked for information about cancer from any source?" were included as a dependent variable in this analysis. all analyses used survey weighting and jackknife variance estimations provided by hints to minimize sampling error [ ] . analyses were first performed using the data from , , and to replicate the findings from chou et al. [ ] to ensure model fidelity. these analyses were then extended to the , , , and hints iterations. maintaining fidelity to the analyses conducted by chou et al., the data was analyzed using the same demographic categories, excluding distress. weighted descriptive statistics were analyzed for respondents who had a history of cancer and answered the question about internet use. chi-square tests were used to test for associations between demographic characteristics and internet use. a weighted multivariable logistic regression model was created to explore the adjusted association between the dependent variables regarding hriu, and the covariates gender, age, race/ethnicity, education, metro status, general health status, insurance status, and time since diagnosis among cancer survivors. respondents excluded from regression analysis were those without any cancer history and/or those that did not answer the question regarding internet use. for the dependent variables pertaining to hriu, the regression analysis excluded those who did not use the internet. additionally, using the full sample of those who responded "yes" to using the internet, we carried out chi-square tests to test for associations between survey year and hriu variables, and associations between cancer survivor status (cancer survivor, family with cancer but no personal history, and no history or family members with cancer) and hriu variables. all analyses were completed using stata version . ( , statacorp lp, college station, texas). a population sample was obtained from the combined , , , , , , and hints data. the response rate in the combined sample was . %. a total of respondents included in these analyses were cancer survivors and answered positively to the question regarding internet use (table ) . while of the survivors used the internet across the sample, responded "no" when asked if they went online for any reason. there were respondents with a history of cancer that did not answer the internet use question (i.e., we used data from respondents for our analysis of hriu among cancer survivors). unadjusted, bivariate analyses produced significant associations between there were significant associations between internet use, emailing doctors, buying medicine online, and seeking any cancer information and survey years ( in the full sample that included all internet users (n = , ), there were significant associations between internet use, emailing doctors, online support group participation, and seeking any cancer information and cancer status (table ) . cancer survivors were the most likely to email doctors, be in online support groups, or seek cancer information online, while those with family who are survivors were most likely to use the internet. those without a history of cancer or family members with cancer had the lowest prevalence of emailing doctors, buying medicine online, being in support groups, or seeking cancer info. figures this study sought to replicate and update the findings of chou and colleagues [ ] , by reporting prevalence, trends, and user profiles of hriu across the survey years from to using data from hints through hints . we found that, despite the widespread availability of the internet, there were cancer survivors who did not utilize the internet. our updated findings indicated that cancer survivors, who use the internet, were younger, white, college educated, metro-dwelling, and more likely to report better self-rated health. it is noteworthy that once online there were no race/ethnic differences in hriu among cancer survivors. overall, our findings are similar to those reported by chou and colleagues suggesting a persistent digital divide despite the altered online health communication landscape. we found that the proportion of cancer survivors reporting internet use increased across the years ranging from . % in to . % in . survivors not using the internet belonged to groups that were older, had a high school education, were from ethnic minorities, resided in non-metropolitan areas, and reported poor or fair self-rated health. as in the original study by chou et al., the time since cancer diagnosis was not associated with internet use and any type of hriu likely due to a healthy volunteer bias in the hints sample. in our multivariable models, the odds of internet use were higher in every year after . with the exception of participation in online support groups, prevalence of all other types of hriu increased across the years from to . in multivariable models, the odds of emailing doctors and buying medicines online were greater for all years in comparison with . our analysis found cancer survivors having college education or higher were more likely to engage in all types of hriu except participation in online support groups. in our study, the proportion of cancer survivors emailing their doctors increased from a mere . % in to . % in . we also found that younger age, metropolitan residence, and college education were significantly associated with emailing doctors in all years as compared with the year in multivariable models. in contrast to the findings by chou et al., we found that being - years old was a significant predictor of emailing doctors among cancer survivors. oppc, of which emailing doctors is a subtype, has emerged as an alternative and a complement to traditional patient-provider communication [ ] . oppc has been associated with enhanced clinical management and self-care, decreased healthcare expenditure, increased accessibility to information, and reduced office visits/phone contacts [ , ] . the increasing trend for emailing doctors found in our study of cancer survivors was also seen in the general us population. a recent study using hints data from to found that oppc among us adults rose from % in to % in as well as a significant increase in the proportion of internet users communicating with their providers using email [ ] . however, our findings also suggest a flatlining of the proportion of survivors emailing their doctors since . this trend could be explained by the variety of other options available to patients like mobile applications, social media, video conferencing, and ehrs that have been increasingly adopted since [ ] , coincident with the meaningful use mandates of the hitech act [ , ] . indeed, % cancer survivors in reported using the secure messaging feature in the ehr and . % used mobile-based applications to communicate with their healthcare providers [ ] . in the same study, jiang and colleagues examined the association of the dimensions of the digital divide with oppc finding that physical internet access was significantly associated for all the years studied, cognitive access (health information overload, self-efficacy in health information seeking, and trust in online information) was not associated with oppc, and the association with over the years, participation in support groups has consistently remained low compared with other types of hriu. we did find that survivors younger than were more likely to participate in online support groups than their older counterparts. sharing experiences online may affect a person's health through finding information, feeling supported, maintaining relationships with others, affecting behavior, experiencing health services, learning to tell the story, and visualizing disease [ ] . research on social media participation, a contemporary platform for support groups, using hints data arrived at a similar conclusion and highlighted the lack of popularity of this medium [ ] . more research is needed to understand the evolution of this medium to gain a better understanding of how social media-type interventions can be used to target age-based diseases such as cancer and how social support is expressed via social media. previous research has indicated that these platforms as a source of support for cancer may be too general and more emotionally oriented (as opposed to focused on resources/needs) [ ] . popular social media outlets such as facebook may also provide information that needs to be appraised for quality to ensure that it is evidence based [ , ] . recent developments in online support group platforms include patient-generated health-data cohorts, where cancer survivors can report health data and experiences through their mobile and wearable health devices [ ] . information collected on such platforms is also used to develop a knowledge base that would help future cancer survivors. the proportion of survivors buying medicines online has steadily risen from . % in to . % in . there was no difference in the prevalence of buying medicines online by cancer status in our analysis. however, our results showed that cancer survivors who purchase medicines online were more likely to be insured after adjusting for sociodemographic factors and time since diagnosis. in a previous study, consumers of online pharmacies were more likely to be male, younger, white, married, more educated, and wealthier [ ] . buying medicine online offers several advantages: convenience to those who are housebound or disabled; privacy that encourages potentially embarrassing questions; variety; and affordable prices [ ] [ ] [ ] . it also allows consumers to compare prices across internet pharmacies; for instance, a study showed that americans can save nearly % per unit of drug if they purchase from an online canadian pharmacy instead of an online us pharmacy [ ] . in regard to cancer information seeking among cancer survivors, in contrast to chou et al., we found that the frequency of cancer information seeking increased over time. it should be noted that the hints survey is altered and no longer asks the participant to distinguish cancer information seeking from the internet or any other source. seeking cancer information online has been shown to be associated with engaging in protective behaviors like using sunscreen and increased adherence to screening guidelines [ ] . in a survey of adult cancer patients at an academic medical center, % reported having internet access and % reported using the internet to search cancer-related information. internet use to seek information about cancer differed by age, race, and education. the leading reason cited for an internet search was to develop questions to discuss with the provider. the top three types of information sought were treatment options, prognosis, and side effects. results of the internet search influenced choice of treatment, choice of doctor, and enrollment in clinical trials [ ] . a recent study examining ihis (i.e., having looked for medical information on the internet for oneself or someone else in the past months) in cancer survivors using data from the , , and iterations of hints [ ] reported a rising trend for ihis from . % in to . % in mirroring the trends in the general us population [ ] . despite an increase in the proportion of cancer survivors engaging in ihis, the study confirmed a digital divide and further examined the dimensions of the digital divide: mental access (trust in online information and self-efficacy in health information seeking), material access (ease of internet access), skill access (skill to seek information online), and usage access (online healthrelated activities like emailing doctors and buying medicines). there was a significant positive association between ihis and material access, usage access, and mental access (trust in online information) across all years [ ] . our study findings have important implications for clinicians, researchers, and public health practitioners in light of the increased adoption of technologies like personal health record portals [ , ] and the paradigm shift underway in response to the covid- pandemic necessitating and fasttracking digital health solutions [ ] . evidence has shown that patient portals can foster patient engagement and improve health outcomes [ ] [ ] [ ] [ ] . there is growing evidence of a digital divide in portal usage as younger, white cancer patients reportedly showed the heaviest use of patient portals [ ] . our updated results show that the persistent digital divide disproportionately affects vulnerable populations. yet, while age, education, and residence were associated with hriu, race and ethnic differences no longer exist among cancer survivors engaging in hriu. as already emphasized by chou et al., these findings underscore the importance of addressing the needs of a vulnerable population that is at risk of being further disenfranchised and potentially disengaged with their health and care [ ] . email has shown to be suited for questions about medications, health status updates, and medical questions [ ] , making it an important tool for patient-centered healthcare. therefore, efforts should be made to encourage both patient and provider adoption of email and other types of oppc. the slow and steady rise in buying medicine online offers several benefits to cancer survivors, such as, variety, better prices, and convenience. however, our findings suggest buying medicine online could contribute to disparities in health. the rising popularity of online pharmacies underscores a need to increase health literacy so consumers may make healthier choices and learn to recognize risks [ ] . a systematic review of outcomes of online support and resources among cancer survivors showed a generally positive, albeit inconclusive, association [ ] . given the multitude of social media avenues available to people, the associated benefits, and the interactive nature of social media platforms, further research is needed to identify the profile of cancer survivors most likely to benefit from this medium. prior evidence showed that the desire for information and the desire for participation in care are independent. cancer patients can be classified as those who are involved in seeking information and healthcare decisions; those who seek information, but defer to their providers to make decisions; and those who prefer to delegate both the information gathering and decision-making to their providers [ ] . however, the rising trend in cancer information seeking brings into focus the role of healthcare providers in guiding their patients' hriu. patients have expressed a desire for guidance from their primary care providers regarding medical information on the internet [ ] ; moreover, over time, there has been an upward trend in guided versus spontaneous use of the internet among cancer patients [ ] . some of the strengths of this study include an examination of all the years of data making the results comparable with previous reports. moreover, the health communication landscape has considerably altered in the past decade influencing the dissemination of health information, hit, and how patients and providers interact [ ] highlighting the importance of updated information on trends in internet usage in the cancer survivor population. however, there are some notable limitations. first, the data are based on self-report and subject to recall bias. second, the surveyed population is subject to selection bias as respondents with poor prognosis and serious disease are less likely to participate in the survey. third, our analysis does not account for type of cancer, which has been shown to be associated with cancer information seeking behavior [ ] . fourth, the data are cross-sectional precluding any inference on causation highlighting the need for other types of studies to complement results from surveys, for example, patterns of information use can be gleaned using internet search data [ ] and mined from social media platforms [ ] to explore questions about the specific topics discussed, relation between promotional information and discussions, and the relationship of these patterns to survey results from hints. lastly, the response rate in hints was low, but it was similar in other national surveys [ , ] . hints administrators have taken steps to minimize bias arising from modality, coverage, and sampling [ , ] . our findings suggest a persistent digital divide in vulnerable segments of the population. compared with the year , when nearly half the american adult population did not use the internet, % of us adults were offline per a report from the pew research center. the offline us adults were more likely to be and older, live in rural areas, and have a high school education [ ] . older adults in the general population have reported physical challenges in manipulating devices and a lack of confidence in using new technologies [ ] . trust in information and ease of searching for information were associated with adoption among older adults suggesting that this group might benefit from navigational assistance [ ] . there is evidence showing significantly worse distress, depression, and overall health-related quality of life among hispanic cancer patients [ ] . in a recent study of online cancer prevention information seeking preferences in an ethnically diverse sample, older age, high school education, and being a spanish speaker were associated with lower odds of seeking cancer prevention information online [ ] . to avoid further exacerbating disparities, it is vital to gain an understanding of the perceptions of and barriers to internet access among older cancer survivors to inform the development of appropriate eliteracy interventions. however, it is noteworthy that once cancer survivors are online and engage in the different activities entailing hriu, there were no significant race and ethnic differences highlighting the need to mitigate the barriers to internet access. in summary, the current analysis using data from iterations of hints showed increasing prevalence of hriu among cancer survivors. differences in internet use across key groups, particularly age, insurance status, and education, have implications for the design of health interventions and communication efforts. is online health activity alive and well or flatlining? findings from years of the health information national trends survey health information | pew research center health online | pew research center patterns of information-seeking for cancer on the internet: an analysis of real world data the social life of health information | pew research center implementation of the federal health information technology initiative cancer prevention and control in the changing communication landscape trends of online patient-provider communication among cancer survivors from to : a digital divide perspective social media outpaces print newspapers in the u.s. as news source | pew research center key ways the u.s. changed in the s | pew research center cancer statistics cancer survivors information seeking behaviors: a comparison of survivors who do and do not seek information about cancer information needs and sources of information for patients during cancer followup why people living with and beyond cancer use the internet health-related internet use among cancer survivors: data from the health information national trends survey the digital divide as a complex and dynamic phenomenon % of americans don't use the internet | pew research center online health information seeking among us adults: measuring progress toward a healthy people objective digital divide and internet health information seeking among cancer survivors: a trend analysis from health information needs and health-related quality of life in a diverse population of long-term cancer survivors cancer patients' decisions about discussing internet information with their doctors patients using the internet to obtain health information: how this affects the patient-health professional relationship national cancer institute. survey instruments. hints. https:// hints.cancer.gov/data/survey-instruments.aspx. accessed use of mobile devices to help cancer patients meet their information needs in non-inpatient settings: systematic review patient-initiated e-mails to providers: associations with out-of-pocket visit costs, and impact on care-seeking and health online communication between doctors and patients in europe: status and perspectives growth dynamics of patient-provider internet communication: trend analysis using the health information national trends survey ( to ) achieving meaningful use of health information technology: a guide for physicians to the ehr incentive programs health and illness in a connected world: how might sharing experiences on the internet affect people's health? the milbank quarterly use of social media in health communication: findings from the health information national trends survey codifying online social support for breast cancer patients: retrospective qualitative assessment cancer communication on social media: examining how cancer caregivers use facebook for cancer-related communication differences between races in health information seeking and trust over time: evidence from a cross-sectional, pooled analyses of hints data social media and the adolescent and young adult (aya) patient with cancer investigation of consumer choice of community, mail order, or internet pharmacies internet pharmaceutical sales: attributes, concerns, and future forecast controversies and legal issues of prescribing and dispensing medications using the internet cyberpharmacies and the role of the us food and drug administration a comparison of brand-name drug prices between canadian-based internet pharmacies and major u.s. drug chain pharmacies cancer information seeking and cancerrelated health outcomes: a scoping review of the health information national trends survey literature a survey of internet utilization among patients with cancer health information seeking in the digital age: an analysis of health information seeking behavior among us adults social disparities in internet patient portal use in diabetes: evidence that the digital divide extends beyond access personal health record use in the united states: forecasting future adoption levels strategies for digital care of vulnerable patients in a covid- world-keeping in touch patients' continuing use of an online health record: a quantitative evaluation of , patient years of access data patient portals: an underused resource for improving patient engagement empowering patients during hospitalization: perspectives on inpatient portal use patients' perceptions of portal use across care settings: qualitative study predictors and intensity of online access to electronic medical records among patients with cancer providing access: differences in pediatric portal activation begin at patient check-in e-mail in patientprovider communication: a systematic review quality of online pharmacies and websites selling prescription drugs: a systematic review outcomes of online support and resources for cancer survivors: a systematic literature review the determinants and consequences of information seeking among cancer patients brief report: what types of internet guidance do patients want from their physicians? a systematic and comprehensive review of internet use in cancer patients: psychological factors differences in information seeking among breast, prostate, and colorectal cancer patients: results from a population-based survey mining twitter to assess the determinants of health behavior toward human papillomavirus vaccination in the united states tracking chronic disease and risk behavior prevalence as survey participation declines: statistics from the behavioral risk factor surveillance system and other national surveys telephone coverage and health survey estimates: evaluating the need for concern about wireless substitution health information national trends survey (hints) picking up the pace: changes in method and frame for the health information national trends survey ( - ) barriers to adoption and attitudes towards tech among older americans | pew research center online health information seeking:the influence of age, information trustworthiness, and search challenges psychological morbidity and quality of life of ethnic minority patients with cancer: a systematic review and meta-analysis predictors of online cancer prevention information seeking among patients and caregivers across the digital divide: a cross-sectional publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations key: cord- - erjf rk authors: maurushat, alana title: the benevolent health worm: comparing western human rights-based ethics and confucian duty-based moral philosophy date: - - journal: ethics inf technol doi: . /s - - - sha: doc_id: cord_uid: erjf rk censorship in the area of public health has become increasingly important in many parts of the world for a number of reasons. groups with vested interest in public health policy are motivated to censor material. as governments, corporations, and organizations champion competing visions of public health issues, the more incentive there may be to censor. this is true in a number of circumstances: curtailing access to information regarding the health and welfare of soldiers in the kuwait and iraq wars, poor health conditions in aboriginal communities, downplaying epidemics to bolster economies, and so forth. this paper will look at the use of a computer worm (the benevolent health worm) to disseminate vital information in␣situations where public health is threatened by government censorship and where there is great risk for those who ‹speak out’. the discussion of the benevolent health worm is focused on the peoples’ republic of china (china) drawing on three public health crises: hiv/aids, sars and avian influenza. ethical issues are examined first in a general fashion and then in a specific manner which uses the duty-based moral philosophy of confucianism and a western human rights-based analysis. technical, political and legal issues will also be examined to the extent that they better inform the ethical debate. censorship in the area of public health has become increasingly important in many parts of the world for a number of reasons. groups with a vested interest in public health policy are motivated to censor material. this may include governments, corporations, professions, and organizations. the censorship may be direct (legal sanctions) or indirect (corporate and individual self-censorship). as experts in the field, ngos and other citizen movements champion competing visions of public health issues, the more incentive there may be to censor. this is true in a number of circumstances. for example, curtailing access to information regarding the health and welfare of soldiers in the kuwait and iraq wars, poor health conditions in aboriginal communities, downplaying epidemics to bolster economies, and so forth. this paper will look at the use of a computer worm (the benevolent health worm) to disseminate vital information in situations where public health is threatened by government censorship and where there is great risk for those who 'speak out'. while there are many examples along the spectrum of public health censorship, this paper's discussion of the benevolent health worm will be limited to that of the peoples' republic of china (china) drawing on three public health crises: hiv/aids, sars and avian influenza. in each of these situations, chinese citizens faced a public health epidemic (which then spread to the international community). in each of these situations the chinese government heavily censored information, allowing the disease to unnecessarily spread faster in an uncontained manner. and in each of these situations individuals who vocalized or published unauthorized news articles on the epidemic (many prominent experts, doctors and activists) were detained without reason serving time in prison. some were threatened or charged with divulging a state secret. the author uses china by way of example due to the extremity of the example, as well as her experience and familiarity with politics, censorship strategy and legal developments in the region. the use of a controversial technology such as a computer worm to disseminate uncensored, sanctioned public health information in china presents contentious ethical issues worth examining. when is the use of an illegal technology ethical? does the dual use of a computer worm for malicious or benevolent reasons play a part in the analysis? if so, at what point? is motivation the determining factor? intended use? actual consequences? is there a moral duty to write and disseminate public health information which differs from authorized accounts? is the duty a general duty or is it specific to certain members of society? does the mode of information delivery play a part in an ethical analysis? are anonymous modes of dissemination less ethical than methods which provide accountability? to what extent does the source of the information factor into the equation? what role does risk of criminal sanction play in ethics? does the risk of criminal sanction depend on the actual use or potential consequences of the technology? does the violation of human rights justify the illegal activity? if so, is the chinese context justifiable? is the use of a benevolent worm compatible with western ethical traditions? chinese ethical traditions? the above questions are raised to illustrate the abundance of ethical queries triggered by the topic, and are not meant to be fully covered in this paper. for the purpose of this paper, an account will be given of the censorship environment in china both in general and in the specific context of public health. this will be followed by an account of technical aspects of the benevolent health worm inasmuch as it will inform and frame the debate on ethical issues. a further account will be given examining the use of an illegal technology such as a computer worm to disseminate non-authorized public health information. the core of the paper will examine ethical issues in a general fashion, and then in a specific manner drawing on the moral philosophy of confucianism and western notions of civil liberties/human rights. confucianism is often thought to be incompatible with western rights-based theories (democracy, civil liberties, human rights and other autonomous rights based theories). the point is not to justify the benevolent health worm through either a 'western' or 'eastern' lens. the point, rather, will be to examine many of the ethical issues from multiple perspectives. in particular, the perspective of chinese moral philosophy -confucianism -which is rooted within the framework of values and duties -will be used along side a western human rights-based analysis. the use of western rights-based theories (human rights) alongside the eastern duty-based theory of confucian moral philosophy provides an interesting platform for an ethical analysis of the benevolent health worm. the author will suggest how human rights and confucian moral philosophy may be used to better understand the ethical issues presented with the use of the benevolent health worm. the paper does not aim to justify the use of the benevolent worm. instead, there is the modest aim to initiate ethical debate on the subject. the application of the analysis could extend to a broader examination of benevolent payloads, and, further along the spectrum, the ethical use of illegal technologies. governments in china have traditionally utilized censorship as a means for control. using censorship as a control mechanism has historically been pitted against the chinese promotion of intellectual growth. the rise of the chinese communist party (ccp) brought with it the continued ideal of control over the dissemination of works and ideas. china continues to censor books, newspapers, and basically, most forms of publications that threaten the governing regime or criticize china's attitude towards human rights. included in this overall censorship strategy is tight control of the media and the internet. all news agencies, including news websites and chatrooms, must be accredited. and sale of foreign news in china may only be purchased and published from the state-run government news agency, xinhua. review and enforcement of laws and regulations is performed by two agencies (one for press, radio, film and television, and the other for written publications including the internet), both of which are run by the communist party's central propaganda department (cpd). against the backdrop of what could best be described as a labyrinth of laws and regulations, the cpd issues weekly informal directives to news agencies and internet service providers (e.g. google and microsoft) on news items requiring restrictive coverage. virtually all statutes and regulations concerned with communications (news or otherwise) contain vague language allowing authorities sufficient flexibility in determining which publications are in breach of the law. navigating through the ever-changing and complex media and internet regulations is a seemingly never-ending process. while the regulations are ever-changing, there is a standard set of vaguely written provisions which appear in all such regulations: divulging state secrets; harming the honour or interests of the nation; spreading rumours which may disturb social order; and inciting illegal assemblies which could disturb social order -all punishable as a criminal offence. state secrets provisions are the most problematic as their wording and interpretation in practice has proven malleable to political will. it is difficult for writers (whether they be journalists or mere bloggers) to determine in advance whether their message would contravene the law. where writers publish illegal content they are subject to a number of punishments such as dismissal from employment, demotion, libel, fines, closure of business, and imprisonment. imprisonment for illegal new stories extends to employees of foreign news agencies. for example, hong kong based journalist ching cheong (singapore's straits times) and zhoa yan (new york times) were arrested and detained for reporting articles about communist party leaders. news around sensitive topics such as a public health crisis is heavily censored and monitored. historically, individuals who reported and disseminated sanctioned public health news were often detained without reason and, in some cases, these individuals were charged with divulging a 'state secret'. many academics and experts have written on the scope of 'state secret' in china. the notion of 'state secret' has traditionally been broad and deliberately ambiguous, while its scope of application is ever-changing. it remains impossible to ascertain whether a person's actions would fall within a 'state secret'. people have been charged with this serious offence for the dissemination of banned information related to human rights, revealing draft laws (white papers), publishing unauthorised news reports, and publishing information critical of governing authorities. the act of circumventing the ''great firewall'' for illicit purpose, and mere research on internet censorship, could conceivably fall within the parameters of 'state secret'. the ccp's unpredictable use of broad, ambiguous laws to deter freedom of expression is heavily criticized in the international arena. while 'state secret' laws remain a potent threat, the ccp has a number of criminal provisions which it regularly uses to curtail the dissemination of sanctioned information. to paraphrase a prominent malaysian journalist steve gan, ''we have the right of freedom of expression. the problem is that we have no rights once such words are freely expressed.'' the same could be said of china. access to information may involve more than freedom of expression; timely information may have repercussions for the health and welfare of individuals. indeed, there are three specific areas where censorship and a lack of accurate information distributed in a timely manner have had unrefuted consequences in china in recent history: aids, sars and avian influenza (often referred to as bird flu). the chinese government has suppressed and continues to suppress information on the spread of c. zissis.''mediacensorshipinchina''councilonforeign relations, hiv/aids. by the government had reported only known cases claiming that aids was a foreigners' disease. the lack of reporting and ineffective preventative measures led a number of people to become aids activists. these activists reported significant rates of people infected hiv and campaigned for the government to take proactive measures to reduce the spread of this debilitating disease. many aids activists, including the famous activist wan yanhai, have been detained and charged with divulging a state secret. infected blood supplies appear to have initially been the main source of the problem. infected blood supplies, however, still taint china with many people in poorer areas donating blood for money while drug use, prostitution and a lack of educative measures continue to exacerbate the situation. the reality today is that china has one of the highest hiv/ aids rates in the world outside of africa. while we will never know the effect that accurate and timely information would have had in this epidemic, it is certainly plausible that access to such important information could have reduced the rate of infection. similar to the hiv/aids crisis, the chinese government withheld critical health information on severe acute respiratory syndrome in . china has a longstanding tradition of curtailing news deemed harmful to society and to china's image. in the case of sars, it was thought that exercising tight media control would reduce public fear and lessen economic damage in the region. the lack of reliable dissemination of information and the underreporting of infected sars patients to the world health organization allowed the disease to spread more readily from guangdong province to other provinces in china, to hong kong and to other countries in the world. the sars health crisis can be partially attributable to nondisclosure of pertinent information. avian influenza (also referred to as bird flu) while avian influenza a has not yet reached the level of crises of hiv/aids or sars in china, historical events give clear signs that any information provided by chinese officials should be perceived with caution. it is believed that quiao songju, a chinese farmer, was arrested and detained for reporting a potentially infectious bird in the anhui province. prominent hong kong virologist, guan yi, was invited by the chinese government to study abf. his account of the disease was vastly different from the official version reported by the chinese government. guan's publications are censored in china while it has been made known to the prominent virologist that he should not return to china. it is rumoured that guan has been threatened with detainment and there is further speculation that he may be in violation of having disclosed a 'state secret'. china's newly drafted censorship rules compound the situation. the newly drafted law states that it is a criminal act to publish any information on 'sudden events' without prior authorization from the chinese government. 'sudden events' are defined as 'industrial accidents, natural disasters, health and public security issues'. the government claims that the law is aimed at irresponsible journalists who report untruths potentially causing panic among the public. critics have claimed that the draft law is aimed at preventing future disclosures of embarrassing news. public health epidemics would fall under the category of sudden events. the internet and wireless technologies have been heralded as vehicles of free expression. it has generally been thought that no government could control information on the internet, hence the expression ''the internet routes around censorship. '' in china, this is increasingly no longer true. the chinese government erected, through its golden shield project, what has become known as the 'great firewall of china'. access, control and censorship of internet content in china is most often attributed to the 'great firewall of china'. this is, however, something of a misnomer; the firewall is merely one path in a maze of controlling technologies and non-technological means in an overall internet censorship strategy. this censorship strategy is comprehensive, incorporating sophisticated technologies, numerous regulatory measures, market influences, and aggressive policing and surveillance of internet activity, resulting in an atmosphere of self-censorship. laws regulating free speech and the internet are implemented out of concern for the potential harm posed by unfettered access to sites that contain political, ideological, social, or moral content that the ccp perceives as harmful. china has adopted a comprehensive internet censorship strategy utilizing a range of control mechanisms. mechanisms of control include laws and regulations pertaining to physical restrictions, regulations of use, ownership and operation of internet service providers (isps), internet access providers (iaps), and internet content providers (icps). similar to media regulations, a series of ever-changing internet regulations are also relevant to the dissemination of information. authorized access entails individuals having to obtain licenses for internet access. in order to obtain a license, individuals are required to register with the local police and provide their names, the names of their service provider, their e-mail addresses, and list any newsgroups in which they participate. this, of course, does not mean that anonymity and pseudonymity cannot be achieved for chinese cybersurfers. users have flocked to cybercafe´s and universities to access the web. the ccp has responded by shutting down many cyber-cafe´s, then later by requiring all cybercafe´s and universities to obtain user identification, and to keep detailed logs of user activities (the regulations are complex and comprehensive). the extent to which such entities have fully complied with the law in practice has not been explored, but the threat of surveillance continues to lead to an environment of self-censorship. the ability to access banned documents and to communicate anonymously is challenged. information flows from the internet subscriber (home, cybercafe´) to the internet service providers (isps) to four gateways controlled by the ministry of posts and telecommunications. isps are regulated through a myriad of laws which are, again, ambiguous and complex. it is difficult for any party to know if they are in compliance with the law. the regulations require isps to restrict and control access to harmful/banned websites, allow surveillance software on their systems, and keep logs of user activity. email is neither private nor anonymous when using an isp regardless of whether a domestic or foreign service is used. isps must and do comply with requests to reveal personal information of the true identity of users as well as information about email content. for example, both yahoo!china and yahoo!hk have disclosed confidential user information of prominent journalists by releasing internet protocol addresses to chinese authorities. many journalists deliberately have email accounts in jurisdictions such as hong kong with a strong rule of law tradition in order to shield their identities from chinese authorities. yahoo!hk handed over the internet protocol address (not the user's name) of journalist shi tao to chinese authorities based merely on an informal request. as such, yahoo!hk circumvented formal judicial requirements of a court order compelling the disclosure of confidential information. yahoo!hk claims that they disclosed the information in compliance with a criminal investigation in mainland china. the dispute has become one of many disputes over the scope of chinese jurisdiction in hong kong. disclosure of an internet protocol address is not classified as personal information under hong kong law making it safe for isps to circumvent their otherwise legal obligation to keep personal information confidential under the privacy ordinance. the securing of shi tao's internet protocol address led the chinese authorities directly to a specific computer port number, and directly to shi tao's computer. the information was classified in china as state secret. shi tao was arrested and sentenced to years in jail. recent popular methods for dissemination of taboo/illegal documents include spam, weblogs and chatrooms -all delivery methods involving the internet which allow for some degree of anonymity or . the precise number of gateways has not been established. some report , others , and others . the author has taken the middle figure as an average only. this ambiguity illustrates the cloud that shrouds accurate information pertaining to the 'great firewall'. liang, footnote . for a short article on the privacy commissioner's ruling see, http://www.reuters.com/article/worldnews/ idussha . pseudonymity. chinese officials have recently begun to crack down on weblog and chatroom use, introducing a host of new regulations directly targeted at information deemed harmful to chinese society. china's filtering/anti-spam technology has likewise greatly evolved so that spam has become a less effective means of communicating information. those who continue to engage in the exchange of banned communications, whether it be via spam, weblogs, text message or other fora, potentially face criminal charges. as the regulations are written in the traditional fashion of ambiguously overbroad provisions, the reality is that merely opening a spam message known to contain harmful material, or forwarding the message could be a contravention of the law. is it possible to route around censorship in china? circumventing the 'great chinese firewall' is achievable using a number of different methods which range from the use of web proxies (tor, anonymizer, dynapass, psiphon) to accessing the internet in peak times (state surveillance requires a large amount of bandwidth), to the use of encryption services. proxies such as tor may still be blocked at the node level (although currently they are not). while state surveillance requires large amounts of bandwidth, the threat of legal and economic sanction plus self-censorship -isps restricting access to potentially contentious sites, cybercafe´s and universities discouraging banned websurfing, individuals refraining from accessing even potentially illegal material -effectively fills the gap left by technological constraints. the use of encryption is able to circumvent filtering and keyword sniffing technology at the router level, but this does not provide a safety net for those wishing to disseminate contraband information. as stated previously, isps must and do comply with requests to disclose personal information. many isps have also built censorship functions into their encryption technology. activists using the encrypted skype technology, for example, have been cautioned against its use due to built-in censorship functions. encrypted messages may arouse further suspicion which may lead from monitoring of general data traffic over the internet to the surveillance of specific individuals and groups. regardless of the method employed, the threat of criminal sanction is always a possibility. the ability to use the internet to publish sanctioned information is a risky proposition. assuming that there are strong ethical arguments in favour of disseminating sanctioned information in times of public health crises, a new mechanism will be required for large-scale information delivery. the benevolent health worm provides one possible solution. a benevolent computer worm is a form of malware. malware typically includes the following types of computers programs: virus, worm, trojan horse, spyware, adware, spam, bot/agent, zombie, exploit, bug, keylogging and so forth. malware refers to computer software which either acts maliciously or whose effects are malicious -the two are not necessarily synonymous. in a wider context, malicious would extend to any type of computer code installed without consent regardless if any damage occurs to the computer. the theory is that the malicious component encompasses the use of bandwidth and again, that there is no consent. the idea of a benevolent virus or worm is not novel. early research and debate focused on the use of a worm to patch existing security flaws in software. expressed more precisely by leading it expert bruce schneier, ''patching other people's machines without annoying them is good; patching other people's machines without their consent is not … viral propagation mechanisms are inherently bad, and giving them beneficial payloads doesn't make things better.'' under this definition, no malware could be construed as benevolent. the weakness of this argument is that its discussion has been limited to patching and similar e-commerce activities, where consent is desirable from a corporate ethics perspective and is necessary in order to conclude a binding legal contract. missing from this discussion is the application of a benevolent worm outside of the e-commerce realm, along with the discussion of the difference between consent and informed consent, the latter being the legal requirement in most jurisdictions. the subject of informed consent in the digital age is contentious. it has been argued that consent is given in most internet applications through checking the ''i agree'' button of end-user license agreements and privacy policy statements. this is not representative of informed consent. most users do not read end-user-license agreements (eula). when they do, such licenses contain onerous obligations unilaterally imposed on them expressed in complex, aggressive legal rhetoric -most of these types of terms remain untested in law and run against the basic tenants of the law of contracts, namely consideration, meeting of the minds, and adequate notice of change of terms. this is perhaps best illustrated by way of example. many corporations, such as sony, release products with an end-user license term authorizing them to utilize rootkits, backdoors and digital rights management systems for a variety of unspecified purposes, all of which may be subject to change without notification to the user. the rootkits, in turn, render computers vulnerable to intruders to install malicious applications onto their computers. digital rights management systems allow monitoring devices which track the use of a work (for example, a music c.d.), which could theoretically be used as evidence to bring legal suits against those who make illegal use of the copyrighted work. the author uses the example of consent to illustrate the discrepancy between tangential concepts of theory and practice. the author agrees that informed consent is a desirable feature in software distribution mechanisms. concluding that consent is required in all contexts is to prematurely rule on an issue which has, so far, only been discussed in the limited context of electronic commerce. if consent is gained, do benevolent payloads become ethical? if there is no consent, are benevolent worms precluded from becoming ethical? it appears as though the debate on consent and malware has inherited the intellectual baggage of assumptions surrounding consent. nowhere is this better articulated than in the famous essay by robin west, ''authority, autonomy, and choice: the role of consent in the moral and political visions of franz kafka and richard posner.'' west exposes the fallacy in posner's theory that choice and consent in a legal system allow for an increase both in morality and autonomy. within the confines of benevolent payloads, there is an assumption that lack of consent is inherently bad or unethical contrasted with acts where a vague notion of consent is obtained, thereby magically summoning the requisites of legal and ethical action. the presence of consent should be regarded as one component in an analysis of all factors contributing to an ethical framework. an effects analysis would look to whether any tangible damage, other than use of bandwidth, has been done to the computer, webserver or user, or in the event that other types of damage are sustained, whether there are compelling reasons to derogate from the principles of user consent and avoiding damaging third party property. more importantly, an effects analysis would address the issues of when it is permissible to utilise bandwidth and install software on a user's computer without their consent. when, if ever, does a benevolent payload become permissible or mandatory as a moral duty? in a larger sense, the issue is one of normative ethics drawing on effects-based analysis in consequentialism with that of moral duty in deontology. while a robust examination of types of malware is not required to understand the benevolent health worm, a basic understanding of the differences between a virus and worm is essential, as the underlying technology of a worm alleviates some of the ethical and legal issues for its intended benevolent use. a virus is a ''block of code that inserts copies of itself into other programs''. viruses generally require a positive act by the user to activate the virus. such a positive act would include opening an email or attachment containing the virus. viruses often delay or hinder the performance of functions on a computer, and may infect other software programs. they do not, however, propagate copies of themselves over networks. again, a positive act is required for both infection and propagation. a worm is a program that propagates copies of itself over networks. it does not infect other programs nor does it require a positive act by the user to activate the worm. in this sense, it is selfreplicating. irrespective of the characterization nearly all computer viruses and worms infect either software or hard-drives without the authorization of the computer owner. similarly, all computer viruses and worms utilize bandwidth imposing a strain on traffic and resource demands. all computer viruses and worms may inadvertently cause unexpected damage to a computer system and may contain bugs. the benevolent health worm is no exception. there are ways to minimize damaging effects of the worm through technical design. such elements include: ( ) slow-spreading, ( ) utilize geo-location technology to limit its propagation within a region (''.cn'' and its equivalent for the internationalized domain name in chinese characters), ( ) installation of short and reasonable shut-down mechanisms to avoid perpetual replication, ( ) low-demand bandwidth, and ( ) undergo professional debugging standards. the benevolent health worm would be an information delivery method with worm-like characteristics. a computer worm is a self-replicating computer program containing a tailor-designed payload. the payload would be programmed to spread from computer to computer in china with the specific function of displaying the information in a pop-up window, or override a user's default web page with one displaying information. in the case of the benevolent health worm, the message would contain vital information relating to a public health crisis otherwise unavailable through traditional media sources. the information would ideally come from a trusted source containing accurate and truthful information (see discussion in following sections). the payload would be carefully programmed to prevent any deliberate or positive technical action by the recipient. the recipient would, therefore, have no knowledge or control of the worm. the latter points require elaboration. the selfreplication method of worms is ideal in this situation as it is only the infected computer which takes part in the dissemination of sanctioned information; the person whose computer is infected is technically prohibited from any deliberate or accidental positive acts, and has no control or knowledge of the worm. in order to achieve this, the pop-up message generated by the worm must have the following features: • must not be a virus in that it must be self-replicating, • not contain links to additional sources of information, • the user would not be able to save the information to his or her computer, • the user could not forward the message to others, and • the information would disappear from the system altogether after a specified amount of time. in the case of the latter, the pop-up would appear for a specified time (e.g. min) and re-appear each time a person turned on their computer for a programmed length of time (e.g. weeks). at the end of a short period of time (e.g. weeks) the worm would completely disappear from the user's computer systemall technically feasible through the programming of the payload. these features greatly reduce if not eliminate any risk to the recipients of the information. all elements necessary to prove a criminal act are removed: positive act, knowledge or foreseeable knowledge, mens rea, and motive. a chief criticism of the use of viruses and worm for benevolent purposes is that there are safer alternative means of achieving the same goal. the same cannot be true with the benevolent worm. alternative means of health distribution would include: illegal news reporting; illegal dissemination of news domestically through a blog, chatroom or spam; spam techniques from a foreign jurisdiction; and access to materials outside of china through anonymizing technologies such as web proxies. a common flaw of these methods is the necessity of a positive act by both the sender and recipient of information. this is especially so for the first two means. a positive act, whether it is through technical (e.g. virus) or manual (e.g. forward spam message) would allow for the possibility of dual criminal charges. meanwhile there are further challenges with the latter two distribution means of foreign spam and web proxies. as pre-eminent human rights activist sharon om has noted, human rights spamming lists are potentially illegal under the united states can spam act. the use of anonymizing technologies such as web proxies is by no means fool proof. such technologies are capable of being blocked (policy choice not a technical feat), even trust-enabled web proxies such as psiphon. in the case of the benevolent health worm, only the sender of the information would perform a positive act. these acts would still be illicit on many fronts but the sender would bear a greater risk than the recipient. the programmer of the human rights worm will be in violation of computer misuse law. in the event that the computer programmer is not necessarily the person or group who distributes the worm, those individuals responsible for ''letting the worm loose'' could face criminal and civil charges. finally, the authors of the actual information appearing in the pop-up screen may be charged with a number of criminal acts including state secret and possibly the new law on disclosure of non-authorized news on 'sudden events.' positive acts are performed by those actors along the sender chain while recipients of information remain removed from the process short of reading the content displayed in the window. stated another way, the benevolent worm potentially offers a way to restore an individual's right to physical and mental well-being through a method that reduces the risk of persecution for those who disseminate un-authorized information and reduces and possibly removes the risk for those who receive the information. the above scenarios, however, envision the propagation of the worm and information writing to be performed by individuals within china. such risks could be greatly reduced by creating the worm outside of china. while it is true that computer misuse is illegal in most jurisdictions, the threat of sanction depends greatly on political will. with open american support of projects which address human rights and democracy in suppressed regions, congressional hearings on internet censorship in china and, more specifically, us corporate compliance and aid in censorship; and the passing of the global internet freedom act, it is hard to believe that, at least in the united states, that there would be political will to prevent the benevolent worm. if anything, there may be available funding. there is a strong psychological and political element to the creator (and disseminator) of the worm. a worm created inside china would have the distinct advantage of appearing to be change from within; a worm created outside china raises issues of external meddling, sovereignty, imperialism, or worse yet, information warfare. these issues will be more fully integrated into the ethical discussion below. the ethical dimensions of the benevolent worm encompass several layers. a more sophisticated approach would be to treat the layers as information branches in the total infosphere. for the purpose of this paper i will adopt a simpler approach referring to the author/producer, sender/distributor, recipient, content, delivery method and medium of communication. one great concern in the propagation of public health information through a computer worm is that of trusted source. trusted sources may be divided into two groups. the first involves the content of the information. the second relates to the information producers -authors and distributor. the 'who' in 'who says what' may be more important than the 'what'. in this sense, a worm released by a national state could be construed as interference with sovereignty and may not carry the same weight as a worm released by a trusted ngo working in the region. the reality, however, is that there is no full-proof method between distinguishing between a trusted and deceptive source. all electronic commerce applications suffer from the same ambiguity of trusted and deceptive sources. the following analysis, therefore, assumes that it is possible to utilize trusted sources. indeed it may be a great leap of faith. the analysis further assumes that the issue of public health endemics is sufficiently grave to warrant deviation from traditional paths of information dissemination (as discussed in the previous sections). western-based rights treatises, in particular human rights frameworks, may provide some justification of a benevolent health worm. human rights or civil liberties frameworks operate on two theoretical models. the first is one related to public international law where states bind themselves to legal obligations contained in treaties. the second involves the universality principle of human rights based on moral rights as opposed to legal rights. a similar dichotomy is to distinguish between what is legal and what is legitimate. the law -that is, what is legal -is premised on the notion that there, ''is a system of enforceable rules governing social relations and legislated by a political system.'' breach of a rule results in an activity being classified as illegal. legitimacy in this context is used in its broadest sense to reflect what is moral which need not be legal. discussions of morality are naturally dependent on the framework of the analysis. the notion of morality as seen through the lens of natural law (e.g. aquinas, aristotle, religion), would likely differ from morality as seen through the lens of normative jurisprudence (e.g. virtue ethics, utilitarianism, deontology). the moral or ethical framework shapes the debate. this section of the paper does not use a specific ethical framework such as deontology to discuss issues of human rights. to the extent that any differentiation is made, there is some delineation between what is legal versus legitimate. the debate will mostly inter-mingle notions of legality with those of legitimacy, as well as notions of binding public international law with the universal principle of human rights. the aim is to initiate ethical debate on the subject, not to fully justify the use of a benevolent worm. l. floridi. information ethics, its nature and scope. in j. van under a legal rights based theory, specific rights and obligations are only provided to the extent of treaty provisions in international law. such rights may or may not be entrenched in domestic/national law. where rights are protected under international law, they may contradict and clash with domestic law. the nexus between national and international law has been discussed using the theories of dualism and monism. as the honorable justice kirby writes: ''for the monist, international law is simply part of the law of the land, together with the more familiar areas of national law. dualists, on the other hand, assert that there are two essentially different legal systems. they exist ''side by side within different spheres of action -the international plane and the domestic plane. '' the clash between national and international law is influenced by whether a court adopts a monist or dualist position. the chinese government and courts use a dualist theory where human rights are viewed as a matter of 'foreign affairs.' as one human rights expert writes, ''the chinese government essentially views these obligations as a matter of foreign affairs, and seeks to insulate the domestic arena from the reach of international human rights law, both in symbolic and practical terms. '' international tribunals and courts also adopt a dualist approach. national law is treated as a fact. an obligation in international law cannot be avoided or excused due to a clash with domestic/national law. other governments and courts adopt a monist approach. this can be seen in the erosion of the dualist approach in many countries such as australia and canada. there have been many court decisions which integrate international law principles into the national landscape. the second level relates to the universality of human rights. universality is not a legal proposition but a moral one; that human rights are naturally acquired at birth regardless of the area of the world where you reside. human rights subsist regardless of international and domestic legal obligations. regardless of the interpretation of human rights, the benevolent health worm represents undisputed legal and moral rights which may be stated in a simple form: everyone has the right of freedom of expression, and the right to the enjoyment of the highest attainable standard of physical and mental health. these rights are legally protected in a number of international, regional, and united nations treaties to which china is party, and, according to the model of human rights one adheres to, are inherently entrenched regardless of the law. the constitution for the people's republic of china (prc) recognizes ''freedom of speech'', however, the concept of free speech is viewed differently in china than in western democracies. reed, an expert on freedom of expression in china, notes: ''the prc believes that rights are only instruments for realizing state objectives. individual rights are merely residual freedoms found within the confines of the law. if necessary, all rights must be sacrificed for the good of the common collective. as a result, china traditionally keeps the dissemination of information and freedom of expression to a minimum. the ccp controls all facets of government, including the freedom of expression granted in the constitution.'' several distinct questions surface as a result of the above passage. is china within its legitimate sovereign right to censor free speech on public health issues on the grounds that such discourse falls under the exemption of ''national security''? is civil disobedience justified in the context of disobeying the law for a higher purpose whether it be construed as a moral obligation or interpreted as for the greater good of the community (emphasis here on worm created within china)? would a worm created outside of china be a deliberate act of interference with a nation's sovereignty? under what circumstances might the benevolent worm be construed as part of information warfare? public health as 'national security' threat according to one view, national security always trumps individual rights because security, on a hobbesian-type view, is necessary for a peaceful society in which persons can enjoy their liberty and rights. this view appears to be gaining adherents at least among legislators. on the moderate viewpoint, free speech rights are defeasible, but only when appropriate justification for m. kirby censorship is available. in order to protect free speech rights, legislative limitations on censorship powers are necessary. in a rights-respecting society, balancing involves prioritizing different rights in the case of conflicts between prima facie rights (e.g. freedom of expression may conflict with freedom of religion). in the case of liberal democracies, there should be strong limitations on violations of freedom of expression and liberty. on the other hand, as we have seen in the quotation from reed, above, from the perspective of the prc, rights are merely instrumental to the common good, and balancing rights can be done by determining what maximizes the common good. rights can be overridden whenever the common good requires. under public international law, governments are allowed to restrict the free flow of information to protect interests such as national security or public morals. national security ideology has, however, been used by authorities to justify human rights infringement. for this reason, international documents and principles were developed to keep rights exemptions confined to narrow determinations. for example, the johannesburg principles on national security, freedom of expression and access to information adopt a standard whereby freedom of expression and access to information may only be restricted where a number of conditions are met: prescribed by law, protects a legitimate national security interest, and is necessary in a democratic society. for example, a legitimate national security interest is incitement to violent overthrow of a government. national security restrictions are not justifiable in the case, for example, of ''embarrassment or exposure of wrongdoing, or to conceal information about he functioning of its public institutions …'' (principle (b)). china's restrictions on free speech and access to information clearly do not adhere to the johannesburg principles or other international standards for protecting the right to information. as human rights watch notes, ''prior censorship in particular is severely disfavoured in international law, and not permitted in many constitutional systems. a decision to block access to online material should be subject to the highest level of scrutiny, with a burden on the government to demonstrate that censorship would effectively avert a threat of irreparable, imminent, and weighty harms, and that less extreme measures are unavailable as alternatives to protect the state interest at issue. at present, it seems apparent that china engages in no such scrutiny … '' moreover, the decision to punish certain speakers merely for exercising their right to speak frankly online (or off) is arbitrary and unpredictable with no opportunity for an individual or group to know in advance whether their actions comport with the law. the inability to comply with the law based on a lack of transparency severely undermines any attempt to posit a law as moral. morality here is used in a procedural sense along the line of natural law theorist, lon fuller's essential principle that the law should be written and promulgated in a fashion which guides behaviour. conscientious objection and civil disobedience i will refer to two general types of civil disobedience. the first is better known as 'conscientious objection' where the moral agent performs or abstains from performing an act to preserve the agent's own moral integrity. there is only the duty to obey a just law. fashioned in a similar vein, there may in fact be a moral obligation to disobey an unjust law. in the case of the benevolent worm, a number of parties including the author/producer, and sender/distributor may feel morally compelled to send what they believe is vital, accurate health information otherwise not available within china. conjecturing on the moral agent's dilemma, the agent is motivated to break the law in order to achieve a number of possible goals such as informing the populace of important news related to the epidemic and to encourage behaviour associated with containing the disease in question. g. hagen and a. maurushat, surveillance, technology, and national security: issues in civil liberties, course materials. asia-america institute in transnational law, . available at http://hei.unige.ch/humanrts/instree/ johannesburg.html. human rights watch, footnote . martin luther king junior, for example, writes: ''how can you advocate breaking some laws and obeying others?'' the answer is found in the fact that there are two types of laws: there are just and there are unjust laws. i would agree with saint augustine that ''an unjust law is no law at all.'' ''… one who breaks an unjust law must do it openly, lovingly … and with a willingness to accept the penalty. i submit that an individual who breaks a law that conscience tells him is unjust, and willingly accepts the penalty by staying in jail to arouse the conscience of the community over its injustice, is in reality expressing the very highest respect for law.' ' ( ) the other type of disobedience, on the other hand, is known as 'civil disobedience' in the sense that it, ''is conscientious disobedience of the law directed primarily … at bringing about a change in a law, policy, institution that is morally unjust or otherwise morally unacceptable ... or a law which may be acceptable in itself but which is disobeyed in order to protest against the offending law.'' the moral agent, in this instance, is motivated to affect change in the law. in the case of the benevolent worm, however, this would likely be a possible ancillary effect rather than a primary goal. conscientious objection and civil disobedience have been justified on a number of grounds. one thought is that disobedience of the law may be justified where there is no disrespect or harm to others. another ground speaks to a utilitarian approach of bringing about useful reconsideration of public policy, respect for human rights, interests of minorities and disadvantages groups, and actual reform of the law. it has been shown that other methods such as news reporting and spam potentially create harm not only for the sender but also for the recipient of sanctioned health information. the benevolent worm has the goal of safely minimizing harm to the sender and attempts to eliminate harm to the recipient (realizing, of course, that unintended consequences are not always foreseeable). many philosophers have specified that justifiable civil disobedience ought to be non-violent with the agent ready and willing to accept punishment as a consequence for breaking the law. this view seeks to disassociate civil disobedience from revolutionary disobedience. the author suggests that this dichotomy is more useful in a democratic state with rule of law whose political leaders and citizens first have respect for their constitutions and second for human rights in general. the dichotomy, therefore, seems less appropriate for autocratic states with documented histories of human rights abuse. would a worm created outside of china be a deliberate act of interference with a nation's sovereignty? the answer to this question may lie in the meaning of sovereignty. in modern international law the notion of sovereignty is ''people's sovereignty rather than the sovereign's sovereignty … [whereby] no serious scholar still supports the contention that internal human rights are ''essentially within the domestic jurisdiction of any state'' and hence insulated from international law.'' the notion of sovereignty in human rights is, therefore, predominantly premised on democracy and rule of law. china is not a democratic nation adherent to the rule of law. it does not, however, follow that china is not entitled to sovereignty but rather, that issues of sovereignty are burdened with additional questions. yet sovereignty has generally been understood as one nation interfering with another nation's legitimate right to runs its affairs. one thinks of the invasion of iraq and not generally of information on public health endemics. sovereignty issues may be affected by the 'who' in 'who says what'. a worm released by the canadian government, for example, could conceivably be construed as intentional sovereign interference. a worm released by a ngo, on the other hand, would be less likely to be perceived as sovereign interference; this would be further buttressed by a trusted ngo with strong links to china. in an extreme circumstance the benevolent worm might be construed as part of information warfare (iw). defined simplistically, information warfare refers to, ''actions taken to affect an adversary's information and information systems while defending one's own information and information systems. '' there are six broad components to iw: physical attack/destruction, electronic warfare, computer network attack, military deception, psychological operations, and operations security. it is difficult to conceive how the benevolent worm in its described application in this paper would fit into any one of these categories. one cannot, however, rule out the possibility of a worm with false and potentially harmful information concerning an epidemic to be released as part of an overall iw strategy. a strategy of disinformation, however, is applicable in a number of contexts including conventional means of information dissemination such as false news reporting, spam, and so forth. careful attention to trusted sources could reduce the risk of the worm being perceived as iw. at its most base conception, 'asian values' emphasize the community as opposed to the individual or self. it has been argued that human rights are incompatible with 'asian values'. expressed more poignantly by samuel huntington: the traditionally prevailing values in east asia have differed fundamentally from those in the west and, by western standards, they are not favourable to democratic development. confucian culture and its variants emphasize the supremacy of the group over the individual, authority over liberty and responsibility over rights. expressed somewhat differently, western human rights-based rhetoric focus on rights of an individual whereas eastern confucian moral philosophy focuses on the duties of an individual to the community. the following analysis places ethical debate on the benevolent worm in the confucian moral philosophy tradition. confucian moral philosophy is often referred to as a duty-based philosophy. confucian ethical teachings are grounded in five moral values: li (ritual), hsia (filial piety, duty to family), yi (righteousness), xin (honesty and trustworthiness), ren/jen (benevolence, social virtue, humaneness towards others), and chung (loyalty to the state). confucius' view of duty was not traditionally extended to all people but was limited to five relationships: ruler to subject, father to son, eldest brother to younger siblings, husband to wife, and friend to friend. there has never been a duty from human to human in traditional confucian thought. two contentious issues are raised in applying confucius' teachings to the benevolent health worm. first, the values of ren, benevolence towards others, may compete with that of chung, loyalty to the state. second, there is no general duty between humans outside of the five relationships. the value of chung requires a person to be loyal to the state but not at any cost. confucius writes, ''if a ruler's words be good, is it not also good that no one oppose them? but if they are not good, and no one opposes them, may there not be expected from this one sentence the ruin of his country?'' [confucius, the analects, book translated by legge]. the most important value as espoused by confucius was ren. a major component of ren involved individual self-cultivation in virtuous action. it has further been suggested that li -norms of social ritual and interaction -is a critical component in analysing ren. li is learned by socializing and interacting with persons who embody ren. as lai writes: the paradigmatic man is a creator of standards rather than a follower … and he possesses a keen sense of moral discrimination. moral achievement reaches its culmination in those who have attained the capacity to assess events and who, being attuned to li, embody a sense of rightness. good governance and social order were derived from a hierarchical chain of individual virtuous action thus it is written that, ''their hearts being rectified, their persons were cultivated. their persons being cultivated, their families were regulated. their families being regulated, their states were rightly governed. their states being rightly governed, the whole kingdom was made tranquil and happy'' [confucius, great learning, translated by legge] . what of the case where ren is not personally cultivated leading to poor governance? loyalty to the state is loyalty to a righteous government who has fulfilled its duties to its citizens in the spirit of ren; loyalty to the state has never been an absolute. by no means does the author suggest that the overall governance of china has been poor under the current administration. china has had to face many problems that other nations, and in particular wealthy democratic nations, have never had to address: starvation, extreme poverty, territory occupation, a devastated economy, and population explosion to name but a few. while china has overcome many hurdles to better provide for its people, its record on factors contributing to human dignity is poor (freedom of expression, protection of minorities, access to important and timely information, and so forth). it is within this limited latter context of human dignity that it is conceivable to characterize governance as poor. the manner in which public epidemics such as hiv/aids, sars and bird flu has been handled is evidence that the government has not fulfilled its duties to the extent required under the spirit of ren. this resembles the notion of obeying just laws, and being further obliged to disobey unjust laws. the formation of a person's character through virtuous action is strongly tied to a sense of community and to one's role in a community. for this similarly, extension of duties beyond the classic five relationships has also been newly interpreted. it could be said that certain members of society may have the duty to disclose information on epidemics which could save lives, reduce the spread of the infectious disease, and perhaps altogether avoid a disease reaching the level of epidemic. certain societal members may include scholars, doctors, journalist, experts, ngos, and other international organizations. this bears a resemblance to justifications of the moral agent in conscientious disobedience. the dissemination of vital information is potentially a virtuous act whether it is through direct means of an internet website, news publications, or whether it is less direct through a benevolent health worm. of course, the opposite could be argued for those who adhere to a traditional view of confucian moral philosophy. it becomes more difficult to justify the benevolent worm in the absence of one of the five relationships giving rise to duties. the most relevant relationship to the benevolent worm is that of ruler and subject. the ruler-subject relationship is predicated on the subject's respect for the ruler. this is similar to chung, loyalty to the state. again, there is the assumption that the ruler will have ren, and will act as an example to his subjects. in the absence of ren, the subject duty to obey the ruler is lessened. regardless of the framework adopted, the use of a benevolent worm does not seem to represent any clear abdication from confucian moral philosophy. water holds up the boat; water may also sink the boat. emperor taizong ( - , tang dynasty) in much the same way, benevolent payloads have the potential to be destructive. they also have the potential to be beneficial. benevolent payloads have in the past been analysed in the context of patches and e-commerce. conclusion has been reached in the wider technological community that benevolent payloads are simply a 'bad idea' because there is no consent, and there are safer methods available. there has been no analysis of benevolent payloads outside of the electronic commerce context. the benevolent health worm provides an interesting case study which undermines and challenges many of the ethical issues of benevolent payloads. this article has attempted to untangle many of the complex ethical issues surrounding the benevolent health worm, and benevolent payloads in general. good' worms and human rights collectivism and constitutions: press systems in china and japan human rights: beyond the liberal vision confucian values and the internet: a potential conflict computer viruses still a bad idea? moral autonomy you've got dissent!: chinese dissident use of the internet and beijing's counter-strategies. rand the limits of official tolerance: the case of aizhixing china's media censorship the analects of confucius. ebooks@adelaide the great learning. ebooks@adelaide a comparison between the ethics of socrates and confucius confucius -the secular as sacred information ethics, its nature and scope will the boat sink the water? public affairs will the boat sink the water? public affairs using predators to combat worms and viruses: a simulation-based study annual computer security applications conference anonymous and malicious human rights and spam: a china case study counter-revolutionaries, subversives, and terrorists: china's evolving national security law. in national security and fundamental freedoms: hong kong's article under scrutiny american democracy in relation to asia: democracy and capitalism: asian and american perspectives bird flu -what is china hiding? battling sars: china's silence costs lives. international herald tribune the growing rapprochment between international law and national law learning from chinese philosophies: ethics of interdependent and contextualised self liberal rights or/and confucian virtues? has china achieved its goals through the internet regulations moral judgment, historical reality, and civil disobedience censorship: a world encyclopedia conscientious disobedience of the law: its necessity, justification, and problems to which it gives rise democracy and confucian values international human rights and humanitarian law new regulations in china target foreign media a theory of justice (rev from the great firewall of china to the berlin firewall: the cost of content regulation on internet commerce reporters without borders. government turns deaf ear to call for ching cheong's release liberalism and the limits of justice benevolent worms, crypto-gram newsletter aids in china: an annotated chronology confucian ethics: a comparative study of self, autonomy and community virtue ethics, the analects, and the problem of commensurability law and ideology the bioethical principles and confucius' moral philosophy learning, and politics: essays on the confucian intellectual. suny is china hiding avian influenza? human rights as ''foreign affairs'': china's reporting under human rights treaties world health organization. who urges member states to be prepared for a pandemic chinese information warfare: a phantom menace or emerging threat? the strategic studies institute media censorship in china'' council on foreign relations key: cord- - szg nbu authors: xie, tiantian; tan, tao; li, jun title: an extensive search trends-based analysis of public attention on social media in the early outbreak of covid- in china date: - - journal: risk manag healthc policy doi: . /rmhp.s sha: doc_id: cord_uid: szg nbu background: a novel coronavirus (covid- ) caused pneumonia broke out at the end of in wuhan, china. many cases were subsequently reported in other cities, which has aroused strong reverberations on the internet and social media around the world. objective: the aim of this study was to investigate the reaction of global internet users to the outbreak of covid- by evaluating the possibility of using internet monitoring as an instrument in handling communicable diseases and responding to public health emergencies. methods: the disease-related data were retrieved from china’s national health commission (cnhc) and world health organization (who) from january to february , . daily google trends (gt) and daily baidu attention index (bai) for the keyword “coronavirus” were collected from their official websites. rumors which occurred in the course of this outbreak were mined from chinese national platform to refute rumors (cnprr) and tencent platform to refute rumors (tprr). kendall’s tau-b rank test was applied to check the bivariate correlation among the two indexes mentioned above, epidemic trends, and rumors. results: after the outbreak of covid- , both daily bai and daily gt increased rapidly and remained at a high level, this process lasted about days. when major events occurred, daily bai, daily gt, and the number of rumors simultaneously reached new peaks. our study indicates that these indexes and rumors are statistically related to disease-related indicators. information symmetry was also found to help significantly eliminate the false news and to prevent rumors from spreading across social media through the epidemic outbreak. conclusion: compared to traditional methods, internet monitoring could be particularly efficient and economical in the prevention and control of epidemic and rumors by reflecting public attention and attitude, especially in the early period of an outbreak. since december , several cases of pneumonia of unknown causes have been reported in wuhan. within just a few months, the novel coronavirus has developed into a global pandemic. as of march , , over , confirmed cases were reported worldwide, of which , were identified in china. immediately after the outbreak in wuhan, several exported cases were confirmed in south korea, japan, thailand, iran, and the eu due to the high transmissivity of the covid- . the most common symptoms of this epidemic were onset of fever, generalized weakness, and dry cough, with clinical presentations greatly resembling viral pneumonia. the incubation period of the virus in the human body is generally - days (within days). this new type of coronavirus is not as fatal as people thought based on the current data. at present, the mortality of covid- in china is . % according to the situation report by china's national health commission (cnhc), compared with . % for sars and . % for mers reported by the world health organization (who). , the covid- pandemic has generated a huge impact on social media behaviors across the world internet community, in particular at the early outbreak in china when global attention was focused on the situation in wuhan and hubei province, the epicenter of the epidemic outbreak. china has exhibited fast internet growth in recent years. according to the latest version of the china internet development statistical report published by china internet network information center (cnnic), there were million internet users ( . % of the total population) and million cell phone users as of june . meanwhile, social media has become the primary source for the vast majority of internet users in accessing news and acquiring real-time information in china and across the world. due to the expansion of wi-fi coverage, the maturity of fourth generation ( g) networks and the popular growth of fifth generation ( g) networks, chinese internet users spent on average . hours per week online. at the global level, there are billion netizens (over % of the whole population) according to the latest internet trends report by mary meeker; the internet penetration is % in asia pacific, % in europe, % in africa and the middle east, % in latin america and the caribbean, and % in north america. with the increase in global internet penetration, people are used to obtaining information and expressing their opinions online. participation and attention of internet users can to some extent represent the public reactions and attitudes to an event. as the leverage of the world's netizens is increasing, public opinion plays an important role in the appeal and participation of many public emergencies, for instance, people spontaneously started using social media to express their dissatisfaction and organize protest marches in france's yellow vests movement in . in recent years, internet search trends have been widely used in scientific researches in european and american countries. there were many previous researches based around these data, for instance, forecasting of tourism demand, prediction of unemployment, monitoring of public health and major mental illness. [ ] [ ] [ ] monitoring public reactions through internet and major social media platforms (twitter and facebook) has become a popular topic, especially in the prevention and control of some infectious diseases, such as measles, ebola, and other diseases. [ ] [ ] [ ] similar to the situation in the regions mentioned above, researchers in many asian countries have also verified that internet search trends and social media data could be considered as an important and effective way for the assessment of public attention, risk perception, and behavioral responses to the epidemic outbreaks, since the outbreak of sars in to the outbreak of covid- . [ ] [ ] [ ] the reaction of netizens can generally represent the public response to the emergency, and people in different countries often have different concerns, which provided us a useful tool to better understand the public concerns as well as to prevent and control the epidemic outbreaks. [ ] [ ] [ ] rational public opinion and correct information transmission from the authorities and authoritative experts can enhance persons' understanding of the outbreaks and relieve their panic. however, the transmission of much contradictory and false information may cause an information overload as well as confusion, which will arouse fear and anxiety that could in turn nourish rumors. for example, it is difficult for people to understand whether face masks can be reused during the epidemic outbreaks, since a thousand posts have a thousand answers. [ ] [ ] [ ] today, people get information in different ways, but web search engines are a common choice for internet users around the world. baidu is the most popular search engine in china, . % of internet users have reported to prefer baidu as their default search engine according to the latest chinese internet search engine usage research report. however, in other countries and regions outside china, people choose google with priority selection incidence of . %, according to the latest search engine market share worldwide report. based on quantitative analyses with data mined from both baidu and google indexes, this study aims to argue that internet monitoring is a convenient and cost-effective way to assess public reactions, which can provide evidence to all governments and the public in the world to handle public health emergency problems in case of epidemic outbreaks. in order to examine the public attention to the outbreak of covid- , we utilize a search index which uses keywords as statistical objects to scientifically analyze and calculate the weighted sum of the search frequency in a specific area and time. in short, the more internet searches the higher the search index, which could illustrate the degree of public concern. baidu attention index (bai) and google trends (gt), two major search indexes, were searched from january to february , , for comparing with some diseaserelated factors. this outbreak has caused widespread concern in the world, especially on the internet. therefore, our study sample was the global netizens (internet users) during the study period of this coronavirus outbreak. through the correlation test, we found that in the early stage of the outbreak, correlation between the keyword "coronavirus" in chinese for bai and the epidemic trends were significant, while "new coronary pneumonia" and "epidemic" were not. similarly, we found that the keyword "coronavirus" for gt compared with other keywords such as "corona" or "covid- " can better reflect the public's concern about the epidemic situation during the study period. therefore, we gathered the bai with " coronavirus" in chinese as the keyword, from the web page of baidu index on march , at the same time, we collected the daily gt as a supplementary data source by using the keyword "coronavirus" from the web page of google trends. besides, we collected rumors identified and confirmed by the chinese authorities during the study period from chinese national platform to refute rumors (cnprr) and tencent platform to refute rumors (tprr). , the former is the chinese official rumor-refuting platform which made rumors verified and confirmed by the authority public, the latter was established by china's largest internet company tencent whose mission is to seek out misinformation by monitoring the top topics ranking on social platforms and mark them so as not to be reposted. the data used in this research are all officially published on the internet and do not contain any privacy information. to represent the public attention, we have graphed the curves of epidemic trends based on collected data. we compared the daily bai and daily gt over time with the disease-related data to explore the changes of public attention. it can be seen from the previous studies, both spearman's and kendall's rank correlation coefficient are appropriate to analyze the correlation between two continuous or discrete ordinal variables. , , the data used in this study is continuous but does not conform to the normal distribution, for this reason, the kendall's tau-b rank correlation coefficient was used to check the statistical correlation. the statistical significance was set at p< . . rumors during covid- outbreaks were classified by date and main content. these correlation analyses were performed by using the commercial software "statistical package for the social sciences" (spss package for windows, v . . , ibm corporation, armonk, ny, us). all data was checked for completeness and accuracy before analysis. on january , , wuhan municipal health commission (wmhc) initially reported cases of human infections with an unknown new coronavirus. afterwards, more cases were consecutively reported in other provinces. due to the approaching of chinese new year, there was a large population movement that promoted the spread of coronavirus. the number of cumulative confirmed cases was on january, and rose alarmingly to days later. subsequently, the chinese government announced the quarantine of wuhan and took a series of measures to reduce population movements. however, the number of infected patients has still increased dramatically. by february , , cases had been reported in provinces of china. by mid-february, the growth of confirmed cases was slowing down gradually and the cumulative case cure rate was continuously rising. by february , , cumulative confirmed cases and , cured cases (cumulative case cure rate= . %), and death cases (cumulative case fatality rate= . %) had been reported in china (figure ). overseas public attention to the outbreak of covid- in china at the beginning, the daily gt for keyword "coronavirus" remained stable. from january , it increased rapidly in the next days due to the outbreak of covid- in china. when the who declared this outbreak to be a public health emergency of international concern (pheic) on january , the daily gt reached a peak of the following day. subsequently, the daily gt gradually declined with small fluctuations until february . from february , coronavirus infections grew exponentially outside china, so that daily gt increased again and reached another peak of on february . we carried out the correlation analysis to investigate the relationship between disease-related indicators and daily gt for the rise and fall of public attention. the analysis illustrated a statistically significant positive correlation between the following sets of data: daily gt and cumulative confirmed cases (kendall's tau-b rank correlation p= . ) . in other words, daily gt increases as each of the indicators above increases. we also found that there was a negative relevant between daily gt and cumulative case cure rate (kendall's tau-b rank correlation coefficient=- . , p= . ). however, we found there was no correlation between the daily gt and cumulative mortality rate (kendall's tau-b rank correlation coefficient= . , p= . ). the results are graphedshown in figures and . in summary, daily gt was associated with multiple indicators, suggesting that overseas netizens were eager to have as much information as possible of covid- through the internet in the early stage of the outbreak in china. public attention of china mainland to the outbreak of covid- prior to january , , the information and news related to the novel coronavirus were not widely disseminated to the public, so the bai remained at a relatively low level. covid- began to spread in areas other than hubei province from january , and the daily bai for the chinese keyword increased sharply, peaking at , , on january . then, the daily bai decreased with fluctuations. as provinces began one by one to launch a first-level response to the major public health emergency, the daily bai remained at a high level of more than , , from january to january . afterwards, the daily bai steadily declined, falling below the median level of about , , on february . we conducted correlation analysis on daily bai and diseaserelated indicators. statistical correlation analysis showed that daily bai was positively correlated with new confirmed cases (kendall's tau-b rank correlation coefficient= . , p= . ) ( figure a ). besides, daily bai has a negative correlation with cumulative cases cure rate (kendall's tau-b rank correlation coefficient= . , p= . ) ( figure b ). it was also positively correlated with new death cases (kendall's tau-b rank correlation coefficient= . , p= . ) ( figure a ). we found there is no correlation between daily bai and other indicators, for example cumulative confirmed cases, new cured cases, cumulative cases cure rate, and cumulative case fatality rate. to summarize, daily bai generally showed a similar trend to daily gt but not exactly in the same way ( figure b ). daily bai was only correlated to three indicators (by comparison, seven indicators were correlated to daily gt). this showed that chinese people who were in the epicenter knew more about the epidemic situation, compared with foreign netizens, they just cared about the new infected cases and their cure rate. in other words, they focused on whether the epidemic was being effectively controlled. the epidemic-related rumors began to circulate as the covid- broke out and spread. from january to february , we collected a total of rumors from cnprr and tprr. we found most of them were spread through group chats and official accounts on wechat, the most popular messaging and social media app in china. in addition, sina weibo blog (chinese microblogging website) is another major site for the spread of rumors. of these rumors, . % ( / ) disseminated incorrect information on how to prevent and treat the new coronavirus; . % ( / ) were about the spread of infections in different cities, and among them, rumors related to wuhan. additionally, . % ( / ) were about the shortage of daily necessities and medical supplies such as food and masks, and . % ( / ) were spoofing in the name of professor nanshan zhong, who is not only head of china's covid- experts team but the key person in the successful fight against sars in . he was the first professor who suggested that covid- could be transmitted from person to person, people did believe that his opinions were authoritative and instructive. through the investigation of rumors, we discovered that the purpose of producing and spreading rumors is on the one hand to attract people's attention or more visits to rumor makers' homepages on social media. on the other hand, the uncertainty of covid- breeds anxiety, which on its own has been linked to rumor spreading. the result of statistical correlation analysis between rumors and disease-related indicators showed that the number of rumors has a significant positive correlation with daily bai (kendall's tau-b rank correlation coefficient= . , p= . ) ( figure a) . furthermore, the number of rumors was positively correlated with new confirmed cases (kendall's tau-b rank correlation coefficient= . , p= . ) ( figure b) . besides, the number of rumors also had a positive correlation with new death cases (kendall's tau-b rank correlation coefficient= . , p= . ) ( figure a ). however, the number of rumors was negatively correlated with cumulative case cure rate (kendall's tau-b rank correlation coefficient=− . , p= . ) ( figure b ). generally speaking, rumors of covid- outbreaks affected daily bai. the growth of new confirmed cases and new death cases seemed to increase the number of rumors. the severity of the epidemic situation also determined the number of rumors. fears and insecurity due to a lack of information at the initial outbreak stage led people to pay much attention to the epidemic, however, as the cure rate increased and more information was made available to the public, individuals were optimistic about the development of the epidemic, so their panic eased and the number of rumors decreased. internet-based analysis is a powerful tool in the new era. baidu, google, and other search engines are used by more and more scholars around the world to carry out academic research. , [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] in this research, we used internet search trends to probe public attention to covid- . this showed that both daily bai and daily gt began to rise sharply on january , after the outbreak, and reached a peak in a short period of time, then the high attention lasted approximately days. afterwards, both indexes have decreased because of the initial positive effect of epidemic control. it is worth mentioning that the daily gt rose again on february , because the epidemic began to spread widely in other countries. our study showed that in the early stages of covid- , public attention increased as new cases were confirmed, however it diminished as the cure rate rose. this suggests a rapidly spreading epidemic would promptly draw public attention which would decrease when its threat decreased. therefore, it is necessary for government to take appropriate measures and provide more information to address public's concerns as early as possible by monitoring the public reactions through the internet and social media. daily gt was directly proportional to the cumulative confirmed cases, cumulative cured cases, and cumulative death cases, but daily bai was not. the reason for this phenomenon is that the early outbreak of covid- was in china, so chinese netizens can naturally learn about the early news of the epidemic more quickly than those from other countries, figure (a) the daily bai for keyword "coronavirus" compared with new death cases from january to february , . (b) the daily bai for keyword "coronavirus" compared with daily gt for keyword "coronavirus" from january to february , . they did not have to deliberately search the internet for information, because they can easily get the latest data from tv, newspapers, community, and family. the outbreak of covid- was accompanied by a variety of rumors. according to our research, the number of rumors is positively correlated with new confirmed cases and new death cases, which also has the same trend with the growth of daily bai. furthermore, the emergence of breaking news can often make the index of rumors reach a peak. this indicates that uncertainty and severity of the new epidemic are major factors in encouraging rumors to which public attention is easily attracted. moreover, internet and social media have become a vehicle of transmitting rumors. the vast majority of rumors that we collected came from social media platforms such as wechat and weibo. it suggests that social media was a two-edged sword. although it can spread rumors, it was also a convenient tool for those refuting false messages if authoritative information could be released in time. in addition, if government invited experts who have received public attention to clarify misinformation during the early period of the outbreak, it would effectively help to dispel unnecessary panic and prevent the spread of rumors. an important policy component in the prevention and control of such infectious diseases as covid- is monitoring. traditional epidemic monitoring mainly relies on data provided by hospitals, medical institutions, and centers for disease control and prevention (cdc) at different levels. [ ] [ ] [ ] yet, this monitoring system has some deficiencies. firstly, the relevant data is collected and processed by each unit level by level, which may lead to a loss of important information during the process and a delay in acquiring analysis results. secondly, a traditional monitoring method consumes a lot of manpower and resources, however the data is rarely published and made available to the public. , internet monitoring has largely improved the data acquisition process and can overcome these deficiencies from three perspectives. first, billions of netizens learn about the latest epidemic situation through the internet. data collected in this way can represent the attention of most people. besides, the search trends were generated based on the daily search behaviors of the public, which can automatically record the public attention. compared with traditional methods such as epidemiological surveys and telephone interviews, [ ] [ ] [ ] [ ] [ ] this data acquisition method can significantly reduce the time and efforts. moreover, the data is always shared, and everyone can download it for free, which can maximize savings. our study has some limitations. for instance, data used in this research was mined from baidu and google, two giants in the search service market, which does not mean that the data from niche search engines are not important. similarly, for people who live in regions with low internet penetration rate, whose attention to the outbreak of covid- was ignored. our research shows that internet users in china and other countries have attached great attention to the covid- outbreak, but their focuses were slightly different. during the first -day period of the outbreak, public attention increased rapidly and remained at a high level. this was a key period for the governments to release relevant and authoritative information to minimize the spread of groundless rumors on social media. publishing accurate information through major social media platforms can quickly clarify rumors and prevent their further spread. in sum, internet monitoring is a quick and convenient way to reflect public attention, which will greatly enhance the ability of governments and the public to address public health emergencies. our study suggests that government should attach more importance to the monitoring of internet search trends platforms such as baidu attention index and google trends during the epidemic outbreaks. in particular, in the early stages of a new coronavirus outbreak, governments need to make proper use of the data of public attention, and take responsible measures in a timely manner for scientific popularization of fighting against the epidemic, in light of relieving people's concerns, and clarifying confusions and misunderstandings among the general public. risk management and healthcare policy is an international, peerreviewed, open access journal focusing on all aspects of public health, policy, and preventative measures to promote good health and improve morbidity and mortality in the population. the journal welcomes submitted papers covering original research, basic science, clinical & epidemiological studies, reviews and evaluations, guidelines, expert opinion and commentary, case reports and extended reports. the manuscript management system is completely online and includes a very quick and fair peer-review system, which is all easy to use. visit http://www.dovepress.com/testimonials.php to read real quotes from published authors. a novel coronavirus from patients with pneumonia in china who. novel coronavirus ( -ncov) situation report- clinical features of cases with coronavirus disease daily situation report of covid- summary of probable sars cases with onset of illness from statistical report on internet development in china. china internet network information center internet trends france's yellow vests: a self-mobilised mass movement with insurrectionist overtones forecasting tourism demand with composite search index predicting unemployment in short samples with internet job search query data monitoring a toxicological outbreak using internet search query data do seasons have an influence on the incidence of depression? the use of an internet search engine query data as a proxy of human affect seasonal trends in restless legs symptomatology: evidence from internet search query data disease detection or public opinion reflection? content analysis of tweets, other social media, and online newspapers during the measles outbreak in the netherlands in ebola virus disease and social media: a systematic review infodemiology and infoveillance: tracking online health information and cyberbehavior for public health infoveillance: framework for an emerging set of public health informatics methods to analyze search, communication and publication behavior on the internet the review on public health media surveillance and risk research the rumouring of sars during the epidemic in china assessment of public attention, risk perception, emotional and behavioural responses to the covid- outbreak: social media surveillance in china. risk perception, emot behav responses to covid- outbreak soc media surveill china the covid- risk perception: a survey on socioeconomics and media attention using twitter and web news mining to predict covid- outbreak importance of internet surveillance in public health emergency control and prevention: evidence from a digital epidemiologic study during avian influenza a h n outbreaks the dark side of information: overload, anxiety and other paradoxes and pathologies rumor surveillance and avian influenza h n rumor-related and exclusive behavior coverage in internet news reports following the h n influenza outbreak in japan rumors of disease in the global village: outbreak verification chinese search engine market research report. china internet network information center search engine market share worldwide report the continuing -ncov epidemic threat of novel coronaviruses to global health -the latest novel coronavirus outbreak in wuhan, china a novel coronavirus outbreak of global health concern chinese national platform to refute rumors tprr tencent platform to refute rumors chinese public attention to the outbreak of ebola in west africa: evidence from the online big data platform assessing ebola-related web search behaviour: insights and implications from an analytical study of google trends-based query volumes detail/ - - -statement-on-the-second-meeting-ofthe-international-health-regulations-( )-emergency-committeeregarding-the-outbreak-of-novel-coronavirus-( -ncov) combating rumor spread on social media: the effectiveness of refutation and warning dengue baidu search index data can improve the prediction of local dengue epidemic: a case study in guangzhou, china early detection of an epidemic erythromelalgia outbreak using baidu search data evaluation of internet-based dengue query data: google dengue trends predicting the present with google trends predicting tick-borne encephalitis using google trends using google hw. trends and ambient temperature to predict seasonal influenza outbreaks using the baidu search index to predict the incidence of hiv impacts of media coverage on the community stress level in hong kong after the tsunami on perceived risk, anxiety, and behavioural responses of the general public during the early phase of the influenza a (h n ) pandemic in the netherlands: results of three consecutive online surveys public perceptions, anxiety, and behaviour change in relation to the swine flu outbreak: cross sectional telephone survey widespread public misconception in the early phase of the h n influenza epidemic the impact of communications about swine flu (influenza a hiniv) on public responses to the outbreak: results from national telephone surveys in the uk all authors made substantial contributions to conception and design, acquisition, analysis, and interpretation of data; took part in drafting the article and revising it critically for important intellectual content; gave final approval of the version to be published; and agree to be accountable for all aspects of the work. this research was funded by china scholarship council, grant number . the authors have no conflicts of interest with any individuals or organizations. key: cord- -ph vrba authors: de’, rahul; pandey, neena; pal, abhipsa title: impact of digital surge during covid- pandemic: a viewpoint on research and practice date: - - journal: int j inf manage doi: . /j.ijinfomgt. . sha: doc_id: cord_uid: ph vrba the covid- pandemic has led to an inevitable surge in the use of digital technologies due to the social distancing norms and nationwide lockdowns. people and organizations all over the world have had to adjust to new ways of work and life. we explore possible scenarios of the digital surge and the research issues that arise. an increase in digitalization is leading firms and educational institutions to shift to work-from-home (wfh). blockchain technology will become important and will entail research on design and regulations. gig workers and the gig economy is likely to increase in scale, raising questions of work allocation, collaboration, motivation, and aspects of work overload and presenteeism. workplace monitoring and technostress issues will become prominent with an increase in digital presence. online fraud is likely to grow, along with research on managing security. the regulation of the internet, a key resource, will be crucial post-pandemic. research may address the consequences and causes of the digital divide. further, the issues of net neutrality and zero-rating plans will merit scrutiny. a key research issue will also be the impact and consequences of internet shutdowns, frequently resorted to by countries. digital money, too, assumes importance in crisis situations and research will address their adoption, consequences, and mode. aspects of surveillance and privacy gain importance with increase digital usage. by late may , at the time of writing of this article, over countries and territories in the world were affected by the coronavirus pandemic. this included most urban clusters and even rural regions. with the spread of the pandemic, almost all regions have implemented lockdowns, shutting down activities that require human gathering and interactions -including colleges, schools, malls, temples, offices, airports, and railway stations. the lockdown has resulted in most people taking to the internet and internet-based services to communicate, interact, and continue with their job responsibilities from home. internet services have seen rises in usage from % to %, compared to pre-lockdown levels. video-conferencing services like zoom have seen a ten times increase in usage, and content delivery services like akamai have seen a % increase of content usage (branscombe, ) . cities like bangalore have seen a % increase in internet traffic. the lockdowns across countries have entailed a rise in the use of information systems and networks, with massive changes in usage patterns and usage behaviour. employees are adjusting to new "normals" -with meetings going completely online, office work shifting to the home, with new emerging patterns of work. these changes have come across all organizations, whether in business, society, or government. the changes have also come suddenly, with barely any time for organizations and people to plan for, prepare and implement new setups and arrangements; they have had to adjust, try, experiment, and find ways that did not exist before. though now, in late may , the pandemic is receding and stabilized in certain countries, it is still on the increase in many others, and with serious threats. experts in most countries are wary of the possibility of the disease spread re-emerging, and that lockdown norms may be relaxed carefully and slowly with social distancing at the core of the new normal. it is in this context that we see the use of information systems to continue in the same vein for some time in the foreseeable future as during the lockdown. we examine the possible scenarios in this surge in information technology usage during and post the pandemic. our estimation of these effects assumes that there was a digital transformation already underway, before the pandemic set in, and it will take certain forms owing to the impact of the lockdowns. in the next section, we examine the impact of the covid- pandemic on the use of digital technologies where we discuss some possible scenarios and research issues of the post-pandemic world. the next section summarizes the implications for research and practice, and in the last section, we present our conclusions. in this section, we discuss some of the most pressing issues regarding the post-pandemic digital surge. these themes reveal the multiple directions in which is research can focus in relation to impacts on technology. as the use of video-and audio-conferencing tools increases significantly, organizations will ramp up their technology infrastructure to account for the surge. this will lead to increased investment in bandwidth expansion, network equipment, and software that leverages cloud services. with employees becoming j o u r n a l p r e -p r o o f acclimatized with the idea of work-from-home (wfh), meeting and transacting online, firms will shift to wfh as a norm rather than as an exception. this is being adopted by many firms (akala, ; bbc news, ; khetarpal, ) , which have the digital infrastructure in place to handle the required load and bandwidth. education is another domain in which there a dramatic shift to the online mode of transacting. since the beginning of the lockdown, schools, colleges, and universities around the world have shifted their classes to video conferencing platforms like zoom and google meet. along with these synchronous modes of teaching, asynchronous platforms like edx and coursera have also seen an increase in enrolments (shah, ) . some institutions are now shifting entirely to the online mode for the forthcoming academic year, with the exception of sessions that require a physical presence, such as the university of cambridge in the uk and the california state system in the us (new york times, ). digital transformation technologies such as cloud, internet-of-things (iot), blockchain (bc), artificial intelligence (ai), and machine learning (ml), constitute a bulk of the of what is being adopted by organizations as part of their transformation effort. blockchain (bc) technology presents an opportunity to create secure and trusted information control mechanisms (upadhyay, ). as education and healthcare services witnesses a shift to the digital domain, bcs enable a way to secure and authenticate certificates, health records, medical records, and prescriptions. research on the design of such systems, along with maintaining their ease-of-use and usefulness will gain importance. another issue is that of designing systems that work with smart contractshow the contracts are authenticated, how these contracts will be designed in a complex chain of processes with many agents involved, and how arbitration related to contracts will be handled. further, is research may point to regulatory aspects of bcs with regard to what must be encrypted and shared (such as for authenticating news and information sources), and how security will be managed. for instance, governments demand access to private keys to view blocks for surveillance and monitoring, versus the requirements of privacy and protection from persecution. the gig economy is driven by online platforms that hire workers on an ad hoc, short-contract, and mostly informal basis. well-known examples of these include uber and airbnb globally and ola and swiggy in india. these platforms have grown immensely since the wide availability of smartphones from onwards. during the lockdown, workers employed by these platforms have suffered heavily, as the demand for their services, taxi rides, rentals, or skill work, has disappeared (bhattacharya, ) . further, since these workers had no guaranteed salaries, their incomes dropped dramatically. in the post-pandemic scenario, there is likely to be, in the short term, a slow return of gig economy workers, as manufacturing and service firms return to their old activities. however, we anticipate that in the longer term as the threat of infection and spread recedes, the gig economy will thrive. this will also be driven by the wfh culture. work-from-home and gig work has received attention in is research, through topics in telecommuting, digital nomads, and virtual teams. one key issue is that of work allocation and collaboration, across and inside teams, and across projects. this issue will face a rise in scale and importance in the post-pandemic world, as the numbers of wfh and gig workers increase. research may focus on aspects of the design of work norms, work contracts, trust-building, and team-building, amongst others. research on telecommuting and virtual teams (belanger, collins, & cheney, ; morrison-smith & ruiz, ) has a long history in is literature. issues include the nature of "distance" whether temporal, spatial, or cultural, and the psychological needs of workers, the technological support and design for this kind of work, and many others. this research is important for the post-pandemic period. we anticipate that the "dark side" of virtual teams and dispersed work also assumes importance in the postpandemic world. substantive issues related to technostress -particularly work overload and presenteeism arise in these situations. research will have to address issues of design of collaborative work, evaluation, team performance and motivation, stress, and the issue of continuous learning. another aspect of digital use by large sections of the working population is that of constant workplace monitoring and being on-the-job continuously. those working from home using video conferencing technology find themselves under intense scrutiny and all interactions are "hyper-focused" (kalia, ) . digital technology makes it easier for bosses and managers to call and locate subordinates at any time, knowing that they can be reached at all times. though there is anecdotal early evidence that this has led to an increase in productivity, it has also led to increased technostress (ayyagari, grover, & purvis, ; tarafdar, tu, ragu-nathan, & ragu-nathan, ) where employees must learn new technologies, be available for work at almost all times, stay with digital devices all the time, and cope with multi-tasking. post-pandemic, it is likely that workers' organizations will demand no-digital hours, where they will find refuge from the constant work pressure. research may address the concerns of work equity, balance, and managing stress. along with the surge in the use of digital technologies, we are now witnessing a rise in online fraud, scams, intrusions, and security breaches. the pandemic has created a scenario of insecurity that is inviting fraudsters to exploit the crisis situation by extracting money or information or by creating vulnerabilities (agarwal, sengupta, kulshrestha, anand, & guha, ) many users are beginning to rely on digital resources extensively, some for the first time, and are becoming targets for fraud and scams. organizations and governments are aware of this threat and are taking countermeasuresfor instance, some governments took a strong stand against zoom sessions for education, forcing the platform provider to upgrade security (yu, ) . it is likely that these scams and frauds will increase in intensity after the pandemic. organizations will implement massive security arrangements, along with extensive information campaigns by government departments. security innovations and firms that offer security services will rise. research will likely focus on managing security, assess the causes of breaches, and the economic and social loss from them. information technology, and particularly the internet, will remain central to the post-pandemic scenario, where innovations will drive the surge in use. a key aspect of this surge will be the management and regulation of the internet itself. though the internet is a global resource and no one country can control its protocols and features, its local access and availability remain an in-country issue. during the pandemic to some countries have restricted access to the internet (chhibber, ) , for certain reasons. the regulation of the internet will become crucial after the pandemic as it will remain a policy tool for governments. they can intercede on aspects of monitoring, bandwidth control, surveillance, intermediary liability, and e-commerce. the pandemic has brought the world to a situation where those not connected to the internet are facing total exclusion. with strict social and physical distancing measures in place, new routines require accessing the internet for most services. hence, those on the wrong side of the digital divide are completely left out. reasons for the divide are many: unaffordable device access, unaffordable internet access, content relevance, access skills or government ordered internet shutdowns (armbrecht, ; scheerder, van deursen, & van dijk, ) in developing countries, the condition is more serious. thus, it becomes extremely important to explore the possibilities of ensuring connectivity. although these issues have been researched and discussed earlier (warschauer, ) , covid- has brought about a situation where internet access seems to have become necessary for survival. as a few studies have suggested, access or no-access to icts may reinforce societal inequalities (ragnedda, ) , where the post-pandemic situation may enhance this further. with the substantial use of technology in accessing basic requirements like health and education, it is imperative to understand the impact of the digital divide on social equality. therefore, it calls for researchers to examine the impact of connectivity to draw policymakers' interest and, perhaps, offer ways to enhance it towards better inclusion. the heavy use of the internet during the pandemic, for various purposes, has raised people's data requirements. with a significant digital divide in societies, this surge in the internet data requirement has revived the discussion on zero-rating plans. zero-rating plans enable firms to let users access data from their sites and services, without having to bear data charges. usually, this is not strictly permitted as it violates the basic principles of net neutrality, where internet traffic has to have the same priority and cost. india, for instance, had an exemplary record of regulating zero-rating plans. although the government did not permit the implementation of such plans, in the aftermath of the pandemic, the telecom regulatory authority of india (trai) decided to allow waiving charges for data and voice for certain websites (coai, ) (coai, ). the list primarily consisted of the sites related to covid- -such as the world health organisation and india's ministry of health and family welfare. the list also included some private players. the principal aim was to allow people, across all socio-economic levels, access covid- related information. given that zero-rating plans can be useful in exceptional circumstances, as is evident from the example of india, research on the conditions on various parameters where allowing zr plans may increase social welfare has enormous practical implications, both for firms as well as regulators. the existing literature on net-neutrality regulations and zero-rating plans (belli, ; cho, qiu, & bandyopadhyay, ) forms the basis to enhance literature in this aspect. issues to be studied include: expanding telecom infrastructure, providing subsidized internet devices, free extra data, or waiving off users' subscription fees (shashidhar, ) in current times, when the productivity of people depends significantly on the internet, its shutdown can be extremely detrimental to societies(isoc, ) however, internet shutdowns are not uncommon even in times like these. the internet was shutdown in kashmir, a union territory in india, since august th, j o u r n a l p r e -p r o o f and continued till may , making it the longest ever imposed in a democracy (masih, irfan, & slater, ) basic internet services, such as filing for driving licenses, were accessed by locals using the internet express, which is a train that shuttles kashmiris to the nearest town where they can get online. the kashmir chamber of commerce estimates $ . billion in losses owing to the internet shutdown (masih et al., ) . similar events are regularly noted across various other countries, arab spring being the significant starting point. with the pandemic, when the internet has become the most important tool available to citizens the impact of internet shutdowns has become grimmer. shutdowns lead to severe implications for all aspects of life, and there are many issues that require research in this regard. the impacts resulting from a climate of uncertainty can potentially discourage foreign investors and spillover on a wide range of sectors, including education, healthcare, press & news media, and e-commerce (kathuria, kedia, verma, bagchi, & sekhani, ) . it is important to understand the far-reaching human rights impact of internet shutdowns, which are exacerbated in the current scenario. shutdowns have deep political reasons and in many cases the consequences are indeterminate. research can focus on aspects of domino-effect consequences leading to grave political crises. digital payments and digital currencies to have a key role in the post-pandemic situation. as digital payments are contact-less they will be encouraged by governments, will likely see a surge. this will also be boosted by the gig economy and wfh situations. there are two distinct phenomena related to digital money that has aided the fight during the pandemic. first, banknotes and coins were suspected to be carrying the virus and digital payment was preferred to the 'dirty money' (gardner, ; samantha, ) . online delivery services were encouraging customers to make payments through digital payment systems like a credit/debit card or mobile payments, with mandates by the government in several parts of india (bhandari, ) . this is likely to result in a surge in digital payment usage, which will lead to work on the diffusion of digital payment technology. second, during the lockdown, there was a loss of jobs, and governments provided aid through payment apps and digital payment modes. these are a convenient mode of fund transfer from donors to recipients, as seen in previous crisis relief cases as well (pollach, treiblmaier, & floh, ) . in various crisis and disaster events, where the mobility of civilians was restrained, many mobile payment service providers (e.g. vodafone in afghanistan, safaricom in kenya, and orange in africa) provided quick funds transfer of remittances from migrants to their homes, and relief aid from the government to victims (aker, boumnijel, mcclelland, & tierney, ; pega, liu, walter, & lhachimi, ; wachanga, ) . this is once again observed in the covid- crisis and needs further examination. issues of surveillance and privacy are gaining prominence with digital usage during lockdowns. commentators, such as yuval harari, have written about the potential for state surveillance "under the skin" (harari, ) as governments rely on digital means to monitor the spread of the pandemic. as many governments have started using apps on smartphones to monitor infected persons and trace their contacts, civil society organizations have raised privacy and state surveillance concerns (pant & lal, ) . postpandemic, these measures of monitoring populations for epidemiological reasons with digital means are likely to continue and become prevalent. though the concerns of privacy and surveillance are valid and have to be addressed, these digital platforms are the most reliable and efficient way of tracking disease spread. "surveillance is a distinctive product of the modern world" (misa, brey, & feenberg, , p. ) , and today we are living in a surveillance society where any internet-based activity using a mobile phone or other electronic gadgets can be monitored and accessed in unfathomable ways (gilliom & monahan, ; lyon, ) . this has resulted in a surge in is research on implications of such web or app-based surveillance in applications including mobile health apps (lupton, ) , environment monitoring and pollution control apps (castell et al., ) , self-tracking apps (barassi, ) , and parental surveillance (ghosh, badillo-urquiola, guha, laviola jr, & wisniewski, ) . covid- has introduced a new application of surveillance for tracking citizens with the symptoms of the virus. this includes the covid- tracker in china (davidson, ) , the aarogya setu app for tracking infectious citizens in india (shahane, ) , and contact tracking apps in the united states (guynn, ) . while these technologies are innovations for fighting the global pandemic today, the issue of government surveillance on citizens has evolved repeatedly. research can focus on the multiple benefits of these apps, but also should not ignore the potential social complications that are possible to arise, including the historic problem of bureaucratic control by the government, using it (gandy, ) . closely related to surveillance is the issue of privacy that mobile apps, including covid- trackers, often tend to threaten users' personal information (gu, xu, xu, zhang, & ling, ; joy, ) . for example, online classes during the pandemic lockdowns have suffered issues of 'intrusion of privacy' as students and teachers are on camera in the private spaces of their homes (garcia, ) . privacy in the digital age has remained a research topic of high priority for is researchers (belanger et al., ; smith, dinev, & xu, ) . privacy has also been considered by is adoption and usage researchers, with privacy risk as a dominantly recurring factor in studies on mobile payments (e.g., johnson, kiser, washington, & torres, ; luo, li, zhang, & shim, ) , location-based mobile services (zhou, ) , and social networking sites (aghasian, garg, gao, yu, & montgomery, ; youn & hall, ) . it would be interesting to examine the different privacy concerns of users while adapting both covid- tracking apps, and online classroom applications. the risks involved in the breach of privacy by these two technologies are unalike and must be investigated with adequate contextual references. in this section, we revisit some of the key issues that are important for research and practice. our discussion is based on the assumptions about the post-pandemic situation and the aspects of is research presented above. . while deploying security technologies like the blockchain, it will be important to understand the implications of smart contracts, their integration in workflows, and their effectiveness in complex resource-constrained settings, as in developing countries. further, understanding the implications of secure and non-erasable technologies like blockchains will become relevant for regulation. . many research issues arise with regard to work-from-home and gig work, which include aspects of trust, measurement of performance, communication effectiveness, and collaboration. j o u r n a l p r e -p r o o f . it can be expected that the dark side of virtual work and gig work, will raise questions of stress, presenteeism, work overload, surveillance, and monitoring. new and severe forms of digital surveillance will have to be understood and their implications gauged. . though much work has been done in understanding the parameters and impact of the digital divide, it will be important to understand how those without access suffer more from the consequences of the pandemic when the world survives on digital communications and operations. . management of the internet within countries is important, and aspects of enhancing networks include regulating zero-rating plans cautiously, seeing their implications for welfare, and how they can enhance access. . internet shutdowns during and after a pandemic lead to severe difficulties for citizens, who have come to depend on these services. research has to examine the direct, second-order, and thirdorder impacts of these shutdown measures. . research on digital payments and their impact in crisis situations, for providing aid and subsidies to affected populations, and for disaster management. . surveillance issues about the extent of data collection by contact tracing apps are important areas of research. issues of persistence and elimination of data, the expanse of data collection, sharing of data between apps, and the multiple trade-offs involved. . design of secure technologies, like blockchain-based applications, for the surge in online education and healthcare activities. . policy for regulating digital infrastructure needed for increased digital transformation. . design of technologies for managing secure online interactionsfor education, healthcare, payments. . design of apps for contract tracing and disease surveillance that balance privacy versus public health. . managers will have to understand resistance to technology and ways to manage change, both among employees as well as customers. . given the significant role which the internet is about to play in times to come, internet intermediaries will work with government and civil society to address privacy and surveillance issues for better adoption of technology. we understand that a pandemic can have severe consequences (keys, ) , including changing the political contour of the world, destroying empires, and creating nations. for the covid- pandemic, we envisage a dramatic shift in digital usage with impacts on all aspects of work and life. how this change plays out remains largely dependent on our responses to and shaping of the emerging trends. in this paper, we have outlined what we see as some key trends and research issues that need to be examined urgently. they will have substantial consequences in the future. internet users to touch million by scoring users' privacy disclosure across multiple online social networks more big employers are talking about permanent work-from-home positions payment mechanisms and antipoverty programs: evidence from a mobile money cash transfer experiment in niger. economic development and cultural change reasons billion people are still offline society for information management and the management information systems … babyveillance? expecting parents, online surveillance and the cultural specificity of pregnancy apps twitter allows staff to work from home "forever technology requirements and work group communication for telecommuters net neutrality, zero rating and the minitelisation of the internet ahmedabad says no to cash on delivery to stop spread of covid- coronavirus lockdown has exposed the serious flaws of india's gig economy the network impact of the global covid- pandemic. the new stack mobile technologies and services for environmental monitoring: the citi-sense-mob approach militancy in kashmir peaked without g, but modi govt keeps forgetting this in court. the print less than zero? the economic impact of zero rating on content competition. the economic impact of zero rating on content competition request-for-non-charging-of-data.pdf.pdf chinese city plans to turn coronavirus app into permanent health tracker. the guardian the surveillance society: information technology and bureaucratic social control it's a new territory": professors and students face struggle to adapt to online classes. the collegian dirty banknotes may be spreading the coronavirus, who suggests. the telegraph safety vs. surveillance: what children have to say about mobile apps for parental control supervision: an introduction to the surveillance society privacy concerns for mobile app download: an elaboration likelihood model perspective apple and google release coronavirus contact tracing technology for public health mobile apps yuval noah harari: the world after coronavirus | free to read limitations to the rapid adoption of m-payment services: understanding the impact of privacy risk on m-payment services coronavirus: centre tones down mandatory clause on the zoom boom: how video-calling became a blessing -and a curse. the guardian anatomy_of_an_internet_blackout.pdf. indian council for research on international economic relations post-covid, % of . lakh tcs employees to permanently work from home by ' ; from % examining multi-dimensional trust and multi-faceted risk in initial acceptance of emerging technologies: an empirical study of mobile banking services m-health and health promotion: the digital cyborg and surveillance society the electronic eye: the rise of surveillance society india's internet shutdown in kashmir is the longest ever in a democracy challenges and barriers in virtual teams: a literature review aarogya setu app: a tale of the complex challenges of a rights-based regime. the wire unconditional cash transfers for assistance in humanitarian disasters: effect on use of health services and health outcomes in low-and middle-income countries the third digital divide: a weberian approach to digital inequalities dirty money: the case against using cash during the coronavirus outbreak determinants of internet skills, uses and outcomes. a systematic review of the second-and third-level digital divide moocwatch : pandemic brings moocs back in the spotlight -class central opinion: does aarogya setu really work? livemint net neutrality in the time of covid- information privacy research: an interdisciplinary review the impact of technostress on role stress and productivity ethnic differences vs nationhood in times of national crises: the role of social media and communication strategies technology and social inclusion: rethinking the digital divide gender and online privacy among teens: risk perception, privacy concerns, and protection behaviors singapore allows schools to resume zoom use for home-based learning examining location-based services usage from the perspectives of unified theory of acceptance and use of technology and privacy risk key: cord- -i n t authors: candela, massimo; luconi, valerio; vecchio, alessio title: impact of the covid- pandemic on the internet latency: a large-scale study date: - - journal: nan doi: . /j.comnet. . sha: doc_id: cord_uid: i n t abstract the covid- pandemic dramatically changed the way of living of billions of people in a very short time frame. in this paper, we evaluate the impact on the internet latency caused by the increased amount of human activities that are carried out on-line. the study focuses on italy, which experienced significant restrictions imposed by local authorities, but results about spain, france, germany, sweden, and the whole of europe are also included. the analysis of a large set of measurements shows that the impact on the network can be significant, especially in terms of increased variability of latency. in italy we observed that the standard deviation of the average additional delay – the additional time with respect to the minimum delay of the paths in the region – during lockdown is ∼ − times as much as the value before the pandemic. similarly, in italy, packet loss is ∼ − times as much as before the pandemic. the impact is not negligible also for the other countries and for the whole of europe, but with different levels and distinct patterns. at the time of writing, the coronavirus disease (covid- ) pandemic is still ongoing and billions of people are under some form of lockdown. the restrictions faced by citizens are more or less stringent, depending on the resolutions adopted by the different governments, but in many cases non-essential activities have been shut down and a large fraction of people is confined at their homes. many activities that are normally carried out in physical presence are now taking place on-line. as a consequence, the amount of traffic on the internet increased significantly during the last months. in this paper, we analyze the impact of the covid- pandemic on the latency of the internet. latency is one of the major properties of the network and it is becoming every day more important, as several internet applications are particularly sensitive to its fluctuations. ex-amples include on-line videogames [ , ] , video calls, voip [ ] , and ip geolocation [ , , ] . we analyzed a large set of measurements, collected by means of the ripe atlas platform [ ] , to better understand the effects on the network caused by this major change in the way we live. the analysis focuses on italy which, in april , has been under lockdown for more than a month, experiencing some of the strictest limitations enforced by authorities: all schools, universities, and non-essential shops are physically closed, and people are authorized to leave their homes only for undeferrable necessities. distance learning and remote working were applied whenever possible, with a significant increase in usage of virtual-meeting and video-conference applications [ , ] . table summarizes the most important events which could have had an impact on the italian internet latency. as can be noticed, limitations to citizens have been introduced progressively. for this reason, the changes caused by the italian lockdown are studied, in the remaining of this paper, by comparing the situation status of the network, as it is antecedent to all restrictive measures; the latter, on the contrary, comes just after the most restrictive limitations. the period in between corresponds to a transitory phase, where partial lockdowns start to impact the network performance. hereafter, we will use w to indicate the baseline week, and w for the week just after the major lockdown event. besides italy, we include a brief analysis also concerning spain, france, germany, sweden, and the whole of europe. spain, france, and germany have been characterized by restrictions similar to the italian ones. sweden instead decided not to impose a mandatory national lockdown. for spain, france, and germany w is shifted according to their major lockdown event (shown in table ). for the entire europe the situation is more heterogeneous, as some countries were less hit by covid- and thus adopted milder restrictions. for sweden and europe, w corresponds to the week, the last of our observation period which goes, overall, from february through march . results show that the impact is not always the same across the considered countries and on a european scale. the contribution of this paper can be summarized as follows: • some statistics have been recently released by internet service providers (isps), and other players of the internet ecosystem, about the increased amount of traffic they have been exposed to because of the covid- pandemic. however, a picture that leaves aside the very specific points of view of the single operators is still missing. this study provides a more global view, not polarized by the single operator's perspective. in addition, most of the statistics released by operators concern the amount of traffic, with limited (or absence of) information about latency. • the amount of measurements analyzed is large, thus providing solid foundations for the included statistics. moreover, besides the sheer number, we decompose the impact on delay according to the most relevant factors, including the time of the day, the type of target (belonging to a content delivery network or not), and the version of the internet protocol. in addition to icmp-based delay, we provide information also on packet loss and path changes, as they are, more or less, related with internet latency, and on http-based latency. • besides italy, results also concerning spain, france, germany, sweden, and the whole of eu-rope are included. since measurements have been collected using a single platform, results obtained for the different countries can be compared without the possible bias introduced by the adoption of multiple and heterogeneous systems. results show that, in italy, lockdown impacted the latency of the network, especially in terms of increased variability. effects are more evident during the evening, suggesting that latency is more negatively affected by the increased traffic due to recreational activities rather than remote working or distance learning. for the other european countries included in the study, the impact is milder in germany and france, and similar to the italian one in spain and sweden. the remaining of the paper is organized as follows: in section , we summarize the most significant work concerning the detection and analysis of large anomalies occurred in the internet; section describes the data collection phase; in section , the method we followed to compute the performance indexes is explained; the main characteristics of the datasets are illustrated in section , together with a preliminary analysis; section contains the results on the italian internet latency from different perspectives (type of measurements, hour of the day, ipv vs ipv , etc), whereas section shows the results concerning the above-mentioned countries and the whole of europe (with less details compared to italy); section concludes the paper. there is an extensive body of literature about anomalies on the internet. however, the focus is, usually, on the design of techniques aimed at detecting the occurrence of an anomaly (for instance [ , , , ] ). in this paper, we do not try to define another detection technique, as the impact of the covid- pandemic on the network is evident. we strive to provide a view of the effects of the pandemic at a large-scale, including a quantitative evaluation of its main characteristics as seen from the perspective of latency. when natural events assume catastrophic proportions, their effects might be observed also on the performance and reliability of the internet. during the last years, several studies have focused on the impact on the internet of a few catastrophic natural events, such as earthquakes and hurricanes. in [ ] , authors studied the effects of the taiwan earthquake on the asian internet, from the viewpoint of interdomain routing and traffic in research and educational networks. the earthquake damaged submarine fiber cables causing the fail- [ ] . the study analyzes the impact on the internet as seen by a local isp, analyzing both traffic and routing. due to link failures, the internet experienced traffic drops and subsequent peaks mainly due to the reconfiguration of content distribution networks. however, thanks to planned backup and redundancy, the internet proved to be resilient to such an event, and was affected only locally and with minimal damage. another study analyzed the latency variations due to the same event [ ] , as seen from pinger monitors [ ] . the study shows that, after the earthquake, latencies from some monitors experienced a significant increase for a limited amount of time. these studies point out the importance of maintaining redundant internet routes, even if at a cost, as they can be extremely useful when disruptive events occur. similar work was conducted on the effects caused by severe weather conditions, such as the hurricane sandy. in [ ] , authors used ping to test the reachability of edge networks, to discover that the number of outages in the areas affected by the hurricane sandy significantly increased during and after the hurricane. from the interdomain routing point of view, aben shows that a significant portion of traffic was rerouted around the affected area, again demonstrating the internet resilience and the importance of a redundant structure [ ] . similar outages can be observed also at a smaller scale, as pointed out in [ ] . the study uses ping to show that reachability issues can occur at residential hosts in case of moderately bad weather conditions (e.g. thunderstorms). however, unlike large-scale outages, on the small scale the degree of redundancy is generally not sufficient to cope with these events. the vast majority of the studies concerning the impact of natural events on the internet found in scientific literature tackle the problem from a reachability, routing, or traffic point of view. very few analyzed the experienced latency increase in such events as we do for the covid- pandemic. however, we believe that studying latency is of paramount importance, as it can give an indication of the perceived quality of service by end users. in addition, none of the previously studied events reached the size of the covid- pandemic, in both space and time. at the time of writing, the covid- pandemic has been impacting the lives of billions of people all around the globe for several months, while in the other analyzed natural events the impact of the internet was limited in time or circumscribed to a relatively small geographic area. since the covid- pandemic has started only few months ago, there is still little scientific literature focusing on its impact on the performance of the internet. however, several network operators, content providers, and internet exchange points (ixps), released reports about the increased usage of their infrastructure, which we review in the following. cloudfare reported statistics about the traffic increase towards their servers placed in seattle, northern italy and south korea [ ] . in particular, cloudfare reports an increase of % of the overall traffic in northern italy and a reduction of the traffic coming from fitness trackers perhaps reflecting the scarce mobility induced by social distancing. similarly, fastly reported traffic and download speeds towards their servers [ ] . various countries were considered, including italy, which had a . % increase in terms of traffic and a . % decrease in terms of download speed on average. de-cix, one of the world's biggest ixps, reported a new traffic world record of . tbps [ ] , as well as a % increase in video conferencing traffic and % of social media traffic. finally, the organisation for economic co-operation and development (oecd) released a report that aggregates the various traffic increase information reported by internet operators in a single document [ ] . such report highlights some important numbers, among which: (i) an increase up to % of traffic reported in many ixps and isps; (ii) an increase up to times higher of the volumes of traffic for video conferencing platforms. all these reports focus on the increase of internet traffic and not on latency. we believe that latency is extremely important as, without information about capacities, traffic itself cannot be used for an estimation of the perceived performance of the internet by end users. moreover, as already mentioned, all the currently available studies are limited to the boundaries of the organizations that provided them. one of the few papers focusing on this topic, from a scientific perspective, is the one by favale et al. [ ] , where the impact of the covid- pandemic is observed from the campus network of an italian university. favale et al. highlight a time decrease in incoming traffic and an increase of . times in outgoing traffic, as a consequence of remote learning activities. moreover, using passive measurements collected using tstat [ ] and application logs, they studied the fruition and performance of their in-house distance learning system. the raw latency data used for this study was collected by ripe atlas [ ] . we then filtered and enriched such data as detailed in the following subsections. ripe atlas is a globally distributed internet measurement platform that produces more than measurements per second [ ] . among the open platforms aimed at measuring the internet, ripe atlas is the one with the largest number of vantage points [ ] , and it has a massive presence in europe. ripe atlas automatically carries out anchoring measurements (ams), where the set of targets is pre-defined. in particular, a large set of devices called probes periodically perform measurements towards other devices called anchors. anchors are usually hosted in ixps, in the operation centers of isps, or in data centers. hence, they enable monitoring of the core infrastructure of the internet. the results produced by ams have been extensively used for both research (see, for instance, [ ] , [ ] , [ ] , [ ] , [ ] ) and operational purposes (for example dnsmon [ ] , a service aimed at monitoring the worldwide core dns infrastructure). additionally, ripe atlas allows its users to collect measurements towards arbitrary targets. results of user-defined measurements (udms) are stored in a database from where they are accessible to the public (access is not restricted to the experimenters who triggered them). ripe atlas, for its measurements, relies on classical network tools. the latency from an atlas node to a target is estimated by means of the ping tool which, as known, makes use of icmp echo requests and echo replies. for ams, ping is launched to collect three round trip time (rtt) values. for udms, the default number of collected rtt values is again three, but this number can also be changed by the experimenter. our analysis of the impact of covid- pandemic relies on the results generated by both ams and udms. starting from the dates of the events reported in table , we considered all ams comprised in the interval from the th of february to the th of march . the set of probes and anchors involved in the measurements is stable, with little variations caused by the possible temporary unavailability of probes. since ams are performed periodically and for the entire period of study, they provide information on the internet latency from a stable point of view. however, to be sure to eliminate measurements occurring in a short time frame and thus not covering appropriately the observation period, we further filtered the selection to contain only the measurements concerning source-target pairs that produced successful results in at least different days ( % of the total time frame). the position of source and target nodes is fundamental to analyze the impact on a country-level basis. for ams, the position of both source and target nodes is well-known, as such information is provided for each node participating in the platform. we use such information to select a subset of the ping measurements having both source and target in europe. from now on we will refer to such a subset as the am-derived dataset (amd). amd is composed of more than . billion rtt values generated by source-target pairs during the monitored time period. users of the ripe atlas platform can define their own latency measurements according to their needs and in-terests. they can select the set of targets to be probed, define the periodicity of probing, and set the time-span of their measurement activities. in udms, targets frequently include the servers of major internet companies, dns servers, or privately owned network resources. as a consequence, the set of targets involved in udms is heterogeneous. sources are always a subset of ripe atlas nodes, but the subset can be different from user to user. for our analysis, we are interested in a subset of udms where measurement activities were possibly scheduled before the covid- outbreak in europe and repeated periodically throughout the observation period. to obtain such a subset, we adopted the following strategies. first, we extracted only periodic latency measurements, i.e. configured to be automatically repeated after a certain amount of time. second, we discarded the measurements configured to collect less than three latency values per ping execution. third, similarly to amd, we filtered the selection to contain only the measurements concerning source-target pairs that produced successful results in at least different days. fourth, we restricted measurements to the ones targeting ip addresses in europe. it is important to notice that in udms only the location of the source nodes is wellknown, as it is provided by ripe atlas. hence, for this step, we estimated the position of the targets by using ripe ipmap [ ] . ripe ipmap uses active geolocation, and it has been reported to be % accurate at continent level, . % at country level [ ] , and . % at city level [ ] . additionally, we used maxmind ge-olite [ ] as a fallback tool, in case of failed ip geolocation with ripe ipmap. measurements where the target was not successfully geolocated using these two tools were discarded. finally, in some cases, we had to limit the amount of extracted information due to the almost unmanageable volume of data in the repository. in particular, when the number of targets in the geographic area of interest was too large, we randomly selected targets and the analysis was restricted to them. even when we had to limit the number of targets, the number of sources from which measurements were started was not subject to any limitation. from now on we will refer to the dataset built according to the above-described procedure as the udmderived dataset (udmd). udmd is composed of more than billion rtt values generated by sourcetarget pairs during the observation time period. let d be the delay of a given internet path. it can be roughly expressed as d = d tra + d pro + d que , where d tra is the transmission delay, d pro is the propagation delay, and d que is the time spent because of queues and processing at intermediate routers and target host. in a wide-area scenario like the one considered, d pro amounts to a significant fraction of the overall delay, as signals travel at approximately km/ms in fiber. moreover, the only component that is going to be affected by increased traffic is d que . to isolate d que from the other terms, the d tra + d pro component can be estimated as the minimum delay observed in a set of latency measurements collected on a given path. the larger the number of collected samples, the better the quality of the estimate: intuitively, the collection of more samples increases the probability of finding lightly loaded network conditions. this approach is not novel and it has been followed in other studies to characterize latency variations on a large scale. in [ ] , for each hostpair the difference between the maximum and minimum rtt observed in a time bin was calculated. then, the evolution of the obtained difference values was used to investigate on transient congestion. the use of the minimum observed rtt as an approximation of the fixed delay associated with a path was adopted also in [ ] , to study the variability in tcp connections. delay variation metrics are discussed in rfc [ ] , where one of the most widely implemented formulations is based on the use of the packet with the minimum delay in the sample as the reference packet. in particular, packet delay variation is defined in rfc as the one-way delay of the considered packet minus the one-way delay of the packet with the lowest value for delay over the current test interval. using the minimum delay as the basis for delay variation is the preferred method also according to the itu-t y. recommendation [ ] (albeit when considering -point packet delays, i.e. one-way delays computed by means of the absolute arrival time at destination minus the departure time at the source host). starting from the above considerations, we defined the following method. each source node, say a, measures the rtt towards a target, say b, using the ping command at time t, which produces a list of delay values d ab t = {rt t , rt t , ..., rt t n }. let us also define d ab min,t = min d ab as the minimum, average, and maximum value found in the execution of the ping command at time t from a to b. first, we found the global minimum value observed for each source-target pair as where o is the entire period of observation. since the observation period we considered is relatively long, it is possible that some path changes occurred, i.e. that the set of traversed routers was not always the same. this is not a problem, provided that some of the above concepts are reformulated appropriately. in a scenario where path changes occur, m ab is an estimate of the transmission and propagation delay of the best path between a and b, among all the paths followed during the observation period. then, we used m as a baseline to estimate the additional time experienced for every single pair of nodes. in particular, we computed for each couple ab of nodes, and for each t ∈ o. the values of q min,t , q avg,t , and q max,t of all source-target pairs in the region of interest were then grouped in buckets with a duration of minutes and averaged. this last step originates from our interest in evaluating the impact of the changed style of life on a large scale, for instance at the country level. more formally, let us call r min,k , r avg,k , and r max,k the average values obtained in the kth bucket t k : r min,k = avg q ab min,t r avg,k = avg q ab avg,t r max,k = avg q ab max,t with t ∈ t k and across all ab pairs located in the region of interest. in practice, r represents the average additional time with respect to the best path ever followed during the whole observation period. note that the average is computed for all measurements occurred in that specific bucket and across all source-target pairs located in the region of interest (e.g. all source-target pairs in italy); as a consequence, the value of r is generally greater than zero. to have r min,k equal to zero, all considered ab couples should experience, almost simultaneously, their best rtt (m ab ) during the kth bucket, which is unlikely. the smallest observed delay found for a source-target pair (m) provides an indication of the "uncompressible" component of the delay for such pair and, as such, can be excluded in a study aimed at evaluating the impact of the pandemic. in this section, we provide a picture of the datasets in terms of geographical extension and overall rtt dynamics. figure a shows the distribution of distances of source-target pairs in italy, for both amd and udmd. in detail, the distances between hosts are computed as great circle distances (gcds), i.e. the length of the shortest path on the surface of the earth. actual distances travelled by packets are generally larger than the corresponding gcds because of circuitousness of internet routes (deviations from shortest physical paths can be significant [ ] ). distances span a couple orders of magnitude, from few kilometers, when source and target are in the same city or metropolitan region, to almost one thousand kilometers. figure b shows the distribution of distances of european source-target pairs limited to targets (but using all sources). for the european data, we had to limit the number of source-target pairs, because to compute the distance both source and target have to be located at city level, and active geolocation is expensive. for both italian and european data, the distances reported in figures a and b are the ones for which we have been able to compute the position of the involved nodes at city level (approximately - % of the considered couples). we believe that the distributions shown in figure are reasonably close to the ones of the complete datasets. the distributions of minimum rtt value observed for each source-target pair (i.e. m) located in italy and in the whole of europe are shown in figure c and d, respectively. the wide range of values, from few milliseconds up to ∼ ms in italy, and up to ∼ ms in europe, originates from the geographical extension of the considered areas. figure a shows the raw rtt values collected in italy, ams, before any processing. more in detail, the scatterplot has been produced using just : of the values to make the image readable. the largest values tend to be slightly higher after lockdown events. however, a trend does not clearly emerge from raw values, probably because of the large amount of values and because they are related to paths with very different lengths. the red curve that is also shown in figure a case, increased variability can be observed in the righthand side of the figure because of lockdown. the period of fluctuations suggests that daily activities may play a role and for this reason they are further analyzed in section . . figure b shows the absolute values, again for the italian ams, but now grouped according to the method defined in section . the d min values have been grouped in buckets and averaged, similarly to r min , but without subtracting m. let us call such values s min . the curve is approximately the same shown in figure a , but with less spikes as the size of buckets is minutes. figure c shows the same s min curve, but now together with r min . the scale is different to have a better view of the phenomenon. the r min curve is very similar to the other one, but translated downwards, as now the m value found for each pair is subtracted. figure d shows the th percentile of q min , again grouped in buckets of minutes. it is thus computed similarly to s min , but now using the th percentile instead of the average. variability of delays is even more visible in this case. we do not provide an analogous discussion about the european dataset and/or udmd for the sake of brevity. in the remaining of this paper, we base the analysis mostly on r, as we are interested in evaluating the impact of the pandemic on the variable component of latency (with the exception of section . where we also include s). the phenomenon is principally observed in its evolution along the line of time, as we wish to understand if, and to what extent, the lockdown measures, enforced in italy and in the other considered european countries, had an impact on the network. we studied the impact of the covid- pandemic on the latency of the italian internet from different perspectives: when both source and target are located in italy or just one of the two, when considering the time of the day and workweek/weekend, and when taking into account the version of the internet protocol. we also studied the observed latency when the target is part of a content provider network. beside icmp-based latency, an evaluation of latency as seen at the http level is provided. finally, a subsection also shows the path changes occurred at the as-level as they can be related to the phenomenon under observation. we start our analysis from the measurements in amd with both source and target in italy. we first analyze the values of s measured in italy. we recall that s is the average of the d values in each bucket, as defined in section . figure a shows the evolution of s min , s avg , and s max for the whole observation period. we notice that all values progressively increase over time. since italy experienced some partial lockdown events before the most restrictive one, there is no step-like increase, rather a continuous one. however, approximately from february , when schools were closed in northern italy, delays start to grow and higher peaks can be observed. the largest increase can be observed on march , which is the date of the complete italian lockdown. the increase appears to be quite significant, for example if we consider the s min line, the peaks during the lockdown are approximately % higher than the ones before the lockdown. besides the generally increasing value of all the three curves corresponding to the three s variations, the higher variability of latency is also evident. to evaluate the increased variation we compared the value of s min in w and w . the average s min experiences a . % increase in w , and the s min standard deviation experiences a . % increase. the measurements in udmd show a similar pattern, but more accentuated, as can be seen in figure b . in fact, in w , the average s min is . % higher than in w , and the s min standard deviation is . % higher. it has to be noticed that the s values in udmd show a little drop around march , which lasts some hours. we investigated on this aspect, and found that one of the targets of the measurements was unreachable in that interval. the target involved multiple source-target pairs, corresponding to approximately / of the measurements, which usually were providing relatively high absolute delays. in other words, this is an artifact due to a temporary lack of measurements, which can happen when measuring real world devices. similar patterns can be observed when analyzing the evolution of r min , r avg , and r max for both amd and udmd, which can be found in figures c and d . in this case however, the curves are translated downwards, as expected. this reflects on the increment of the average r min value in w with respect to w , which is . % in amd and . % in udmd, but not on the increment of the r min standard deviation value, which remains almost unchanged: . % in amd and . % in udmd. similar considerations can be made about measurements with sources in italy and targets spread all over europe (excluding italy), as shown in figures a and b for amd and udmd respectively. also in this case, the generally increased variability during lockdown is clearly visible. when considering measurements with sources in europe and targets in italy, as shown in figures c and d , the situation is a bit different between amd and udmd. in particular, for amd the pattern is approximately the same as in figure a . for udmd, instead, the pattern is quite different. an increase of the overall variability is still noticeable, however the three curves appear to be much more squeezed on top of each other, i.e. the r min and r max curves are much closer to each other than in the other scenarios. it is also worth to notice that, especially for amd, the local minima of the r min curve tend to get higher during the the transitory period but then they start to get lower. this is particularly visible in figures a and c . the local minima (the troughs) correspond to night hours, when the network is lightly loaded. this phenomenon could be explained by the infrastructural enhancements introduced by network operators to respond to the crisis. for example, during the transition period, tim (the italian incumbent) started peering again in public peering lans of italian ixps for the first time since the end of [ ] . also, ixps reported an increase of traffic of - %, which pushed them to introduce upgrades in the capacity of their peering lans, as reported during the italian network community meeting held for the occasion [ ] . such improvements could also jus- tify the situation in figures a and c , where the troughs during the lockdown reach smaller values compared to the one of the first analyzed week. we estimated the impact of the lockdown in terms of packet loss, as the fraction of unsuccessful echo request/reply. figure shows the evolution of the packet loss for the whole observation period, for both amd and udmd. both datasets show an increase of packet loss over time, especially after the lockdown start. this is consistent with the previous results, and indicates a generally increased congestion due to the lockdown. we computed the average packet loss in w and w . the increase is significant: from . e- to . e- (+ . %) for amd, and from . e- to . e- (+ . %) for udmd. also the standard deviation of the packet loss rate in w increases significantly compared to w : from . e- to . e- for amd (+ . %), and from . e- to . e- for udmd (+ . %). we notice that in amd there is an increase of the packet loss that lasts approximately days, around february -march . we investigated this behaviour and found that the increase is not attributable to a narrow subset of the source-target pairs. we did not register any disconnections of sources nor targets, and the same packet loss pattern can be seen in measurements directed to % of the targets, although with different intensities. these targets are geographically spread all over italy. in particular there are two targets that contribute in a significant manner, which are located in milan (at the milan internet exchange) and monopoli (puglia region), respectively (we point out that measurements towards one target involve multiple source-target pairs). the temporary increase of packet loss is not reflected in the r curve, as for the involved source-target pairs the packets that are correctly delivered experience a negligible increase of the rtt. the three curves shown in figures and show repeated peaks and troughs. this is particularly evident during lockdown, and suggests that latency gets more influenced by weekly or circadian rhythms. to evaluate the influence of the day of the week, we separated the italian measurements in those run during work days and those run during weekends for w and w . results can be seen in figure , which shows the boxplot of r min in w and w for work days and weekends. in both w and w the weekend values are slightly higher and show higher variability. in addition, a significant increase can be observed in w compared to w . to evaluate the influence of the time of the day on latency, we divided italian measurements in one-hour slots and aggregated them when executed in the same hour. then, we calculated for each slot the ratio between the average r values collected during w and during w . results are represented in figure a form across the time of the day. night hours show no considerable increase. this is not surprising, as human activity is very limited at night, thus also the congestion of the internet. morning hours show some little increase. afternoon hours show a more evident increase, but the highest increase occurs between : utc and : utc, with a peak between : and : utc. this is interesting, as, combined with the results of weekend highlighted in figure , it suggests that remote working and distance learning have some impact on the italian internet latency, but the major effects can be attributed to leisure activities which typically occur in the afternoon/evening and in weekends, such as gaming or video streaming (the reader has to keep in mind italy is utc+ in the analyzed period). this could be due to the lack of other recreational activities during the lockdown. to deepen on this aspect we computed the empirical cdfs of the r min values in peak and off-peak hours for w and w (figures a and b) . for peak hours, we considered the interval : - : , for off-peak we considered the interval : - : . we can notice that in both peak and off peak hours the cdfs for w and w are clearly separated, and the cdf for w shows the highest values. however, it is clearly visible that while in off-peak hours the two cdfs are very similar, in peak hours the cdf for w shows a substantial increase of the r min values (note that the x axis scale in the two figures is different). in section . , we highlighted as in night hours the values of r min are particularly low during the lockdown, hour of the day hour of the day figures a and c the local minima during the lockdown seem lower than the ones before the lockdown. to evaluate this effect we computed the ratio between w and w for the two cases depicted in the aforementioned figures. the results are shown in figures b and c , respectively. in both cases the ratio goes slightly below one during night and morning hours. especially surprising is figure c where the the ratio goes below one for the entire morning. this could happen as in figure c we consider sources outside italy, thus the access network is still not involved in a lockdown phase, and targets in the italian infrastructure, which as mentioned has been improved to cope with the traffic increase. during night and morning hours the load on the network is still light, so in this particular configuration the performance could increase. to conclude, in evening hours the increased internet us- age during lockdown generally produces larger delays, but in periods of lighter load the network is sometimes more efficient than before the lockdown. we further analyzed amd taking into account the version of the ip protocol. to perform a fair comparison we only selected measurements run by dual stack probes (i.e., probes which have both ipv and ipv connectivity). this means that for both ipv and ipv we consider measurements run between the same sources and targets. we found . million ipv rtt values and . million ipv rtt values with both source and target in italy. figures a and b shows the three r variations for ipv and ipv , respectively. ipv latency, in italy, seems to be characterized by larger variability compared to ipv latency, independently from the lockdown period. such variability increases even more during the lockdown period. however, for ipv a reduction of minimum r min values is particularly evident, probably due to the network improvements introduced by operators. in particular, a significant drop around march is visible. we investigated on this aspect and found that a subset of measurements were initially originated by source-target pairs which shared a common upstream provider and were flowing through non national paths (mainly via germany, switzerland, and netherlands). in correspondence of the observed drop, the measurements involving these source-target pairs start to flow through local paths via other providers. these paths show considerably smaller latencies. after few days, the original configuration is restored. we further studied the latency experienced between : utc and : utc (the peak hour previously identified) by the two protocol versions. in particular, we compared the values collected during w with the values during w . for ipv , values of the r min average show a . % increase in w . for ipv instead, the increase is more modest, . %. both show a similar increase in variability in w : . % ipv , and . % for ipv . to conclude, ipv in italy is characterized by a generally higher variability than ipv , but in peak hours the former has been impacted less than the latter by stayat-home orders. this is not surprising, as ipv and ipv are generally served by different infrastructures, and follow different paths [ ] . in addition, ipv is not as common as ipv in domestic connectivity, and this could justify the minor impact of the lockdown on latencies observed in ipv measurements. the same analysis was not repeated using udmd, as the relatively limited amount of ipv -based udms does not allow us to produce statistically sound results. since a large fraction of traffic is nowadays directed towards content providers, which we suspect also being related to most of the evening traffic (e.g. video entertainment), we investigated the impact of the lockdown on the latencies towards youtube. we collected measurements towards the youtube content delivery networks (cdns), which is used to serve video content. youtube operates elaborated server selection strategies [ ] which could lead to inaccurate results. to avoid this, we operated as follows: (i) we first mapped the names associated with the ad-hoc youtube cdns (googlevideos.com and ggpht.com) to the ip addresses that are used to serve con- tent in italy. for this purpose, we used all the ripe atlas probes in italy to run multiple dns queries in order to obtain the ip addresses associated with the youtube servers. (ii) after obtaining all the visible ip addresses of the youtube's servers ( addresses), we used the anycast detection service offered by ripe ipmap, which uses active measurements from ripe atlas probes to detect if an ip is anycast or not. (iii) then, we used again ripe ipmap to geolocate the ip addresses of the servers. the servers not in italy were discarded. (iv) finally, we used these ip addresses to extract ripe atlas measurements targeting them. we only selected icmp ping measurements (which are not subject to http or application layer redirects). we found approximately million rtt values towards youtube's servers located in italy. figure shows the three r variations we obtained. an increase of the overall latency and its variability is visible during the days of the lockdown. the standard deviation of r min is approximately . % higher in w with respect to w . the average r min instead increases by . %. our hypothesis that the increase of rtt registered in the evening hours is due to people forced to stay home and using the internet for entertainment, is thus strengthened by the results of the measurements towards youtube. figure a shows measurements collected between : utc and : utc while figure b shows measurements collected between : utc and : utc, for the entire observation period. also in this case, the r min during night hours slightly improves during the lockdown. the r min in the evening gets moderately higher during the transitory period and abruptly increases after the first day of complete lockdown. we acknowledge that youtube is not fully representative of all content providers networks, in fact our initial purpose was to collect measurements towards facebook and netflix as well. however, we did not find in ripe atlas enough measurements towards facebook and netflix servers to cover the whole observation interval and obtain statistically significant results. to further improve our analysis, we looked for anomalous patterns in the level of path changes between the sources and targets of our measurements. we consider path changes at the autonomous system (as) level, as such level is the one impacting the most on the geography of the paths, while the ip level is subject to various known artifacts and intra-as load balancing [ ] . from ripe atlas, we collected icmp traceroute measurements to match the italian sourcetarget pairs of our analysis. for amd, this was rather straightforward as, besides icmp ping measurements, ams also include other types of measurements such as icmp traceroutes. for udmd, instead, we could not find enough traceroute measurements to match the source-target pairs of the ping measurements, thus we restricted our analysis to the amd source-target pairs. to extract the as paths from the traceroute ip paths, we use the following methodology: initially, we match all the ip addresses against the internet topology data kit [ ] produced by caida. this step converts different ips belonging to the same router to a single one. then, we remove the first ip hop inside the network of the probe's host, as it is usually fixed. this step allows us to avoid most of the artifacts introduced by the icmp rate limiting happening close to the probes [ ] . if an ip path includes series of wildcards (i.e., non responding hops), these are squashed into a single one. then, each ip is converted to the as number originating the prefix such ip belongs to. for this step we use the longest-prefix matching on ripe ris data [ ] . some addresses cannot be mapped to an as because they are not available in the ripe ris data (e.g. not announced publicly), or, more frequently, they are part of the private address space (rfc ) [ ] . we discard such unmappable addresses. this approach has been previ-ously described by hyun et al. [ ] . more sophisticated approaches are of difficult application due to the size of our dataset. moreover, our main interest is to quantitatively estimate the variation of the path changes over time, and not to analyze the internet as-level topology. we then divided the observation period in buckets of one day each, and for each bucket we computed the fraction of source-target pairs that incurred in at least one change of as path in that bucket. figure shows the fraction of path changes over time for amd. the fraction of source-target pairs experiencing path changes each day is rather uniform on all the observation period, between . and . , which means that each day just - % of source-target pairs experienced a change of as path. however, by comparing w and w from a numerical point of view, we found that the average fraction of path changes per day is . % higher in w . thus, there seems to be a slight increase of the number of path changes per day due to the impact of lockdown. in the previous sections, we reported results obtained via ping measurements based on icmp echo requests. this type of traffic is used for network diagnostics and sometimes routers can treat it differently from traffic generated by end users, even if they usually behave similarly in terms of rtt [ ] . even if the trends highlighted in the previous sections by icmp show an evident increase of the network congestion caused by the covid- lockdown, we also analyze the latencies produced by ripe atlas http anchoring measurements (based on tcp). these kinds of measurements are quite different from the icmp measurements which we analyzed above. in particular, each http measurement collects just one sample (instead of at least three for the icmp ping ones). in addition, the latencies that can be extracted from this measurements include the time for sending an http request to a server and waiting for the http response body to arrive. this time includes the time needed for the server to build the response. the time needed for the dns resolution of the server address, instead, is not included. for this reason, it must be noted that the latencies extracted from http measurements are not directly comparable with the latencies obtained by means of icmp ping. we analyzed the latencies extracted from http measurements with the methodology described in section , however, being each measurement composed by just one sample, we will not have d min , d avg , and d max , but just d, and therefore just q and r. figure a shows the results for measurements with source and target located in italy. the figure shows an increase in the latency due to the lockdown. like in the previous results, the increase is progressive and not step-like, as we recall that italy experienced multiple partial lockdowns before entering in the strictest one. a step-like increase is noticeable in correspondence of the increase of the packet loss showed in section . . in this case the packet loss affects the performance of http as, differently from icmp, the packet loss triggers retransmissions of the tcp protocol. consequently, the end-toend latency of the tcp connection increases. besides this anomaly, the overall increment is noticeable also by comparing average and standard deviation values of r, which are respectively . % and . % higher in w with respect to w . in figure b , the failure rate of http measurements is shown. such failures are due to a combination of network errors which include connection timeout, host unreachable and network unreachable problems. also in this case an increment of the failure rate can be noticed as long as the italian lockdown becomes more restrictive. these results allow us to con-firm that the increased latency, in italy, during lockdown is not limited to network diagnostic traffic but also to end user traffic. in this section, we show how the latency in other european countries was affected by the covid- pandemic. we considered spain, france, and germany, which had their major lockdown events respectively five, seven, and twelve days after italy. in addition, we considered sweden, as an example of a country that adopted a less formal lockdown policy. finally, we considered all european countries together. overall results are depicted in figure . table reports the increment of the average and standard deviation of r min and packet loss after stay-at-home orders. to obtain the increment, for each country we compared the first week of measurements (w ) and the first week of lockdown (w ). for germany, the considered periods are four day long instead of seven: germany was put on lockdown at the end of our observation period and a full week was not covered by the collected data. since there is not a unique date for lockdown in the whole of europe, and since some countries did not even enter a lockdown phase, for sweden and europe we compared the first and the last week of the observation period. the two weeks used for comparison are highlighted in light grey in figure . in the following we analyze the considered countries in detail. spain shows high variability of latencies also before the lockdown. however, an increase due to enforced restrictions is noticeable in amd. similarly, circadian patterns become more evident (figure a ). this is confirmed also by the summary statistics, which show an in-crement of . % of the r min average, and of . % of the r min standard deviation, in w compared to w . the visual analysis of udmd shows a progressive increase of latency which starts much earlier than the lockdown, with a generally spread variability (figure b ). in fact, results in table show a significant increment in the average r min of . %, but a modest . % for the standard deviation. on march , a temporary increase can be observed in figure b . we investigated this phenomenon and found that it is due to a considerable increase in the latency towards one target, observed from multiple sources. we believe this to be an anomalous behavior due to the target itself or the network in its proximity. the analysis of the packet loss shows opposite behaviours between amd and udmd. in amd, the packet loss decreases during the lockdown, while in udmd increases greatly. in france, the situation is different, as shown in figures c and d, which show amd-and udmd-based results respectively. in amd, the overall increase of latency is barely noticeable. lockdown seems to accentuate the periodic fluctuations due to circadian rhythms. in fact, the summary statistics included in table show even a decrement of r min average and standard deviation in w compared to w , in the amd. however, it must be noticed that the first week of measurements in amd appears as particularly noisy if compared to the other pre-lockdown weeks. udmd instead show much higher variability, which is the attribute that shows the most evident impact of the lockdown. in fact, the analysis shows an increment of just . % for the r min average and . % for the standard deviation of r min . the packet loss increases during lockdown for both amd and udmd, in both average and standard deviation. in germany the situation is different from both spain and france. in amd, the effect of lockdown is noticeable only in terms of amplified circadian patterns (figure e) . the analysis confirms this, by showing just a . % increase of the r min average, and a . % increase of the r min standard deviation. in udmd, a more significant increase of latency is visible (figure f) , which corresponds to a . % higher value for the r min average and a . % higher value for the r min standad deviation in w compared to w . it is worth noticing that in germany the degradation of latency starts more than one week before the lockdown. this happened because germany, like italy, proceeded to some partial lockdowns and school closures before the major restrictions. the average packet loss slightly increases for udmd, and decreases for amd. as mentioned, sweden adopted less formal restrictions. for this reason, results shown in figures g and h are definitely interesting. both amd and udmd show a progressive increase of r and of its variability in the considered time period. this can mean either that swedish people autonomously increased social distancing and implemented stay-at-home policies as suggested by the swedish government, or that the performance of the swedish internet infrastructure has been affected by the lockdown imposed in other countries. also the comparison of w and w show a significant increase of r min average and standard deviation: . % and . % for amd and . % and . % for udmd. the packet loss increases just for amd, but decreases for udmd. figure i and figure j show respectively the amd-and udmd-based results for all of europe. additional latency is generally smaller than the individual countries we analyzed. the response to national lockdowns seems to be fairly good even if we notice an accentuation of the variability due to the circadian activities, starting from the italian lockdown, but not a significant increase of the overall latencies. this is confirmed by the statistics reported in table : the r min average is subject to a modest increase, equal to . % and . % in amd and udmd respectively, while the r min standard deviation experiences a significant increase, equal to . % and . %. these results seem to indicate that, on a continent-level scale, the impact of lockdown is still noticeable but without dramatic changes in observed performance. the packet loss instead decreases for amd and is almost unchanged for udmd. it is worth to notice that, after analyzing the packet loss in the different countries, we cannot conclude that an increase in latency is coupled with an increase in packet loss. in fact, in some cases we found increasing latency and decreasing packet loss, and vice-versa. it is well-known that computer viruses may cause a slowdown of the internet [ , ] . in we all learned that also biological viruses may affect the global internet performance, because of the changes they bring in the way we live. in this paper, we analyzed the impact of the covid- pandemic on the latency of the internet on a large scale. latency is particularly important not only because it has a profound effect on some classes of applications, but also because it is, by itself, an excellent indicator of the health status of the network. results, which have been obtained from the analysis of a large amount of measurements, show that the impact of the increased on-line activities is relevant, especially in terms of higher variability. the major changes have been observed in the evening, the time of the day when most of the on-line activities are related to entertainment. this suggests that distance learning and remote working contributed to a lesser extent in terms of additional network latency. results obtained for the considered countries show relevant differences, which can be due to the resilience levels of their network and/or to the non-uniform restrictions imposed by authorities. we believe that the provided numbers and the related analysis, despite being limited to a portion of the internet, definitely help in better understanding this previously unseen event in the history of the internet. latency and player actions in online games the effects of latency on player performance in cloud-based games assessing voice quality in packet-based telephony geolocation of internet hosts: accuracy limits through cramér-rao lower bound smartphone-based geolocation of internet hosts using ripe atlas for geolocating ip infrastructure ripe atlas: a global internet measurement network our commitment to customers during covid- zoom cfo explains how the company is grappling with increased demand detecting anomalies in network traffic using maximum entropy estimation on the detection of network traffic anomalies in content delivery network services measurement of large-scale bgp events: definition, detection, and analysis a basisevolution framework for network traffic anomaly detection experience with restoration of asia pacific network failures from taiwan earthquake quaking tables: the taiwan earthquakes and the internet routing table the japan earthquake: the impact on traffic and routing observed by a local isp the pinger project: active internet performance monitoring for the henp community a preliminary analysis of network outages during hurricane sandy hurricane sandy and how the internet routes around damage pingin' in the rain covid- impacts on internet traffic: seattle, northern italy and south korea how covid- is affecting internet performance de-cix experiences record traffic amid coronavirus and call of duty - vgoglwswy&title=keeping-the-internetup-and-running-in_times-of-crisis campus traffic and e-learning during covid- pandemic traffic analysis with off-the-shelf hardware: challenges and lessons learned a survey on internet performance measurement platforms and related standardization efforts vantage point selection for ipv measurements: benefits and limitations of ripe atlas tags pinpointing delay and forwarding anomalies using large-scale traceroute measurements disco: fast, good, and cheap outage detection joint minimization of monitoring cost and delay in overlay networks: optimal policies with a markovian approach systems for characterizing internet routing visualization and monitoring for the identification and analysis of dns issues tracing cross border web tracking ripe ipmap active geolocation: mechanism and performance evaluation why is the internet so slow?! proceedings of the rd acm sigcomm conference on internet measurement, imc ' , association for computing machinery packet delay variation applicability statement internet protocol data communication service -ip packet transfer and availability performance parameters on the spatial properties of internet routes italian networking community meeting summary measuring the deployment of ipv : topology, routing and performance dissecting video server selection strategies in the youtube cdn proceedings of the th acm sigcomm conference on internet measurement, imc ' periodic path changes in ripe atlas routing information service (ris) traceroute and bgp as path incongruities, tech. rep., cooperative association for internet data analysis (caida) on evaluating the differences of tcp and icmp in network measurement the internet worm program: an analysis worm epidemics in high-speed networks alessio vecchio is an associate professor at the university of pisa, italy. he has been involved in several national and eu-funded projects (congas, neutmon, mecperf). he is currently serving as associate editor of pervasive and mobile computing and ieee access. he co-launched and co-organized five editions of the permoby workshop (an ieee percom workshop focusing on human mobility). he has been tpc chair of the seventh ieee international workshop on sensor networks and systems for pervasive computing (ieee percom persens) and in the technical committee of many other international events. key: cord- -o i hy authors: holliday, ian; tam, wai-keung title: e-health in the east asian tigers date: - - journal: int j med inform doi: . /j.ijmedinf. . . sha: doc_id: cord_uid: o i hy objective: the article analyzes e-health progress in east asia's leading tiger economies: japan, hong kong, singapore, south korea and taiwan. it describes five main dimensions of e-health provision in the tigers: policymaking, regulation, provision, funding and physician-patient relations. methods: we conducted a series of fieldwork interviews and analyzed key healthcare websites. results and conclusion: our main finding is that the development of e-health in the region is less advanced than might be expected. our explanation focuses on institutional, cultural and financial factors. the application of information technology (it) to public sector operations sometimes captured in the notion of e-government is starting to have an impact on developed healthcare systems the world over. as time goes by, that impact is expected to become even more pronounced. ''the consensus seems to be that new information technologies will significantly affect almost every aspect of health care,'' wrote blumenthal [ ] . in this article, we examine the progress of e-health in the five leading economies of east asia: japan, hong kong, singapore, south korea and taiwan. each seeks to place itself at the forefront of the information revolution and has high levels of internet access and usage. each also has a sophisticated healthcare system dedicated to securing maximum healthcare benefit at minimal cost. by standard outcome indicators, these systems all have very good records. the tigers, therefore, form a cluster in which e-health might be expected to be notably advanced. however, our finding is that although some progress is being made, it remains limited. it is also variable across the five societies. the article begins by reviewing some of the literature on e-health taking from it a series of critical dimensions and issues. it then briefly analyzes the two relevant contextual aspects of the east asian tigers: their participation in the information age and the nature of their healthcare systems. on these bases, it examines their e-health progress, focusing on the major themes unearthed in the contemporary literature. finding limitations and variations, it concludes by thinking through possible explanatory factors, focusing on institutional, cultural and financial issues. much of the existing e-health literature has been developed in the context of the united states, reflecting both us leadership in the information age and the continuing search for solutions to us healthcare problems. five main themes are prominent. four of the five address distinct dimensions of the broad policy and management framework for healthcare, examining internet impacts on policymaking, regulation, provision and funding. the fifth theme looks inside the healthcare sector and inside the surgery, at the implications of the internet for physician-patient relations. eventually, this may have policy significance, but for now, it is best treated separately. the major argument made about healthcare policymaking is that the us government has been slow to engage with the numerous issues generated by the it revolution [ ] . the core features of that revolution, notably enhanced information flows, increased networking possibilities and novel commercial opportunities, are now well documented [ ] . however, it is said that in us healthcare, most policy actors in both congress and the executive branch continue to focus on pre-information age agendas. although, the bush administration has started to address these concerns, the result remains something of a lack of internet-related policy activity and only a limited number of perspectives on the internet's potential to transform the us healthcare system. clearly structural features of the us system, including fragmentation both of government and of the healthcare sector, play key roles. looking at the narrower sphere of regulation, concerns are expressed about the failure of regulatory agencies to keep pace with internet-related developments. goldsmith notes that the internet generates many potential regulatory problems, ranging from licensing e-health practitioners to monitoring information quality in a virtual world with no boundaries [ ] . fried et al. detail some of the obstacles placed in the way of e-health by existing regulations, holding that individuals and organizations must navigate a maze of rules and codes, old and new, if they wish to implement fresh ideas and approaches [ ] . kassirer's prediction is that the courts will play a role when substandard medical advice provided through web sites or e-mail yields poor medical outcomes. he believes that courts will be especially important when professional advice is given without a direct patient encounter, or when state lines are crossed [ ] . some regulatory issues are us-specific, but many have much wider relevance. partly building on the regulatory theme, analysts have also debated the limitations currently imposed on healthcare provision through the internet. kleinke argues that the internet will not contribute to a solution to the administrative redundancies, economic inefficiencies, and quality problems that have long plagued the us healthcare system. instead, it will exacerbate the cost and utilization problems of a system in which patients demand more, physicians are legally and economically motivated to supply more, and public and private purchasers are expected to pay the bills [ ] . goldsmith holds that the challenges of standardizing coding and formats for clinical information, and protecting patient privacy, will hinder the realization of network computing potentials in healthcare [ ] . the problems to which these and other authors point are structural. economic, organizational, legal, regulatory, and cultural conflicts rooted in the us healthcare system are all barriers to e-healthcare provision. further problems are found in the sphere of healthcare funding. shortliffe criticizes congress for focusing on short-term benefits, arguing that research investment for e-health must be balanced between basic and applied analyses [ ] . robinson examines the effect of distinct forms of capital on the development of the healthcare internet. in the late s, venture capital flooded into the ehealth sector, rising dramatically from us$ million in early to us$ million in late . in the same period, e-health firms went public, raising us$ . billion at their initial public offerings. however, the technology-sector crash in late hit the e-health sector especially hard, prompting an extended period of consolidation between e-health and more conventional firms [ ] . us funding problems thus relate to both the public and the private sectors. finally, analysts have looked inside the surgery at physician-patient relations. existing survey data show that citizens make considerable use of the internet for healthcare information and services, mostly of a generic kind [ ] . indeed, anderson reports that in , % of us adults with internet access did so [ ] . as more patients go online, increasing numbers will turn up in surgeries with internet-fueled questions and concerns. meeting the growing expectations of these individuals will be a significant challenge for physicians [ ] . assessing the likely impact, kassirer argues that the internet will change the physician-patient relationship in unpredictable ways, with some aspects of electronic communication strengthening the bond, and others undermining it [ ] . goldsmith believes patients have most to gain from the emergence of the internet, arguing that it will rebalance the steeply asymmetrical medical knowledge held by patients and physicians [ ] . using information gained through internet searches, patients can now open their dialogues with physicians at a much higher level than before, and thereby gain leverage in the care process. ball and lillis also discuss the potential challenges the internet presents to physicians. as internet searches often generate as many questions as answers, physicians are likely to find themselves under increased workload pressures [ ] . the variable quality of healthcare information accessed through online searches [ ] , a matter that is being actively addressed by bodies such as the health on the net foundation (www.hon.ch) and the internet healthcare coalition [ ] , can only reinforce those pressures. zupko and toth hold that physicians sometimes encounter a form of cultural shock when confronted by well-informed patients [ ] . it is therefore perhaps not surprising that an april survey found that physicians are much more reluctant than patients to use the internet for healthcare interactions. while % of patients wanted to exchange e-mail with their doctors, only about % of doctors actually did so. physician-patient confidentiality, time concerns and increased exposure to malpractice liability were cited as primary reasons for doctors' wariness [ ] . in the face of this mounting speculation and evidence, lumpkin is sanguine, however, contending that the physician-patient encounter is little changed, despite widespread internet usage in healthcare [ ] . though focused on the us, the existing e-health literature generates key themes for an analysis of progress in other parts of the world, including the east asian tigers. however, before exploring those themes, we first present some basic contextual information about our five societies. two features of the east asian tigers are particularly relevant to this analysis: their participation in the information age and the nature of the healthcare systems in which their application of it needs to be assessed. in this section, we examine both features. looking at broad social participation, it was exploited more rapidly in the five east asian tigers than in any other global cluster. for many years, the nielsen//netratings global internet index has ranked all five societies in the top worldwide for personal computer (pc) connections and internet access and usage. the four smaller societies, hong kong, singapore, south korea and taiwan, are particularly advanced. furthermore, the severe acute respiratory syndrome (sars) crisis that hit the region in spring gave a major boost both to internet usage in general, and to e-health in particular [ ] . in hong kong, for instance, the number of active internet users increased by % from february to april , before falling back by % from april to june . the overall increase was %. also at the start of , the time spent online by hong kong people first increased by % and then fell back by %, registering an overall increase of % [ ] . consistent with the image of economic and social dynamism, they have projected for many years now, the east asian tigers are among the most advanced it societies on earth. to some extent, this strong it orientation is the product of developmental state strategies. with the partial exception of hong kong, the east asian tigers have long placed considerable faith in stateled growth strategies. furthermore, for many years they have frequently focused those strategies on it and it-related sectors. in japan, in the s, the fabled ministry of international trade and industry targeted supercomputers and the fifth generation as a major development project [ ] . despite a long period of economic stagnation in the s, the japanese it industry remains a significant global force. in singapore in the s, the state took the lead in nurturing wafer fabrication, the most sophisticated ''front end'' of the semiconductor industry. chartered semiconductor manufacturing, established by the government in , is now the third largest silicon foundry in the world. in south korea, in the early s, the state reorganized the public-sector telecommunications system by closing inefficient firms and allocating profitable segments to major chaebol like samsung and goldstar, enabling them to establish specialized chip businesses. by the early s, samsung had become the world's number one producer of dynamic random access memories for pcs and workstations. in taiwan since the s, the ministry of economic affairs and the state-controlled electronics research service organization have played crucial roles in developing the semiconductor industry. today, it is the fourth largest in the world, and firms within it have entered into strategic alliances with leading industry players in the west [ ] . even in hong kong, where a developmental state took longer to emerge, the government is currently overseeing the construction of a flagship cyberport, intended to host a strategic cluster of companies and professional talents, specializing in it applications, information services and multimedia content creation, and designed to project a hi-tech international digital city image. the east asian tigers are also leading players in the development of e-government. the un/aspa benchmarking survey of all un member states placed singapore at number (a long way behind the us, but only fractionally behind australia and new zealand), south korea at number and japan at number . all three states featured in the top category of high e-government capacity. the survey did not assess hong kong and taiwan, neither of which is a un member state. the report noted that singapore ''demonstrated a balanced and citizen-centric e-government program, while possessing the benefits of a high technological infrastructure and human capital measures''. it held that south korea ''made perhaps the most dramatic advances in its e-government program by successfully implementing several new online transaction features''. it was more critical of japan, arguing that it had ''yet to live up to its rather significant potential''. ''japan's e-government program has not yet reached a comparable level of sophistication as that of the regional leaders due primarily to achieving only a limited interactive presence among national government websites'' [ ] . a january analysis of e-government in east and southeast asia reached similar conclusions, identifying the five tigers as regional leaders [ ] . accenture's survey looked at only two of the five east asian jurisdictions analyzed here. it ranked singapore number in the world after canada, and japan number [ ] . looked at from many different perspectives, then, the east asian tigers are leading participants in the emergent information age. healthcare systems in the tigers share a basic orientation, but are otherwise quite varied. the orientation is best termed productivist, in that in each society social policy has usually been subordinate to economic objectives. while the governments of all five tigers certainly get involved in social policy, they usually do so either for economic reasons or after they have made provision for their various economic goals. the main stimuli to this strong focus on economic development were, in all cases, the devastation brought by the second world war, and the uncertainties of the international order constructed thereafter [ ] . this shared orientation has fed into healthcare policy in three main ways [ ] . in japan and its two former colonies, south korea and taiwan, healthcare was initially left chiefly to the market. only once economic policy was on track and a measure of growth had already been attained, did these societies turn their attention to planning their healthcare systems. in doing so, they concerned themselves chiefly with healthcare finance, creating social insurance systems by gradualist means. now, in all of these societies, the health insurance scheme is universal in aspiration and near universal in fact. across all three societies, healthcare provision remains privatesector-driven, with the state performing a chiefly regulatory role. traditional medicines are significant in all three societies and covered by national health insurance schemes [ ] . however, they are not consistently brought within the planning frame. in hong kong, until the early s, the colonial government took a strictly reactive and incremental approach to healthcare. its major interventions focused on subventing charitable organizations in the healthcare business, though in time, it also built hospitals and delivered care directly through them. throughout, government activity was funded out of general government revenue. the major and to date, only step change came in , with the formation of the hong kong hospital authority (hkha). this imposed state control and state funding on the secondary sector and gave hong kong a miniature version of the british national health service. however, there has never been any attempt to bring primary care within the planned healthcare system. only in , was traditional chinese medicine subjected to anything more than minimal government regulation [ ] . in singapore, the early post-war experience was similar to that of hong kong. here, however, separate initiatives were taken in the spheres of provision and funding. in , much provision was integrated at the secondary care level through creation of the state-run hospital corporation of singapore. this body subsequently sought to drive private-sector disciplines into state provision through ''corporatization''. in , in an attempt to generate integrated pathways of care, it was broken into two territorial clusters focused on the secondary sector but also having primary and tertiary elements. however, as most of the primary sector remained outside the state sector, the extent of integration was limited; in singapore, the state provides % of secondary care but only % of primary care. on the funding side, singapore in created a compulsory savings system, medisave, within the wider central provident fund scheme. it added the insurance schemes medishield and medishield plus in and and created a basic social safety net, medifund, in . these various schemes partially fund secondary care provision. there is also some direct state subsidy. funding of primary care takes place mainly through out-of-pocket expenses. the traditional sector stands outside all state planning and, as in hong kong, has only recently been brought within the regulatory framework. these healthcare systems have enviable records. not only did they make a rapid post-war transition from the contagious disease characteristic of thirdworld countries to the chronic disease characteristic of first-world societies, but also they register very favorable health outcomes as measured by standard input and outcome indicators (table ) . healthcare systems in the east asian tigers thus share a productivist orientation and strong performance. they exhibit varied state roles, with much healthcare activity lying outside the public sector and some of it falling beyond the planning horizon. in japan, south korea and taiwan, state involvement is extensive in finance but limited in provision. in hong kong, the government both funds and directly provides care in the secondary sector, but not elsewhere. in singapore, the state provides a large amount of secondary and some primary care. the funding regime is complex, comprising direct state subsidy, forced individual saving, state-run and private-sector insurance, and out-of-pocket expense. in all five tigers, both the public and private sectors play important roles and face clear incentives to take an interest in harnessing the internet for healthcare gain. note: taiwan data are from . sources: [ ] . against the dual backdrop of sophisticated it societies that make extensive use of the internet and cost-effective healthcare systems driven in variable ways by actors from the public and private sectors, we now turn to a survey of e-health in the east asian tigers. to frame the survey, we begin by providing a brief descriptive overview of the major state-run healthcare websites in the region. we then structure our analysis using the five main analytical spheres that dominate the existing e-health literature: policymaking, regulation, provision, funding and physician-patient relations. all ministries or departments of health in the east asian tigers have their own website. throughout the region, the major quasi-autonomous state agencies, such as the national health insurance agencies in japan, south korea and taiwan, the hkha in hong kong and the two big healthcare clusters in singapore, also have sites. here, we look only at the main government healthcare sites ( table ) . the overall quality is high. all have clickable links to organizational objectives and tasks. most also offer detailed information about subsidiary divisions. all contain links to the government homepage and related healthcare sites so that visitors can conduct further searches and collect additional information. all provide feedback channels. in singapore, the ministry of health (moh) offers online feedback opportunities. in taiwan, citizens can make online appointments with the director of the department of health (doh). in japan, the ministry of health, labour and welfare (mhlw) uses e-mail to solicit however, within this generally strong showing, there are also significant differences, with japan's mhlw and to a lesser extent, hong kong's hwfb lagging behind their regional counterparts in key respects. firstly, the mhlw fails to provide contact details for named officials on its website. this is standard practice in the other four tigers. singapore's moh, for example, gives address, telephone number and e-mail details for key officials. secondly, while healthcare professionals and officials in singapore, south korea and taiwan can communicate with each other through the internet, their counterparts in japan and hong kong cannot. thirdly, the range of options available to users is more restricted in japan and hong kong than in the other three tigers. in south korea, for instance, it has played a role in the surveillance system for communicable disease since . through electronic data interchange and regional database management systems, notifying and reporting systems have been computerized, and an electronic record of all notified and reported cases is kept. using the super-highway communication network, physicians and public health centers can access the notifying and reporting system, disweb, anywhere and anytime through the internet (http://dis.mohw.go.kr). in singapore, the moh site within the government's ecitizen portal enables healthcare professionals to download application forms for license renewal, approval to perform a pregnancy termination, and so on. the policymaking strand of the e-health literature castigates us policymakers for being slow to grasp the potential of the internet. such a charge is less easy to sustain in the east asian tigers, though again experience is variable. singapore and taiwan are the regional leaders. singapore's ecitizen portal addresses many aspects of citizen interaction with government, with healthcare being a prominent theme. the internet is used to reinforce the public health messages that have been disseminated by the singaporean government through other media for many years. behind the scenes, e-mail links pervade the healthcare system and enhance the cohesiveness of policy networks. in a controlled city state, those networks are in any case very tight. in taiwan, the doh in launched an ambitious e-health project, with a timeline stretching to . the health information network that is central to this initiative has a backbone funded by central government and permits local users in both the public and private sectors to participate on a self-paying basis. drawing on us experience, it seeks to promote electronic medical records, based on a smart card system, so that information can flow to all parts of the healthcare sector. a healthcare certification authority, created in , oversees promotion of this initiative. in the other three tigers, progress is less impressive. japan launched an e-japan strategy in january , designed to make it ''the world's most advanced it nation within years'' [ ] . the strategy had an explicit e-government strand. in september , the mhlw followed up by issuing a ''grand design'' for promotion of it in the healthcare sector. the aim was to computerize the entire sector by and to introduce an electronic medical record system covering % of clinics and % of hospitals with plus beds by . progress towards targets appears to be on track. however, japanese performance in the e-health domain is poor by regional standards. hong kong is also quite slow to place healthcare online. the hwfb site contains standard bureaucratic information, such as current policy initiatives and recent speeches, plus public health information that has been developed particularly since the sars crisis. here, the major networking initiative is being taken by the dominant public-sector delivery agency, the hkha. while its primary focus is provision, the networking links being created among hospitals are likely to have policy consequences. as in singapore, e-mail links also bolster ties within policy networks that are already quite cohesive. south korea is making an aggressive attempt to exploit the internet across all areas of government, but in the healthcare sphere, currently remains an average performer. turning to regulation, three main issues are raised in the literature. the first is that e-health generates a number of regulatory problems. the second is that excessive regulation may impede e-health progress. the third is that the courts are likely to have to step in when administrative regulation fails. in the east asian tigers, regulation is clearly a major concern and an evident constraint on ehealth development, often for good reason. one instance is limitations placed on consultations, which in all five tigers quite properly mandate face-toface physician-patient contact before any specific healthcare information or advice can be given. for the foreseeable future, online consultation, though technically feasible, is likely to be restricted by professional concerns. another instance is limitations placed on information sharing and exchange, which in all the tigers are again very properly re- stricted by privacy considerations. however, there is some variation in regional regulatory practice. in singapore, patients requiring repeat prescriptions can place an order online and have the medications delivered to their homes. only after months, do they have to return to the healthcare system to consult a physician. elsewhere, this practice is illegal. in japan, physicians are prohibited from answering specific questions about healthcare or disease by e-mail or telephone. regarding provision, assessments in the us literature are mainly negative. on the one hand, the argument is made that it cannot be expected to solve structural problems in healthcare systems. on the other, barriers even to less ambitious networking initiatives are held to be substantial. these are fair points, but they should not be allowed to obscure the real progress being made by healthcare systems around the world, and in our case in east asia. among the five tigers, taiwan's healthcare websites, both public and private, provide the most comprehensive services to patients. singapore ranks second, and hong kong third. japan and south korea are somewhat behind the regional pace. an overview is given in table . in taiwan, the doh operates a taiwan e-hospital site to provide free online medical advice to patients (http://taiwanedoctor.doh.gov.tw/). currently, medical practitioners and nutritionists from public hospitals form a consulting team to answer questions about specialties. patients seeking general medical advice can send questions to a particular practitioner and receive feedback online or by e-mail. in the private sector, a number of hospitals, such as the chang gung memorial hospital, have online question-and-answer services for patients. the kingnet second opinion webhospital (www.webhospital.org.tw) and the taiwan physician's net (www.doctor.com.tw) are two prominent sites providing free online medical advice to patients. established by kingnet entertainment (www.kingnet.com.tw) in , the webhospital has some voluntary physicians answering questions from the public. the taiwan physician's net brings together about physicians, whose information and advice are posted on the web. apart from getting online medical advice, patients can search for a particular physician and visit his or her office for treatment. in taiwan, patients can also make medical appointments online with many public and private hospitals. looking to the future, the doh is planning to develop a medical information exchange center to promote information sharing and enhance treatment quality. in singapore, health is one of a number of cluster points within the ecitizen site. to date, the internet is mainly used to provide general healthcare information, with the healthcare portal containing comprehensive information about healthcare providers, the healthcare establishment, healthy lifestyles and public health issues such as sars. many searches are possible. the site also allows individuals to submit complaints and feedback. only a few transactions can be undertaken online. as in taiwan, appointments can be made and altered online. through singapore's e-pharmacy services, recurrent prescription items can be ordered online and delivered throughout the island. in one of its two main healthcare clusters, patients can register online and access summary medical records. inside the healthcare system, information flows are starting to change as polyclinics and gps gain access to hospital records online. the likelihood is that enhanced integration of the public and private sectors will result. in hong kong, the hkha, which oversees almost the entire secondary sector, is currently introducing online networking in hospitals. its clinical management system is an integrated clinical workstation giving clinicians access to departmental information and patient records. it will soon develop into a longitudinal electronic patient record within the public hospital system, enabling records to be accessed by many parties simultaneously anywhere, anytime. the system will also actively support clinical decisions by offering alerts, reminders, links to medical knowledge and other aids. it is expected to play an important role in reducing medical errors and improving the quality of patient care. over the next years, the hkha is planning to create a hong kong health information infrastructure, with the aim of networking all healthcare providers in the public, private and social welfare sectors. it also intends to build an electronic medical record for every hong kong resident and provide citizens with an electronic gateway to healthcare information and evidence-based medicine [ ] . these initia-tives are likely to enhance information flows within the public healthcare system. compared with taiwan and singapore, however, hong kong lags behind in developing internet services for patients. individuals cannot register and access summary medical records online. lacking an e-pharmacy service, the hong kong system does not allow recurrent prescription items to be ordered online. japan and south korea are falling behind their regional counterparts in providing online health services to patients. their official health websites do not deliver any electronic service to individual patients. with the exception of initiatives taken by a small number of private hospitals in south korea, like the yonsei eye and ent hospital, neither public nor private hospitals in these two tigers allow patients to register online. however, in , japan's mhlw established telemedicine networks to provide specialized care to people in remote areas. the government will provide us$ million a year to form networks consisting of one large hospital and three clinics working together to supervise patients. each patient will be equipped at home with a computer that can monitor heart rate, blood pressure and other indicators, as well as a phone capable of transmitting video. they will be linked to physicians through an isdn digital phone connection, thus enabling physicians to diagnose illness by electronically transmitted data. from june , the mhlw started to establish such networks a year, so that all districts will have at least one by [ ] . in , south korea's semipublic seoul national university hospital founded ezhospital, which is business-oriented instead of patient-oriented. with three main business elements, education (content services), e-trading and system integration, ezhospital is starting to alter purchasing arrangements for both medical and non-medical supplies. as the south korean system is highly fragmented, the purchasing consortia that can be built through the internet could one day become significant. at present, however, e-purchasing is at an early stage of development. analyses of e-health funding focus on one main issue, the short-termism of us initiatives. in this domain, it is difficult to reach an overall assessment of the tigers' performance. on the one hand, their developmental state orientations make the general climate for it industrial emergence very different from the climate found in the us. in this regard, the tigers look to the long term in a systematic fashion that has no us equivalent. on the other hand, it is hard to find evidence that the tigers are investing heavily in e-health applications. moreover, because the private sector plays such a large regional role in healthcare, as it does in the us, many of the relevant initiatives fall outside the state sector and are hard to capture. there are undoubtedly many small commercial initiatives in east asia as, again, there are in the us. furthermore, like other commercial websites, private healthcare sites throughout the region rely heavily on advertising and sale of products for income. to take just two taiwanese instances, the kingnet webhospital and the taiwan physician's net offer online sales not only of healthrelated products, but also of cinema tickets. the very fragmented nature of private-sector healthcare operations throughout the tigers means that few summary assessments can be made. looking finally at physician-patient relations, the existing literature contains variable forecasts of unpredictable change, little change, and so on. however, there is a clear belief that patients have most to gain from e-health and physicians correspondingly have most to lose. in general, physicians in the tigers have tended to be wary of exploiting the internet for patient interactions. this partly reflects the tight regulatory climate in which many find themselves, with many modes of physician-patient contact outlawed. it may also reflect a certain reluctance on the part of both physicians and patients to engage in the informalities of online contact. until recently, then, the emergence of virtual physician-patient relations was highly limited. since the spring sars crisis, however, the pattern may have started to change. although it is too early to register the longterm impact of the crisis, it is clear that during the sars outbreak, many individuals sought to shift to online interactions with healthcare professionals. the fear of visiting surgeries and, in particular, hospitals that gripped the region in has certainly not disappeared and seems likely to provide a lasting stimulus to virtual delivery of healthcare. furthermore, the generic healthcare information found in great abundance on english-language websites is paralleled on regional websites operating in chinese, japanese and korean. there is also some official encouragement for patients to migrate to e-health. in may , hong-jen chang, ceo and president of taiwan's bureau of national health insurance, argued at an oecd forum that e-health could make a major contribution in informing patients. as evidence, he cited taiwanese experience in confronting hiv/aids and the role of the internet in educating patients about the disease. in the long run, he contended, patients equipped with information gained from online searches ''will translate into quality improvement and efficiency gains for the system'' [ ] . overall, east asian societies retain many traditional features, which generate some resistance to change in established modes of physician-patient contact. nevertheless, there are also factors operating in the opposite direction. one long-term impact of the sars crisis seems likely to be heightened caution about visiting healthcare facilities, for fear of contracting infectious disease, and a consequent boost for e-health. the east asian tigers form the most wired cluster of societies found anywhere in the world. moreover, they have long had a developmentalist orientation that has seen their states become involved in many aspects of economic and social development. in the sphere of e-health, however, their performance is strong at the level of basic web provision, but otherwise not particularly advanced. on the whole, their health ministries or departments have good sites covering all the fundamentals of online provision. outside central government agencies, they often have a wealth of additional sites in the public and private sectors. beyond that, they do not make pioneering use of the internet in healthcare. there are many possible reasons for this slightly disappointing performance, some of which apply to all of the tigers and others which are specific to a particular society. in japan, the structural problems that mired the economy in stagnation for more than a decade from the early s also form part of the explanation for its sluggish e-health performance. a notable feature of the japanese healthcare system is the considerable power of the japan medical association and its extensive links to the liberal democratic party that has governed the country for almost all of the post-war period. in hong kong, the sovereignty transfer was quite disruptive, and only several years on is the political system taking a settled shape on the developmental state model. looking beyond the specific circumstances of individual tigers, however, the major explanatory factors appear to be institutional, cultural and financial. institutionally, east asian healthcare systems tend to be highly fragmented, notably in japan, south korea and taiwan. in consequence, policymakers in healthcare ministries and departments have rather few levers that they can use to direct change. in the e-health sphere, they can quite easily construct official government websites, but generating reform in the wider healthcare system is more difficult and depends on their success in building consortia of interest among many private-sector actors. in part, they seek to do this by offering ring-fenced seed money for specified development projects. in part, they resort to exhortation, calling on all members of society to engage in the project of securing and maintaining regional and/or global leadership in the information age. in these many respects, the east asian tigers have a great deal in common with the us. in the additional domain of culture, they differ from the us. while capitalism is certainly a dynamic force in east asia as in north america, it also co-exists with still vibrant cultural underpinnings. the confucian heritage that characterizes all five east asian tigers has many complex strands. among them is considerable respect for authority, hierarchy, status and so on. in the medical sphere, one consequence is that doctors tend still to be accorded considerable professional status. this may make it difficult for full commercialization to take place and for the market drive that characterizes ehealth in the us to work its way through the system. finally, the financial dimensions of healthcare in the east asian tigers should not be overlooked. these are healthcare systems that deliver the excellent outcomes already mentioned at a fraction of the cost registered in the us and, indeed, in most developed societies. as a proportion of gdp, east asian tigers spend between and % on healthcare, with most coming in at around %. this is far below the us figure of - %, and also below the highincome country standard of almost %. one result of the tigers' success in holding down healthcare costs is that the incentive to experiment with new initiatives is reduced. clearly, there still are some incentives, but they are not as strong as in the us. e-health in the east asian tigers remains at an early stage of development. all have attained a good basic standard, but few are engaged in pathbreaking initiatives. alongside institutional factors that are similar to those found in the us, cultural and financial factors help to explain this rather unsatisfactory level of performance. doctors in a wired world: can professionalism survive connectivity? milbank q the internet promise, the policy reality the information age: economy, society and culture. vol. ii. the power of identity the information age: economy, society and culture. vol. i. the rise of the network society the information age: economy, society and culture. vol. iii. end of millennium how will the internet change our health system? ehealth: technologic revolution meets regulatory constraint patients, physicians and the internet com: the failed promise of the healthcare internet networking health: learning from others, taking the lead financing the healthcare internet rethinking communication in the e-health era consumers of e-health: patterns of use and barriers the impact of cyberhealthcare on the physician-patient relationship e-health: transforming the physician/patient relationship health care web sites: are they reliable? internet healthcare coalition, e-health quality partners named exclusive education and outreach affiliate of the internet healthcare coalition physicians get on line, aspen publishers doctor-patient e-mail slow to develop, international herald tribune nielsen//netratings, internet provided vital information and alternative access to shopping, banking and education for people in hong kong as sars took hold online shopping and banking sites soared in popularity as people in hong kong shunned the crowds sars stimulates ongoing growth in internet usage in hong kong divided sun: miti and the breakdown of japanese high-tech industrial policy tiger technology: the creation of a semiconductor industry in east asia united nations/american society for public administration building e-government in east and southeast asia: regional rhetoric and national inaction high performance, maximum value productivist welfare capitalism: social policy in east asia welfare capitalism in east asia: social policy in the tiger economies traditional medicines in modern societies: an exploration of integrationist options through east asian experience agenda-setting for the regulation of traditional chinese medicine in hong kong networking health: dawning of the e-health era. paper presented to apami-mic conference at the hong kong convention and exhibition centre e-health and the informed patient, paper presented to oecd forum taiwan council of economic planning and development, taiwan statistical data book world development indicators the work described in this article was substantially supported by a grant from the research grants council of the hong kong special administrative region, china [project no. cityu / h]. initial seed funding was provided by the governance in asia research centre, city university of hong kong.we are grateful for the research support we received. we thank academics, officials and practitioners in east asia for talking to us about e-health. the usual disclaimer applies. key: cord- -snvq ol authors: castano, adriana mejia; hernandez, javier e; llanos, angie mendez title: kids today: remote education in the time of covid- date: - - journal: nan doi: nan sha: doc_id: cord_uid: snvq ol with the recent covid- breakup, it became necessary to implement remote classes in schools and universities to safeguard health and life. however, many students (teachers and parents, also) face great difficulties accessing and staying in class due to technology limitations, affecting their education. using several nationally representative datasets in colombia, this article documents how the academic performance of students in their final high school year is affected due to technologies, aggregated by municipalities. we conclude that internet access strongly affects these results, and little improvement on the internet/computer access will reflect better academic performance. under these conditions, belonging to an ethnic group or high rurality (non-geographic centralized municipalities) has a negative impact. policy implications are discussed. in particular, these implementations could increment the gap in access to quality education in low-income or geographically remote populations, which keeps countries from achieving sustainable development goals from the united nation development programme (zhao, lu, huang & wang, ) . despite the need to establish these remote programs, there is little empirical evidence on how technologies in the last years affect their performance (zhao, lu, huang & wang, ) and what is the effect of its absence. this is a critical question given that in pandemic times almost all academic activities require internet access and all economic activities resulted affected (presented reductions) due to the pandemic (chetty, friedman, hendren & stepner, ) . in colombia, data sets like those provided by the icfes (colombian institute for the evaluation of education) have long comparisons over time about the academic achievement of students in all stages. most of the studies over these data indicate that the covariables that most affect the academic performance (have a good score), according to chica, galvis, and ramirez ( ) are socioeconomic status, parents scholarship, the number of hours in the school, school type (private or public) and gender; but there is little work on how technologies affect it. the current article aims to fill this gap. using nationally representative samples of children in their final high school year from to , this article addresses two related research questions: ! which municipalities have the most difficulty in implementing the remote academic continuity program? ! what should be the municipalities' prioritization (in each state) for information technology programs investment? we decide to proceed using three classification algorithms and compare them to predict academic performance. solving these questions will help decisionmakers focus on certain geographical areas to develop specific intervention programs to improve the learning experience and results of high school and middle school students. background. in this era of high internet connectivity, which brings positive and negative impacts in our lives, many facets are linked to its access; and education is not an exception. since the early s, it has been noticed (pereira, pleguezuelos, meri, ; greenhow, robelia, hughes, ) that there were necessary changes to take advantage of the online resources. parents and teachers know the importance and necessity of computer usage and internet access in all stages of academic formation (bassok, latham, ) . in kindergarten students in the united states, bassok, and latham ( ) , found that computer access increases academic children skills. the results of amorim, jeon, abel, felisberto, barbosa, and martins ( ) , in brazil, show that technologies, specifically using games, improves reading scores. in high school students, specifically in china, the relation between internet self-efficacy, internet accessibility, behavior, academic performance between others was investigated (zhao, lu, huang & wang, ; penaranda, aragon-muriel, micolta, ) . there is strong evidence that digital inequality exists on the internet, and students with internet access at home had a better academic performance. in the united states, rickles, heppen, allensworth, sorensen, and walters ( ) , found no statistically significant differences in longer-term outcomes between students taking online and face-to-face courses; this could indicate that online courses have the same benefits in some populations (a majority of hispanic and african american students). in college students, junco and cotten ( ) , and sazili and khafidhah ( ) found a negative correlation between academic performance and internet usage, and a weak relationship with computer ownership. in university students, sushma, draus, goreva, leone, and caputo ( ) pointed out that time spent on the internet could be a measure of academic achievement under certain conditions, however, some online media could have a negative effect on university students (junco, ; asemah, okpanachi & edegoh, ) . according to paulsen and mccormick ( ) , student engagement is necessary to have positive benefits. in countries like malaysia and taiwan, the use of the internet is apparently a distractor. indeed, the results of xu, wang, peng, and wu ( ) indicate that discipline plays a vital role in the academic success of undergraduate students; in particular, the results show that the use of internet data could differentiate and predict student academic performance. therefore, the evidence suggests that in any stage of learning, technology strongly affects academic performance across the world (donald, foehr, ) and it is most notorious in pandemic times since all economic activities were seriously affected (chetty, friedman, hendren & stepner, ) and internet access became a necessity (chiou and tucker, ) . going deeper, it looks like there is a way to predict academic performance based on some covariables (hellas et.al, ) , and technologies could be one of them. in colombia there is an academic test provided by icfes, proctors on standardized tests, called saber , that scores students in their final high school year, and also has self-reported socio-demographic information. analyzing data from , chica, galvis, and ramirez ( ) found that high school graduates who own a computer are more likely to obtain a high grade on saber , although internet possession was not statistically significant. however, analyzing more recent data, internet access became an interest covariable (see appendix). in particular, family-related variables are the best to predict academic performance (garcia-gonzalez, skryta, ). the dataset from icfes also includes information of undergraduate students, and an analysis of academic efficiency about these students shows that social covariables play an important role (rojas ; delahoz-dominguez, zuluaga, fontalvo-herrera, ). during pandemic times it is necessary to understand how the internet or computer access affects the score of saber since that can offer a way to comprehend the effect of these technologies in each municipality and which are the best improvements according to each necessity. the actual existing evidence, while limited, does support the notion that internet/computer access strongly affects the academic performance of saber . the findings suggest that, as we notice in literature, technologies strongly affect academic performance. however, in this scenario, there are two covariables that need to be in consideration: rurality and belonging to an ethnic group. then upgrades in technology could improve academic performance, but it is not enough: teachers, parents, and the social environment play an important role. there are necessary policies about technologies and, like zhao, lu, huang, and wang ( ) and sazili and khafidhah ( ) pointed out, about increasing learning skills in remote environments. data. saber tests have been modified during the last years, we have chosen the period from onwards since it has the same standardized content, taking tests on different subjects: critical reading, citizenship skill, english, written communication, and quantitative reasoning; score from to . their sum gives a global score from to (icfes, ) . this is the independent variable that we consider in this study. this data also has socio-demographic information such as internet accessibility, computer, family income, habitability conditions, belonging to an ethnic group, among others. there are municipalities that in this period did not take the saber test, which are not included in this work. this data was aggregated by municipality into numerical covariables. given that data about internet ownership is self-reported by students, we also decided to use the data reported by the ministry of information technologies and communications mintic, about the total number of internet subscribers, the type of subscription and technology, indicators of internet reach by region, information on internet providers by demand and participation statistics (mintic, ). additionally, there is relevant information by region reported by the dane (national administrative department of statistics), including accurate population values ( census). with this information it is possible to know how rural a municipality is, to compare if this rurality is correlated with scores, internet access, and other important covariables (dane, ). in the appendix, we collect descriptive statistics from a consolidated data frame aggregate by municipalities. from this exploratory data analysis and since we need covariables that allows governmental investments, we conclude that the main numerical covariables to use in this study are: • code: numeric code that identifies the municipality, • year: year of the proctored test, • internet: percentage of students with self-reported internet, • computer: percentage of students with self-reported computers, • ethnic: percentage of students self-reported as belonging to an ethnic group, • school: percentage of public schools, • global_score: student global score average on saber , • population: censused population during the test year, • connectivity: access to the internet per inhabitants, • rural_index: percentage of the population that lives in rural sectors. a first finding of the exploratory data analysis, see appendix, is that if the family owns a computer, the average score is higher but rural index decreases, coinciding with the findings of chica, galvis, and ramirez ( ) . the results showed that scores have a mainly rural composition. comparing this with geographical location (figure ) , we obtain that good scores are geographically centralized, there are many families without the internet, and it looks like these two variables are correlated. in this kind of study, the variables are commonly analyzed via correlation (sazili & khafidhah, ; sushma et.al., ) . in figure classification. there are many algorithms, from linear and logistic regressions to neural networks, to help identify and classify covariables/results. however, we wanted to choose interpretative models, this is, models that classify but also give us a statistically significant set of covariables. in the literature, it is common to use linear or logistic regression (bassok & latham, in particular, in the work of xu, wang, peng, and wu ( ), they have covariables as online duration, internet traffic volume, connection frequency, among others. and using decision trees, neural networks, and support vector machines predict academic performance from these features. in this work, we want to do something similar using another set of covariables with logistic regression, decision forest, and regression forest. therefore, the initial model is a logistic regression given that this model shows statistically significant covariables and how they affect (positive or negative) the independent variable. going deeper into the list of models, we found the random forests, which is an ensemble learning method for classification, regression, and other tasks that operates by constructing a multitude of decision trees at training time and outputting the class that is the mode of the classes (classification) or mean prediction (regression) of the individual trees. random decision forests correct for decision trees' habit of overfitting on the training set. we construct a classification algorithm over the municipalities considering the initial set of covariables, which gives a list of municipalities at different vulnerability levels and a few actionable features on which governments (national, state, and/or municipal) can act upon. to use these algorithms, we need a risk threshold that we define for each year as the global score average minus k times the global score standard deviation (where k can be any real number between and ; for optimal purposes we choose k= ). also, we split the data frame into training data (from to ) and validation data ( ), to decide about the accuracy. the main results of the logistic regression are: • considering the p-value for each covariable (less than . ), we conclude that the relevant variables are internet, computer, ethnic, connectivity, and rural index. • ethnic and internet have positive coefficients which imply these covariables have a positive impact on the global score. • computer and rural index have negative coefficients which imply these covariables have a negative impact on the global score. with this new set of covariables, we perform a regression random forest with depth m (with m> ; we choose m= ) and a classifier random forest of depth l. we compare the three models using the auc, see table (area under the curve roc): the roc curve is a performance measurement for classification problems at various thresholds settings. roc is a probability curve and auc represents a degree or measure of separability. it tells how much a model is capable of distinguishing between classes. the higher the auc, the better the model is at predicting 's as 's and 's as 's. the choice of parameters k, m, l was made looking for auc between . and . since lower auc implies that the model is underfitting and bigger implies that it is overfitting. notice that logistic regression, according to auc, is the best classificatory up to these conditions. if we increase the depth to any random forest, we get greater auc, table . but the complexity of the algorithm increases. each one predicts over each municipality if it will be at risk or not, that is, if the average global score of the municipality is greater or not than the risk threshold. to compare and use at the same time the results of the three algorithms, we construct a vulnerability level called total_risk as the sum of the predicted risk of these three models. this analysis allows to define levels of vulnerability: in figure , we can see a map of colombia with all levels of total_risk. to determine how good the estimations are, we can compute a confusion matrix, which is a specific table layout that allows visualization of the performance of the algorithm. each column of the matrix represents the instances in a predicted vulnerability while each row represents the instances in actual risk, and determines if the algorithm is confusing classes. in this case, the first row of the matrix is ( , , , ) and the second row is ( , , , ) . for example, the last column ( , ) implies that municipalities are at serious vulnerability ( of them are false positives, that is, municipalities are labeled as 'at vulnerability' but they are not). we notice that the quantity of false positives and false negatives ( ) is little; therefore, the predictions are accurate. beyond classification. in figure , we also notice some interesting municipalities in the center of the country, in high vulnerability, contradicting the commonly believed that remote locations are always vulnerable. this implies that not necessarily being surrounded by no vulnerability affects completely this covariable. one of the findings was coyaima in tolima, in the center of the country. in tolima, during the last years, the percentage of the population belonging to an ethnic group is near to %, and the percentage of rurality is around %, but these zones can be a reference to improve on their neighbors. listing states with the count of municipalities in each vulnerability we found that vaupés, chocó, guainía, and amazonas are states that need serious intervention as soon as possible since almost % of its municipalities are in serious vulnerability. running a bonferroni to understand each vulnerability level, we notice that low and no vulnerability have significant differences in self-reported internet, computer ownership, belonging to an ethnic group, and connectivity. low and medium vulnerability in connectivity; and medium and serious vulnerability in ethnic group and connectivity. since connectivity and belonging to an ethnic group are common variables, in table , we analyze the average number of connections per habitants and percentage of students belonging to an ethnic group, in each vulnerability level, obtaining that municipalities at serious vulnerability have a high percentage of students belonging to an ethnic group, opposite to no vulnerability municipalities. with covid isolation measures banning students from classrooms, most learning has moved online, but even those pupils who can successfully connect are likely to fall behind if not possessed of the self-direction and motivation needed for remote education. moreover, research has shown that poorer students perform worse in online courses (chetty, friedman, hendren & stepner, ) than face-to-face ones. our analysis found that the most relevant covariables related to an increased academic vulnerability in colombia, related to technologies, are connectivity per inhabitants (as a measure of reliable broadband internet access) and belonging to an ethnic group (associated with low income). amazonas, vaupés, guainía, and chocó need serious intervention, given the high values shown for the vulnerability factors measured. geographic centralization (associated with increased urbanization, as measured by the rural index), is also a strong predictor of better scores for students. we found some centralized municipalities with high vulnerability as coyaima in tolima and altos del rosario in bolívar, surrounded by municipalities of zero vulnerability. in particular, these municipalities have low access to the internet and computer and high rurality and percentage of belonging to an ethnic group, showing an interesting feature of colombia. according to hartley and bendixen ( ) , improving the academic results and reducing the large gap in academic achievement requires the interventions to focus on maintaining the students' interactions (among themselves and with their teachers), also learning skills in the students are necessary as selfregulatory skills, epistemological beliefs, motivation, self-efficacy, ability, physical challenges, and learning disabilities. for this scenario to be realized, a series of factors have to line up: schools need to have the resources to implement remote learning, students need to have access to computers and reliable internet connections, and parents need to have the ability, time, energy, and patience to turn into home-school instructors. a concerted effort between the state government and parents seems to be the most effective strategy. complementary projects might include: • optimizing accessible solutions to mobile devices given that students have some access to them. • offering limited data plans (access granted only to academic sites, to avoid misuse). • focused policies on the rural ethnic minorities, the communities with higher vulnerability. • lending computers to families. • delivering financial stimulus to the municipalities most at vulnerability, conditional on the improvement of academic results compared with the previous year. • support to parents so they can step up as temporary teachers. the challenges are related to providing students at vulnerability the necessary tools for success, not only laptops and reliable broadband internet access but also to motivate parents to help their children succeed academically. cundinamarca has good scores in almost every municipality, we assume that this happens because bogotá (the country capital) is in this state. we now analyze the levels of vulnerability predicted by our algorithm, remember that the choice of the threshold was made to detect the municipalities most at risk. we found that % of municipalities in the country obtain no risk. this could be an indicator that, for future works, the threshold risk could be the average in every year. at low risk is %, in medium risk %, and in high risk %. as mentioned before, the municipalities in high vulnerability are mainly in remote locations, the percentage of computers across time is stable, the percentage of internet access has increased since , but the percentage of students belonging to an ethnic group and rurality are high (near % using escribo play video games to improve phonological awareness, early reading, and writing in preschool influence of social media on the academic performance of the undergraduate students of kids today: the rise in children's academic skills at kindergarten entry how did covid- and stabilization policies affect spend and employment? a new real-time economic tracker based on private sector data determinantes del rendimiento académico en colombia: pruebas icfes saber social distancing, internet access and inequality censo nacional de población y vivienda dataset of academic performance evolution for engineering students". data in brief trends in media use predicting academic performance based on students' family environment: evidence for colombia using classification trees learning, teaching, and scholarship in a digital age: web . and classroom research: what path should we take now? educational research in the internet age: examining the role of individual characteristics predicting academic performance: a systematic literature review transparencia y acceso a la información pública, resultados pruebas de estado en colombia no a u: the relationship between multitasking and academic performance in-class multitasking and academic performance automatically measuring question authenticity in real-world classrooms internet dedicado, accesos fijos a nivel nacional reassessing disparities in online learner student engagement in higher education description and preliminary evaluation of a program for improving chemistry learning in high school students effectiveness of using blended learning strategies for teaching and learning human anatomy online credit recovery and the path to on-time high school graduation does stem stand out? examining racial/ethnic gaps in persistence across postsecondary fields class relations in the higher educational system and its effects in academic performance: the case of bogota the influence of internet usage on student's academic performance the impact of internet addiction on university students and its effect on subsequent academic success: a survey based study internet inequality: the relationship between high school students' internet use in different locations and their internet self-efficacy prediction of academic performance associated with internet usage behaviors using machine learning algorithms is an adjoint professor of mathematics and statistics at universidad del norte, km via puerto colombia, barranquilla -colombia key: cord- -zzc a id authors: otoom, mwaffaq; otoum, nesreen; alzubaidi, mohammad a.; etoom, yousef; banihani, rudaina title: an iot-based framework for early identification and monitoring of covid- cases date: - - journal: biomed signal process control doi: . /j.bspc. . sha: doc_id: cord_uid: zzc a id the world has been facing the challenge of covid- since the end of . it is expected that the world will need to battle the covid- pandemic with precautious measures, until an effective vaccine is developed. this paper proposes a real-time covid- detection and monitoring system. the proposed system would employ an internet of things (iots) framework to collect real-time symptom data from users to early identify suspected coronaviruses cases, to monitor the treatment response of those who have already recovered from the virus, and to understand the nature of the virus by collecting and analyzing relevant data. the framework consists of five main components: symptom data collection and uploading (using wearable sensors), quarantine/isolation center, data analysis center (that uses machine learning algorithms), health physicians, and cloud infrastructure. to quickly identify potential coronaviruses cases from this real-time symptom data, this work proposes eight machine learning algorithms, namely support vector machine (svm), neural network, naïve bayes, k-nearest neighbor (k-nn), decision table, decision stump, oner, and zeror. an experiment was conducted to test these eight algorithms on a real covid- symptom dataset, after selecting the relevant symptoms. the results show that five of these eight algorithms achieved an accuracy of more than %. based on these results we believe that real-time symptom data would allow these five algorithms to provide effective and accurate identification of potential cases of covid- , and the framework would then document the treatment response for each patient who has contracted the virus. since its discovery in late december of , there have been more than . million confirmed cases of covid- reported in countries, as of july , [ ] , with approximately a % daily increase. among these cases there have been more than thousand deaths, which represents an approximate . % mortality rate. this novel coronavirus was characterized on march , as a pandemic by the world health organization [ ] . unfortunately, there is no successful treatment procedure or vaccine yet. it is expected that the development of an effective vaccine will take more than a year, especially since the nature of the virus has not yet been completely characterized [ ] . currently, the only way that the world can deal with this coronavirus is to slow down its spread, (i.e. "flatten the curve") by using measures such as social distancing, hand washing and face masks. however, technology could also help slow its spread, through early identification (or prediction) and monitoring of new cases [ ] , [ ] . such technologies include big data, as well as cloud and fog capabilities [ ] , the use of data gathered through remote monitoring, such as mhealth, telehealth, and real-time patient status follow-up [ ] . this paper proposes a covid- detection and monitoring system that would collect real-time symptom data from wearable sensor technologies. to quickly identify potential coronaviruses cases from this real-time data, this paper proposes the use of eight machine learning algorithms, namely support vector machine (svm), neural network, naïve bayes, k-nearest neighbor (k-nn), decision table, decision stump, oner, and zeror. this detection and monitoring system could be implemented with an iot infrastructure that would monitor both potential and confirmed cases, as well as the treatment responses of patients who recover j o u r n a l p r e -p r o o f from the virus. in addition to real-time monitoring, this system could contribute to the understanding of the nature of the virus by collecting, analyzing and archiving relevant data. the proposed framework consists of five main components: ( ) real-time symptom data collection (using wearable devices), ( ) treatment and outcome records from quarantine/isolation centers, ( ) a data analysis center that uses machine learning algorithms, ( ) healthcare physicians, and ( ) a cloud infrastructure. the aim of this framework, is to reduce mortality rates through early detection, following up on recovered cases, and a better understanding of the disease. this work conducts an experiment to test these eight machine learning algorithms on a real dataset. the results show that five of these eight algorithms achieved accuracies of more than %. using these five algorithms will provide effective and accurate prediction and identification of potential cases of covid- , based on real-time symptom data. this paper is organized as follows. section reviews the relevant literature. section details the proposed framework, including the five components. section focuses on the identification (or prediction) of new cases, using machine learning algorithms. lastly, section concludes the work. there is considerable work in the literature regarding the use of the internet of things (iot) to deliver health services. usak et al. conducted a systematic literature review of the use of iot in health care systems. that work also included a discussion of the main challenges of using iot to deliver health services, and a classification of the reviewed work in the literature [ ] . wu et al. proposed a hybrid iot safety and health monitoring system. the goal was to improve outdoor safety. the system consists of two layers: one is used to collect user data, and the other to aggregate the collected data over the internet. wearable devices were used to collect safety indicators from the surrounding environment, and health signs from the user [ ] . hamidi studied authentication of iot smart health data to ensure privacy and security of health information. the work proposed a biometric-based authentication technology [ ] . rath and pattanayak proposed a smart healthcare hospital in urban areas using iot devices, inspired by the literature. issues such as safety, security and timely treatment of patients in vanet zone were discussed. evaluation of the proposed system was conducted using simulators such as ns and netsim [ ] . darwish et al. proposed a cloudiot-health paradigm, which integrates cloud computing with iot in the health area, based on the relevant literature. the paper presented the challenges of integration, as well as new trends in cloudiot-health. these challenges are classified at three levels: technology, communication and networking, and intelligence [ ] . zhong and li studied the monitoring of college students during their physical activities. the paper focused on a physical activity recognition and monitoring (parm) model, which involves data pre-processing. several classifiers, such as decision tree, neural networks, and svm, were tested and discussed [ ] . din and paul proposed an iot-based smart health monitoring and management architecture. the architecture is composed of three layers: ( ) data generation from battery-operated medical sensors and processing, ( ) hadoop processing, and ( ) application layers. because of the limited capacity of batteries to power the sensors, the work employed an energy-harvesting approach using piezoelectric devices attached to the human body [ ] . otoom et al. developed an iot-based prototype for real-time blood sugar control. arima and markov-based statistical models were used to determine the appropriate insulin dose [ ] . alshraideh et al. proposed an iot-based system for cardiovascular disease detection. several machine learning algorithms were used for cvd detection [ ] . nguyen presented a survey of artificial intelligence (ai) methods being used in the research of covid- . this work classified these methods into several categories, including the use of iot [ ] . maghdid proposed the use of sensors available on smartphones to collect health data, such as temperature [ ] . rao and vazquez proposed the use of machine learning algorithms to identify possible covid- cases. the learning is done on collected data from the user through web survey accessed from smartphones [ ] . allam and jones discussed the need to develop standard protocols to share information between smart cities in pandemics, motivated by the outbreak of covid- . for instance, ai methods can be applied to data collected from thermal cameras installed in smart cities, to identify possible covid- cases [ ] . fatima et al. proposed an iot-based approach to identify coronavirus cases. the approach is based on a fuzzy inference system [ ] . peeri et al. conducted a comparison between mers, sars, and covid- , using the available literature. they suggested the use of iot in mapping the spread of the infection [ ] . to our knowledge, no one has developed a complete framework for using iot technology for the identification and monitoring of covid- . equipped with sensors to collect heterogeneous data. these sensors have a limited computational capacity, and a limited lifetime. the more data that they collect, the more helpful decisions can be made. however, data processing complexity becomes a bottleneck [ ] . connectivity can be used to cope with the limited computational power of these sensors. several different communication technologies have been employed, including lowpan, bluetooth, ieee . . , rfid and near-field communication (nfc) [ ] . the network layer is not just used to upload collected data, for analysis. it is also used to facilitate communication between heterogeneous iot objects, at the physical layer. in doing this, the network layer should support scalability, as the number of the objects increases, as well as device discovery, and context awareness. significantly, it should also provide security and privacy for iot devices [ ] . the data uploaded from the iot devices can be deeply analyzed, to generate insights and help make decisions. currently, machine learning and deep learning (ml/dl) algorithms are used for this purpose, and are replacing more traditional methods because of their ability to deal with big data [ ] . al-garadi et al. provided a thematic taxonomy of ml/dl used for iot security [ ] . there are a wide range of applications where iot can be effectively used, including healthcare, smart cities, smart buildings, agriculture, and power grids. in healthcare, iot is sometimes called internet of medical things (iomt) [ ] . it has largely displaced traditional ict-based methods, such as telemedicine or telehealth. iomt can provide more advanced features than these traditional methods. for example, while traditional methods can connect patients with medical doctors remotely, iomt also supports machinehuman and machine-machine interactions, such as ai-based diagnosis. one important issue in designing iomt is the balance between data privacy/security and patient safety [ ] . examples of cyber threats that challenge such designs are eavesdropping on communication channels (to sell the collected data), intervention, disruption, or even modification of the service. however, in cases where the patient's life is at risk, breaking some security measures to access the iomt might be needed to save the patient's life [ ] . ml/dl methods can be used to support this balance. this section depicts and discusses our envisioned iot-based framework, which could be used to monitor and identify (or predict) potential coronaviruses cases, in real time. equally important, this framework could be used to predict the treatment response of confirmed cases, as well as to better understand the nature of the covid- disease. fig. shows the framework of our proposed iot architecture. it consists of five main components: symptom data collection and uploading, a quarantine/isolation center, a data analysis center, an interface to health physicians, all of which are interconnected through a cloud infrastructure. j o u r n a l p r e -p r o o f a. symptom data collection and uploading. the aim of this component is to collect real-time symptom data through a set of wearable sensors on the user's body. in our earlier study [ ] , the most relevant covid- symptoms were identified, based on a real covid- patient dataset. these identified symptoms were: fever, cough, fatigue, sore throat, and shortness of breath. there are several biosensors available to detect these symptoms. for instance, temperature-based sensors can be used for the detection of fever [ ] . cough and its classifications for different ages can be detected using audio-based sensors with acoustic and aerodynamic models [ ] . motion-based and heart-rate sensors can be used to detect fatigue [ ] . sore throat can be detected using image-based classification [ ] . finally, oxygen-based sensors can be used to detect shortness of breath [ ] . other relevant datasuch as travel and contact history during the past - weeks, can be collected in an ad-hoc manner through mobile applications. center. this component collects data records from users who have been quarantined or isolated in a health care center. these records include both health (or technical) and non-technical data. for health (or technical) data, each record includes time-series data of the above-mentioned symptoms, while for non-technical data, each record includes travel and contact history during the past - weeks, chronic diseases, age, gender, and any other relevant information, such as family history of illness. each record would eventually also include the treatment response for each case. c. data analysis center. the data center hosts data analysis and machine learning algorithms. these algorithms are used to build a model for covid- , and to provide a real-time dashboard of the processed data. the model could then be used to quickly identify or predict potential covid- cases, based on real-time data collected and uploaded from users. the model can also predict the patient's treatment response. over time, the disease models developed from this data will provide useful information about the nature of the disease. physicians will monitor suspected cases whose real-time uploaded symptom data indicates a possible infection by our proposed machine learning based identification/prediction model. the physicians will then be able to respond swiftly to these suspected cases by following up with any further clinical investigation needed to confirm the case. this allows the confirmed cases to be isolated and given appropriate health care. e. cloud infrastructure. the cloud infrastructure is interconnected through the internet, and ( ) allows upload of real-time symptom data from each user, ( ) maintains personal health records, ( ) communicates prediction results, ( ) communicates physician recommendations, and ( ) provides for storage of information. j o u r n a l p r e -p r o o f . the system non-invasively collects real-time user symptom data through wearable devices and sensors. again, these symptoms are: fever, cough, fatigue, sore throat, and shortness of breath. further, the user submits information via a mobile application about living in (or travel to) infected areas, as well as possible contact with covid- infected persons. the quarantine/isolation center also periodically submits data from their isolated and quarantined patients who are housed in the center. the content of that data is similar to the real-time data collected from users. . the sensed symptom data are uploaded to the data analysis center using a smartphone, through the cloud infrastructure. digital records from the health care center are also regularly sent to the data analysis center through the cloud infrastructure. the data analysis center hosts machine learning algorithms, which use the data received from the health care center to continuously update its models. the models are then used to identify potential cases, based on the real-time symptom data from each user. the data center also analyzes all its data, and presents the results on a real-time dashboard. that dashboard can be informative to physicians about the nature of the virus. . if a potential case is identified, it will be sent to the relevant physician to follow up with the patient. the patient will then be called and encouraged to visit the health care center for clinical tests, such as the polymerase chain reaction (pcr) test, which is used to identify positive cases. if it turns out that the case is confirmed, the patient can be isolated, and all contacts will be contacted and quarantined. a complementary and integral component to this framework is the use of the same mobile application to educate users, by including useful information on how they can avoid illness, and how to avoid being exposed to the virus. this section further discusses the predictive models, and the machine learning algorithms that will be employed in the data center component of the proposed iot-based framework. in particular, an experiment was conducted to investigate the possibility of using machine learning algorithms for quick identification (or prediction) of potential covid- infections. the rest of this section describes that experimental setup, and presents and discusses the results. a dataset of confirmed covid- cases from the covid- open research dataset (cord- ) repository [ ] was used. the data contains different types of information about each case. our work focused on symptoms, travel history to suspicious areas, and contact history with potentially infected people. however, some of this information was missing for many of the cases documented within the database. moreover, the data was not well structured for use by machine learning algorithms. in our previous work [ ] , the data was preprocessed and structured to be better suited for machine learning. the cases with documented symptoms were collected. this resulted in a list of symptoms. however, many of these symptoms were judged to be synonyms. thus, the number of symptoms was reduced to . this merging of synonymous symptoms was done in an ad-hoc manner by two medical doctors, who are co-authors of this work. for example, "anorexia" and "loss of appetite" were merged together. our previous work also determined the relative importance of these symptoms. the following six different statistically-based feature selection algorithms were employed in that work, to rank the symptoms, based on their importance: spectral score, information score, pearson correlation, intra-class distance, interquartile range, and our variance based feature weighting [ ] . the first five of these methods had been proposed earlier in the literature [ ] . the sixth method was a new one. it not only ranks the symptoms, but also assigns importance weights to each of them. it was found that the most important five symptoms (ordered from most important to least important) are: fever, cough, fatigue, sore throat, and shortness of breath. based on the findings of that earlier work, this work uses those five most important symptoms. in addition, two extra features were added: live and contact. the first feature (live) represents whether or not the person lived, travelled to, or passed by a potentially infected area. the second feature (contact) represents whether or not the person was known to be in contact with a potentially infected person. this resulted in a preprocessed dataset of × data records. among which of those records were from confirmed covid- cases, and records were for non-confirmed cases. this work used this preprocessed dataset to build a predictive model for our identification (or prediction) system. the function of this model is to estimate the likelihood that a given person is infected by covid- . several learning algorithms (i.e. classifiers) could have been used for this purpose. those classifiers can be categorized into multiple categories. weka software [ ] , (which we used in this work) categorizes the classifiers into six categories: ( ) functionbased classifiers, such as support vector machines, ( ) lazy classifiers, such k-nearest neighbors, ( ) bayes based classifiers, such as naïve bayes, ( ) rule-based classifiers, such as decision tables, zeror, and oner, ( ) tree-based classifiers, such as decision stump, and ( ) meta classifiers, such neural networks. in this work, at least one classifier from each category was selected. specifically, this work compares the performance of eight machine learning algorithms: ( ) support vector machine (svm), using radial basis function (rbf) kernel, ( ) neural network, ( ) naïve bayes, ( ) k-nearest neighbor (k-nn), ( ) decision table, ( ) decision stump, ( ) oner, and ( ) zeror [ ] . this work used weka software to run all of these algorithms on our dataset [ ] . the default parameter values were used for each of the eight algorithms. below is a brief description of the eight algorithms: svm is a supervised learning method. given a set of training examples that are labeled (i.e. each instance in the training set either belongs to the positive or negative class), svm learns the hyperplane that best separates the instances from each class, and maximizes the margin between the data instances and the hyperplane itself. this learnt hyperplane is then used to assign (or predict) a class label for any new test instance. ann is a supervised learning method. the learning process tries to mimic the learning that takes place inside the human brain. to do so, multiple layers of nodes are connected through edges. the edges connecting between the nodes are represented as numerical weights. the output of each node is computed as weighted sum of its inputs. given a set of training examples that are labeled (i.e. each instance either belongs to the positive or negative class), the ann learns the numerical weights that best classify the instances from each class. this learnt model is then used to assign (or predict) a class label for any given test instance. the test instance drives the inputs to the nodes of the first layer. then a threshold is applied to the outputs of the final layer, to determine the label for that test instance. naïve bayes is a supervised learning method. the learning process follows a probabilistic approach. it uses bayes theorem to compute the model parameters. given a set of training examples that are labeled (i.e. each instance either belongs to the positive or negative class), naïve bayes computes multiple model parameters, such as the probability of each class label to occur. these parameters are then used to assign (or predict) a class label for any given test instance. this is done by computing the probabilities of the test instance to be assigned to each of the possible class labels. the maximum value among these probabilities decides the label of that test instance. k-nn is a supervised instance-based learning method. the learning process follows a lazy approach. it does not compute a model. given a set of training examples that are labeled (i.e. each instance either belongs to the positive or negative class), k-nn computes distances between a given test instance and all the training instances. these distances are then used to assign (or predict) a class label for the test instance. this is done by aggregating the class labels of the k closest training instances to the test instance. table decision table is a decision stump is a supervised learning method. given a set of training examples that are labeled (i.e. each instance either belongs to the positive or negative class), this method computes a model by building a decision tree, with only one internal node. in other words, it makes the prediction for any given test instance using only one feature of that instance. this feature is determined by computing the information gain for all features across all training instances, selecting the one with the maximum information gain value. oner is a supervised learning method. given a set of training examples that are labeled (i.e. each instance either belongs to the positive or negative class), this method computes a model by generating one rule for each feature in the data set. it then selects the one with the minimum total error. zeror is a supervised learning method. given a set of training examples that are labeled (i.e. each instance either belongs to the positive or negative class), this method computes a model by using only the target feature (i.e. class) while ignoring all other features. it is considered the simplest classification method. it assigns any new test instance to the majority class. usually, it is used as a benchmark to determine baseline performance. to evaluate the performance of the eight learning algorithms, four performance measures were used: accuracy, root mean square error, f-measure, and roc area. these measures can be computed using a confusion matrix and cross validation methods. the confusion matrix is used to visualize the performance of a binary ( -class) supervised learning problem by creating a -by- matrix. each row in the matrix shows the instances in the predicted (or computed) class, while each column shows the instances in the actual class. the resulting matrix consists of four values (see table cross validation is a statistical method used to measure the performance of learning and classification methods. this is done by splitting the available labeled data instances into k folds. one of these folds is used for testing, and the rest are used for training. this work used -fold cross validation. the data instances are divided into folds. for iterations, one-fold was used for testing and folds for training, such that in every iteration a different fold is used for testing. the accuracy of a classifier is computed as the number of correctly classified instances to the total number of instances. it is given by: the root mean square error (rmse) is computed as the square root of the average of squared differences between the predicted classes (or labels) and the actual ones. it is given by: the f-measure is computed by combining the two measures of precision and recall. it is given by: the receiver operating characteristic (roc) is another way to measure the performance of a classifier. this is done by plotting the true positive rate against the false positive rate. the area under the resulting roc curve is then used to measure the performance of the classifier. the closer the area to is, the better the classifier is. the true/false positive rates are given by: fig. shows the confusion matrices that resulted from applying -fold cross validation to the eight selected classifiers. (large numbers in the upper-left and lower right boxes of these matrices represent good scores. large numbers in the lower-left and upper right boxes of these matrices represent bad scores.) fig. shows the roc curves that resulted from applying -fold cross validation to the eight selected classifiers. fig. shows the roc curves that resulted from applying -fold cross validation to the eight selected classifiers. table and fig. compare the performance of the eight algorithms. it shows the accuracy, root mean square error, f-measure and roc area of each algorithm, which were calculated using the well-known -fold cross validation method [ ] . the results presented in table and fig. suggest that the models built using svm, neural network, naïve bayes, k-nn and decision table algorithms are effective in predicting confirmed and potential cases of covid- . taken together, this suggests that our proposed iot-based framework could use a combination of these five effective models. this could be done by aggregating the results of these five learnt models, based on majority votes. this paper has proposed an iot-based framework to reduce the impact of communicable diseases. the proposed framework was used to employ potential covid- case information and health records of confirmed covid- cases to develop a machine-j o u r n a l p r e -p r o o f learning-based predictive model for disease, as well as for analyzing the treatment response. the framework also communicates these results to healthcare physicians, who can then respond swiftly to suspected cases identified by the predictive model by following up with any further clinical investigation needed to confirm the case. this allows the confirmed cases to be isolated and given appropriate health care. an experiment was conducted to test eight machine learning algorithms on a real covid- dataset. they are: ( ) support vector machine, ( ) neural network, ( ) naïve bayes, ( ) k-nearest neighbor (k-nn), ( ) decision table, ( ) decision stump, ( ) oner, and ( ) zeror. the results showed that all these algorithms, except the decision stump, oner, and zeror achieved accuracies of more than %. using the five best algorithms would provide effective and accurate identification of potential cases of covid- . employing the proposed real-time framework could potentially reduce the impact of communicable diseases, as well as mortality rates through early detection of cases. this framework would also provide the ability to follow up on recovered cases, and a better understanding the disease. j o u r n a l p r e -p r o o f coronavirus covid- global cases by the center for systems science and engineering who director-general's opening remarks at the media briefing on covid- - the most promising coronavirus breakthroughs so far, from vaccines to treatments digital technologies and disease prevention digital technology for preventative health care in myanmar a fog-computing architecture for preventive healthcare and assisted living in smart ambients personalized telehealth in the future: a global research agenda health care service delivery based on the internet of things: a systematic and comprehensive study an internet-of-things (iot) network system for connected safety and health monitoring applications an approach to develop the smart health using internet of things and authentication based on biometric technology technological improvement in modern health care applications using internet of things (iot) and proposal of novel health care approach the impact of the hybrid platform of internet of things and cloud computing on healthcare systems: opportunities, challenges, and open problems internet of things sensors assisted physical activity recognition and health monitoring of college students erratum to "smart health monitoring and management system: toward autonomous wearable sensing for internet of things using big data analytics artificial intelligence in the battle against coronavirus (covid- ): a survey and future research directions a novel ai-enabled framework to diagnose coronavirus covid- using smartphone embedded sensors: design study identification of covid- can be quicker through artificial intelligence framework using a mobile phone-based survey in the populations when cities/towns are under quarantine on the coronavirus (covid- ) outbreak and the smart city network: universal data sharing standards coupled with artificial intelligence (ai) to benefit urban health monitoring and management iot enabled smart monitoring of coronavirus empowered with fuzzy inference system the sars, mers and novel coronavirus (covid- ) epidemics, the newest and biggest global health threats: what lessons have we learned real-time statistical modeling of blood sugar a web based cardiovascular disease detection system intelligent multi-dose medication controller for fever: from wearable devices to remote dispensers a mobile cough strength evaluation device using cough sounds heart rate monitoring system during physical exercise for fatigue warning using non-invasive wearable sensor novel image processing method for detecting strep throat (streptococcal pharyngitis) using smartphone extraction and analysis of respiratory motion using wearable inertial sensor system during trunk motion cord- ). . version - - introduction to data mining. pearson education india the weka workbench. online appendix for "data mining: practical machine learning tools and techniques internet of things: a survey on enabling technologies, protocols, and applications fog computing: helping the internet of things realize its potential a survey on the internet of things security context aware computing for the internet of things: a survey internet of things: architectures, protocols, and applications the role of big data analytics in internet of things a survey of machine and deep learning methods for internet of things (iot) security security and privacy issues in implantable medical devices: a comprehensive survey security tradeoffs in cyber physical systems: a case study survey on implantable medical devices a novel computational method for assigning weights of importance to symptoms of covid- patients filter feature selection for one-class classification the authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. key: cord- -op hiqw authors: lee, joseph g. l.; leprevost, catherine e.; harwell, emery l.; bloss, jamie e.; cofie, leslie e.; wiggins, melinda f.; firnhaber, gina c. title: coronavirus pandemic highlights critical gaps in rural internet access for migrant and seasonal farmworkers: a call for partnership with medical libraries date: - - journal: journal of the medical library association : jmla doi: . /jmla. . sha: doc_id: cord_uid: op hiqw migrant and seasonal farmworkers, who are essential workers in the coronavirus global public health emergency, face unique risks to their health as well as longstanding health inequities. this commentary highlights these risks and argues that internet access represents an underappreciated but critical part of the public health response. the authors first discuss the unique risks farmworkers face. we note the importance of internet access in the time of physical distancing, the fact that many health outreach workers are no longer visiting camps, the need for telemedicine infrastructure, and the role of internet access in providing connections to families in communities of origin. we describe existing efforts that have been implemented in north carolina to raise awareness among public health and health promotion practitioners and researchers. the current coronavirus pandemic demands the attention of medical libraries, public health practitioners, and policy makers to address the digital divide for farmworkers and their families. the global public health emergency from severe acute respiratory syndrome coronavirus (sars-cov- ), which we refer to as "coronavirus" for simplicity, has caused global disruptions and substantial morbidity and mortality. many communities and states, including the authors' state of north carolina (nc), have mandatory stay-athome orders in place at the time of this writing. essential workers, however, continue to be on the job, and, as paul farmer warned in infections and inequalities: the modern plagues [ ] , the consequences of existing structural inequalities are beginning to manifest in higher rates of infection and death for those with fewer resources [ ] . one group of essential workers has unique vulnerabilities, which are created and maintained by having few legal protections: migrant and seasonal farmworkers [ , ] . migrant and seasonal farmworkers, who represent a critical part of rural economies and us food systems, face health inequities, poor and overcrowded housing conditions, limited access to personal protective equipment and handwashing facilities in the fields, and lack of access to health information [ , ] . these issues required attention before the pandemic, and they are even more urgent now. farmworkers' living and working conditions may make it impossible for them to adhere to physical distancing guidelines, given existing minimum standards. many workers, who arrive from mexico on h a temporary agricultural workers visas or with a crew of migrant workers from another state in the united states, spend long hours together on a bus. upon arrival, protections are limited: beds in employer-provided migrant housing must be placed a minimum of three feet apart. in shared kitchens and bathrooms, one sink must be available for every six people, one stove for every ten, and one toilet for every fifteen [ ] . these limited protections can make it difficult for farmworkers to comply with health recommendations. indeed, the farm labor journal of the medical library association ( ) october jmla.mlanet.org organizing committee (floc), a union of farmworkers, issued an emergency call on april , , for donations of basic supplies such as cleaning products, handwashing soap, telephone cards, and toilet paper. floc's members reported arriving for the growing season to nc housing camps that lacked these critical supplies. work conditions also involve risk. depending on the crop and task, farmworkers work in large groups and in close proximity (e.g., riding on a tractor implement to transplant tobacco) in fields where the minimum washing facilities are set to be one per twenty people [ ] but are sometimes altogether unavailable [ ] . connections to health care for farmworkers are limited by rural geography, language, and barriers to internet access. most farmworkers in nc live in isolated areas in rural, agricultural communities that have limited connectivity. there is longstanding evidence of inequities in access to the internet by race, ethnicity, and rurality [ , ] , and these inequities extend to farmworkers [ ] [ ] [ ] . access to the internet has been declared a basic human right by the united nations in article of the universal declaration of human rights, with economic and political reasons such as the rights to freedom of opinion and expression [ ] . it is also important for accessing health information, connecting with health outreach workers during physical distancing, and maintaining connections with families who have remained in communities of origin. the authors focus this commentary on the consequences of limited internet access for farmworkers and their families during a pandemic and we note the need to substantially expand efforts that are underway to address this problem. in a - pilot project with youth from seasonal farmworker families in central nc, half of participants did not have internet access at home [ ] . the importance of having internet access is highlighted during this global pandemic, given that access to up-to-date information from reputable sources can mean the difference between stemming the spread of coronavirus or accelerating the rate of illness and death among groups who are most vulnerable [ ] . during the coronavirus outbreak, there has been a wealth of information online for those who have internet access. yet, many farmworkers have no access to these important resources from state and local governments, health department websites, and worker organizations, and no way to keep abreast of new health information as it is discovered and disseminated. the quality of society's response to a global pandemic depends on meeting communication needs of all populations, with particular attention given to the most at-risk and vulnerable populations [ ] . there are approximately farmworker health outreach workers (i.e., community health workers or promotoras de salud) each agricultural season in nc. outreach workers provide culturally sensitive and occupationally relevant health education, health assessments, and case management services, as well as organize access and transportation to clinical services [ ] . outreach is typically conducted in person as most employer-provided housing does not have phones or internet. to minimize risk to outreach workers as well as the spread between farmworker housing sites, many of nc's outreach workers are effectively grounded at the time of this writing. some services can be administered virtually, such as sharing health information and organizing appointments with physicians or specialists, but outreach workers rely heavily on their ability to meet physically with farmworkers. in focus groups that were recently conducted with funding from the national library of medicine (award number g lm ), outreach workers in nc reported that in-person, interactive health education activities yielded the best results for engagement and information retention among farmworkers. routine health assessments and clinic visits traditionally necessitate in-person interaction. the current pandemic and the resulting physical distancing requirements have removed a fundamental mode of service provision from outreach workers. turning to the existing virtual infrastructure, attempts by outreach workers to maintain connection with farmworkers are frustrated by limited and inconsistent access to internet and jmla.mlanet.org ( ) october journal of the medical library association cellular service for both farmworkers and outreach workers. without regular site visits, many farmworkers' only point of connection to outreach workers is via phone or, if available, text messaging. under pandemic conditions, facilitation of telemedicine appointments and health information sharing are the only enabling services that outreach workers can administer. telemedicine services require strong internet connection and audiovisual hardware that, due to geography and economic conditions, are often unavailable to farmworkers. even when the infrastructure does exist, online health information that is relevant to farmworkers is diffuse and not always culturally or linguistically sensitive. as outreach workers move to adapt service offerings and sustain connections with farmworkers in response to physical distancing requirements, the need for a more robust virtual infrastructure to support rural outreach work is clear. for many, internet access provides now ubiquitous virtual meetings with friends, family, and coworkers in this time of physical distancing and stay-at-home orders. previous work has shown that providing internet access to migrant farmworkers allowed better connection with family members who remained in communities of origin [ ] . indeed, the most common applications (apps) used by farmworkers who were provided internet in farmworker housing were social and communication apps like facebook and messaging apps like whatsapp [ ] . these apps can serve as a means to gather and exchange vital information with loved ones. they are also a way for farmworkers to mobilize their social ties both in the united states and countries of origin to provide badly needed resources to their relatives and friends in the most vulnerable situations during a crisis. without internet access in this time of crisis, farmworkers are deprived of the ability to sustain connections to friends and families. the existing challenges caused by lack of internet access have been recognized by researchers and advocates. nonprofit organizations like student action with farmworkers (saf), based in durham, nc, have pioneered early efforts to connect farmworkers with internet access. funded in part by the nc farmworker health program, saf's computers and connectivity project piloted efforts to place desktop computers and cellular wifi hot spots in migrant housing at four labor camps across nc [ ] . an external evaluator found many positive benefits for farmworkers in terms of emotional support from being able to connect with family [ ] . in a separate project funded by the national network of libraries of medicine, saf partnered with east carolina university's department of health education and promotion and joyner library to provide health information literacy training, ipads, and cellular data service to middle and high school students from farmworker families who were participating in a leadership development and college access program. the evaluation noted the substantial impact on the students of gaining internet access for homework and access to health information [ ] . our team is using funding from the national library of medicine (award number g lm ) to extend these two pilot projects with partnerships across nc using a train-the-trainer approach, development of best practices, identification of materials for wider use, and assessment of the sustainability of efforts to provide better access to the internet. our work will not, however, be able to fill the substantial gaps in internet access for migrant and seasonal farmworkers in nc, much less nationally. thus, it is critical to recognize the imperative prescribed by the coronavirus pandemic: rural and farmworker health, medical libraries, emergency preparedness, education, rural economic development, and broadband infrastructure must be brought together to address barriers to internet access in ways that include migrant and seasonal farmworkers. research reported in this publication was supported by the national library of medicine of the national institutes of health under award number g lm . infections and inequalities: the modern plagues black americans face alarming rates of coronavirus infection in some states overview of immigrant worker occupational health and safety for the agriculture, forestry, and fishing (agff) sector in the southeastern united states latino farmworkers in the eastern united states: health, safety, and justice the human cost of food: farmworkers' lives, labor, and advocacy perceptions of housing conditions among migrant farmworkers and their families: implications for health, safety and social policy. rural remote health occupational safety and health standards: general environmental controls: temporary labor camps occupational safety and health standards: general environmental controls: field sanitation observed and self-reported pesticide protective behaviors of latino migrant and seasonal farmworkers navigating the digital divide: a systematic review of ehealth literacy in underserved populations in the united states the digital divide: percentage of households by broadband internet subscription, computer type, race and hispanic origin providing health information to latino farmworkers: the case of the affordable care act hispanic migrant farm workers' attitudes toward mobile phone-based telehealth for management of chronic health conditions mobile and traditional modes of communication among male latino farmworkers: implications for health communication and dissemination is internet access a human right? linking information and communication technology (ict) development with global human rights efforts a pilot project to increase health literacy among youth from seasonal farmworker families in rural eastern north carolina: a qualitative exploration of implementation and impact a human right to access the internet? problems and prospects. human rights law rev effective health risk communication about pandemic influenza for vulnerable populations the health and well-being of north carolina's farmworkers: the importance of inclusion, accessible services and personal connection student action with farmworkers. computers and connectivity in farmworker housing: evaluation report nc: student action with farmworkers key: cord- -nd ozu w authors: furr-holden, debra; carter-pokras, olivia; kimmel, mary; mouton, charles title: access to care during a global health crisis date: - - journal: health equity doi: . /heq. . .rtl sha: doc_id: cord_uid: nd ozu w nan access to care has been an ongoing health care issue for socially and economically disadvantaged populations in the united states for many decades. the recent covid- pandemic has highlighted these disparities as people of color suffer disparate mortality and face growing inequities in care. this moderated panel discussion provides a broad insight into these issues and discusses the need for greater attention to the access to care problem for many u.s. communities. dr. debra furr-holden: thank you all for joining on this roundtable discussion about health equity and access to care. to start our conversation, i would love to hear from each of you about what you see as the current highest priorities regarding access to care. dr. olivia carter-pokras: in preparation for this discussion, i reached out to other members of the latino health steering committee from montgomery county, maryland, because we are a mix of academics as well as health care providers and others, so i wanted to bring in voices other than just my own. at the time of this discussion, we are dealing with the covid- pandemic. there are many factors that can result in lower access and quality of care received by racial and ethnic minorities, in general, and specifically for those with covid- infection. [ ] [ ] [ ] these factors can include not knowing where to go for testing or where to get care, for instance, as well as the ability to pay, transportation to get there, and whether or not the person has a usual health care provider (referral is often required to receive testing). the lack of interpretive services continues to be a big issue. only about half of foreign-born latinos in maryland have health insurance, even though maryland is ranked second in the nation in median household income. , health issues are interrelated with other problems such as unemployment and stress. these are long-standing concerns that have only been exacerbated by the coronavirus pandemic. on top of that, we have individuals who are undocumented and afraid to seek care. as we learned after / , many are ''frightened because we don't know what is going to happen (with immigration).this is an emergency.'' that fear certainly is coming into play with people who, even if they have the signs and symptoms, even though they are very ill, are refusing to go in for care. many of them were afraid to leave their homes even before all of this happened because of that fear. immigrant workers are a significant share of essential workers these days, from staffing grocery stores to cleaning hospital rooms and transporting food. although % of civilians working in were foreign born, it is a much larger percentage when you look specifically at workers on the front line. immigrant workers are also overrepresented in jobs that are currently shuttered, such as hotels and restaurants, cleaning services, personal services, and hair and nail care. lack of income right now is a big issue and a major disparity during this crisis. language is still an issue when trying to figure out where to go for testing; information on testing has mainly been available only in english. and finally, inconsistent messaging continues to be a major challenge. i think we learned a lot in the past about not sufficiently dealing with health literacy. when we think about who should get tested, we get one message from the cdc, one message from the state, and one message from the county health department, but none of them are consistent, and, frankly, misinformation within our communities has been shared by the mainstream media as well as through social media. for instance, using neosporin in the nose to help prevent getting the virus was one piece of misinformation that was spread early on. these issues existed before covid- , such as the lack of insurance and the lack of interpretation and translation of materials, but the pandemic has compounded the problem significantly. dr. mary kimmel: i want to talk about mental health priorities regarding access to care. building upon what dr. carter-pokras just talked about, the same stressors are still there, and the same issues are still there, but now things are magnified. i actually hope that one silver lining would be that we can learn from these issues and figure out how to handle some of them. we are seeing that with mental health. we already had many people who did not feel able to talk about their mental health or mental wellness. and i like the term ''mental wellness,'' because when i say ''mental health,'' people often think of that as illness. i really want us to start thinking about mental health and wellness as our mental life, which includes for all us all the things we manage, our relationships, the stress we manage such as from limited finances and access to food and housing and health care. so those things were already there, and people already did not feel they could always talk about how they were feeling and their emotions and how they were managing stressors and afraid what people would think if they did talk about how they are managing stressors. but now, i think the good thing about this crisis is that people are acknowledging that this is highly stressful and that this is something that is going to make other people very stressed too. going forward, i anticipate that we are only going to see increases of this reflection on some of the things in the news and thinking about the number of people who are being impacted by this in different ways, either by family members being ill, being afraid that they are ill and not knowing, not knowing whether they are going to infect their families, and just the trauma of all this is going to be longstanding. one nice thing is this has really propelled us forward in terms of telehealth. i am the medical director for a program called north carolina maternal mental health matters, and that program provides education to frontline providers, such as obstetricians and family practice physicians and pediatricians, to help them support mental health in their patients. but through that, we also had intended to pilot a telehealth assessment that we could do to support primary care providers with psychiatric assessment. owing to new support for telehealth, we have been able to move forward with these telehealth assessments. whereas this was previously taking a while to get going, are now being pushed forward, and payers are now saying this is something they are going to pay for. going forward, we need to work to say that those things are important, and we should continue those things even after the immediate need is gone. but i also, from that, have found some of the inequity. i certainly have worked with patients who are supposed to be doing our visits through our medical records system but are experiencing access challenges. some patients cannot access the internet, and they do not have the bandwidth to do that kind of visit. i have done phone visits with some patients, and i really hope that we can also move that forward as an option when video visits for mental health support are not possible. that has historically been something that has not been compensated as well, to do phone visits, and i really hope that is something now that payers will start to think about as important. the telephone visits have gone very well. but even there, again, a patient may experience inequity, because there are some patients who do not have a reliable phone or a secure private place to talk. this is particularly true as families are being forced to stay home. so still some patients are not able to talk as openly about their mental health needs and about stress they are facing if they cannot find privacy. for example, this is certainly relevant as we are seeing the increases in intimate partner violence some individuals are facing in their ''shelter-inplace'' housing situations. so, by moving all these things to telehealth, i think we have to be cognizant that some patients are not going to be able to access those things, and so in some ways, we are taking away barriers from people getting access to mental health care, but i also think we are also highlighting some increased barriers for some individuals. dr. charles mouton: yes. thank you. i think that some of the things that dr. kimmel has mentioned are also true in primary care. it looks like this health crisis is going to usher in a new way to approach health care and primary care, but i think in a lot of respects, these possible advancements leave behind some of the most vulnerable. with the covid- crisis, there have been closures of many primary care clinics, federally qualified health centers, and urgent care centers, with many of them switching to telemedicine visits. the issues highlight that many of the most needy do not have access to appropriate internet or computers to do any kind of video visits; this applies in the primary care setting as well. the inability to get adequately reimbursed for telehealth visits, with the requirement that it must be video captured, means that these telehealth visits may not get reimbursed. these are things that the centers for medicare & medicaid services (cms) is going to need to change if telehealth will move forward. the impact of the cms regulations on telehealth for some of the community providers that are not affiliated with large provider networks is even more profound; these are essentially private practices. so now you put them in a position where they do not have the necessary protective equipment to allow patients to come through their doors, they cannot see their patients face to face, and you force them to employ a new telehealth capability that they have to now get up and running. these limitations and loss of patient care revenue have many looking at closure of their practices. so now an already distressed underserved population becomes even more underserved and facing a real stressful time in terms of trying to get access to health care. in this access to care vacuum, you are seeing people trying to take care of themselves and family members with home remedies. we have heard about the unfortunate incidents with people using hydroxychloroquine inappropriately. we have instances of people, because of the misinformation that is being spread, using neosporin and other things. i think that people are becoming a little bit more desperate because they are unable to get the care that they feel that they need. on the opposite end of access to care is the issue, when they do wind up getting hospitalized, they then face problems with the distribution of the current health care resources. this reflects my concern about potential inequities that may arise when the decision has to be made about who gets a ventilator and who does not. we already know that implicit bias influences health care decision making. we have shown that over and over again. how does implicit bias influence a critical resource decision for people who are the most vulnerable? that question has yet to be explored and adequately addressed. i do not know of any effective strategies for eliminating implicit bias. and now we are faced with trying to come up with a strategy to deal with critical lifesaving decisions in stressed health care environments. will these biases have an untoward influence on not only access to care, but also people's general willingness to seek care? these are areas we need to have some attention paid, in particular, to alleviate. dr. furr-holden: i have been talking to a colleague who is a physician, and he said he had an ''aha moment.'' he is an anesthesiologist, so he is there when patients are being intubated, et cetera. and he said he watched a provider have a conversation with a family that said, ''this person has a zero percent chance of survival on a vent.'' so, you know, in essence, he talked the family into a ''do not resuscitate'' (dnr). there was really no medical foundation for that. so you talk about implicit bias, and it sounded a little bit like you were talking about bias on the side of the patients. what about bias on the side of providers? you know, we have been looking at and we have been experiencing these tremendous disparities. what we now know and are seeing is that it is not just the disparity of people dying on the vent, but also what is happening in the system of care where they are not even making it to the vent? how do you all think implicit bias is playing into the conversations that have no criteria for anybody else in the room? so some well-meaning provider potentially is steering black and brown and colored and other disparate populations toward dnrs. what do you think about that? dr. mouton: i think that implicit bias is, as they say, a real and present danger. i think providers are well meaning, but they are still making judgments. when thinking about people's ability to recover when they have comorbid conditions, it is important to bear in mind that those comorbid conditions are often a reflection of social inequities and society. and now, providers are having to tell these patients with certain comorbidities that when they need a health care resource, they may be ineligible. is that compounding the inequities that already exist? i think it is a difficult question. ethicists continue to debate the proper distribution of resources during times of scarcity. but, again, i am convinced that our own implicit biases influence how we make judgments and decisions. i worry that these biases will potentially lead to greater inequities as this crisis moves forward. more importantly, some of the policies and processes that are being put in place to deal with resource scarcity caused by the covid- crisis may wind up becoming the new standard that we turn to when making these kinds of decisions, so it is concerning. dr. carter-pokras: differential access and quality of care and implicit bias have been discussed in numerous reports over the past few decades, for example, with the task force report on black and minority health and the unequal treatment report from the institute of medicine. after taking into account health insurance coverage and other issues, you still see that difference in access to certain treatments when you look at many cardiovascular and other health conditions. the comment that dr. mouton just made is absolutely right. when it comes down to it during times of really tight constraints like what people are experiencing right now, decisions are indeed being made about who is going to receive care and who is not, and who is going to receive what could be considered lifesaving care and who is not. unfortunately, racial and ethnic minorities are more likely to have conditions that put them at greater risk for severe complications. right now, with covid- , underlying conditions that could prove to make outcomes worse could include diabetes, cardiovascular disease, hypertension, or asthma. we already know that racial and ethnic minorities have a disproportionate burden of illness and death. the ethicists certainly have their jobs cut out for them. even though this is the first major pandemic we have had since , it is not to say that we will not be dealing with another issue in the future wherein we are going to have to deal with major constraints and burdens to our existing health care facilities. the mental health of our providers is highly stressed and strained. important decisions are having to be made right now in almost a fight-or-flight mode, and those may tend to be even more likely wrought with implicit and unconscious bias. because there is already so much bias against mental health, that is just adding another layer of complexity. as much as people are saying that we understand how stressful this is, i do not know that people always still really do fully understand. i just read an article on a website about finding a therapist during this time, and it tries to give a number of resources no matter your insurance or need. however, our system is still very much biased against those needing mental health support and set up for self-pay. although telehealth opens new avenues, it still is going to be more accessible for those who can afford to pay a therapist online through numerous services popping up to take advantage. those who cannot afford these services will still have to deal with the same biases such as mental health care not always as covered as other health care needs. we are all bringing bias about what is needed and how to use resources appropriately. we have talked a bit about telehealth and telemedicine. what do you think are some of the best practices that you think are being underutilized, and what do you see as really the stopgaps or policies that we need to implement as a part of the standard of care? dr. mouton: there is a major economic gap within the health care sector. inner city and rural provider practice differ substantially in the financial resources they have available to reinvest in new practice technologies. i think that as we try to use technology to provide access to health care, without policy guidance and financial support, particularly from cms, there will be a number of obstacles. if telehealth is here to stay, which most people say it is, then coupled with the expansion in digital health care monitoring, such as using the internet of things in people's homes to support better monitoring, we can start to see an intensification of the inequities for the communities served by these practices. what does that mean for someone who has no internet and no things? so, you talk about the internet of things, but many of our patients who are disadvantaged in terms of access and health care provision also lack those tools and devices. more and more, health care is going to move in that direction, and unless the funders of health care decide this is of big enough value to distribute to those who no longer can afford it or do not have it, to allow them to reap the benefits of this, we will again create a broader and deeper gap in terms of what people are able to receive and benefit from. so, to me, that is something that, if we would address, would go a long way to bridging that gap. but it is going to take an investment to make that leap forward. dr. carter-pokras: one of the things that has been brought to bear is that people are recognizing now how good or how poor their internet access is in a way that i do not think they really fully understood previously. we are understanding that in under-resourced communities where children have been expected to continue their schooling at home, they are unable to continue their online studying because they do not have enough bandwidth, even if they do have a computer. , schools have been making computers available to the children who do not have computers at home, but you have to stand in line within a certain period of time to go pick up the computers, and even then there may not be enough computers. and those who do finally get their hands on a computer are not given the online homework that they were supposed to have been given, or their parents cannot help them, whether it is because of language issues or other issues. so, in addition to universal health care, universal access to the internet is something that we can put on our wish list. on top of everything else, some people have been asked to vacate their housing. , can you imagine? forget about access to the internet, right? they have been asked to vacate their housing. they do not have a way to pay for their housing because they lost their job, and they have been asked to vacate their housing. we are also seeing an increase in domestic violence and calls to mental health lines with people experiencing desperation, anxiety attacks, suicidal tendencies, et cetera, because of this. so universal health care and universal access to the internet are two things we definitely want to think about. in the meanwhile, we cannot have everything available only online. it has to be available in other ways that people can access because not everybody has access to the internet right now. it is interesting that you mention these points, like the mental health concerns and people being evicted, because again, we are thinking about, what are the policies to support some of these things? as an example, michigan has issued a stay on all evictions, so people are not actually allowed to be evicted during this period. now, i think about homeowners who maybe moved and are renting out their house but are dependent on that rental income to pay their mortgage. well, their mortgage is still coming due, so all these pieces are connected. if you think about it from a policy perspective, i love when you say not just universal health care, but also universal access to the internet, especially if we are talking about telehealth and telemedicine and some of these other things. if we can identify the gaps, if we can see the problems, then how should we respond in terms of policy? what happened in michigan is then they started adding to the stay of eviction order to prevent people from being evicted, because, of course, housing is so essential, if the whole order is to stay put to stay safe. but then they started to put resources in place for noncommercial owners of homes who are providing housing so that they are also solvent during this time. sometimes the one policy creates another problem that is just one step down the road. we need to be asking what these comprehensive policies are that will actually get us somewhere and create solutions, not displace or move the problems one tick down the road. dr. kimmel: along those lines, i have talked to some pregnant women who already had housing instability before all this, and they had been saving up to get more stable housing of their own before the baby comes. but now they cannot go out and must continue to make difficult cramped housing situations continue to work. many services individuals may have once accessed at a physical office, such as working with someone to get a housing voucher, cannot be done during this time when so many things have to remain closed. online is now sometimes the only way to access things. we need internet, better internet, but we also need better and more case management and services to help people navigate, because the system was always really fractured, but i feel like now it is even more unclear who can help and who cannot. when we are all allowed to go back to work, there are still going to be these downstream effects of when a person did not find housing at the critical time before the baby came. dr. furr-holden: i have what i call a ''man-onthe-moon'' idea. what i mean by that is, when they said ''man on the moon,'' that just seemed like it was purely aspirational and almost impossible to achieve at the time, but despite that, it was well resourced, and it happened. what do you see as the kind of the man on the moon? where should we really be setting our sights if we are actually and truly going to deal with inequity in access to care? dr. mouton: i think because of the covid- crisis, we now have a renewed focus on the need for an adequate public health infrastructure. i would like to see us set up a framework of a solid adequately funded public health and preventative care infrastructure that assures a baseline of health care access across all populations and all classes. we couple that with key access provisions for communities across this country, both rural and urban, that gives access both to ''brick and mortar'' clinical practices as needed plus access to primary care, either by telehealth or in-person, which assures that these communities are able to maintain an adequate level of health that, in turn, allows the members of the community to enter the workforce, maintain income, and maintain social stability. that would be my moon shot. dr. carter-pokras: i already talked about my wish for universal health care, but i would also stretch that to not having access to health care be tied to our workplaces because there are so many people who lost their jobs as a direct result of covid- , and that also means they lost their health insurance. health in all policies, also, i think that is really important. health is created by a number of factors that go well beyond the scope of health care and public health activities. we gave the example about what happens when somebody gets evicted during this pandemic and the need for housing interventions at this time. i also think we need to have better appreciation for essential workers who have previously gone unnoticed. there are so many people who are keeping us going, whether it means our water coming through in our tap, or somebody picking up our garbage, or the people who are stocking the shelves at the grocery store, and hopefully providing them with the benefits that they need going forward. we could also benefit from having a better understanding of our supply chain, especially the food chain, and what we need to do to protect and maintain that. dr. furr-holden: i love that universal health care not being tied to work has been mentioned, and universal access to the internet. dr. carter-pokras: and also, protecting our supply chain, especially our food chain, and protecting those who are essential employees, including those who up to this point many people did not even really see them. they were almost invisible in a sense. dr. kimmel: all of these things are in line with what i would love to see. i think really trying to get people care in a way that meets their needs and is more individualized and personalized is a priority. for some people, that is going to mean being able to use the telephone to access care. for some people, that is going to mean that their internet access should be improved. then for some, these improvements still are not going to be the answer, so how can we also have safe places patients can go and use the internet to do an appointment or that someone comes to them and takes them to a safe place? to have this kind of individualized support will require thinking about how to expand our workforce for mental health support and use our workforce more effectively to ensure patients have time with supports and also those supports have time to personalize their approaches to the resources of each patient. this does not necessarily apply only health practitioners but also peer supporters and other community members who can be there to support mental health needs. i grew up in a more rural region and now many of the patients i care for are from rural areas. we have so many rural families in north carolina who have to drive hours to get to care. is there a way that we can have someone who can go out to them and do a visit, who can bring the ipad to access their specialist? can a psychiatrist provide supervision to a care manager who can go out to them and do that visit? dr. carter-pokras: thank you, dr. kimmel, for bringing that up, because our previous surgeon general, dr. vivek murthy, has really felt that loneliness is a crisis, an epidemic in our country. he is working on trying to get more information out about this crisis. i went through a -month training for a certificate in positive psychology and i am also a resilience trainer. it seems people are really desperate for finding ways to strengthen resilience skills now more than ever. in today's digital world, resilience is something that we would lift everybody's boats and help them identify how they can strengthen those social connections going forward. and i just want to let you know, in addition to all of the other issues that i had mentioned earlier for my wish list, i would also add that language interpretation, improving transportation, and improving the ability to pay for care are also important. all of these things still need to be addressed in terms of improving access to care and quality of care. dr. furr-hoden: dr. kimmel, you said something about expanding and increasing the health care workforce. i would be curious to hear your thoughts about how the public health infrastructure was not prepared to meet the demands of the covid- pandemic. what are the immediate needs you see? dr. kimmel: i think right now, so much of the focus is understandably on the immediate needs because so many people are getting seriously ill. and on top of that many health care workers are becoming ill and not able to come into work. with so much immediate need, we are focusing on the immediate needs and realizing the workforce does not have as much plasticity as we need. but then even as covid- cases slow and people are not getting so ill from the coronavirus, we are going to be left with large mental health needs. we will need to focus on building the workforce to cover those areas, too. we need to be building a workforce that can also help communities develop their resilience and help people build resilience. part of managing trauma is helping people realize how their stories are important, and what they have been through is important. this can help them see the strength that they have gotten from their experience. so we need that kind of workforce, as well, to build resilience. building that kind of resilience requires time for people to sit down with people and talk, and it is hard to do that in a -min visit. how do you build a workforce wherein the work does not have to be done in that short of a visit? how do we have other people who can also fill that role? when i think about the workforce, i have kind of a contrarian view, particularly when you talk about the physician workforce. if we talk about the physician workforce, i think the issue substantially is a maldistribution problem more than an adequate numbers problem. and again, that is driven by economic factors and reimbursement issues rather than, i would argue, individual physician desires and practice hopes. i see economic factors related to where physicians choose to practice as a major driver. now, if we look at the health care workforce at large, i think we still have needs for many health professions in the context of covid- , particularly from nursing and respiratory therapy, to meet this acute need. but i even wonder sometimes about that, because when you look at what is happening to the hospitals, due to the covid- -related economic forces, many of them are laying off nurses and looking at a future wherein there will be far fewer of these health care workers. so, to me, it begs the question, if we have a true health care workforce shortage, why would economics be the sole driver in the decision to lay off a substantial workforce, because you still would need them to run your enterprise? so, again, i wonder how much of this is driven by a distribution issue, and where people are being deployed, and where they are being located, as opposed to an absolute numbers issue. and it may be both. if it is a distribution issue, how do we solve that problem? how do we build the kind of incentives so we can have people where the needs are versus where the resources are? dr. mouton: well, right now the issue is that incentives are driven by the economic forces. as you move the economic forces out of the industry and say that we will deploy people where the need is the greatest, and they will be paid substantially the same or with some increased benefit for taking on those assignments, then that would be different. now, i am not saying that a universal one-payer health care system will necessarily do that, but it might. i have to study the distribution issues in canada and in england to see how their systems work, but i think they are better prepared. people may argue about whether or not that is true. i think part of our issue is that economic factors drive where physicians live, practice, what specialty they choose, and what geographic area they live in. and without economics being the main motivating factor, figuring out how to distribute the health care workforce in a manner that takes that away, you could have a situation wherein people will choose to work in environments that give them better economic fulfillment. dr. furr-holden: fantastic. this was an important conversation and i am glad you were all able to join me. thank you all for participating, and for sharing your thoughts about what impacts equitable access to health care in the united states. latino health steering committee improving health insurance coverage for latino children: a review of barriers, challenges and state strategies healthcare disparities and models for change role of acculturation research in advancing science and practice in reducing health care disparities among latinos available at www.pewhispanic.org/states/state/ md/ accessed table r : median household income american community survey year estimate emergency preparedness: knowledge and perceptions of latin american immigrants immigrant workers: vital to the u.s. covid- response, disproportionately vulnerable call for more resources in spanish in montgomery. activists warn latinos in county need more information and help what do health literacy and cultural competence have in common? calling for a collaborative health professional pedagogy verify: rubbing antibiotic cream in your nostrils won't prevent coronavirus. wusa . fair allocation of scarce medical resources in the time of covid- available at https:// collections.nlm.nih.gov/catalog/nlm:nlmuid- -mvset accessed us) committee on understanding and eliminating racial and ethnic disparities in health care centers for disease control and prevention. covid- in racial and ethnic minority groups how to find a therapist during the covid- pandemic coronavirus lockdown, lack of broadband could lead to 'education breakdown educators tell van hollen: digital divide doesn't just impact distance learning. maryland matters landlords are illegally evicting tenants during the coronavirus pandemic. lawyers fear a 'tsunami' of evictions when state moratoriums end council freezes rent hikes but omits undocumented immigrants from covid- relief bill. the washington post domestic violence cases surge amid stay-at-home orders. the hill americans increasingly reaching out for mental health help during coronavirus pandemic centers for disease control and prevention. health in all policies association of state and territorial health officials. covid- intervention actions: supporting individuals experiencing homelessness publish in health equity -immediate, unrestricted online access -rigorous peer review -compliance with open access mandates -authors retain copyright key: cord- - avgxkir authors: hii, aurysia; chughtai, abrar ahmad; housen, tambri; saketa, salanieta; kunasekaran, mohana priya; sulaiman, feroza; yanti, nk semara; macintyre, chandini raina title: epidemic intelligence needs of stakeholders in the asia–pacific region date: - - journal: western pac surveill response j doi: . /wpsar. . . . sha: doc_id: cord_uid: avgxkir objective: to understand the global outbreak surveillance needs of stakeholders involved in epidemic response in selected countries and areas in the asia–pacific region to inform development of an epidemic observatory, epi-watch. methods: we designed an online, semi-structured stakeholder questionnaire to collect information on global outbreak surveillance sources and limitations from participants who use epidemic intelligence and outbreak alert services in their work in government and nongovernment organizations in the asia–pacific region. results: all respondents agreed that it was important to remain up to date with global outbreaks. the main reason cited for following global outbreak news was as an early warning for serious epidemics. mainstream media and specialist internet sources such as the world health organization (n = / ; %), the program for monitoring emerging diseases (promed)-mail (n = / ; %) and the united states centers for disease control and prevention (n = / ; %) were the most common sources for global outbreak news; rapid intelligence services such as healthmap were less common (n = / ; %). only % (n = / ) of respondents thought that their sources of outbreak news were timely and sufficient for their needs. conclusion: for those who work in epidemic response, epidemic intelligence is important and widely used. stakeholders are less aware of and less frequently use rapid sources such as healthmap and rely more on validated but less timely traditional sources of disease surveillance. users identified a need for more timely and reliable epidemic intelligence. results: all respondents agreed that it was important to remain up to date with global outbreaks. the main reason cited for following global outbreak news was as an early warning for serious epidemics. mainstream media and specialist internet sources such as the world health organization (n = / ; %), the program for monitoring emerging diseases (promed)-mail (n = / ; %) and the united states centers for disease control and prevention (n = / ; %) were the most common sources for global outbreak news; rapid intelligence services such as healthmap were less common (n = / ; %). only % (n = / ) of respondents thought that their sources of outbreak news were timely and sufficient for their needs. for those who work in epidemic response, epidemic intelligence is important and widely used. stakeholders are less aware of and less frequently use rapid sources such as healthmap and rely more on validated but less timely traditional sources of disease surveillance. users identified a need for more timely and reliable epidemic intelligence. e merging and re-emerging diseases are significant threats to global health security. the asia-pacific region has been the global epicentre for many emerging infectious diseases, including some with pandemic potential. the emergence of new diseases such as severe acute respiratory syndrome and avian influenza, the threat of diseases external to the region such as ebola, and recurring outbreaks of endemic diseases highlight the ongoing threat that infectious diseases pose to national, regional and international health security. - the asia-pacific region encompasses two world health organization (who) regions: south-east asia and the western pacific, home to . billion people, or over % of the world's population. the region is one of the most diverse areas in the world in terms of socioeconomic development, geography and geopolitical influence. it is also particularly vulnerable to emerging and re-emerging infectious diseases due to several factors including increased population growth and movement, urbanization, globalization, limited access to health care, changes in food trade, land degradation and encroachment on natural habitats and antimicrobial resistance. , , this rapidly changing landscape, along with weak health systems, limited health infrastructure, resource constraints (financial, human, technical), geographical isolation and poor population health, challenge countries' abilities to adequately prevent, detect and respond to public health threats. [ ] [ ] [ ] [ ] the ability to rapidly detect and respond to infectious diseases is critical to global health security. the international health regulations, or ihr ( ), provide the legal framework to protect the international community from these threats, requiring member states to develop core capacities to detect, assess, notify and respond to public health threats and events of national and international concern. wpsar a semi-structured stakeholder survey was developed and administered electronically using surveymonkey (san mateo, california, usa) between june and october. the survey questions pertained to respondents' employment characteristics (organization location and type, occupation and position level) and global outbreak surveillance sources (automated outbreak alerts, reasons for following outbreak news services, types of sources and services accessed, limitations of outbreak sources, timeliness and adequacy of outbreak news sources, types of journals accessed at least once a month and preferred format to receive information). responses to questions consisted of pre-defined single and multiple choice options and a free text "other" option. the survey was piloted in june on five individuals with infectious disease experience in government and academic institutions in australia. minor changes to the survey were made following feedback to improve the consistency and clarity of questions. pilot participants were not included in the survey sample or results. the final survey was offered in english, french and bahasa indonesia. the survey questionnaire was forward-translated into french and bahasa indonesia. we invited participants to complete the survey from the following countries and areas: australia; picts (american samoa, cook islands, fiji, french polynesia, kiribati, marshall islands, new caledonia, niue, commonwealth of the northern mariana islands, samoa, tokelau, tonga, vanuatu); indonesia; and malaysia. our ihr ( ) emphasize the importance of incorporating event-based surveillance with traditional systems to detect public health risks. event-based surveillance is "the organized and rapid capture of information about events that are a potential risk to public health". information can be reported through official or unofficial channels such as media reports, healthcare workers and nongovernment organizations. , while traditional indicator-based surveillance systems are essential for collecting and analysing information on known diseases, event-based surveillance systems use broad definitions to detect rare or unusual events and are more timely and sensitive. , , they are an essential tool for the rapid detection and assessment of events that could pose serious risks to public health. increased availability and reliance on the internet has driven the development and acceptance of eventbased internet surveillance as a key tool and source of epidemic intelligence. , this method brings together disparate sources of data from the internet to provide a comprehensive overview on the current state of global infectious disease events in near real-time for public health action. there are three types of eventbased internet surveillance methods for rapid epidemic detection: ( ) existing internet-based surveillance systems and news aggregators that use event-based reporting and syndromic surveillance; ( ) search query surveillance using web-based search engines; ( ) social media. understanding countries' needs to detect and respond to infectious disease risks is relevant to common frameworks such as ihr ( ) and the asia pacific strategy for emerging diseases that require cost-effective surveillance tools to coordinate health security activities. there are limited studies on the epidemic intelligence needs of end-users. a review of evaluations of global electronic event-based biosurveillance systems found that evaluations focused on the quantitative analysis of system performance. the authors recommended that future evaluations assess the usefulness of systems for public health action for end-users. stakeholder engagement in all stages of surveillance system development from planning to implementation is important to create a successful and useful system that meets end-users' needs. , as part of the development of a new epidemic observatory, epi-watch, we sought to understand the global outbreak surveillance needs of stakeholders hii et al epidemic intelligence needs considered part of a single language group, the malay language) were also grouped together. ethics approvals were obtained from the following committees: university of new south wales ( there were responses to the survey and a % ( / ) completion rate. of the surveys emailed to participants, we received a completed response rate of % ( / ). five responses were excluded because respondents did not meet the study inclusion criteria, completed only the first section of the survey or selected a country from which ethics approval was not obtained, leaving ( %) eligible responses. of the respondents, % ( / ) worked in organizations based in australia, % ( / ) in organizations in picts and % ( / ) worked in malaysia or indonesia. table shows the employment characteristics of survey respondents by region. all respondents agreed that it was important to be up to date with global outbreaks. when asked about sources of automated global outbreak alerts (such as google alerts or program for monitoring emerging diseases [promed]-mail updates), % ( / ) reported receiving automated alerts, % ( / ) followed outbreak news sample was targeted to selected countries so that results would be relevant to inform development of an epidemic intelligence system for use within the region. malaysia and indonesia were selected in particular because of ongoing, separate research on epidemic surveillance in the malay and indonesian languages. we used several methods to recruit participants. eligible participants were those who use epidemic intelligence and outbreak alert services in their work across government and nongovernmental organizations. purposive and snowball sampling methods were used to select individual participants. representatives of all picts were invited to participate through the pacific community (spc). in australia, participants were identified through the communicable diseases network of australia, federal and jurisdictional health department websites, an existing list of public health contacts held by the study team, colleagues and organization websites. malaysian and indonesian participants were identified through ministries of health. participants were chosen based on their role and field of employment meeting the study inclusion criteria. the survey was emailed to participants from australia, participants from picts, four from malaysia and three from indonesia. participants were asked to forward the survey link to relevant colleagues. three email reminders to complete the survey were sent to countries with a low response rate to meet our overall target sample size of . in addition to emailing eligible participants, a stakeholder workshop was organized by iser in october to explore in more depth the outbreak surveillance needs of stakeholders. workshop attendees were required to complete the survey as a prerequisite for attendance. eligible attendees at the communicable diseases control conference in melbourne, australia from to june were also invited to complete the survey. * the total number of respondents by country/region was used as the denominator to calculate percentages separately by country/region. † peak body refers to an expert group that provides information, support, advocacy, coordination and strategic guidance to government or nongovernmental organizations. ‡ senior decision-maker: manages a section/branch/division/head of an organization, has significant and/or final decision-making authority. § mid-career: manages a small team, has some decision-making authority and/or influence. || junior: no management role, has limited authority to make decisions. ** categories were not mutually exclusive as respondents could select more than one option. just over half of respondents, % ( / ), thought their usual sources of global outbreak news were timely enough for their needs, % ( / ) did not find their sources timely and % ( / ) were unsure. fifty-one * categories were not mutually exclusive as respondents could select more than one option. † total number of respondents (n = ) was used as the denominator to calculate percentages to identify the most common reasons for following global outbreak news among all respondents. respondents overwhelmingly preferred email as a mechanism to receive global outbreak news. eightyseven per cent ( / ) of respondents selected this option; % ( / ) of respondents preferred websites; % ( / ) chose a weekly video presentation; and one each opted for the use of short message service (sms), social media and other formats. this question did not allow for multiple responses, and feedback from some respondents indicated that they may have had several preferred methods for receiving information, depending on the nature of the outbreak. a final question asked respondents to provide any other feedback. answers included needing information for different purposes such as preparation of emergency plans, border health control and advice to traveller consultations; a need to better inform health officials for preparedness, planning and response; and a need for systematized unified surveillance. our survey provides insight into the epidemic intelligence needs of a diverse range of stakeholders from across the asia-pacific region. there was consensus that timely and easily accessible global outbreak notifications are essential to plan for and respond to public health risks. respondents' professional needs are consistent per cent ( / ) of respondents thought that their usual sources of global outbreak news were sufficient enough for their needs. twenty-four per cent ( / ) found their sources were insufficient, and an equal proportion were unsure. one respondent ( / ) reported that timeliness and sufficiency were not personally relevant. the timeliness and sufficiency of outbreak news sources were cross-tabulated by respondent's usual sources of global infectious disease outbreak news ( table ) . sixty-two per cent ( / ) of respondents thought that specialist internet sources such as eventbased internet surveillance systems were timely enough for their needs, and % ( / ) found these sources sufficient ( table ) . when asked about the limitations of global outbreak news sources, % ( / ) of respondents reported that there was not enough critical appraisal, and % ( / ) did not have enough time to read/watch or listen to information. thirty-two per cent ( / ) of respondents identified that there was not enough information, % ( / ) that the sources were not timely enough, and % ( / ) that there were too many different sources and did not know which one was best. twelve per cent ( / ) reported other reasons, such as a delay in or no reporting of events at the country level and lack of local relevance. nine per cent ( / ) reported no limitations in their sources. multiple responses were allowed for this question. * total number of responses for each global outbreak news source was used as the denominator to calculate percentages for timeliness and sufficiency for each source separately, as not all respondents used all sources. † categories were not mutually exclusive as respondents could select more than one when selecting global outbreak news sources. ‡ one respondent who used these sources reported that timeliness and sufficiency were not relevant to their needs. hii et al epidemic intelligence needs outbreaks and the cdc's current outbreak list were more commonly used by respondents over other services such as healthmap but are less timely. previous studies have documented significant delays in official reporting of outbreaks compared to unofficial reports. , research has identified that the majority of event-based internet surveillance systems are generated from north america and europe; few local systems in the asia-pacific region and event-based surveillance systems in general are not well understood in developed and developing countries. , increased awareness of the availability and operability of systems providing timely, relevant and reliable information to professionals in the region could address some of these concerns. unofficial reports are key sources of information for internet-based systems, but they can be subject to noise and false alerts, potentially causing unnecessary investigation or alert fatigue among responders. our findings suggest that reliability and accuracy are important considerations in the choice of global outbreak surveillance sources; however, many respondents were unable to identify the best sources to use. who outbreaks and promed-mail were the most commonly accessed sources by many respondents. promed-mail is qualitative, but it uses human moderators to review alerts for relevance and accuracy before dissemination, increasing the reliability of reports. a service that can provide critical appraisal, including risk assessment within the broader context of the region, could address the need for more reliable information and help facilitate countries' abilities to assess risks and inform decisionmaking for the response required. this study had several limitations. due to the crosssectional online survey design, we were unable to monitor trends in responses/behaviour over time, and findings may not be representative because of the snapshot nature of the timing of the survey and possible non-response bias. as we were interested in stakeholder views at a point in time, this design was appropriate. the online nature of the survey meant that questions could not be explored in-depth; however, a free text option was provided for most questions. limited access to the internet and computers in remote and resource-constrained areas could have affected the response rate. compared to positing surveys, this was the most feasible option, and with some of the most remote picts participating, we do not believe access was a major barrier. the study with the key attributes of successful event-based surveillance systems: to be simple, flexible, timely and sensitive. with automated alerts being the predominant information-seeking strategy employed by respondents, internet-based services that provide this function can support the rapid and timely identification of events to limit the spread and severity of disease outbreaks. a limitation of event-based surveillance systems is that new information is not necessarily disseminated efficiently. while healthmap is a rapid intelligence source, it was only used by % of participants, possibly reflecting low awareness of this resource. consumers preferred global outbreak alert systems be flexible in the way information is accessed and disseminated. email was identified by respondents as the preferred communication method to receive global outbreak news; however, these needs may change depending on the context of the outbreak and over time (reflecting generational change in the use of communication technology); systems should consider a range of media such as sms and social media. communication technologies such as social media can be harnessed for rapid access and dissemination of information to support emergency preparedness and response. the use of mainstream media and specialist internet sources for global outbreak news is not surprising given the increased accessibility and reliance on the internet for information and acceptability of event-based internet surveillance systems. approximately % of initial reports to who about infectious disease events come from informal sources such as the internet. a systematic review of event-based internet biosurveillance systems identified systems, of which were online and fully functioning at the time. many of these systems use mainstream media as a key source of information. , the finding that the same proportion of respondents used both mainstream media and specialist internet sources for global outbreak news suggests that internet-based services are not meeting end-users' needs, and other media sources are required to supplement information leading to duplication of effort. timeliness of global outbreak news sources was a limitation identified by % of survey respondents. one study explored end-users' perceptions of the attributes of seven publicly available event-based internet surveillance systems and found that timeliness scores ranged from % to %. official sources such as who epidemic intelligence needs hii et al employed purposive sampling instead of probability sampling because of the small and highly specialized pool of eligible participants. while this approach ensured participation of professionals from a wide range of backgrounds and levels who use epidemic intelligence, it can create researcher bias because of the judgmental nature of sample selection. epidemic response is a small and specialized field, so the sample frame from which we could draw was small, making purposive sampling the most appropriate. limited inclusion of other large asian countries, differences in participant selection across countries and low numbers of respondents meant that results could not be compared between countries and may not be generalizable to other countries or representative of the whole asia-pacific region. finally, survey versions in languages other than english were not back-translated, which may have affected the quality of these responses. as % (n = ) of respondents completed the survey in a language other than english, translation inaccuracies are unlikely to have any impact on the overall validity of the survey. further research on language-specific needs for epidemic surveillance is warranted. for those who work in epidemic response, epidemic intelligence is important and widely used. the choice of sources for global outbreak news varies, and there is less use and awareness of rapid sources such as healthmap and more reliance on less timely, traditional sources such as who and public news media. we identified a need for more timely and reliable epidemic intelligence in the asia-pacific region. more effective and efficient sources and methods to deliver user-friendly intelligence to end-users should be explored. there are several global outbreak surveillance systems available; development of a new system should take into consideration how it can integrate into and add value to already established systems within the region. the authors declare no conflicts of interest. this study was funded by the national health and medical research council centre for research excellence, integrated systems for epidemic response app . manila: who regional office for the western pacific and new delhi: who regional office for global surveillance, national surveillance, and sars. emerg infect dis asia pacific strategy for emerging diseases and public health emergencies. manila: who regional office for the western pacific emerging infectious diseases from the global to the local perspective: workshop summary securing our region's health: the asia pacific strategy for emerging diseases. manila: who regional office for the western pacific and new delhi: who regional office for south-east asia emerging infectious diseases in southeast asia: regional challenges to control infectious diseases and their outbreaks in asia-pacific: biodiversity and its regulation loss matter trends and directions of global public health surveillance challenges to global surveillance and response to infectious disease outbreaks of international importance early warning epidemic surveillance in the pacific island nations: an evaluation of the pacific syndromic surveillance system canberra: department of foreign affairs and trade geneva: world health organization global incident map early detection of disease outbreaks using the internet social media and internet-based data in global systems for public health surveillance: a systematic review social media and disasters: a functional framework for social media use in disaster planning, response, and research hot spots in a wired world: who surveillance of emerging and re-emerging infectious diseases digital disease detection: a systematic review of event-based internet biosurveillance systems ga: united states centers for global capacity for emerging infectious disease detection timeliness of nongovernmental versus governmental global outbreak communications promed-mail: years of digital surveillance of emerging infectious diseases. int health a guide to establishing event-based surveillance. manila: who regional office for the western pacific early detection, assessment and response to acute public health events. geneva: world health organization early alerting reporting project of the global health security initiative. evaluation of epidemic intelligence systems integrated in the early alerting and reporting project for the detection of a/h n influenza events a review of evaluations of electronic event-based biosurveillance systems utility and potential of rapid epidemic intelligence from internet-based sources use of unstructured event-based reports for global infectious disease surveillance. emerg infect dis new technologies for reporting real-time emergent infections can internet search queries be used for dengue fever surveillance in china? public health surveillance systems: recent advances in their use and evaluation the pacific community [website disease outbreak news geneva promed international society for infectious diseases international society for infectious diseases outbreak news today florida the global dispatch global public health intelligence network [website]. ottawa: public health agency of canada we would like to acknowledge the support of dr jerico pardosi (university of new south wales), dr elizabeth kpozehouen (kirby institute) and mr dillon adams (kirby institute) for their contribution to this study. we also thank and acknowledge ministries of health from all participating countries for their support and agreement to participate and all respondents for taking the time to complete this survey. key: cord- -zrvv l p authors: saadi, abdulghani; kanmanthareddy, arun; anantha-narayanan, mahesh; hardy, karen; williams, mark; alla, venkata m title: access to smart devices and utilization of online health resources among older cardiac rehabilitation participants date: - - journal: world j cardiol doi: . /wjc.v .i . sha: doc_id: cord_uid: zrvv l p background: newer models of cardiac rehabilitation (cr) delivery are promising but depend upon patient participation and ability to use technological media including internet and smart devices. aim: to explore the availability of smart devices, current utilization and proficiency of use among older cr program attendees. methods: study participants were enrolled from four cr programs in omaha, nebraska united states and completed a questionnaire of items. results: of participants approached, responded ( %). mean age was . (sd ± ) years. demographics were . % males, . % caucasians, % with college degree and . %, with income of k$ or more. smart device ownership was . %; desktop computer was the most preferred device. average internet use was . h/d (sd ± . ); . % of participants indicating for general usage but only . % pursued health-related purposes. utilization of other health information modalities was low, . % used mobile health applications and . % used wearable devices. of all participants, % reported no barriers to using internet. education and income were associated positively with measures of utilization and with less perceived barriers. conclusion: among an older group of subjects attending cr, most have access to smart devices and do not perceive significant barriers to internet use. nonetheless, there was low utilization of health-related resources suggesting a need for targeted education in this patient population. have read the strobe statementchecklist of items, and the manuscript was prepared and revised according to the strobe statement-checklist of items. open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. it is distributed in accordance with the creative commons attribution noncommercial (cc by-nc . ) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. see: http://creativecommons.org/licen ses/by-nc/ . / manuscript source: invited manuscript introduction cardiac rehabilitation (cr) is an important component in the management of patients with heart disease [ ] . however, despite abundant evidence demonstrating its benefits and strong recommendations from multiple international and national associations, it remains underutilized [ , ] . newer models of cr delivery such as home-based cr, which can be implemented and monitored using internet, have emerged as potential solutions to address some of the barriers to widespread adoption of cr. furthermore, there is increasing evidence that such models can be equally effective and can be used to complement or extend traditional cr [ ] . highlighting this gap and opportunity, the american heart association (aha) issued a presidential advisory for improving access and utilization of cr. in this advisory, aha emphasized the importance of adopting newer models and chronic disease management interventions that can be delivered and monitored via telephone, internet or other means of communication [ ] . nonetheless, implementation and success of such models outside the research setting can be limited by participant access to internet and "smart" "devices", as well as proficiency and ease of utilization [ , ] . approximately % of the general population uses internet, with % of these reporting use for digital health information (dhi). in contrast, among seniors, approximately % report using internet and less than half of these use internet to access dhi [ ] [ ] [ ] . these observations highlight the barriers to access and underscore the need to understand the reasons for such barriers so as to enable effective delivery of smart models for cr in older adults. the purpose of this study was to explore the availability of internet/smart devices, current utilization patterns, and proficiency in using them for health-related issues in an older cohort of attendees within an urban outpatient cr program. study participants were enrolled from four american association of cardiovascular and pulmonary rehabilitation-certified, hospital-based cr programs within a single healthcare system in omaha, ne, united states. subjects were recruited over a period of six months in to participate in a survey consisting of items. indications for cr were: stable angina, myocardial infarction, heart failure, percutaneous coronary intervention and cardiac surgery. internet utilization was measured as usage time per day in hours. on-site subject recruitment and survey administration were conducted by a single investigator. participation was voluntary and there were no financial or other incentives offered for participation. the survey was only available in english. subject education status was categorized as "college" (completion of a four-year college degree or greater) and "no college" for participants with some college education, a high school degree, or less. data were described using averages and percentages. potential relationships between various items were assessed using chi square tests for categorical data and pearson's coefficient for continuous data. a significance level of . was used, and all data analyses and graphics were developed using the stata statistical package (college station, tx, united states). a total of ( %) of the eligible cr participants consented and completed the survey. patients wishing to forgo participation reported lack of interest as the primary reason. demographic characteristics of participants are displayed in table . study participants were % male, % caucasian, and % had college degrees or above. among the participants, % did not own any smart device and % owned a smart device (any of smartphone, tablet or personal computer). of those who owned a smart device, % owned ≥ devices. smartphone ownership was %, and this is consistent with previously published data in this age group [ ] . approximately twothirds of survey participants ( . %) were daily users with an average internet time of . hours (sd ± . ). although % of participants used internet for general purposes such as emails, paying bills, shopping and social media, only % used internet for health-related purposes. of those who used online health resources, % reported accessing online reviews of doctors, % used patient portals offered by their electronic health record, and % reported that they cross checked or verified information provided by their providers. however, % reported either difficulty or they had never tried searching for information related to their medical condition. the utilization of other dhi was low; . % watched health-related videos online, . % used mobile health applications, . % used wearable devices, and . % used smart fitness tools. as for barriers to utilization, % of participants reported perceiving no barriers to using internet, . % did not have internet, . % could not afford internet, and . % either found internet difficult to operate or preferred not to use it. mean use time was higher in participants who were < years ( . h/d) compared to those who were ≥ years of age ( . h/d) though this was not statistically significant (ns, p = . ). age had a weak negative correlation with use time (pearson's r = − . , p = . ). perceived barriers to internet use were not different between the groups (< years: % vs ≥ years: %, p = . ). participants who had a college education or above had a higher mean use-time of . h/d vs . h/d in the "no college" group (p = . ) ( figure ). of all participants without a college education, % perceived barriers to internet use versus only % of participants with a college degree who perceived a barrier to internet use (p = . ). finally, a significantly greater proportion of participants ( %) with an income of < $ k, perceived barriers to internet use compared to % of those with an income ≥ $ k (p = . ). the major findings of our study are: ( ) the majority of cr attendees had internet access and device ownership was high ( % in general, and % for smart phone); ( ) despite three quarters of cr attendees reporting no perceived barriers, only % used the internet for dhi; and ( ) consistent with the general population, younger age, college education, and higher income predicted greater use of the internet and less perceived barriers [ ] . in a randomized trial of patients (mean age of years), widmer et al [ ] demonstrated that digital health interventions significantly improved weight loss and reduced cardiovascular-related emergency department visits following acute coronary syndrome. additionally, the positive impact of "homebased or self-delivery" cr models using a variety of smart technologies has also been demonstrated in patients with chronic heart failure (hf) [ , ] . however, patient education is absolutely essential to the success of such programs. while utilization of internet and smart devices is common in younger individuals, adoption of technology is less pronounced among older individuals [ ] . however, recent trends are encouraging and demonstrate that an increasing number of older individuals are adopting these tools. our survey indicates that the internet and device-use in this patient population is consistent with the general population as reported by the pew research center and us census bureau [ , ] . given the findings in our survey, device ownership and access to internet are unlikely to be limiting factors in delivering home-based cr using smart technology. however, despite device availability and perception of ease of use, the actual utilization of devices and internet for health-related applications was low with less than one third of study participants using these various devices for accessing dhi. multiple factors likely contribute to this discrepancy between degree of use of smart tools for general purpose applications vs health-related applications. some of these factors could be attitudinal, i.e., older patients trust their health care providers much more than younger patients and may prefer to maintain this trust by not seeking online resources and tools [ ] . generational beliefs pertaining to religion, social norms, and preference for personal interactions could also contribute to the reluctance of older adults to use smart devices for dhi [ , ] . furthermore, older people are more likely to have negative views about the societal impact of internet and smart devices. finally, people older than years have been shown to have a greater concern over internet privacy and this may add to their reluctance to use smart devices for dhi [ ] . aside from the aforementioned attitudinal factors, a number of physical and cognitive impairments that are frequently seen in this demographic group are known to contribute to the lower use of smart devices among older adults. these may include visual or hearing impairments, small joint arthritis, tremor, and impaired fine motor skills or coordination, all of which impair ability to use keyboard, mouse or touch screen functions etc [ , , ] . in addition, memory issues and cognitive dysfunction can hinder ability to retrieve passwords, impair learning ability (necessary for learning new skills), and ability to recall information, thereby resulting in disinterest and disengagement with technology and smart tools [ , , ] . it is also worth mentioning that a number of devices and apps are designed with the younger user in mind and there is an increasing appreciation of the need to redesign these interfaces to improve technology engagement among senior citizens [ ] . not surprisingly, interfaces that provide healthcare-related information are often more complex and not as user friendly as commercial platforms that cater to entertainment or shopping, etc. this results in technology anxiety and lower self-efficacy in the interactions of older adults with health-related smart apps. all of the above age-related factors interact in a complex fashion and result in decreased technology adoption for health-related applications among older individuals. interestingly, although younger persons reported greater internet time ( . h) vs older persons ( . h) per day, this difference was not statistically significant although this could be a result of the small sample size of our study. a major strength of our study is its novelty and an assessment of the ownership of smart devices, usage patterns and barriers has not been previously identified in the cr population. our study has some limitations that include small sample size and lack of sample diversity. our subjects were mostly urban caucasian males thus limiting the generalizability of our findings. only % of eligible patients responded to our survey and there is a possibility of responder bias. our survey was not pilottested for validity and should be considered exploratory. finally, variables such as usage time were self-reported rather than measured and could be impacted by recall bias. in conclusion, our study demonstrates that most older patients attending cr in an urban metropolitan area have access to internet/smart devices and do not perceive significant barriers to use. despite this, most participants did not utilize these devices for health-related applications. further studies involving larger diverse groups of patients, sampled to account for geographic, racial and gender differences, are needed to add to the existing evidence on the impact of smart technology-based cr although two such studies are currently ongoing [ , ] . in addition to building more evidence to support potential clinical benefits, further research to enhance our understanding of the barriers, our ability to design supportive educational programming and develop protocols that can enable effective and efficient delivery of home-based smart cr are urgently needed. newer models of cardiac rehabilitation (cr) delivery are promising and there is increasing evidence that such models can be equally effective and can be used to complement or extend traditional hospital-based cr. highlighting this opportunity, the american heart association issued a presidential advisory emphasizing the importance of adopting these newer models for improving access and utilization of cr. however, effective use of these smart models depends upon patients' ability to use technological media including internet and smart devices. there is a dearth of knowledge on the availability of internet, ownership of smart devices, usage patterns and barriers to use specifically among cr attendees. cr attendees tend to be older than the general population or patients attending routine chronic disease management clinics. the purpose of this study was to explore the availability of such technology, current utilization and proficiency of use among older cr program attendees. this knowledge can help us understand the feasibility of such smart home-based cr programs in routine clinical practice outside of research trials. underuse are the need for patients to travel significant distances multiple times in a week, lack of transport, and inflexible schedules. proposed solutions include newer models of cr delivery such as home-based cr using smart device-based instruction and monitoring. to be able to implement and deliver these home-based cr regimens, we would need to know whether cr attendees who are generally elderly have access to such tools and whether they can use them proficiently. hence this study was designed to address some of these gaps in knowledge. the objectives of this study were to assess access to smart devices, predictors of their use and perceived barriers to the use of smart devices among cr attendees. this was an observational study assessing access to internet, smart device ownership and usage among attendees of american association of cardiovascular and pulmonary rehabilitationcertified, hospital-based cr programs in omaha, nebraska, united states. this was a voluntary survey using a pilot survey tool consisting of items. subjects were recruited over a period of six months in . on-site subject recruitment and survey administration were conducted by a single investigator. the survey was only available in english. data are described using averages and percentages. potential relationships between various items were assessed using chi square tests for categorical data and pearson's coefficient for continuous data. a significance level of . was used, and all data analyses and graphics were developed using the stata statistical package (college station, tx, united states). there has been no such study focusing on cr attendees in united states with most data currently available coming from general population surveys done by the pew research center. we approached attendees of our program, of which responded ( %). patients as expected were relatively older with a mean age of years, % were caucasians and ≈ % were males. approximately half of the respondents had college education and had a household income of ≥ usd. smart device ownership was . % with desktop computer being the most common and preferred device for connecting to the internet. approximately half of them owned a smart phone and / rd owned multiple devices (phones, tablets etc.). on average, internet use was . h/d. only about % used their smart devices and computers for healthrelated purposes. utilization of other health information modalities was low, . % used mobile health applications and . % used wearable devices. of all participants, % reported no barriers to using internet. education and income were associated positively with measures of utilization and with less perceived barriers while age had a negative correlation. in this survey, we did not address the medical comorbidities that may impact patients' ability to use smart devices for health-related applications and patients' attitudes towards such use. our study demonstrates that most older patients attending cr in an urban metropolitan area have access to internet/smart devices and do not perceive significant barriers to use. unlike data from prior decades where elderly patients did not have access to smart devices, our study proves that access is no longer an issue. despite this, the majority of participants did not utilize these devices for health-related applications. we hypothesize that attitudinal factors such as concern about internet privacy, physical and cognitive impairments that make it difficult to interface with smart devices such as small joint arthritis or memory impairment and lack of education on how to use the devices may be contributing to the low rates of use of smart devices for health related applications. patient income, educational attainment and age correlated with use of smart devices in our study confirming the findings of prior studies across different age groups. our findings have significant implications for the efforts to transition cr away from hospitals and closer to home and to create hybrid models of cr. these models not only increase access for patients, increase participation and engagement but may also prove economical and more sustainable for prolonged periods of time. more importantly, the ability to deliver cr in this fashion may be the only way to ensure safety of our patients in this current time of the corona virus pandemic. our study demonstrates that access to smart devices is no longer a limiting factor to the implementation of smart models of home-based cr. limited use of smart devices for healthcare applications in our elderly patients was likely a result of attitudinal factors, cognitive impairments and lack of proper education. further research is necessary to confirm our findings in larger diverse groups of patients, sampled to account for geographic, racial and gender differences using validated survey tools. future avenues for research include investigation into the impact of smart device and apps' design as well as the impact of targeted education to improve technologic proficiency among older adults on the adoption of these technologies. the ultimate success of these smart models of cr will depend on their ability to improve clinical outcomes and their comparative efficacy and cost effectiveness vis-à-vis traditional hospital/center-based cr. american association of cardiovascular and pulmonary rehabilitation. cardiac rehabilitation and secondary prevention of coronary heart disease: an american heart association scientific statement from the council on clinical cardiology (subcommittee on exercise, cardiac rehabilitation, and prevention) and the council on nutrition, physical activity, and metabolism (subcommittee on physical activity), in collaboration with the american association of cardiovascular and pulmonary rehabilitation referral, enrollment, and delivery of cardiac rehabilitation/secondary prevention programs at clinical centers and beyond: a presidential advisory from the american heart association participation and adherence to cardiac rehabilitation programs. a systematic review home based versus centre based cardiac rehabilitation: cochrane systematic review and meta-analysis feasibility of a smartphone-enabled cardiac rehabilitation program in male veterans with previous clinical evidence of coronary heart disease a review of wearable sensors and systems with application in rehabilitation technology use among seniors the us census bureau reports: computer and internet use in the united states trends in seniors' use of digital health technology in the united states digital health intervention during cardiac rehabilitation: a randomized controlled trial home-based cardiac rehabilitation for people with heart failure: a systematic review and meta-analysis smartphones in the secondary prevention of cardiovascular disease: a systematic review factors predicting the use of technology findings at a glance: medical doctors how older adults meet complexity: aging effects on the usability of different mobile phones mobile applications in an aging society: status and trends security and password protection rank high for the + older adults and mobile technology: factors that enhance and inhibit utilization in the context of behavioral health design of mobile phones for older adults: an empirical analysis of design guidelines and checklists for feature phones and smartphones smartphone and social media-based cardiac rehabilitation and secondary prevention (smart-cr/sp) for patients with coronary heart disease in china: a randomised controlled trial protocol smartphone cardiac rehabilitation, assisted self-management versus usual care: protocol for a multicenter randomized controlled trial to compare effects and costs among people with coronary heart disease key: cord- -mgq a t authors: mead, darryl; sharpe, mary title: aligning the “manifesto for a european research network into problematic usage of the internet” with the diverse needs of the professional and consumer communities affected by problematic usage of pornography date: - - journal: int j environ res public health doi: . /ijerph sha: doc_id: cord_uid: mgq a t the manifesto for a european research network into problematic usage of the internet was published in may . it was written from the perspective of the cost action network, a programme of the european cooperation in science and technology ca and is expected to have significant influence on research funding priorities over the next decade. the manifesto identified nine key research priorities to advance understanding in the field. our analysis shows that while at the most general level it identified problematic usage of pornography (pup) as a key research priority, it then barely mentioned it again within the body of the report. this paper uses the manifesto’s framework to suggest research areas into the problematic usage of pornography which are of particular relevance to clinicians and other professionals working in the field who want to develop approaches to assist individuals and target groups affected by pup. it also looks at potential research opportunities inspired by the lived-experience of users withdrawing from pup. a large number of opportunities are identified for new work on pup across all nine key research areas of the manifesto. the publication of the manifesto for a european research network into problematic usage of the internet (pui) [ ] in may was a milestone in planning the roadmap for the behavioural addictions' community. it provided an international focus for developing policy across different elements of the research landscape for behavioural addiction over the next decade. the manifesto was written from the perspective of the framework of the cost action network, a programme of the european cooperation in science and technology ca . the manifesto identified nine key research priorities to advance the understanding of pui (table ) . the manifesto identifies pornography usage as a potential pui. we have called this activity problematic usage of pornography (pup). table . summary of key research priorities to advance the understanding of pui [ ] . "reliable consensus-driven conceptualisation of pui (defining main phenotypes and specifiers, related comorbidities and brain-based mechanisms) . age-and culture-appropriate assessment instruments to screen, diagnose and measure the severity of different forms of pui . characterise the impacts of different forms of pui on health and quality of life . define the clinical courses of different forms of pui . reduce obstacles to timely recognition and interventions . clarify the possible role of genetics and personality features in different forms of pui . consider the impact of social factors in the development of pui . generate and validate effective interventions, both to prevent pui, and to treat its various forms once established . identify biomarkers, including digital markers, to improve early detection and intervention". since pornography consumption has transitioned from the model where a market supplied physical media such as magazines and dvds to consumers through retail networks, to a transnational, internet-based system mainly operating on a freemium trading model [ ] [ ] [ ] . in the process it has scaled from predominantly national businesses reaching an audience of millions through shops and mail-order, to a world-wide online one with perhaps a billion customers [ ] [ ] [ ] [ ] . today, pornography is generally accessed on smartphones and other devices. its consumption has the potential to develop into a large-scale form of pui [ ] [ ] [ ] . this technology-driven progression has massively increased the number of people potentially exposed to pup, while simultaneously removing barriers moderating individual levels of consumption. some consumers uncomfortable with the apparent impacts of their pornography use have now gathered together in large-scale, online recovery communities [ , ] . many of these are self-help groups who try to support individuals to manage their consumption, or to end it all together. a growing community of professional therapists [ ] [ ] [ ] [ ] , coaches [ ] and mental health experts [ ] has developed to support these self-help groups, as well as people accessing mental and physical health services through more mainstream medical, psychological and psychiatric routes. academic research and public policy to support consumers, recovery communities and professionals is still at an early stage of development. much of the motivation for writing the current paper comes from the opportunity the manifesto offers when it says: "additionally, through an interactive dissemination plan (including website, social media, blogs) we are reaching out to relevant stakeholders at international and national levels with an emphasis on encouraging people with the lived experience of pui to become involved in the action" [ ] p. . this paper is a result of the authors heeding this call to action. first, it discusses the extent to which the topic of pornography usage is carried through in the nine key research themes. next it considers how the manifesto tries to address the diverse needs of the professional and consumer communities affected by pup. it then goes on to suggest additional areas of research to bridge any identified gaps. our research question is "what topics should be included within future research proposals under the manifesto to meet the diverse needs of consumers, recovery communities and professionals impacted by the problematic use of pornography"? the current study has used the content of the manifesto [ ] as its target document. its authors created the manifesto both in the hope of influencing future research directions around pui and to improve the availability of funding to support those aspirations. as far as the authors are aware, the manifesto is the only research planning and policy document of its type currently available, which is relevant to the field of pui. rather than developing a mechanism to choose to include some parts of the manifesto and to exclude other portions, the target document is sufficiently short for it to be analysed as a single entity. each and every paragraph referencing pornography within the manifesto has been identified and is reproduced in this paper to provide context to allow consideration of both what has been said, and just as importantly, what has been omitted. the manifesto began by setting a level playing field for different problematic internet usage behaviours. it also introduced the central focus on health impacts. "we use the umbrella term pui to encompass all potentially problematic internet related behaviours, including those relating to gaming, gambling, buying, pornography viewing, social networking, "cyber-bullying," "cyberchondria" among others. pui may have mental and physical health consequences" [ ] p. . the nature of the health impacts is then developed further. the next step was to contextualise the behaviours according to their form, their place within the international medical diagnostic frameworks and as issues affecting identifiable user populations. "as noted, pui envelops a wide range of activities including video gaming, pornography viewing (and other compulsive sexual behaviours), buying, gambling, web-streaming, social media use and other behaviours. some of these behaviours may fall into an existing mental disorder in psychiatric nomenclature (e.g., gambling disorder), whereas others are likely to be formally recognized in future dsm/icd revisions, notably internet gaming disorder (kim et al., b). different types of pui often start in childhood or adolescence (volpe et al., ), but broad age ranges can be affected (ioannidis et al., ). age and gender relate importantly to pui behaviours, with younger people typically having problems with gaming and media streaming, males with gaming, gambling and pornography viewing and females with social media and buying (andreassen et al., )" [ ] p. . the manifesto then moves into a process of exploring the individual research priorities according to the list above. the clinical aspects of some internet-related behaviours appear phenomenologically much like addiction (e.g., gambling or viewing pornography), and demonstrate impaired control (unsuccessful attempts to reduce or cease the behaviour), preoccupation (craving), associated functional impairment ( for problematic online sexual behaviour (e.g., cybersex), three structural elements have been highlighted as being important contributors per the so-termed triple a model involving: accessibility, affordability, and anonymity (cooper, ; cooper et al., ) , though more research is needed on this topic (brand et al., a; wery and billieux, ) . another similar proposed framework is the ace model (anonymity, convenience, and escape) (young, ). for excessive streaming (watching videos excessively), important structural features may include the ability of given programmes to grab attention by activating a biological 'orientating response', mediated through techniques including the use of attention-grabbing noises, zooming/panning, and presentation of rewarding stimuli (e.g., of a sexual or thrilling nature) (flayelle et al., (flayelle et al., , . collectively, public research into the structural elements that may promote pui in different contexts is lacking" [ ] p. . however, this is the only mention of pornography after section . so what is missing and what areas of research would be useful to the consumer, recovery and professional communities affected by problematic usage of pornography? this paper sets out to determine "what topics should be included within future research proposals under the manifesto to meet the diverse needs of consumers, recovery communities and professionals impacted by the problematic use of pornography"? the manifesto tells us that pup is a part of pui and provides some discussion on phenomenologies, comorbidities and brain-based mechanisms. it introduces pornography in the context of how to classify it as a disorder. some early models of pornography consumption are discussed in research priority . beyond these initial pointers, direct reference to the manifesto provides little specific guidance for policy-makers considering the research environment for pup. unlike other behavioural disorders such as gambling and gaming, which were referred to repeatedly in the manifesto, with pornography the manifesto was mostly silent. it did not try linking pup to assessment instruments, characterisation, defining clinical forms and removing obstacles to recognition and treatment. equally it did not recommend directly relevant research on issues around genetics, personality features, interventions for prevention and/or treatments, or the field of biomarkers. is the absence of more mentions of pornography viewing as a pui in the manifesto accidental or driven by other factors? we will not attempt to answer this. we recognise that the manifesto's authors covered a great deal of ground in words and some omissions were always likely. several of the authors have published widely in the field of pup. while still limited in comparison with research in areas such as gambling and gaming, the scale of literature investigating pornography viewing as a pui is growing rapidly. during we read over papers on pornography usage, about % of them published in - . new work which has appeared since the manifesto was published revealed previously unquantified consumer patterns. the diagnostic category of compulsive sexual behaviour disorder (csbd) was published in icd- , two months after the manifesto appeared [ ] . over % of people now seeking treatment for csbd have a pornography-use related issue, a statistic not widely known when the manifesto was published [ ] . subsequent pornography-consumer behaviour has also been significantly influenced by entirely new factors such as the covid- pandemic [ ] . so, where does an analysis of the manifesto leave policy-makers, funders, researchers and other interested parties? our discussion now approaches this from two separate, though linked, directions. first, we consider how pup can be researched to meet the needs of the different stakeholder communities. we then consider the potential for future research, as seen through the lens of the nine priorities set out in the manifesto. pornography accessed for free via the internet is now used on a frequent basis by hundreds of millions of people. for most users any problems only emerge slowly over time. however, the scale of the consumer population suggests that problematic pornography usage has the potential to impact the health of millions of people across the nine key research priorities set out in the manifesto. to make wise choices for research resource allocation, policy-makers and funders need to have sight of a research landscape that addresses each priority. what might it look like? the following is a pragmatic assessment based on a consideration of the existing literature, mediated by the constraints set within the manifesto. this assessment is supported by the author's wide-ranging dialogues with professional, recovery and consumer communities. since we have had face-to-face discussions on pornography consumption with over people in the united kingdom, the republic of ireland, france, germany, croatia, hungary, the ukraine, turkey, japan, australia and the united states [ , ] . what else do therapists, medical practitioners, counsellors and sex educators want to know about the mental and physical health implications of problematic pornography usage? what are the issues that their clients, the problematic pornography users, want to have investigated? it is unhelpful that the diagnostic manuals do not mention the word "pornography" when over % of people seeking treatment for compulsive sexual behaviour have a pornography-related issue [ ] . further, the nature of the condition for many meets the criteria for addictive disorders and should be clearly categorised as such to enable healthcare professionals to respond with appropriate treatments. both groups have an interest in distinguishing between classifying pornography usage as a potentially addictive disorder versus it being considered as an impulse control disorder. at any rate, there is clear desire in both lay people and professionals to have the word "pornography" appear in any given classification of a future edition of the international classification of diseases [ ] and the diagnostic and statistical manual of mental health disorders [ ] . it is apparent from our experience in meeting researchers from all parts of the world [ , ] that the therapists' own knowledge, attitudes and prejudices are very important issues in terms of delivering access to appropriate treatment services. while the internet brings pup to all parts of the world, different cultures have different taboos and even blind spots about what is pup, what is unacceptable behaviour and what sorts of treatment might be considered appropriate. the legal systems of different countries are also relevant, and there are no universal standards for most sexual behaviours. for the last four years the authors have been teaching healthcare professionals about internet pornography and behavioural addiction. these audiences have consistently expressed a desire for the health and social implications arising from pornography usage to be taught as a component of general medical education. clinicians have also expressed interest in integrating the management of compulsive sexual behaviour disorder into the regular functioning of national healthcare systems. it would be desirable for the manifesto to address the concerns of the online pornography recovery communities and members of -step programmes such as sex addicts anonymous [ , , ] . so far, we know of no quantitative research that has investigated elimination of digital porn use to reverse sexual dysfunction and a variety of mental health disorders reported by those who quit [ ] . the focus within recovery communities begins with recognition/identification/diagnosis of problematic pornography usage. the first question is "do i have a problem?" if they do have a problem, this quickly becomes "how do i stop?" with a focus on tools, techniques and support. it then becomes "how do i maintain my desired level of sexual sobriety?" here we need more longitudinal research on the mechanisms and techniques for overall programmes of quitting pornography usage, on abstaining from pornography viewing, and the issue of managing masturbation. unlike most puis such as gambling, shopping or gaming, pornography viewing is linked to very deep biological sexual and reproductive drives, with the option of the interactive reinforcement by masturbation. the pornography, masturbation and orgasm (pmo) cycle reinforces the viewing behaviour, making simple abstinence strategies difficult to maintain for many former users. when you stop gaming you can get rid of your game console or you can have yourself banned from gambling venues, but you cannot stop being a sexual being. there are some specific characteristics of problematic pornography viewing which merit more research, particularly issues around triggering and flatlining. a deeper understanding of pup triggering factors would be useful to help individuals and clinicians build better strategies and models for avoiding problematic usage. flatlining is the term for the severe and sustained, short to medium-term loss of libido some users report upon quitting a cycle of problematic pornography viewing. while widely reported in community forums such as nofap [ ] and rebootnation [ ] , it is not well-covered by the academic literature [ ] . the medical establishment has an ongoing interest in the development of new approaches to support quitting pornography viewing by identifying effective drug treatments. naltrexone is a medication that is used often in drug and alcohol addiction. it blocks the effects of drugs known as opiates. it competes with these drugs for opioid receptors in the brain. naltrexone can reduce a patient's desire for drinking thereby helping them remain abstinent. three papers have looked at the effect of naltrexone on problematic pornography use [ ] [ ] [ ] . given the similarity of brain changes in substance abuse disorders and behavioural addictions, this could prove a fruitful area for further research. very recently transcranial magnetic stimulation has been suggested as a novel treatment [ ] . investment in research to try to demonstrate the value of such approaches to clinical populations at scale would be very helpful. therapists operating within the context of organisations such as the association for the treatment of sexual addiction and compulsivity (atsac) in the united kingdom [ ] and the society for the advancement of sexual health (sash) in the united states [ ] would benefit from more research into behavioural modification techniques to help their clients stop internet-based behaviours. this might include the creation of apps and other internet tools. very recently metacognitive therapy has claimed to be three-times as effective as traditional cognitive behavioural therapy. it deals successfully with 'desire thinking' and rumination, a behaviour that often predicts relapse. extending new work on this to clinical populations of recovering pornography users would be helpful [ , ] . pornography viewing is now a widespread behaviour in most countries where there is good quality access to the internet [ ] . to have a protective effect on health and wellbeing, consumer-facing academic research on pup needs to have impact in the wider community. this is a science communication issue. little published research has looked at the ecology of spreading this type of message. there is scope for developing a very diverse array of effective communication strategies and toolkits. more research is needed on what would help adults to avoid situations and behaviours that might lead to pup. there is more active research focused on schools and adolescents. the manifesto could support the sort of educational work the reward foundation does in schools to help equip young people with the knowledge about the reward system of the brain and the skills they need to avoid not just pup, but pui in all its forms [ ] . other educational programmes are available from an array of non-governmental organisations such as culture reframed, echildhood, and the naked truth project. common to all programmes is a lack of longitudinal studies evaluating their effectiveness over time, across cultures and among varying age groups. we also need to recognise that pup is not just a first-world issue. the reward foundation website operates in english, but uses the machine translations system gtranslate to make all pages available in over languages. as a result, a significant proportion of our traffic is now in african languages such as hausa and somali. these people have a need for information about pup, but it is not readily available from within their own language communities. world-wide data from pornhub shows that heaviest usage of pornography is by people in the to years age brackets. it is strongly tilted towards male consumers, but supply-side drivers are slowly pushing consumption towards gender parity [ ] [ ] [ ] [ ] . research over the past decade has shown that pornography viewing correlates with higher levels of sexual assaults and dating violence, pathways into domestic violence and reduced levels of bystander intervention [ ] [ ] [ ] [ ] . the next generation of research should help move us from noting these as correlations to investigating questions of causation (or not) in these fields. the categories in the following discussion are directly based on the manifesto's nine research priorities, simply substituting 'problematic usage of pornography' for the wider pui category. they map out some of the ways the manifesto could respond to the needs of the professional and consumer communities. research area : reliable consensus-driven categorization of problematic usage of pornography (defining main phenotypes and specifiers, related comorbidity and brain-based mechanisms) [ ] . separate and interlocking strands of research would be useful in the two main fields analysing sexual behaviour; compulsive sexual behaviour towards people and the usage of pornography. in particular, the overlap between them needs more work. there would also be considerable value in research on the overlap of pornography viewing with other pui's involving sexual behaviour-related activity, especially dating apps and social media. research area : age-and culture-appropriate assessment instruments to screen, diagnose and measure the severity of different forms of problematic usage of pornography [ ] . the number of screening, diagnostic and measuring tools for pornography usage keeps expanding. there would be considerable value in testing and calibrating the best of the existing crop of tools, such as the brief pornography screener and problematic pornography use scale, across genders, age groups and societies [ , ] . extending field trials would materially improve the ease with which these tools can then be rolled out for widespread adoption by clinicians and other people who are in the first line of contact by individuals concerned about pup. even among the experts the brief pornography screener is not universally known [ ] . separately, major international reviews of erectile dysfunction have not referred in any way to the behavioural addiction field for data [ ] . tolerance and escalation are classic behavioural symptoms of the journey towards an addictive disorder. each contributes to some consumers developing problematic usage. more research would be useful in relating the roles of tolerance and escalation in the pup of viewers who escalate to viewing child abuse material. this seems to have become a wide and growing aspect of cybercrime around the world [ , ] . more knowledge of effective crime-fighting strategies could lead to better policy-making on prevention and to better risk-based advice for the courts and judges around sentencing and disposal. standardised and tested risk assessment tools would be particularly useful. problematic usage of pornography seems to be creating a new separate sub-population of consumers who choose to view child abuse material without having any particular sexual interest in children in real life. they appear to be a separate clinical population to individuals with a paedophilic sexual orientation. another area where screening tools are woefully lacking is in the categorisation of the use of pornography by people on the autistic spectrum. whereas about % of the population have autism spectrum disorders, crime statistics suggest that a disproportionately large number of the people (circa %) who escalate to viewing child abuse material come from specific vulnerable communities such those with as autism spectrum disorders, adhd and those with learning difficulties [ ] . these vulnerable populations are often considered 'asexual', and not requiring any sex education. this is wrong as many use the internet as a substitute for relationships that they find so difficult to negotiate in real life. people with asd are rarely represented in other crime categories. their emotional immaturity and inability to interpret social rules can result in an attraction to younger people online who match their emotional age better. more research needs to be done on this group to improve prevention of access to child abuse material and to consider their special needs in terms of disposal following a conviction. for example, they do not function well in groups, which are the typical types of setting for sex offender rehabilitation [ ] . this question can be explored by research into the pornography, masturbation and orgasm cycle. in turn, this line of work could lead to consideration of the extent to which pmo may be driving depression, social isolation and other mental health issues in the wider population. is it a bi-directional process -to what extent are these mental health issues powering the pmo cycle? work in this space to test causation was recommended as far back as [ ] , but has yet to gain funding. bi-directional work would also be helpful for studies exploring health links or problematic pornography usage and the big- personality factors. first, factors such as narcissism are regularly shown as correlates of heavy pornography usage [ ] , but few studies look at what happens to personality factors after people end problematic pornography use. do they remain constant or is there an individual and/or population tendency for them to revert to what they were before the individual began engaging in pup [ ] . other recent studies have highlighted the need for longitudinal studies in this area [ ] . similarly, conference reports [ , ] and our informal discussions with both individual problematic pornography users in the community [ ] and among sex therapists [ , ] , suggest that there are co-morbidities with other addictions, both to substances and behaviours. developing new and/or improved strategies for treating complex webs of problematic usage are always of value. more literature would be helpful on the impact of problematic pornography use at different points in the human life-cycle. most research published to date has focused on teens or university students, though around half of all pornography viewers are older [ ] . this makes case studies or more specific work on older people very rare [ , ] . it would also allow us to look at the potentially very different trajectories of younger populations who began with access to unlimited, free streaming videos in or around , and older users who may have transitioned through several successive modes of access to pornography such as magazines and dvds, to internet use. the many roles of pornography in forming, sustaining and ending couple relationships would benefit from further research. the majority of research to date has been on users, and to a lesser extent their sexual partners. there would be value in unpacking the factors relating to pornography consumption as they impact on various aspects of an individual's health and quality of life in the context of a dyad [ , , ] . this focus on couple relationships naturally tends to exclude work on the lack of relationships among young heavy users, around % of whom are virgins and not in relationships [ ] . the other area needing more work is on the way pornography creates expectations of "how partnered sex will be" [ , ] . the literature of wider family and community impacts of pornography use across the human lifespan is also limited [ ] . additional research into the impact of accessibility would be valuable, as problematic pornography usage would seem to be a case where the pui strongly links to private viewing. this is especially true for mobile platforms which can be used alone or with sexual partners. smartphones are also now used widely for technology-enabled flirting including "sexting" with diverse relationship, social and potentially legal implications. how does pornography consumption relate to the health and quality of life in a couple or a family? the authors frequently train social workers and criminal justice professionals who report that consumption of pornography, particularly violent pornography, is a consistent negative factor in the out-of-control domestic situations they encounter in their work. unlike mainstream television where intimate partner violence in programmes tends to be simulated, most violence depicted in pornography is real [ , ] . it is not simulated and may offer viewers ethically challenging issues while they are viewing. it may also lead to post traumatic stress disorder and related symptoms after they stop. once seen, real violence and intimate partner violence shown in pornography cannot be unseen. where the real rape of minors is being streamed as entertainment by major commercial pornography sites, the legal and ethical challenges can be even more complex [ ] . to what extent does acting out violent, coercive or controlling pornography scripts with real partners drive domestic violence and a growing appetite for dangerous practices such as sexual strangulation [ ] ? these questions are key in many different cross-disciplinary approaches to understanding pui and pup. the manifesto noted that: "remarkably little prospective research has been conducted on the courses of different forms of pui and we remain relatively ignorant of key factors affecting long-term out-comes. such data are of crucial importance in understanding aetiology, planning treatment and improving prognostication. for example, for some individuals pui may represent a temporary phenomenon and spontaneously resolve (e.g., in some young people as brain systems mature), whereas for others pui may become chronic." [ ] p. . these observations apply very strongly to pup. there are many areas where we have little data. in simple terms, pup can lead to the individual failing to stop their use when faced with consequences, following the model set out in the diagnosis for compulsive sexual behaviour disorder [ ] . however, icd- is careful not to set out the details of the specific sexual behaviours that are causing the distress. equally it is vague about the progression of the disorder from the perspectives of screening, assessment and diagnosis. this gives considerable scope to clearly define the clinical courses of each and every form of pup. we suggest that there are two fundamental classes of pup, and each needs more research. the first is where the consumption of pornography leads to problems directly attributable to the consumption process. the user cannot stop their use and it is leading to negative consequences. this is applicable to csbd as an impulse control disorder [ ] and would also apply if a future version of the international classification of diseases gave it an entry in the addictive disorders section. we propose that the second category is where the development of a pup comes from the ideas it introduces into a viewer's mind and life. these can range across a whole spectrum from requiring to replay sexual scripts from pornography in your head to get aroused during partnered sex, to learning to ignore the need for consent in sexual activities, to promoting rape myths, the diminution of bystander intervention and the promotion of practicing dangerous activities such as autoerotic asphyxiation. other problematic uses can encompass escalation to criminal activities around viewing or creating images of child sexual abuse. it can also lead individuals to engage in sextortion, 'capping', and other illegal activities towards minors [ ] . other new populations are the technology-enabled voyeurs whose illegal activities are also directed at adults, encompassing upskirting, hidden cameras, revenge porn and sharing sexual material without permission. pornography has also appeared that was secretly recorded on baby monitors lacking password protection. pornography-induced erectile dysfunction (pied) [ , ] has been identified as a key factor that causes male sufferers of pup to seek treatment [ ] , but this area still lacks clear before and after brain-imaging studies [ ] . pied has also been linked to escalation to child abuse material. as desensitisation, tolerance and hypofrontality develop, a problematic pornography user needs more shocking material to feel any arousal. therapists need to know how long it can take for a pup to develop and what external or internal factors can act as triggers. reports from the recovery communities talk about the triggering influence of everything from life events and rites of passage, to social networks and the manipulation of viewer's taste by the algorithms used by the commercial pornography suppliers. the withdrawal process from pup, both physical and mental, seems generally similar to that of other puis. the only unique characteristic of pup identified so far is 'flatlining'. the literature does not adequately characterise the features of flatlining as a symptom across the population of users withdrawing from pup. this means that when it happens, problematic users can relapse to try to feel aroused again and overcome their feeling that their sexuality might be permanently 'broken'. more studies of co-morbidity, both between pup and other behavioural or substance addictions, and with other compulsive sexual behaviours, would also be welcome. there are other elements research could reveal about the general characteristics of pup. what are the characteristics of the incubation periods of pup at both the individual and population levels? do they have characteristics driven by gender or other factors? this in turn leads to the need for wider studies looking at gender, lifestyle and sexuality-based approaches to tease apart key elements in the pup process. what commonalties and differences exist across the lgbtqi++ spectrum? in parallel, is pup in the chemsex community driving addiction, or is its influence bi-directional [ ] ? existing brain-imaging studies [ , ] hint that both the quantity of pornography viewed and the period the viewing is spread across can have impacts on the development of pup. however, the current crop of brain studies on pornography are restricted to people in their s and s and they also draw on very narrow demographic and cultural samples. in a world where an average age for boys' first exposure is years or younger, brain development issues need investigation [ ] . more women-focused studies would be beneficial. how is pup the same for women as for men, and how is it different? are the genders the same in the etiology of compulsive use? to what extent is sex toy used [ ] a component of women's pup? some research has suggested that pornography usage by women has influenced their behaviour in couple relationships [ , ] . are there differences according to ages and points in the lifecycle? we are unaware of any pup studies in women controlling for stages across the menopause as a variable. age and stage of life should also be addressed from other perspectives. virgins make up perhaps % of the recovery community [ ] . what are the longer-term implications of early pup across their lives as sexual beings? separately, how do the challenges faced by older addicts, who used a lot of pornography before the internet, compare to those of purely internet-based consumers? what are the impacts of pup on relationship formation, sustainability and sexual health? a study of korean men in stable relationships suggested they had developed a preference for viewing violent pornography as an alternative to real sex with their dyadic partner [ ] . this also raises the issue of the nature of the content of the pornography being viewed. in line with the self-reports from the recovery community, a swedish and italian study of university students demonstrated that unusual sexual content creates more excitement. it also concluded that lots of the acts considered "unusual" by the researchers in , already had become mainstream to many viewers and did not seem "unusual" at all [ ] . research area : reduce obstacles to timely recognition and interventions [ ] . obstacles around the timely recognition of pup and in delivering interventions exist both within the user population and among professionals. the interests of the groups overlap where there is a desire to develop, test and deploy more public health intervention strategies to draw the attention of pornography users across society to indicators of problematic usage. at the widest level there is value in developing strategies to have the issue of pup accepted by national healthcare systems. it can now be diagnosed as compulsive sexual behaviour disorder, coded c under the parent category of impulse control disorders [ ] . some clinicians are using this categorisation, but there will be considerable lags before it is adopted by nations world-wide. in some countries such as the united states, the need for clinicians to have a code for csbd to allow patients to claim on medical insurance is paramount. a separate issue revolves around making effective training for professionals more widely available. there is a need for the right balance of availability of practitioners (numbers, skills, geographical spread) to the scale of needs of people with issues. training can be built into continuing education programmes for professionals, as it is in the united states through the society for the advancement of sexual health and the international institute for trauma and addiction [ , ] . in the uk this is done through the association for the treatment of sexual addiction and compulsivity and the college of sexual and relationship therapists [ , ] . however, csbd and pup are very new ideas and they have not become part of general medical training in most tertiary institutions. proof of concept was demonstrated by a paper [ ] examining links between gene methylation and hypersexual disorder. the study found that the epigenetic state in the crh gene may contribute to explaining the biological mechanisms of hypersexual disorder. thus, some genes appear to play a role in hypersexual disorder. within the recovery community there is a strong desire to find answers to the 'why me' question. what is the relationship between predisposition to pup versus acquiring it through consumer behaviour in a changed technological environment? epigenetic factors need to be further researched. informal evidence from the online recovery communities indicates that certain character traits or personality types thought to be fixed actually change when regular use of a supernormal stimulus such as hardcore pornography ceases. for example, in some users out of control hypersexual or narcissistic or sensation seeking or aggressive behavioural traits remit or disappear when the stressor is removed. several separate, but overlapping strands of research around social factors would be useful. five are highlighted here as requiring more investigation: the technologies of the attention economy; women's changing role as consumers; environmental impacts; social acceptability; and the potential for early exposure to pornography to be considered as an adverse childhood experience. first, the manifesto in research area , in section above, identified "for excessive streaming (watching videos excessively), important structural features may include the ability of given programmes to grab attention by activating a biological 'orientating response', mediated through techniques including the use of attention-grabbing noises, zooming/panning, and presentation of rewarding stimuli (e.g., of a sexual or thrilling nature)" [ ] p. . this neatly summarises the core behaviour of people engaged in pup, where watching is generally accompanied by the viewer masturbating. this is absolutely an area where public research is required into the structural elements of commercial pornography sites as a part of the attention economy. commercial sites seem to have adopted the structural concepts pioneered by b.j. fogg at stanford university and the behavioural design work of nir eyal [ , ] . this allows them to build websites and apps that change users' thoughts and behaviour without their knowing. in particular, the pornographers target the unconscious mind via dopamine pathways in the reward centre to stimulate cravings that keep users coming back for more, resulting in pup for increasing numbers of consumers. artificial intelligence is being deployed by the commercial pornography suppliers to learn about user's sexuality to deliver the most engaging experience possible. this has implications for privacy. as some research suggests that pornography viewing can change a person's sexual tastes, this is a very slippery area. do we really want a big-porn machine algorithm to reshape our sexual templates [ ] ? another area ripe for research is the issue of intentional manipulation of consumption patterns by commercial interests in the pornography industry. in particular, since there has been a sustained increase of usage by women as a world-wide phenomenon [ ] [ ] [ ] [ ] ] . to what extent has this resulted from deliberate female-focused promotional strategies of the commercial pornography suppliers and what are its implications for dyadic relationships, family structures and society in general? pornography sites get heavy, repeat traffic from users. the largest supplier, pornhub, streamed billion videos in to a world-wide audience of perhaps billion individual viewers [ ] . it would be extremely helpful to have reliable calculations for the world-wide use of pornography. the industry is now so large that it has been calculated to consume % of the energy used by the internet and is currently responsible for . % of all greenhouse gas emissions world-wide, roughly equivalent to the domestic energy use of every household in france [ ] . the environmental impact is large enough for researchers to calculate a quantifiable contribution to rising sea levels. across cultures and within cultures, the phenomenon of "normalisation" of pornography consumption as a socially acceptable activity has grown rapidly. since the arrival of the iphone in june , supported by high-speed network services, the smartphone supplying "free" commercial pornography has removed barriers and increased temptation through its triple-a model of effectively unlimited access, affordability and anonymity [ ] . the speed of this adoption has outstripped the capacity of researchers to place it in context and to include it in theoretical models. it is clear that the arrival of smartphones, wi-fi and g networks leads to most societies adopting pornography as a more acceptable form of recreation, especially among the young. the trajectory of this change can be unique to each country [ ] . many stories circulate within the recovery community suggesting that childhood exposure to pornography in general, and internet pornography in particular, should be considered an adverse childhood experience (ace). however, at present this cannot happen. the list of definitions of aces pre-dates the arrival of problematic usage of pornography from the internet. it would mean revisiting the research field of defining, testing and verifying the validity of pup as an ace. this would be a very worthwhile project. these tend to revolve around whole-population approaches to changing the ease of access to internet pornography. after three false starts, will age verification legislation in the uk prove to be a successful intervention? how and where might it be tried in other countries? how will it be validated and will it achieve the intended outcomes? there is an established and diverse commercial ecosystem of blocking technologies now available for individuals and groups to prevent access to internet pornography. their characteristics tend to be specific to their local geographies. they can operate at the individual level, usually by subscription. many internet service providers (isps) offer filtering on either an opt-in or opt-out basis. academic analysis of the effectiveness of these measures would be helpful. pup research could also look to public health models used for other areas of addiction, particularly online gambling, where the business models and delivery mechanisms are often quite similar. policy research into new models of control, monitoring and product delivery limitation involving the government, commercial and voluntary sectors could be developed. this remains minimally charted territory. a recent development in pornography delivery (and sometimes production) is the increasing role of social media as a point of access. while games delivery platforms are now a significant source of pornography for some consumers, we have also encountered adolescents getting around parental controls via the internet of things. a poorly protected smart refrigerator could get a grounded teen online. individual or community approaches to treatment from counsellors, schools, churches or other local units can be successful, especially when linked to demand-side reduction measures. accountability partners are popular in some communities [ ] . again, more external validation of success, or not, of different approaches would be valuable. are all pornography consumers at risk of developing pup? the simple answer is probably not, but we do not know for certain. if the general characteristics of pup are broadly similar to other better studied behaviours with a propensity to lead to behavioural addiction, the answer is 'no'. a proportion of users will be likely to move first into problematic use and then addictive use. some studies suggest that all pornography consumption generates brain changes, even at quite low durations [ , ] . so far, the world health organization only considers the research base is strong enough to support the diagnosis of compulsive sexual behaviour disorder under the parent category of impulse control [ ] . at the same time, few pornography consumers have been exposing themselves to large volumes of online erotica for longer than two decades, so we have only explored it as a factor for, at most, half of the duration of a human beings' sexual life-cycle. we do know that the more pornography some people consume, the more likely they are to develop compulsive behaviours as a result. it is the sort of issue that could respond well to the development of a risk-managed framework. such a framework could balance benefits of pornography consumption at the individual and society levels with the inevitable side-effects from some individuals being unhappy with the consequences arising from their use. where consumers do develop concerns, many practitioners believe that the solution to pup is rooted in consumer education. as a society we can choose to try to reduce the strength of demand for access to pornography. some individuals who see pornography as a valuable part of a sex positive culture would be uncomfortable about this, but it is up to each society to consider what restrictions may or may not be appropriate if some individuals may be harmed by excessive consumption. more systematic assessments of the success of educational strategies would show how effective they can be as risk-reduction strategies. what role will learning about the addictive potential of pup play in reducing demand? does learning about porn-induced erectile dysfunction reduce consumption? thousands of anonymous self-reports on the recovery websites show that the availability of knowledge of the effects of internet pornography on the brain was sufficient to encourage users to experiment with quitting porn [ , ] . when they did, a range of mental and physical health problems remitted or cleared up. only when heavy users quit porn did they realise that those conditions were related to their porn use. formal research into this phenomenon would be valuable. a separate investigation into the issues of prevention versus cure would be interesting. we are suggesting research to compare the benefits of never beginning pornography usage versus treating the problem when it has been 'cured'. at a population level, are people who quit ultimately as mentally and physically healthy as those who never engaged in pup in the first place? what methodologies might help unpack this question? the development of possible drug treatments is still at an early stage. in theory adjusting brain chemistry with substances could reduce libido and cravings for pornography. the most common approach is to look at substances that are already approved for clinical use and to begin the process of taking them through stages from case reports and on to small and then large-scale clinical trials. what drugs might discourage pup and what pathways should be targeted? a major review was published by sniewski and colleagues in [ ] . early work focused on possible similarities to obsessive compulsive disorder and suggested trying drugs known as selective serotonin re-uptake inhibitors (ssris). recent case reports have been published for paroxetine [ ] , nalmefene [ ] and aripiprazole [ ] , with mixed results. research area : identify biomarkers, including digital markers, to improve early detection and intervention [ ] . this is a new area, but it may have real potential to help improve health at the individual and population levels. as far as we are aware, biomarkers are not currently in use for detecting or diagnosing pup. would there be value in their introduction? research might be able to separate out the underlying tendency of some individuals or groups to be susceptible to problematic usage. this could be contrasted to the potential for internet pornography to lead to problematic usage as a natural characteristic in and of itself. the issue here is "are you the problem?" versus "is pornography the problem". co-morbidity studies involving other puis would also be helpful-many users find themselves trying to unhook from gaming or recreational drugs and pornography at the same time. biomarkers are a high-tech intervention. in face-to-face situations, such as a doctor's surgery or a counsellor's office, are they a better solution than simply asking a person about their level of pornography consumption? the brief pornography screener with five questions on a single a sheet of paper can be administered in three minutes and seems to be both sensitive and reliable [ ] . a different line of research would be to look at the potential for widescale interventions using internet data. it might even extend to artificial intelligence algorithms at commercial pornography suppliers or the isp level. it would mean monitoring people for signs of problematic usage and then invoking a warning or even a cut-off protocol. this has privacy and data management implications, but the reality seems to be that monitoring rather like this is already at the core of the business model of major commercial pornography suppliers. could governments consider introducing national level intervention programmes? how would they operate and what would be the success criteria? at the same time, the pornography supply companies are very reluctant to share their data. in this they follow the lead of other industries supplying addictive substances like tobacco and alcohol. the commercial pornography suppliers are quite vulnerable to a 'smoking gun' which showed that were trying to induce addictive behaviour as a part of their business model. when you consider the conclusions in the manifesto [ ] in the light of the current paper, there is nothing recommended for research into pui that does not equally apply to the sub-discipline of pup. to come back to our research question, "what topics should be included within future research proposals under the manifesto to meet the diverse needs of consumers, recovery communities and professionals impacted by the problematic use of pornography"? the comments in this section have been grouped to answer these questions individually, but there is, and should be, overlap between them, so we begin by looking at general questions about the field of research. the management and treatment of pup requires the right approaches to meet the specific needs of everybody in this mental and physical health ecosystem. given the massive scale of use of pornography and its potentially negative outcomes for some users, it is important for the european union's problematic usage of the internet (eu-pui) network to have a visible and trusted champion for pup, just as it should for other behaviours. linked to this, europe should hold a planning conference under the umbrella of cost action to design its response. the opportunity for pup data to be made accessible to researchers through shared multinational databases is extremely attractive. to assist the world health organization in improving the classification of compulsive sexual behaviour disorder in future to include the word pornography, research is required to separate out the natural history of pup from compulsive sexual behaviours focused on people. experimental work to investigate causation through longitudinal studies would provide a better strategic foundation for pup. this would be stronger if it included large-scale brain imaging studies to gather evidence before and after people have quit or have been given treatment. it is essential to develop a wider evidence-base covering all ages and stages of the human life-cycle for people engaging in pup. there should be much more research focus on women as they are now the main sector of growth for pup. equally our data should embrace sexual diversity and investigate the usage among lgbtqi++ communities where levels of mental health disorders and chemical dependencies are higher than in the mainstream heterosexual population. could pup be a contributing factor? the s is the decade when training of health and social care professionals should begin to incorporate a deep and broad understanding of all puis as a matter of course. this will be a communication and public relations programme. it should be recognised that within the field of pup we are starting from a low-base with modest levels of training or resources available for people working as professionals and therapists. new research should be leading towards giving professionals a wider toolkit to support treatment of all the different forms of pup. ideally the next ten years will see a rising quality of research into pup, moving us towards a landscape with more double-blind placebo trials for drugs and longitudinal studies covering time periods that allow us to take a broader view on pup. it is a disorder that is slow to develop and potentially also slow to heal. we should encourage better and deeper validation of a few assessment tools across diverse populations. standardisation is helpful and surely two proven ones are sufficient. link this to a strong communication programme to make them known widely in communities from general practitioners (mds) and urologists to sex therapists, relationship counsellors, school counsellors and parents. the next decade should see the debate around pornography use and erectile health fully resolved. dialogues about the impact of pornography use in the context of violence against women and children also require more experimental approaches if the discussion is going to be able to move from correlation to causation. the next few years will clearly be a time of great advances in linking genetic data and behaviour and we will monitor the contributions that this form of high-technology can make to epigenetic changes in humans. at the low-tech end, the most important thing is to initiate a research and development programme to determine if early exposure to internet pornography can be recognised as an adverse childhood experience (ace). success in that project could have a profound impact on raising awareness of the need to reduce early exposure and thereby reduce the risk of subsequent pup. work monitoring the success, or otherwise, of age verification legislation around the world will be important for policy development. for individuals and communities, it is important that we learn which interventions are sustainable and cost effective. could commercial pornography suppliers be encouraged to take a pro-active role in heading off the development of pup, rather than following their current laissez faire view that a customer can never have too much pornography? it will be important to have research to boost the accountability of suppliers. accountability can be achieved through government intervention or industry self-regulation. so far the latter does not seem to be working in the consumer's favour. in general the access to pornography is free, but the consumption risks are all transferred to the user. the powerful position of commercial pornography suppliers within the context of the attention economy leaves the average consumer poorly informed about how their tastes and behaviour are being influenced by 'free' pornography. commercial pornography suppliers use the data that gather to develop a deep understanding of consumer behaviour. commercial players also benefit from the access to artificial intelligence and other data-mining tools to build market share and develop new markets. good academic studies of the recovery communities, be they nofap, sex addicts anonymous or others, could be revolutionary. recovery communities are a cheap solution to a potentially very expensive medical problem. however, often the communities do not really want to be studied. members just want to be healed or to help others to heal. there is some acceptance that good research could help. there is also a recognition that for some pup members, recovery communities are only part of the answer and that they may also need help from therapists and other professionals. quantitative research looking at the elimination of pornography is the highest priority. does it really reverse sexual dysfunctions and have a positive impact on a variety of mental health disorders? the field currently has extensive anecdotal evidence to support these suggestions, but that is not a substitute for large-scale, well-grounded studies. it is now nearly two years since the manifesto was published. in that time there has been a continuing growth world-wide in the volume of research appearing on pornography use in general and pup in particular. however, most of the work being done tends to be at the easier, cheaper end of the scale. as far as we can tell as frequent observers of this field, there remains a general lack of vision and large-scale co-ordination between the many research groups around the world. the future needs to be more about designing investigations to test causation, not just to demonstrate yet more correlations. the time is right for the cost action team and the eu-pui network to create, fund and promote a visible and trusted champion for pup research. europe has sufficient resources to support such a role. perhaps the research ideas sketched in this paper could be a positive contribution towards cost action accepting it should deliver a leadership role in the field of pup. focusing on the manifesto is a limiting strategy, in that many researchers in the pornography field around the world are unlikely to be familiar with it. our approach has been to work with what is already available to build a more coherent research environment. at the same time, focusing on the manifesto is essential. it is the only major published document which deals with the whole policy context of problematic use of the internet. it is europe-wide in its reach and is designed to endure for a decade. by writing a response homing in on the issues around pup, we hope to encourage researchers to have conversations to advance the field in ways that will benefit the continent's taxpayers and publicly funded research programmes. our methodology of simply analysing all of the manifesto as a whole through the particular lens of pup is a limitation. this approach was taken to balance achieving the clearest level of understanding of what readers could infer from the document with an open agenda to build the strongest possible set of recommendations to shape the future of pup research in europe over the next decade. manifesto for a european research network into problematic usage of the internet the risks young people face as porn consumers. addicta-turk the internet's impact on sexuality: a critical review of years of research online sexual activity experiences among college students: a four-country comparison pornhub insights year in review year in review the year in review when pornography becomes a problem: clinical insights a current understanding of the behavioral neuroscience of compulsive sexual behavior disorder and problematic pornography use college of sexual and relationship therapists clinical characteristics of compulsive pornography users: a military sample society for the advancement of sexual health association for the treatment of sexual addiction and compulsivity international institute for trauma and addiction professionals addicted to internet porn understanding and treating sex and pornography addiction: a comprehensive guide for people who struggle with sex addiction and those who want to help them international classification of diseases, th revision (icd- ) revisiting the role of impulsivity and compulsivity in problematic sexual behaviors pornography use in the setting of the covid- pandemic pornography and sexuality research papers at the th international conference on behavioral addictions. sex. addict. compulsivity pornography and sexuality research papers at the th international conference on behavioral addictions sex addicts anonymous; international service organization of saa eliminate chronic internet pornography use to reveal its effects is internet pornography causing sexual dysfunctions? a review with clinical reports internet sex addiction treated with naltrexone naltrexone for the treatment of comorbid tobacco and pornography addiction treatment of compulsive pornography use with naltrexone: a case report positive response of compulsive sexual behavior to transcranial magnetic stimulation desire thinking: a new target for treatment of addictive behaviors? advances in metacognitive therapy the reward foundation -love, sex and the internet the impacts of sexual media exposure on adolescent and emerging adults' dating and sexual violence attitudes and behaviors: a critical review of the literature pornography and sexual assault. in handbook of sexual assault and sexual assault prevention pornography and sexual violence problematic pornography use and physical and sexual intimate partner violence perpetration among men in batterer intervention programs the development of the problematic pornography consumption scale (ppcs) online porn addiction: what we know and what we don't-a systematic review epidemiology update of erectile dysfunction in eight countries with high burden promotion and protection of all human rights, civil, political, economic, social and cultural rights, including the right to development motivational pathways underlying the onset and maintenance of viewing child pornography on the internet family and abuse characteristics of british child and adolescent sexual abusers are adolescents who consume pornography different from those who engaged in online sexual activities? problematic pornography viewing from a big- personality perspective diverse sexual behaviors and pornography use: findings from a nationally representative probability survey of americans aged to years nalmefene in the treatment of internet pornography addiction-a case report and review of literature aggression and sexual behavior in best-selling pornography videos: a content analysis update. violence women florida man arrested after porn videos, photos link him to missing underage teen girl characteristics of individuals who participate in autoerotic asphyxiation practices: an exploratory study erectile dysfunction: a review of comprehensive treatment options for optimal outcome your brain on porn: internet pornography and the emerging science of addiction bdo): a literature review with a focus on uk fatalities related to non-medical use brain structure and functional connectivity associated with pornography consumption: the brain on porn neural correlates of sexual cue reactivity in individuals with and without compulsive sexual behaviours characteristics of vibrator use among women individual differences in women's pornography use, perceptions of pornography, and unprotected sex: preliminary results from south korea exploring new measures of online sexual activities, device use, and gender differences korean men's pornography use, their interest in extreme pornography, and dyadic sexual relationships unusual online sexual interests in heterosexual swedish and italian university students methylation of hpa axis related genes in men with hypersexual disorder tiny habits: the small changes that change everything hooked: how to build habit-forming products online sexual activities: an exploratory study of problematic and non-problematic usage patterns in a sample of men the official xhamster trend report climate crisis: the unsustainable use of online video; the shift project surfing into the new millennium the assessment and treatment of adult heterosexual men with self-perceived problematic pornography use: a review paroxetine treatment of problematic pornography use: a case series hypersexuality and new sexual orientation following aripiprazole use this article is an open access article distributed under the terms and conditions of the creative commons attribution (cc by) license the authors would like to thank julius burkauskas, the science communications manager of the cost action dissemination plan for encouraging the preparation of this paper. the authors declare no conflict of interest. key: cord- - fjlz lf authors: libório, matheus pereira; ekel, petr iakovlevitch; lyrio, renata de mello; bernardes, patrícia; soares, gustavo luís; machado-coelho, thiago melo title: expand or oversize? planning internet access network in a demand growth scenario date: - - journal: j netw syst manage doi: . /s - - -w sha: doc_id: cord_uid: fjlz lf internet network design specialists are looking for technologies and strategies to deliver network service under increased demand conditions. the choice of strategies is based on applying optimization and decision-making methods to select the most appropriate cable network design considering criteria established by the problem definition. however, this definition is itself a decision problem that has not received analysis in the literature. in particular, one of the most important questions is the necessity to define an expansion strategy. the first alternative (expansion) is to design a network to serve consumers with internet demand equal to or greater than the predefined one to expand the network annually to serve consumers that reach the predefined internet speed. the second alternative (oversizing) is to design a network to serve consumers with future internet demand (after years) at or above that the predefined one. considering this, the objective of this research is to define the most advantageous strategy of expansion planning to attend a years forecasted internet demand, considering: ( ) the possibility of utilizing a gigabit-capable passive optical network technology; ( ) the application of the minimal steiner tree and dijkstra algorithms in planning procedures; ( ) the influence of economic and technological factors on the demand forecast; ( ) the aggressive, moderate, and conservative scenarios in decision-making. the results show that the over-dimensioning strategy reduces network investment by between and %, but that this reduction does not always lead to a market investment ratio higher than that observed in the expansion strategy. in the early s, the oversizing of internet networks left equipment and cables underused for several years [ ] . however, the ability to service these networks has not kept pace with the growth of internet demand [ ] . this hindered the recovery of investments made, especially as these networks approached their physical limitations [ ] and were left without margin to generate new revenues and higher operating costs [ ] . this was because the technologies of these networks were not fully flexible [ ] . that is, if, on the one hand, the exchange of equipment allowed the increase of the bandwidth capacity, on the other hand, the cable network architecture limited the attendance of new customers, especially customers outside the cable network coverage area [ ] . for these reasons, network planners have increasingly considered in their network planning strategies: i) forecast of internet demand [ , [ ] [ ] [ ] [ ] ; and ii) the use of flexible technologies [ , , , ] . taking brazil as an example, forecasts based on historical data show that internet demand will reach . kbps per person by [ ] . however, research also shows that forecasting models based on historical data distort long-term internet demand results [ ] . such distortions occur because internet demand information on emerging technologies [ ] (e.g., online games, video streaming [ , ] , cloud computing [ ] , and systems of iot and sc [ ] ) is not present in historical data. in brazil, researchers forecast that the interconnectivity between emerging technologies and iot and sc innovations will boost internet demand per person, making demand nearly double between and ( . kbps per person to . kbps per person) [ ] . this forecast is % higher than forecasts based on historical data [ ] and reinforces the understanding that emerging technologies and iot and sc systems can impact both long-term internet demands forecasting results and planning strategy of new networks. this is because these networks must be designed to support billions of parallel sensor data streams [ ] , enable big data applications, process integrated real-time, high-speed data exchange [ ] , oversee the physical components of the system, ensure high levels of security [ ] , prevent spying and/or external attacks [ ] and withstand high flash rates [ ] . that is, internet network planners should consider technology solutions that are adaptable to dynamic internet demand scenarios [ ] which suggest that: (i) % of internet traffic will not originate from personal computers; (ii) . billion devices will be connected, and iii) the devices will consume in h all internet traffic currently consumed in one week [ ] . among current technologies, studies suggest that the gigabit-capable passive optical network (gpon) [ ] is a technology solution that is adaptable to the scenario of increasing internet demand [ , , ] , and can be considered a proven solution of the future [ ] for at least five reasons. first, gpon networks are immune to electromagnetic interference and do not require active components between a switch and a client [ ] . second, gpon networks allow service traffic originating from both mobile ( g) and cable networks [ ] . third, because gpon networks offer abundant bandwidth capacity to support bandwidth-intensive services traffic [ ] . fourth, because gpon networks enable network traffic capacity to be increased by upgrading and/or adding equipment [ , ] . fifth, because gpon networks make it possible to serve new customers by laying cables from splitters [ , ] . in short, the characteristics of gpon technology favor the expansion of network capacity (traffic and/or new customers) to meet increased internet demand regardless of the planning strategy (expand or oversize). but what is the most advantageous network planning strategy? in general, network planning includes internet demand forecast, geoprocessing the characteristics of the local analysis unit (city, neighborhood, etc.), and cable network design. internet demand forecast involves variables that represent social conditions (e.g. macroeconomic aspects [ ] , technologies and innovations [ ] , the internet availability [ , ] ) and, the procedures adopted in this relatively fickle activity [ ] . the geoprocessing of the physical characteristics of the unit of analysis involves variables that represent the occupation of the city (e.g., distribution and population concentration [ ] , the layout of streets and avenues [ , , , ] ) being procedures adopted in this activity relatively constant [ ] . network design involves cost-related variables (e.g. cable routing and equipment installation), not the relatively constant procedures adopted in this activity [ ] . telecommunications network design studies generally focus on minimizing investments, i.e., minimizing the number of equipment and the size of the cable network. in this sense, such studies are restricted to the use, adaptation, and development of optimization methods (e.g., integer programming [ , , , ] , combinatorial optimization [ ] , fast fourier transform heuristic algorithms [ ] , dijkstra algorithm [ , ] , and other, for instance, [ ] ). in the case of minimizing cable network size, planners look for the shortest path between network vertices (streets and customers) to establish a minimal network. establishing this minimal network is an np-hard problem known as steiner's minimum tree [ ] . however, path optimization algorithms [ ] used to design the steiner's network are heuristics that can yield suboptimal results [ , ] . although no overall minimization of investments is guaranteed, these algorithms are widely used in network planning [ , ] and are found both for planning the expansion of existing cable networks [ , , , , , ] as well as for the planning of new cable networks [ , , , ] . however, although some of these studies are concerned with the internet demand forecast models [ , , ] and/or the geoprocessing of the features of the analyzed sites [ , , ] , these studies ignore the expand strategy and merely design an oversized network. in other words, these works focus exclusively on minimizing investments, ignoring the fact that that the most advantageous network strategy is defined by the return of investment ratio [ ] , that is, the ratio between the market and investments. in this footnote (continued) therefore, it is considered in this research that: the number of mobile devices in homes is increasing [ ] ; users are increasingly adopting home wi-fi [ ] ; home wi-fi is used to offload traffic from mobile devices over fixed broadband, reducing costs and improving the quality of mobile services [ ] . in short, internet networks are to be designed to support some traffic from g network services. sense, this research checks whether lower investments are reflected in higher return of investment ratio. this check is based on the cable network planning scenario presented in this section and briefly illustrated in the fig. . as illustrated in fig. the goal of this research is to answer which is the most advantageous cable network planning strategy using the market investment ratio information. answering this question is a key element in the decision on investing in new networks because they are directly related to the return on investments made. this answer may be especially important for countries that rely on investments in new cable networks to increase their bandwidth capacity and, especially, to reduce their internet access inequality. in brazil, data from the national telecommunications agency [ ] and the brazilian institute of geography and statistics or ibge [ ] reveal that % of customers served by cable networks are concentrated in the largest cities ( . % of the total) while cities smaller than , inhabitants ( % of brazilian cities) concentrate , customers or . % of total accesses. the most advantageous cable network planning strategy was defined from three sets of procedures: estimating the megabits per second (mbps) in brazil, estimating investments in kilometers (km) of cables, and analyzing the market investments ratio or mbps/km. the procedure one "estimating the megabits per second (mbps) in brazil" includes the following steps: i) internet demand forecasting (mbps per person), executed on ninna-pca and microsoft excel software; ii) geoprocessing of the characteristics of the city under analysis; iii) creation of conservative, moderate, and aggressive decision scenarios; and iv) estimation of the internet demand [ ] . pca was applied to reduce the dimension/number of variables related to emerging technologies and the iot and sc systems in order to avoid collinearity of independent variables in the forecast model [ ] . the pca model must satisfy three criteria [ ] . first, average variance extracted (ave) measures the variance of the variable that is explained by the principal component. second, kaiser-meyer-olkin (kmo) test measures whether the sample is suitable for the model, indicating the proportion of variables with common variances. third, bartlett's test of sphericity compares the pca correlation matrix with the identity matrix, verifying the degree of redundancy between the model variables. the pca model must include enough components to overcome the ave threshold of . [ ] . the kmo test greater than . indicates that sampling is adequate and the bartlett's test less than . indicates that the correlation matrix is not an identity matrix [ ] . in turn, the mlr was applied to find the internet demand forecast function [ ] . the mlr is a technique used to predict a variable (dependent) such is the internet demand from a set of explanatory variables (independent). the mlr model is reliable when the adjusted r-squared (adjusted-r , significance test (f-test), and p-values criteria are satisfied [ ] . the adjusted-r indicates how much of the dependent variable can be explained by the independent variables, taking into account the number of independent variables in the model. in other words, it indicates the power of the forecasting model, with adjusted-r equal to indicating that the independent variables do not explain the dependent variable and that an adjusted-r equal to indicates the dependent variable is perfectly explained by the independent variables. the test-f determine the statistical significance of the model. test-f values less than . shows that the regression is statistically significant. the p-values tests the statistical significance of each independent variable of the model and must be less than . [ , ] . the dependent and independent variables used in the mlr to forecast the internet demand are listed in table . the internet mbps/person forecast was made for a period of five years. the internet market is estimated based on this forecast and the results of the second step: the geoprocessing of the characteristics of the city under analysis. this step was based on the georeferencing of population census data [ ] . these data were updated and projected (five years ahead) using the respective estimates [ ] and population projections [ ] produced by the brazilian institute of geography and statistics (ibge). in turn, the georeferencing of the data was based on: i) the proportion of the population present in each census tract [ ] ; ii) number of vertices in each census tract, and; iii) latitude and longitude of the vertices of the census tracts. the geoprocessing of the characteristics of the city under analysis yields the number of people in each census tract vertex or market point. the internet market in mbps is obtained by multiplying the internet demand per person (mbps/person) obtained in the forecast model by the number of people in each market point (person). knowing the internet market in the city, the question "expand or oversize?" is analyzed in the third step from the creation of different scenarios. the investment decision scenarios (conservative, moderate, and aggressive [ ] ) take as a criterion the minimum internet market in mbps accepted by the offerer. thus, the decision scenario is considered conservative when the offerer requires a higher minimum mbps, that is when the offerer imposes more restrictions to serve customers. in turn, the decision scenario is considered aggressive when the offerer accepts a lower minimum mbps, that is when the offerer imposes fewer restrictions to serve customers. finally, the decision scenario is considered moderate when the offerer is strategically positioned between the conservative and aggressive posture. an example of the use of minimum mbps demands as a constraint investment by the offerer can be found in the work of ljubic, putz, and salazar-gonzález [ ] . in the study, the minimum mbps of conservative, moderate and aggressive scenario was defined respectively from the third, second and first quartile of internet market per point (obtained by multiplying the results of steps and ). this gives the minimum market point mbps definition of each of the three scenarios. in this example, the internet demand per person (mbps per person) for each scenario is the same. the difference between the scenarios is the minimum market (mbps) that the offerer is willing to invest. when the mbps required by the offerer is the same, the difference between the scenarios would become the mbps per person itself. in this case, the higher the mbps per person, the greater the number of market points. as we will see, the scenario with the highest number of market points is the aggressive one followed by the moderate and conservative ones. in this sense, the aggressive scenario may represent the increase in mbps per person caused by the covid- [ ] as more people access the internet from home [ , ] . the conservative scenario may represent a future retraction of mbps per person as the independent variables used in the forecast model: ur and gdp will be affected in the medium term by the covid- [ ] . the international monetary fund estimates that the world gdp will retract by . %, and that the brazilian ur will reach . % of the population. thus, there is an opportunity to interpret the aggressive scenario as the internet demand that occurs during the covid- and to interpret the conservative scenario as the internet demand that occurs after the coronavirus. the procedure two "km investments in cables estimates" includes the following steps: i) design the internet network for each strategy and scenario by applying the dijkstra [ ] and steiner [ ] algorithms; and ii) measure the cable network for each strategy and scenario in km, executed in the qgis software [ ] . the cable networks planned were designed considering the market points in mbps (obtained in the step of procedure one), and the physical barriers (e.g. set of streets, squares, and avenues) of the unit of analysis (obtained from digibase digital maps [ ] ). to design the oversize strategy network, we used steiner's minimum tree algorithm [ ] and the market points in mbps from year . to design the expansion strategy network, two procedures were performed. the first procedure employs steiner's minimum tree algorithm [ ] and year mbps market points to design an initial network. the second procedure employs the dijkstra algorithm [ ] , the initial network and the year mbps market points to design the first network expansion. then the second procedure is repeated successively, employing dijkstra [ ] , the initial network, the expanded networks and the mbps market points of the year in question. thus, six cable networks were designed: three cable networks for each of the two strategies. the procedure three "market investments ratio or mbps/km" analyzes the market investment ratio of the six designed cable networks. the answer for which is the most advantageous network planning strategy is obtained by comparing the market investment ratio (mbps/km) of each strategy for the three scenarios. this subsection includes the results of the following steps: (i) internet demand forecasting (mbps per person); (ii) geoprocessing of the characteristics of the city under analysis; (iii) creation of conservative, moderate, and aggressive decision scenarios; and (iv) estimate the internet demand in mbps. the first results of the internet demand forecasting step are related to the inclusion of the emerging technologies and iot and sc systems in the future internet demand. the tec variable, created from the pca, had the ave of . and kmo of . , and bartlett's test of less than . . this means that the variance explained by the principal component represents almost entirely the variables: (i) ip traffic; (ii) internet traffic; (iii) mobile data. in turn, the kmo found indicates that the pca sample size is appropriate for the analysis [ ] . finally, the bartlett's test showed that the correlation matrix is not an identity matrix and therefore the variables are correlated. thus, the tec variable allows incorporating the effects of emerging services [ ] and iot and sc systems [ ] as well as to avoid collinearity between independent variables in the forecasting model [ , ] . considering the presence of these effects in the forecast model, it can be seen in fig. that internet demand in brazil will grow % in years, reaching kbps per person in years. in the forecast presented in the graph, it is observed that the internet demand in brazil is % and % higher than the forecasts presented in previous studies, respectively [ , ] . in this model, the adjusted-r was . , showing that the internet demand can very strongly predicted by the independent variable. f-test was less than . , showing that the regression is statistically significant. the p-values were the supply scenarios, defined on the basis of the first, second, and third quartile values obtained from the internet demand forecast result ( kbps per person) and the population at each demand point (ibge [ ] ), were as follows: i) in the aggressive scenario (first quartile), the offerer, agrees to meet demands equal to or above . mbps, restricting the cable network to meet % of the market points; ii) in the moderate scenario (second quartile), the offerer agrees to meet demands equal to or above . mbps, restricting the cable network to meet % of the market points; iii) in conservative scenarios (third quartile) the offerer agrees to meet demands equal to or above . mbps, restricting the cable network to meet % of the market points. table shows how the internet market evolution for each scenario has evolved over the years. the market points are the vertices of the cable network. if, on the one hand, the number of market points indicates what the investment in the cable network will be, on the other hand, the internet market signals the return on these investments. it should be noted that the difference of % between the internet market in the moderate ( mbps) and aggressive ( mbps) scenarios is close to the percentage of increase in internet traffic observed as a result of covid- [ , ] . that is, although superficially, the aggressive scenario can be useful to understand how the changes caused by covid- influence the planning of internet networks. from the market points and applying the minimum path algorithms, the internet networks were designed for the strategies of expanding and oversizing. figure compares the km of cables for the aggressive scenario for expanding and oversizing. in this scenario, the internet provider, by imposing a minor service constraint, plans the network to meet market points in the fifth year. in the fifth year of the aggressive scenario, the strategy of expanding implies km of cables . % higher ( . km) compared to the oversize strategy. in turn, the strategy of expanding implies higher km of cables during the first years. figure brings the comparison of the km of cables in the moderate scenario based on the networks designed by the strategies of: expansion and oversizing. in this scenario, the internet provider invests in a cable network to meet % of the in the moderate scenario illustrated in fig. , it is possible to notice a reduction in the number of market points compared to the aggressive scenario. with fewer market points to meet, the km of cables difference from networks designed by the expansion ( . km) and oversizing ( . km) strategies decreased from . km (aggressive scenario) to . km. however, the km of cables difference from networks designed by expand and oversizing increased from . % (aggressive scenario) to . % in the moderate scenario. figure brings the comparison between the km of cables for the conservative scenario in the strategies: expand and oversize. in this scenario, the internet provider invests in a cable network to meet % of the market points. this percentage represents a scenario of greater restriction of supply, that is, the fulfillment of a smaller number of market points. therefore, cable networks are designed to meet the points that have mbps market equal to or above . mbps. in conservative scenario, the lower number of internet market points compared to aggressive ( %) and conservative ( %) scenario, are reflected in the smaller km-cables km-cables ---- expand oversize fig. aggressive scenario-internet market points > . mbps km-cables - km-cables ----expand oversize fig. moderate scenario-internet market points > . mbps cable network. while in the strategy of expanding the network designed in the conservative scenario was . km of cables, i.e. . % and . % smaller than the network designed by the same strategy in the moderate and aggressive scenarios, respectively. in the strategy of oversizing, the network designed in the conservative scenario was . km, . %, and . % smaller than the networks designed by the same strategy in the moderate and aggressive scenarios, respectively. among the three scenarios, the conservative scenario presents the smallest difference in cable investment (km of cables = . km). on the other hand, expanding the network in this scenario implies a . % higher investment in cables than oversizing the network. this is the largest percentage difference in cable investment of the three analyzed scenarios. these results show that oversizing the network is the most advantageous strategy for reducing km cable costs, especially when the constraint on meeting demand (minimum mbps) is less. this is because oversizing the network implies investments of . km, . km, and . km smaller when the minimum mbps are respectively . mbps, . mbps, and . mbps. however, planners should consider that the percentage of cost savings in km of cables is lower when the demand constraint is lower, . %, . %, and . %, respectively. in summary, the results of the km of cables suggest that oversizing the network implies lower investments regardless of the internet supply scenario. these results confirm that oversize is an optimal solution while the expand strategy is a kind of greedy optimization. however, it is still necessary to verify whether lower investments are reflected in a better return of investment ratio. this verification is performed by analyzing the market investment ratio of each strategy by scenario. the mbps/km of the expand and oversize strategies in the aggressive scenario equals mbps/km in the fourth year. it means that the difference between the market investment ratio in the fourth year is equal to zero. before that, the expand strategy presents an average of % higher mbps/km. in the fifth year, the mbps/ km-cables km-cables ----expand oversize km of the oversize strategy is % higher. in turn, the sum of the mbps/km differences in the expand and oversize strategy is positive at mbps/km. this result suggests that expanding is the most advantageous strategy for the aggressive scenario when taken into account all the five years. figure presents a graph that summarizes the analysis of the market investment ratio for the aggressive scenario. the lines represent the mbps/km of each strategy per year. the points represent the mbps/km difference between the expand and oversize strategies. positive points indicate that the expand strategy is more advantageous, as they generate a higher mbps/km. the rhombus is the sum of the mbps/km differences that indicates which is the most advantageous cable network planning strategy. in the moderate strategy the results are not so different. the network designed in the fifth year by the oversize strategy is % ( mbps/km) higher than the network designed by the expand strategy. however, in this scenario, the oversizing the cable network is the best planning strategy not in the fifth year, but in the third year. as a result, the sum of the mbps/km differences of the expand and oversize strategies is negative at mbps/km. this means that oversizing the network is the most advantageous strategy when taking into account the market investment ratio of the five years analyzed. figure summarizes the return of investment analysis of the moderate scenario which supports the understanding that the first years of the network's life are the most essential to return the investments made [ ] . in fact, in the conservative scenario, the mbps/km of the oversize strategy is also better than the strategy of expanding in the fifth year. but, just as in the aggressive scenario, the sum of the mbps/km (rhombus) differences is positive, suggesting that expanding is also the most advantageous strategy for the conservative scenario. the comparisons of the mbps/km per scenario shows that oversizing the cable network is the most advantageous strategy for meeting fifth-year demand in any scenario. however, the mbps/km results also show that major supply constraints (high minimum mbps) favor the strategy of expanding the network. this can be seen in fig. , as the values of the mbps/km in the oversize expand, are larger (positive results) than the mbps/km of the oversize strategy (negative values). the mbps/km analysis suggests that oversizing networks are the best strategy for meeting medium to long-term demand and expanding is the best strategy for meeting short-term demand. however, in the conservative scenario, the strategy of expanding the network proves to be the most advantageous strategy until the fourth year, the mbps/km of the oversizing ( mbps/km) and expansion ( mbps/km) strategies have the largest difference ( mbps/km) between the three scenarios. in short, the results of this research suggest that cable network planning strategies drive changes into the network designs, altering (positively or negatively) the volume of network investments. the results show more specifically that cable networks, if planned by the oversizing strategy, may be between % and % smaller than if planned by the expansion strategy. however, the results show that a smaller network is not always reflected in a higher market investment ratio (mbps/km). in the aggressive and conservative scenario, oversizing the network results in higher mbps/km only in the fifth year. considering the sum of mbps/km over the five years, expanding the network is the most advantageous strategy in these two scenarios. these results show how important the market investment ratio in the early years influence the network planning strategy definition, and consequently the return the investments made [ ] . thus, this research contributes to open up new perspectives to increase the return the investments made in internet networks. this means that network planners should not limit themselves to applying optimization methods, adding that expansion strategies are to be incorporated in their planning activities. besides, the results not only reinforce the understanding that flexible network technologies [ , , , ] are [ ] , adding that these technologies offer different possibilities of network expansion strategies and design. these contributions may be especially important to increase the supply of networks in developing countries. thus, the results may be limited to the unit of analysis (city), leaving a question: what is the most advantageous network planning strategy in other cities? thus, it is suggested to analyze the possible changes in the decision of the planning strategy when changing: the configuration of physical barriers (streets, avenues, etc.) present in each city, the demographic density found in each city, and the socioeconomic potential (presence of computer at home, income, etc.) of households. this research shows that internet demand growth maybe % higher than predicted in previous studies [ , ] . this evidence reinforces the understanding that emerging technologies [ , ] and iot and sc (tec) systems [ , ] stimulate the development of bandwidth-intensive applications and services [ ] . this forecast of internet demand is an important contribution for network planners as they rely on these forecasts to design internet networks [ ] . in this case, knowing that internet demand will be greater than predicted contributes to network planners designing networks with the minimum cable network and the highest possible bandwidth and service capacity. besides, this research contributes to network planners being able to design smaller cable networks with sufficient bandwidth and/or service capacity to meet growing demand through different cable network planning strategy. firstly, regardless of the planning strategy, flexible technologies (e.g., gpon [ , ] ) allow for increased network bandwidth and service capacity while optimizing methods (e.g., dijkstra [ ] and steiner [ ] algorithms) make it possible to minimize investments in cable networks [ , ] . second, regardless of the scenario, oversizing the network reduces the cable network (km) by at least . %. third, oversizing the network tends to be the most advantageous strategy when it has higher mbps/km from the third year. fourth, expanding tends to be the most advantageous strategy when it has higher mbps/km up to the fourth year. although these results are limited to the analyzed city, this research opens new perspectives for studies that seek to minimize investments in the cable network. in this sense, future studies may explore the effects of planning strategies in defining the positions and number of optical network splitters [ ] or the amount of fiber in the network cables [ ] . another possibility is to explore possible changes in the definition of the most advantageous network planning strategy when using other network design optimization methods [ , , , ] . finally, it is suggested to analyze the hypothesis that changes in the configuration of the physical barriers of cities (e.g. avenues, streets, and bridges) or the population distribution/population density alter the decision of the most advantageous planning strategy. conflicts of interest the authors declares that there is no conflict of interest. a novel long term telecommunication network planning framework dinâmica espaço tempo da demanda por internet: as cidades do futuro no cenário brasileiro de telecomunicações physical layer design of optical networks with practical considerations planning and operating flexible optical networks: algorithmic issues and tools cost minimization planning for greenfield passive optical networks forecasting internet demand using public data: a case study in brazil value-added services in next-generation sonet/sdh networks research challengesˇ in optical communications towards and beyond towards end-to-end integrated optical packet network: empirical analysis optimization techniques for incremental planning of multilayer elastic optical networks a techno-economic network planning tool for pon deployment including protection strategies impacts of planning on telecom access network developmental project: a case study of ptcl the internet of things as an accelerator of advancement of broadband networks: a case of thailand mip model for effi-˙ cient dimensioning of real-world ftth trees metaheuristics in telecommunication systems: network design, routing, and allocation problems cost estimation of a fixed network deployment over an urban territory towards achieving practical gpon ftth designs a note on two problems in connexion with graphs steiner tree problems profitable˙ areas in large-scale ftth network optimization gigabit-capable passive optical networks (gpon): physical media dependent (pmd) layer specification . : gigabit-capable passive optical networks (gpon): general characteristics applications of big data to smart cities faster return of investment in wdm networks when elastic transponders dynamically fit ageing of link margins visual networking index: forecast and trends domestic demand for bandwidth an approach to forecasting requirements for the period forecast analysis: carrier network infrastructure, worldwide forecasting financial and macroeconomic variables using data reduction methods: new empirical evidence ftth-based broadband access technologies: key parameters for cost optimized network planning long-range dependence-ten years of internet traffic modeling an information framework for creating a smart city through internet of things the internet of things (iot): applications, investments, and challenges for enterprises security and privacy in the internet of things anatomy of threats to the internet of things the next generation of passive optical networks: a review enabling technologies for effective deployment of internet of things (iot) systems internet of things technology and its applications in smart grid research on the mutual relations between isp and asp efficiency changes for the sustainable growth of the internet industry design theory for dynamic complexity in information infrastructures: the case of building internet service in the platform context: a review of the state of the art and future research," in collaborative value co-creation in the platform economy geo-intelligence based terrainaware mobile backhaul planning for rural india roll-out of reliable fiber to the cabinet: an interactive planning approach a profile of visual analytical toolkits for understanding the spatiotemporal evolution of broadband provision a mip-based´ approach to solve the prize-collecting local access network design problem evolutional algorithm based cascade long reach passive optical networks planning grass gis," in open source approaches in spatial data handling opensource tools for environmental modeling europe's g field of dreams: if we build it, will they come? digital policy towards g: scenario-based assessment of the future supply and demand for mobile telecommunications infrastructure technologies for convergence of fixed and mobile access: an operator's perspective how, when and where can fixed wireless access compete with ftth? g consumer potential: busting the myths around the value of g for consumers tracking the evolution and diversity in network usage of smartphones acessos -banda larga fixa brazilian institute of geography and statistics multivariate data analysis practical multivariate analysis discarding variables in a principal component analysis. i: artificial data world economic and financial surveys: world economic outlook database measuring the information society report brazilian institute of geography and statistics practical risk management for the cio coronavirus disease (covid- ): what we know? covid- pushes up internet use % and streaming more than %, first figures reveal global internet traffic spiking? add bandwidth with cisco's flexible consumption model world economic outlook, the great lockdown advanced dynamic migration planning toward ftth qgis geographic information system mapa digital das ruas de são sebastião do paraíso, mg brasil publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations key: cord- -maym iah authors: rogala, anna; szczepaniak, maria; michalak, natalia; andersson, gerhard title: internet-based self-help intervention aimed at increasing social self-efficacy among internal migrants in poland: study protocol for a randomized controlled trial date: - - journal: internet interv doi: . /j.invent. . sha: doc_id: cord_uid: maym iah background: migration is a challenging life transition that may be a source of various problems related to well-being and mental health. however, the psychological adaptation of migrants may be potentially facilitated by social self-efficacy—the beliefs in one's ability to initiate and maintain interpersonal relationships. previous research suggests that social self-efficacy is positively related to adjustment and negatively related to loneliness, depression, and psychological distress. research also confirms that self-efficacy beliefs can be effectively enhanced using internet-based interventions. these results served as a background for creating the new in town, a self-help internet-based intervention for internal migrants in poland that aims at increasing social self-efficacy. exercises in the intervention are based on the principles of cognitive behavioral therapy and relate to sources of self-efficacy beliefs: mastery experiences, vicarious experiences, verbal persuasions, and emotional and physiological states. users complete increasingly challenging tasks that encourage them to interact with their environment. the aim of this trial was to investigate the efficacy of the new in town intervention. methods: the efficacy of the new in town intervention will be tested in a two-arm randomized controlled trial with a waitlist control group. social self-efficacy will be the primary outcome. secondary outcomes will include loneliness, perceived social support, and satisfaction with life. additionally, we will measure user experience among participants allocated to the experimental group. we aim to recruit a total of n = participants aged at least years who have changed their place of residence in the last months and have an internet connection. participants will be assessed at baseline, -week post-test, and -week follow-up. discussion: the trial will provide insights into the efficacy of internet-based self-help interventions in increasing social self-efficacy. given that the intervention works, new in town could provide an easily accessible support option for internal migrants in poland. trial registration: the trial was registered with clinicaltrials.gov (identifier: nct ) on th september . internet-based self-help intervention aimed at increasing social self-efficacy among internal migrants in poland: study protocol for a randomized controlled trial the number of migrants worldwide has been growing rapidly over the past years [ ] . most of them migrate inside their own country [ ] . according to the estimations of the united nations development programme, in there were million internal migrants worldwide, compared to million international migrants [ ] . king and skeldon [ ] suggest that "the age of migration is therefore also an age of mass internal migration" (p. ). migration-both internal and international-is a source of potential specific stressors that could threaten migrants' well-being and mental health, e.g., communication difficulties, cultural differences, socioeconomic as well as employment status change [ ] . it is also a challenging life transition that may be related to problems in the social area. research has shown that students and visiting scholars from china in canada report more communication problems, difficulties in making friendships, loneliness, and lower subjective adaptation, compared to non-chinese canadian and chinese-canadian students [ ] . research also suggests that depression symptoms are more prevalent among internal migrants in china compared to the general population [ ] . what is more, internal migrant adolescents report lower levels of self-esteem, higher levels of depression, and have less social connections than local adolescents [ ] . however, psychological adaptation of migrants may be facilitated by personal resources, such as self-efficacy beliefs [ ] . social cognitive theory [ ] defines selfefficacy as ''belief in one's capabilities to organize and execute the course of action required to produce given attainments''(p. ). self-efficacy has been shown to be one of the important factors that help an individual to deal with stressful life transitions and adjust to a new situation [ ] . maciejewski, prigerson, and mazure [ ] found that self-efficacy mediates the j o u r n a l p r e -p r o o f relationship between dependent stressful life events and symptoms of depression among americans with a prior history of depression. moving to a new residence (internal migration) was the most often reported life event in the aforementioned study (n = ), along with serious financial problems (n = ) and life-threatening illness or injury (n = ) [ ] . jerusalem and mittag [ ] suggested that "within this stressful transitional adaptation to the new societal living conditions, self-efficacy can function as a personal resource protecting against deleterious experiences, negative emotions, and health impairment" (p. ). numerous studies have confirmed the positive effects of self-efficacy beliefs among migrants. research on young east german migrants suggests that individuals high in selfefficacy perceive changes in their lives more as challenges and less as threats, and report lower anxiety and better health than migrants low in self-efficacy [ ] . american expatriates in europe with high levels of self-efficacy express greater degrees of cultural adjustment than those with low levels [ ] . wang and sangalang [ ] found that among filipino immigrant employees in canada self-efficacy is positively related to work adjustment. a study among afghan and kurdish refugees in new zealand and australia showed that self-efficacy beliefs were associated with higher subjective well-being and lower psychological distress [ ] . selfefficacy was also found to be a predictor of mental health among malawian returning refugees [ ] . moreover, research among somali adolescents resettled in the united states suggests that self-efficacy is positively related to a sense of school belonging and negatively to ptsd and depression symptoms [ ] . not only self-efficacy beliefs but also social support from friends and significant others is positively related to psychological adjustment [ ] and subjective well-being of migrants [ ] . therefore, the constitution of a new social network is very important for this population. there is one type of self-efficacy beliefs that may especially help migrants to establish new social bonds-social self-efficacy [ ] . although self-efficacy can be defined j o u r n a l p r e -p r o o f [ ] more globally as a "traitlike general sense of confidence in one's own capabilities to master different types of environmental demands" (p. ), it is usually conceptualized as a context-specific construct [ ] [ ] . research suggests that using domain-specific self-efficacy measures allows predicting the outcomes more successfully [ ] . social self-efficacy beliefs can be defined as [ ] "confidence in one's ability to engage in the social interactional tasks necessary to initiate and maintain interpersonal relationships in social life and career activities" (p. ). research has shown that social self-efficacy is negatively correlated with attachment anxiety, depression, and loneliness among internal migrants in the usafreshman college students [ ] . in addition, social self-efficacy mediates the relationship between attachment anxiety and loneliness [ ] . social self-efficacy is also negatively related to acculturative stress, depression, and self-concealment among international college students in the united states [ ] . fan and mak [ ] found that first-generation migrant students in australia reported lower social self-efficacy than second-generation ones. they also reported more social difficulties and fewer shared interests with people in the host society, compared to second-generation migrant students [ ] . international students in canada also reported lower levels of social self-efficacy than canadian and secondgeneration migrant students. moreover, their low self-efficacy predicted low academic satisfaction and high psychological distress [ ] . similar results were obtained in australia-chinese migrant and overseas students reported lower self-efficacy than anglo-australian and second-generation migrant students from southern europe. social self-efficacy was also positively related to academic satisfaction and negatively to psychological distress. overall, the results of the aforementioned study suggest that social relationship issues impact the academic satisfaction of overseas students and migrants to a greater extent than the satisfaction of non-migrant students [ ] . therefore, it may be concluded that social selfefficacy is potentially beneficial for the psychological adjustment of migrants helping them j o u r n a l p r e -p r o o f establish new connections in the social environment. however, most evidence for the positive effects of social self-efficacy is derived from non-experimental research. our study addresses this gap. not only social self-efficacy, but also other factors, such as age, gender, educational level, and occupational background, socio-economic position, and availability of social support are related to adjustment and well-being of migrants [ ] [ ] . however, not all of these factors can be easily changed. on the other hand, self-efficacy beliefs in various domains, including initiating and maintaining interpersonal relationships, are malleable and can be intentionally enhanced [ ] . in line with bandura's social cognitive theory [ ] , selfefficacy beliefs can be altered by interpreting information from four different sources. first, perceived as the most powerful is mastery experience rooted in the interpretation of one's own performance in a specific domain as successful. second, the vicarious experience is related to observing the accomplishments of others which can be interpreted as a proof of specific goals attainability. encouraging feedback about one's own skills, abilities or performance given by significant others serve as a third source of self-efficacy beliefs, verbal and social persuasion. psychological and emotional states are the fourth source. individuals treat their emotional and physiological reactions like stress or anxiety while performing particular tasks as indicators of their capabilities [ ] . bandura's theoretical framework found support in empirical research results [ ] [ ] . therefore, we can distinguish not only theoretically but also empirically the aforementioned sources in the context of social selfefficacy beliefs [ ] . furthermore, previous research showed that an internet-based intervention drawn upon social cognitive theory could be an effective means of self-efficacy beliefs reinforcement [ ] . human services professionals exposed to indirect trauma who took part in an internet-based intervention displayed significantly greater improvements in self-j o u r n a l p r e -p r o o f efficacy beliefs related to managing secondary traumatic stress compared to an active control group. the internet-based sessions included modules focused on recalling past personal successes, the cognitive reappraisal of situations perceived as failures, an individually tailored plan of boosting self-efficacy beliefs and positive thoughts, and emotions reinforcement [ ] . internet-based interventions are a promising approach to reinforcing the self-efficacy of participants in a variety of specific contexts. teachers who accomplished an internet-based problem-solving training displayed significantly greater improvements in general and work-specific self-efficacy compared to waitlist control group members [ ] . an online positive psychology intervention to promote positive emotions, self-efficacy, and engagement at work consisting of modules targeting happiness, goal setting, and resource building, has shown its positive impact on participants' self-efficacy [ ] . the web-based intervention aimed at promoting healthy eating has proven its effectiveness in enhancing selfefficacy for total dairy intake among college students [ ] . moreover, the internet-based program aimed at reducing cannabis use has shown a positive effect on participants' userelated self-efficacy [ ] . to sum up, empirical evidence suggests that self-efficacy beliefs are malleable and can be effectively enhanced using internet-based interventions [ ] [ ][ ] [ ] .the aim of this study will be to evaluate the efficacy of a newly developed self-help internet-based intervention, new in town, in enhancing social self-efficacy when internet-based interventions by encompassing theory-driven content. we expect that an experimental intervention group will be superior to the control group regarding primary outcome-social self-efficacy-and secondary outcomes such as loneliness, perceived social support, and satisfaction with life. furthermore, we will evaluate the user experience of the intervention. the study is a two-arm randomized controlled trial in parallel design. participants will be randomized into two groups: a self-help internet-based intervention (new in town) and a waiting list control group. the trial was registered at clinicaltrials.gov (identifier: nct ) on st september . the study protocol and informed consent have been approved on th january by the ethics committee of the faculty of psychology at the swps university of social sciences and humanities in warsaw, poland (ref. no. / ). the new in town is an internet-based intervention created by the researchers at the swps university of social sciences and humanities in warsaw. it is available in the polish language version. the mode of internet recruitment and data collection enables potential participants from all over poland to apply for trial enrollment. months. targeted sample is prone to adjustment disorders development. in line with dsm-v, adjustment disorder became chronic when the symptoms persist more than six months from the moment when the trigger has occurred [ ] . as the new in town intervention is focused on soft skills development related to social self-efficacy it cannot be treated as professional help for those individuals who faced clinical symptoms of chronic adjustment disorder. therefore, we have chosen the criterion of months. when explaining participants this criterion we use internal migration definition of statistics poland-"change of place of residence (…) in the territory of poland, related to crossing the administrative border of a gmina (polish administrative unit), including-in case of urban-rural gminas-changes of the place of residence within a gmina, i.e. from rural to urban areas and vice versa" [ ] . the exclusion criterion is the lack of internet access. applicants who meet inclusion criteria will be automatically directed to electronic informed consent to enroll in the study. all study j o u r n a l p r e -p r o o f participants will be informed of their right to delist from the study without any consequences. after giving consent, participants will be asked to complete baseline questionnaires. the intervention aims at increasing social self-efficacy and consists of cognitive behavioral therapy (cbt)-based exercises related to sources of self-efficacy beliefs: ) mastery experiences, ) vicarious experiences, ) verbal persuasions, and ) emotional and physiological states [ ] . intervention contains modules (see table ). users complete increasingly challenging tasks that encourage them to interact with their environment. psychoeducation on social support (e.g. buddy support exercise). psychoeducation on emotional and physiological states as a source of self-efficacy beliefs (e.g. self-care exercise). planning initiating and maintaining interpersonal relationships (e.g. goal setting exercise). all data including informed consent will be stored in cloud-based software, surveymonkey. this platform uses an encrypted connection, which guarantees the security of data transmission. assessments will not require any personal data, except an email address. it is necessary in order to send an online questionnaire link at post-test and follow-up. only authorized persons will have access to the stored data. the sample size will be determined a priori using g*power [ ] . empirical evidence suggests a medium effect of internet-based interventions aimed at increasing self-efficacy beliefs [ ] . in our analysis we will include four outcomes. therefore, we will apply bonferroni's adjustment that will compensate for multiple comparisons and lower the probability level. [ ] . therefore, we plan to include participants at baseline ( participants per condition). after completion of the informed consent and baseline questionnaires, all participants will be randomized and assigned with a : ratio to one of two groups. information about group allocation will be provided within days to participants. blinding allocation is not possible due to the study design. participants assigned to the new in town group will be provided with login details as soon as possible. the waiting-list group will be given login instruction weeks after baseline. both groups receive access to the intervention for a period of weeks. in order to log in to the intervention website, all participants will get a unique password. after logging for the first time each participant is automatically asked to create own password. the participants can change the password to an account any time they wish. the post-test assessment is scheduled at weeks after baseline, follow-up assessment at weeks after baseline. all assessments (baseline, post-test, and follow-up) are self-reports and will be conducted online. research has confirmed the reliability and validity of the internet administration format of self-report psychological questionnaires [ ] . participants will be informed via email to complete assessments and will receive links to online questionnaires. two email reminders will be sent if the questionnaires stay incomplete for one week. eligible participants will be randomized by an independent researcher after the baseline measurement to either an experimental group or a waitlist control group ( : allocation ratio) using an online randomization program (www.randomizer.org). to ensure an equal number of participants in both study conditions we will use non-stratified block randomization with two participants per block. all outcomes will be measured in each assessment; baseline, post-test ( weeks after baseline), and follow-up ( weeks after baseline). exceptions are demographic data and user experience. demographic data will be collected at baseline, while user experience will only be assessed at the post-test among participants allocated to the experimental group. in a feasibility study, the reliability of measures was acceptable, with cronbach's alpha ranging from . to . . the average completion time of online questionnaires was minutes. general self-efficacy scale. social self-efficacy is the primary outcome studied. it will be measured with the general self-efficacy scale (gses) [ ] . the gses consists of two subscales for measuring ) generalized beliefs about self-efficacy ( test items) and ) beliefs about self-efficacy in establishing and maintaining relationships with others ( test items). the remaining test items ( ) are buffer theorems. the respondents give answers on a scale -point likert scale ( = strongly disagree, = strongly agree). in the trial, we will use one subscale, which measures beliefs about self-efficacy in establishing and maintaining relationships with others. social self-efficacy will be indicated by the total sum of items scores, e.g. "i have acquired my friends through my personal abilities at making friends". j o u r n a l p r e -p r o o f higher scores reflect a higher level of self-efficacy. the scale demonstrates good psychometric properties [ ] . de jong gierveld loneliness scale [ ] will be used to measure loneliness. the scale is composed of items; of them are formulated negatively and positively. each item is measured using on -point likert scale ( = definitely yes, = definitely no), e.g. "i often feel rejected". positive items should be reversed. the sum of scores is counted on separate subscales: emotional and social loneliness. the scale can be applied in different data collection modes; in face-to-face interviews as well as in online questionnaires. it is a wellvalidated measure with a bifactor structure [ ] . the berlin social support scale (bsss) [ ] will be used to assess social support. the bsss consists of subscales: perceived available support, need for support, support seeking, actually received support (recipient), provided support (provider), protective buffering scale. the scale contains items that are scored on a -point scale ( = strongly disagree, = strongly agree). before scoring items, negative ones need to be reversed, e.g. "i get along best without any outside help". scores can be counted either on a general scale or subscales. the measure was validated in several studies and showed good reliability [ ] [ ] . in the trial, only of subscales will be applied: perceived available support ( test items), need for support ( test items), and support seeking ( test items). internal consistency and test-retest reliability are highly satisfactory [ ] . swls is widely used to measure satisfaction understood as a component of well-being [ ] . demographic data questionnaire. this short questionnaire contains questions about gender, age, education, profession, tenure, the old and the new place of residence (rural area, city up to , residents, city up to , residents, city up to , residents, city with more than , residents) questionnaire. an overall impression of the intervention will be assessed by the user experience questionnaire (ueq) [ ] . it is a self-reported measure that contains subscales: attractiveness, perspicuity, efficiency, dependability, stimulation, novelty. participants respond to items (e.g. annoying/enjoyable) using a -point scale (- = the most negative answer, = neutral, + = the most positive answer). the ueq presents high internal consistency and good validity of scales [ ] . generalized estimating equations (gee) approach will be used to examine changes in primary and secondary outcomes over time and to assess differences between experimental and waitlist control group. this approach allows the correlated structure of data from repeated measures. what is more, gee is nonparametric and does not assume that the dependent variable is normally distributed [ ] [ ] [ ] . we will include group (experimental/waitlist control), measurement time and interaction between group and time as independent variables. dependent variables will include social self-efficacy, loneliness, perceived social support, and satisfaction with life. qic coefficients will be used to choose the best assumption for the working correlation matrix. we will follow intention-to-treat j o u r n a l p r e -p r o o f (itt) principles and use model-based imputation to handle missing values [ ] . adverse events will be assessed through the categorization of primary outcome change. we will recognize change that: ) is higher than - % from baseline as deterioration events, ) is between - % to % from baseline as non-response events, ) is between - % from baseline as minimal response events, and ) is higher than % from baseline as remission events. this classification scheme will enable us to compare outcomes, such as worsening with improving symptoms or non-response to intervention [ ] [ ]. to assess possible measurement bias we will compare the outcome change with and without the effect of missing data. all analyses will use intention-to-treat (itt) principles. we will additionally perform per-protocol analyses to examine the robustness of the effects. individual characteristics of participants can become confounding variables. random assignment to groups minimizes the potential for confounding. nevertheless, to assess possible assignment bias we will compare demographic and baseline characteristics between experimental and waitlist control group. the choice of sensitivity analyses will be, therefore, to some extent, data-driven [ ] . migrants go through a stressful process of transition and adaptation which may be related to problems in the area of well-being and mental health [ ] . however, research suggests that adaptation of migrants may be facilitated by social self-efficacy beliefs [ evidence also suggests that self-efficacy is malleable and can be effectively reinforced using internet-based interventions [ ] [ ][ ] [ ] . because the number of migrants is steadily increasing over the past years [ ] and more than % of polish citizens are internal migrants [ ] , the current challenge is to develop an effective and easily accessible intervention targeting this population. with our study, we hope to gain insight into the efficacy and acceptance of the new in town-self-help internet-based intervention aimed at increasing j o u r n a l p r e -p r o o f social self-efficacy among internal migrants in poland. the study is a randomized controlled trial with a waitlist control group. the primary outcome is social self-efficacy. secondary outcomes include loneliness, satisfaction with life and perceived social support. we will also explore user experiences. the study is not free from limitations. firstly, the online platform does not collect the data of the way participants will use the intervention (duration and frequency of being logged in to it). it is crucial to explore participants' patterns of accomplishing modules and identify the most crucial modules. each intervention module is activated one by one, but participants have always the possibility to return to previous ones. dates of module activation are not set in advance. therefore, the chance that some participants might complete the whole intervention at once still exists. on the other hand, open access to non-blocked modules allows participants to use the intervention at own pace and to adapt exercise completion to the daily life duties. at the post-test, we plan to collect user experience data that might be helpful in enhancing the intervention. secondly, the new in town and assessments were not provided on the same platform. this inconvenience may have an impact on drop-out rates. relatively high dropout rates are found in a number of internet-based interventions studies [ ] [ ] . therefore, we plan to adjust for missing data using the gee approach [ ] . the other concern is recruitment. participants will be recruited via social and traditional media campaigns. it will be a self-selected sample that might be highly motivated for internet-based interventions. this fact could affect the outcomes and become the main concern that should not be ignored without considering further research directions. for this reason, we plan in the near future to transfer the intervention to a more traditional medium and conduct it in the form of workshops or e-learning courses dedicated to freshman college students and assess its efficacy as well. moreover, the control group in our study is a waitlist control. research has shown that using this type of control condition may lead to bigger j o u r n a l p r e -p r o o f treatment effect sizes estimates compared to no treatment and psychological placebo [ ] . however, the aforementioned research was on cbt for depression and therefore generalizability of this evidence may be limited. additionally, one of the inclusion criteria is having changed the place of residence in the last months. therefore, participants in a waitlist control group can wait longer to access the intervention than they were between changing the place of residence and study enrolment. because of that study could yield more conservative results. future studies should focus on comparing the new in town with a different intervention as an active control condition. there will also be several strengths of this study. firstly, the content of the new in town intervention is theory-driven. exercises in the intervention are based on the principles of cognitive behavioral therapy and relate to four sources of self-efficacy beliefs. according to social cognitive theory, these beliefs can be enhanced by interpreting information from mastery experiences, vicarious experiences, verbal persuasions, and emotional and physiological states [ ] . secondly, we will use well-validated measures of primary and secondary outcomes. the strong aspect of the study design is also using two follow-up measurement points that enable us to track long-term effects. finally, the new in town is designed specifically for migrants, who are potentially at greater risk of difficulties in the social area [ ] [ ] . mind the gap!" integrating approaches to internal and international migration overcoming barriers: human mobility and development common mental health problems in immigrants and refugees: general approach in primary care psychological adaptation of chinese sojourners in canada jin li xm. depression and associated factors in internal migrant workers in china the effects of social connections on self-rated physical and mental health among internal migrant and local adolescents in self-efficacy in stressful life transitions self-efficacy: the exercise of control self-efficacy as a mediator between stressful life events and depressive symptoms differences based on history of prior depression the relationship between cross-cultural adjustment and the personality variables of self-efficacy and self-monitorin work adjustment and job satisfaction of filipino immigrant employees in canada learning to fit in : an exploratory study of general perceived self efficacy in learning to fit in : an exploratory study of general perceived self efficacy in selected refugee groups returning refugees : psychosocial problems and mediators of mental health among malawian returnees belonging and connection to school in resettlement : young refugees , school belonging , and psychosocial adjustment self-efficacy, perceived social support, and psychological adjustment in international undergraduate students in a public higher education institution in malaysia negative affect in east german migrants: longitudinal effects of unemployment and social support the selfefficacy scale: construction and validation guide for constructing self-efficacy scales self-efficacy specificity and burnout among information technology workers: an extension of the job demand-control model sources of social self-efficacy expectations : their measurement and relation to career development adult attachment , social self-efficacy , self-disclosure , loneliness , and subsequent depression for freshman college students : a longitudinal study acculturative stress, and depression in african, asian, and latin american international college students.pdf measuring social self-efficacy in a culturally diverse student population c the psychological adaptation of international and migrant students in canada the psychological adaptation of overseas and migrant students in australia migration, social mobility and common mental disorders: critical review of the literature and meta-analysis. ethn health migration and mental health self-efficacy: a theoretical analysis of its determinants and malleability sources of self-efficacy in school: critical review of the literature and future directions effects of internet-based self-efficacy intervention on secondary traumatic stress and secondary posttraumatic growth among health and human services professionals exposed to indirect trauma efficacy of an internet-based problem-solving training for teachers: results of a randomized controlled trial flourishing students: a longitudinal study on positive emotions, personal resources, and study engagement web-based nutrition education intervention improves self-efficacy and self-regulation related to increased dairy intake in college students a controlled trial of an internet-based intervention the national census of population and housing . . . who. who announces covid- outbreak a pandemic american psychiatric association. diagnostic and statistical manual of mental disorders. american psychiatric association terms used in official statistics power : a flexible statistical power analysis program for the social, behavioral, and biomedical sciences self-guided psychological treatment for depressive symptoms: a meta-analysis public library of science computers in human behavior internet administration of self-report measures commonly used in research on social anxiety disorder : a psychometric evaluation manual of the loneliness scale validating the polish adaptation of the -item de jong gierveld loneliness scale soziale unterstützung bei der krankheitsbewältigung: die berliner social support skalen (bsss) long-term effects of spousal support on coping with cancer after surgery the satisfaction with life scale is the shift in chronotype associated with an alteration in well-being ? the satisfaction with life scale and the emerging construct of life satisfaction further validation of the satisfaction with life scale: evidence for the cross-method convergence of well-being measures construction and evaluation of a user experience questionnaire longitudinal data analysis using generalized linear models to gee or not to gee: comparing population average and mixed models for estimating the associations between neighborhood risk factors and health an overview of longitudinal data analysis methods for neurological research dealing with missing outcome data in randomized trials and observational studies measurement of symptom change following web-based psychotherapy: statistical characteristics and analytical methods for measuring and interpreting change a comparison of the characteristics and treatment outcomes of migrant and australian-born users of a national digital mental health service a tutorial on sensitivity analyses in clinical trials: the what, why, when and how adherence in internet interventions for anxiety and depression dropout from internet-based treatment for psychological disorders waiting list may be a nocebo condition in psychotherapy trials: a contribution from network we thank aleksandra bis for contributing to the pilot study. key: cord- -rdlfm v authors: durak batıgün, ayşegül; Şenkal ertürk, İpek; gör, nağme; kömürcü akik, burcu title: the pathways from distress tolerance to cyberchondria: a multiple-group path model of young and middle adulthood samples date: - - journal: curr psychol doi: . /s - - -y sha: doc_id: cord_uid: rdlfm v the use of the internet for medical information elicited a recent term called “cyberchondria”. this study aimed to scrutinize the mediating effects of health anxiety (ha), anxiety symptoms (as), and internet addiction (ia) in the pathway from distress tolerance (dt) to cyberchondria by using a bootstrapping method. in order to examine the role of age in the proposed model, multiple-group path analysis was used to evaluate differences between young and middle adulthood groups. the final sample consisted of both young adult (n = ) and middle adult (n = ) internet users located in ankara, turkey. the results of path analyses for both age groups showed that dt is negatively associated with as and ha; as and ha are positively associated with ia; ia and ha are positively associated with cyberchondria. mediation analysis for both age groups demonstrated that as and ha significantly mediated the relationship between dt and ia; ia significantly mediated the relationships of as and ha with cyberchondria; ha significantly mediated the relationship between dt and cyberchondria. the results of the multiple-group path analysis showed that the relationship between ia and cyberchondria is significantly stronger in middle adulthood than young adulthood. the results of the current study are consistent with the relevant literature and provide crucial contribution especially by focusing on the role of age. the internet has been evaluated as the largest medical library in the world (morahan-martin ) as people reach web sites that include the symptoms of diseases, information on risks and benefits of drugs, treatment methods or the opinions of professionals, and they can also have contact with other internet users through chat rooms or social media (de choudhury et al. ; epstein ; greene and kesselheim ; starcevic and berle ) . hence, the internet cannot be underestimated when researchers conduct studies on health. the use of the internet for health elicited a phenomenon called as "cyberchondria" which is characterized by an excessive and escalated seek for health-related information on the web while feeling increased distress or anxiety, and a need to get reassurance (starcevic ; starcevic and berle ; white and horvitz ) . for people suffering from cyberchondria, this repeated medical information search for themselves or for others would lead them to gain considerable amount of information that is hard to interpret on one's own, and then would result in reading about serious health outcomes with an increased anxiety (aiken and kirwan ; lutwak ; starcevic and berle ) . cyberchondria is closely related to hypochondria (starcevic and berle ) in which the misinterpretation of bodily symptoms leads to the belief that one has a severe illness (apa ) . health anxiety (ha) (i.e., tendency to misinterpret uncertain health-related information as a proof for a severe physical illness; salkovskis et al. ) was evaluated as a distinct (mathes et al. ) and pivotal feature of cyberchondria (starcevic and berle ) . previous studies has shown that as the severity of ha increases, individuals could conduct more online health-related research (te poel et al. ) , and they could feel worse or more anxious after symptom checking on the web (doherty-torstrick et al. ; muse et al. ; singh and brown ) . as cyberchondria is linked to distress or anxiety related to health (starcevic and berle ) , distress tolerance (dt) should be considered when studying cyberchondria. dt was defined as "the capacity to experience and withstand negative psychological states" (simons and gaher , pp. ) . in the literature, the components of dt (i.e., intolerance of uncertainty, ambiguity, frustration, negative emotional states, and physical sensations) were found to be significantly correlated with ha . especially, intolerance of uncertainty was viewed as an outstanding risk factor for anxiety (carleton ) , and the positive correlation between intolerance to uncertainty and ha was found by kraemer et al. ( ) . furthermore, anxiety sensitivity and intolerance of uncertainty were evaluated as the probable risk factors for cyberchondria (fergus ; fergus ; norr et al. ; zangoulechi et al. ) . another precursor of cyberchondria could also be internet addiction (ia) (e.g., ivanova ) . internet addiction has been perceived as an impulse control disorder (young ). in this regard, studies have shown that as the time spent for online health-related search gets longer, functional impairment and anxiety during/after the search increases (doherty-torstrick et al. ) ; and individuals with higher ia also reported higher scores on ha and two aspects of cyberchondria (i.e., escalations and persistence of concern) (ivanova ). in another study, the participants who experienced increased ha after online medical information search reported significantly greater problematic internet use (piu) (fergus and dolan ) . consistently, the positive correlation between cyberchondria and piu was supported by fergus and spada ( ) . it has also been reported that individuals with higher ha use the internet for health-related searches for a longer time and more often (baumgartner and hartmann ; muse et al. ) . research regarding the role of demographic variables on online health information search is still relatively nascent. considering age, in the study of german version of the cyberchondria severity scale (css), the total score and age were not correlated, but "the mistrust of health professionals" subscale scores have increased as people get older (barke et al. ) . in parallel, no significant relationship was found between age and laypeople's evaluations on their activities and experiences while using the internet for their symptoms (white and horvitz ) . the current study proposes that a decrease in dt is associated with an increase in as and ha, which in turn leads to an increase in ia, and ia is associated with an increase in cyberchondria in young and middle adulthood. additionally, the current study suggests that a decrease in dt is associated with an increase in ha, which also in turn leads to an increase in cyberchondria in young and middle adulthood. furthermore, by considering the changes in health-related problems as people age, this study will contribute to the related literature by comparing the young and middle adulthood samples in terms of proposed model and thus exploring which associations are significantly stronger in the development of cyberchondria across samples. to sum, the aim of the current study is to investigate the hypothesized mediation model in which (a) anxiety symptoms (as) and ha would mediate the relationship between dt and ia, (b) ia would mediate the relationship of as and ha with cyberchondria, (c) ha would mediate the relationship between dt and cyberchondria in both samples, (d) suggested paths are different across the two samples. the study sample consisted of participants in their young and middle adulthood (n: , age: - , m = . , sd = . ). data was recruited through convenience sampling technique (dörnyei ) from ankara, turkey. participants of the young (n: , age: - , m = . , sd = . ) and the middle (n: , age: - , m = . , sd = . ) adulthood sample could use the internet and had no psychiatric diagnosis. since participants did not report their age, their data was not included in the analyses. after outliers were excluded, statistical analyses were run for participants. the required permission was obtained from ankara university ethics committee, and informed consent was obtained from all participants. descriptive statistics were demonstrated in table . demographic information form it is a form prepared by the authors to acquire information about the participants' sociodemographic characteristics (e.g., age, gender, education, any psychiatric diagnosis within the last six months). cyberchondria scale (cs) cs is a -item self-report and point likert-type scale developed in turkish (durak-batigun et al. ) to assess health-related information seeking behavior from the internet. the cs includes five subscales: "anxiety-increasing factors", "compulsion/hypochondria", "anxiety-reducing factors", "physician-patient interaction" and "non-functional internet use". sample items are "i think myself as a hypochondriac / valetudinarian", "internet has helped me to understand the terminology / explanations provided by the physician" and "the internet was useful in the initial diagnosis of the disease". cronbach alpha for subscales was . , . , . , . , . , respectively and . for the total score. higher scores indicate greater cyberchondria. internet addiction test (iat) iat is a -item with a -point likert type scale developed (young ) and adapted into turkish (bayraktar ) to assess ia. a sample item is "how often do you find that you stay online longer than you intended?". cronbach alpha was . . higher scores of the scale indicate higher ia. the scale has sufficient psychometric properties. health anxiety inventory (hai) hai is an -item with a point likert type scale developed (salkovskis et al. ) and adapted into turkish (karapıçak et al. ) to evaluate ha independent from the current health status. a sample item is participants are asked to select the one which best describes their feelings, over the past six months. cronbach alpha was . . aydemir et al. ( ) also performed a reliability and validity of the scale in individuals with somatoform disorder, panic disorder and major depressive disorder. in their study, cronbach alpha was . . hai was found valid and reliable. distress tolerance scale (dts) dts is a -item with a -point likert-type self-report scale developed (simons and gaher ) and adapted into turkish (sargın et al. ) . the scale consisted of three subscales called "tolerance", "regulation" and "self-efficacy" (sargın et al. ) . a sample item is "feeling distressed or upset is unbearable to me". cronbach alpha for the total score is . and for the subscales it ranged between . and . . higher scores indicate higher ability in dt. dts was a valid and reliable instrument. brief symptom inventory (bsi) bsi is a -item self-report scale with a -point likert type scale developed (derogatis ) and adapted into turkish (Şahin and durak ) with three different studies to evaluate a variety of psychological symptoms. these studies reported that the scale consisted of five subscales called "anxiety", "depression", "negative self", "somatization" and "hostility". sample items are "feelings of guilt", and "the idea that something is wrong with your mind". cronbach alpha for these subscales ranged between . and . . higher scores indicate higher frequency of symptoms of the individual. in the current study, anxiety subscale was used. all data analyses were performed using ibm spss statistics and amos software. independent samples t test was calculated to examine differences between young and middle adulthood groups in terms of all study variables (field ). pearson's correlation analysis was carried out to investigate relationships between the study variables. path analysis was used to test the hypothesized mediation model for both young and middle adulthood samples. the mediation effects were analyzed for both groups, using a bootstrapping method ( resamples) with % bias-corrected confidence intervals (bc ci) (shrout and bolger ) . it is considered that if zero is not included in the ci for the estimate of the indirect effect this effect is statistically significant at p < . level (shrout and bolger ) . furthermore, a multiple-group path analysis was employed to test whether structural weights were invariant across distinct groups (byrne ) . the results of independent samples t-test, there is significant difference between young and middle adulthood sample in terms of as (t = . , p < . ), ia (t = . , p < . ), and cyberchondria (t = . , p < . ). pearson's correlation coefficients for each sample are presented in table . a single-group path model was tested for young and middle adulthood samples separately. the fit indices for the model of each sample and the baseline model in which the two path models were combined into one model before and after the error association are presented in table . baseline model and the models of both samples had acceptable goodness of fit values after the error association (see fig. and fig. ) . firstly, for the young adulthood sample, dt is significantly associated with as (β = −. , p < . ) and ha (β = −. , p < . ). as (β = . , p < . ) and ha (β = . , p < . ) are significantly associated with ia. ia (β = . , p < . ) and ha (β = . , p < . ) are significantly associated with cyberchondria. secondly, in the middle adulthood sample, dt is significantly associated with as (β = −. , p < . ) and ha (β = −. , p < . ). as (β = . , p < . ) and ha (β = . , p < . ) are significantly associated with ia. ia (β = . , p < . ) and ha (β = . , p < . ) are significantly associated with cyberchondria. to test the mediating effects in the pathway from dt to cyberchondria, % bias-corrected confidence intervals (bc ci) were calculated for both groups, using a bootstrapping method with re-samples (shrout and bolger ) . estimates, standard errors, and ci's of mediation models are presented in table . multiple-group path analysis was used to test whether there were differences in the final model between young and middle adulthood groups. the significance of the differences in the structural weights was investigated across the two samples. the unconstrained model (no constraints group : young adulthood sample, group : middle adulthood sample; cfi: comparative fit index, gfi: goodness-of-fit index, agfi: adjusted goodness-of-fit index, rmsea: root mean square error of approximation, ecvi: expected cross validation index (byrne ) . thus, it was tested whether there would be a significant difference between young and middle adulthood samples on each significant association in the path model. the chi-square difference test was used to compare models (see table ). results showed that only the parameters coefficient in the path between ia and cyberchondria had a statistically significant difference between two samples [Δχ ( ) = . , p < . ]. for the middle adulthood group, the standardized regression weight was β = . , p < . , whereas for the young adulthood group, it was β = . , p < . . this suggests that the relationship between ia and cyberchondria seems to be significantly stronger in middle adulthood than young adulthood. the other path coefficients were found to be invariant across the samples. the aim of the current study was (a) to investigate the hypothesized mediation model in which as and ha would mediate the relationship between dt and ia in both young and middle adulthood samples (b) to investigate the hypothesized mediation model in which ia would mediate the relationship of as and ha with cyberchondria in both samples (c) to investigate the hypothesized mediation model in which ha would mediate the relationship between dt and cyberchondria in both samples (d) to test whether study variables were different across the two samples. firstly, the results of path analyses for both age groups indicated that dt is negatively associated with as and ha. dt was defined as the capacity to withstand repulsive situations such as perceived or actual negative affect or physical discomfort (leyro et al. ) . since as and ha could be considered as unpleasant feelings, our findings are coherent with the literature. in addition, our findings are consistent with the previous studies suggesting that dt is associated with several as (keough et al. ) and ha . the second association was that as and ha are positively associated with ia. consistently, previous studies showed that anxiety positively predicted ia in an adolescent sample (li et al. ) and anxiety was predicted by ia (akin and iskender ; tang et al. ) . it was also found that ha was correlated with some aspects of health-related internet use, frequency and proportion of health-related online ) noted that previous studies that conducted with older participants found a stronger association between ha and cyberchondria. the results of these studies are consistent with our findings demonstrating that ia and ha are positively associated with cyberchondria. our result is also consistent with the findings of the aforementioned studies (fergus and dolan ; fergus and spada ; ivanova ) . briefly, it was found that all variables in the model were significantly associated with each other in the expected direction. although there is no inference about the causality of the relationship, it can be said that there is a bidirectional relationship between using the internet to seek health-related information and ha. additionally, the interaction of ia with an anxious pattern seems to be a predisposition for cyberchondria in both age groups of our study. secondly, according to the mediation analyses, the results indicated that as and ha mediated the relationship between dt and ia for both samples. it can be stated that low level of dt will increase the ia when as and ha have mediator role in this relationship. a person with low level of dt may not experience ia, however, one becomes predisposed to ia when accompanied by high scores of as or ha. in addition, it can be inferred that individuals with higher as and ha will have increased cyberchondria when ia has a mediator role. taken together, these results support previous research findings showing that individuals with symptoms of as or ha use the internet to search for health-related information which may result in cyberchondria (e.g., mathes et al. ; mcmullan et al. ; muse et al. ; te poel et al. ; white and horvitz ). in addition, our results are consistent with the findings of the previous studies conducted on ia/ piu and cyberchondria (fergus and dolan ; fergus and spada ; ivanova ) . furthermore, ha mediated the relationship between dt and cyberchondria. this result may indicate that people with low dt could develop cyberchondria when they have also high ha. similarly, having difficulty in enduring uncertainty would result in increased ha and then health-related search on the web (fergus ; zangoulechi et al. ). as our result shows, searching for certainty on the internet could eventually lead to intense ha (fergus ; fergus and spada ) . thirdly, the results of multiple-group path analysis indicated that all direct associations between dt, as, ha, ia (except for the association between ia and cyberchondria), and cyberchondria are invariant across young and middle adulthood samples. only the association between ia and cyberchondria is stronger in the middle adulthood sample than young adulthood sample. this finding could suggest that individuals in middle adulthood use the internet for different purposes than young adults, resulting in cyberchondria. in our study, since most of the middle adulthood sample use the internet - h in a day ( . % of the middle adulthood and . % of the young adulthood), we only compared people who use the internet - h in a day. in addition, . % of the middle adulthood and . % of the young adulthood sample searched online health-related information an average of - times per month. hence, it was calculated that . % of the middle adulthood sample and . % of the young adulthood sample were searching online health-related information per month by using the internet - h in a day. it may be interpreted that most of the participants in their middle adulthood spare more time to health-related search on web during their relatively limited internet usage time period. the fact that the middle adulthood group is more likely to face healthrelated problems (e.g., hooker and kaus ) may also contribute to this result. considering their general health, they may be at risk in terms of cyberchondria. although the internet usage/addiction rates are high in young people (e.g., cao and su ; lam et al. ), in our study the association between ia and cyberchondria is stronger in the middle adulthood sample compared to young adulthood sample. it should not be concluded that ia is higher in the middle adulthood group. in terms of age, our findings provide additional evidence for the role of ia in the onset and persistence of cyberchondria. the studies examining the relationships between cyberchondria and demographic variables (especially age) are quite limited and have conflicting results (fox ; white and horvitz ) . therefore, our findings provide profound contribution to the literature about the role of age in cyberchondria. one of the strengths of our study is that it is the first study to test whether young and middle adulthood samples differ in the associations between ia and cyberchondria. thus, considering age could be useful in intervention studies. furthermore, given its association with distress and potential economic costs (fergus ) , identifying variables related to cyberchondria could be crucial for developing interventions and prevention strategies to address cyberchondria and factors that predispose to cyberchondria. according to starcevic ( ) , treatment approaches targeting cyberchondria should enable people to use the internet for healthrelated purposes without increasing health concerns and reducing the time spent on health-related information online. when we investigated anxiety as a precursor of cyberchondria, consistent with the results of our study, it is considered important that individuals' levels of as, ha, dt and ia should be also taken into account when developing cyberchondria interventions. in other words, our results spoke to the importance of taking into account psychological symptom levels, and age in any intervention efforts to reduce cyberchondria. in addition, other potential mediator or moderator variables that may influence the relationship between study variables and cyberchondria should be considered in interventions. especially during covid- pandemic, it is suggested that ha and cyberchondria serve as risk factors for coronavirus anxiety (asmundson and taylor ; jungmann and witthöft ). our findings that point to roles of as, ha, ia and dt in cyberchondria have significant importance for the context of pandemics. nevertheless, there are several limitations of our study. firstly, more research is required to provide additional evidence whether the relationship between ia and cyberchondria is variant in different age groups. at this point, it is important to replicate the results with different samples to confirm these findings. secondly, longitudinal study designs are needed to better understanding of the causal relationships. we defined structural paths among variables, but our study still does not provide any cause-effect relationships. as third, although the frequency of the internet usage to obtain healthrelated information has been inquired in our study, a comprehensive investigation of the internet usage purposes could help to have a deeper understanding about the differentiated paths in terms of age groups. finally, since the study is based on self-reports only, structured interviews or behavioral measures could be helpful to understand cyberchondria from different perspectives. authors' contributions all authors contributed to the study conception and design. material preparation, data collection and analysis were performed by İpek Şenkal ertürk, nağme gör, and burcu kömürcü akik. the first draft of the manuscript was written by ayşegül durak batıgün, İpek Şenkal ertürk, nağme gör, and burcu kömürcü akik and all authors commented on previous versions of the manuscript. all authors read and approved the final manuscript. all authors contributed equally to editing of the final manuscript. funding this study was supported by ankara university scientific research projects (project no: b ). data availability the datasets analysed during the current study are available from the corresponding author on reasonable request. conflicts of interest/competing interests the authors declare that they have no conflict of interest. ethics approval this study was performed in line with the principles of the declaration of helsinki. approval was granted by the ethics committee of ankara university. consent to participate informed consent was obtained from all individual participants included in the study. disclosure of potential conflicts of interest on behalf of all authors, the corresponding author states that there is no conflict of interest. research involving human participants and/or animals all procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee (ankara university ethics committee) and with the helsinki declaration and its later amendments or comparable ethical standards. informed consent informed consent was obtained from all individual participants included in the study. code availability not applicable. the psychology of cyberchondria and 'cyberchondria by proxy internet addiction and depression, anxiety and stress diagnostic criteria from dsm-iv-tr how health anxiety influences responses to viral outbreaks like covid- : what all decisionmakers, health authorities, and health care professionals need to know reliability and validity of the turkish version of the health anxiety inventory the cyberchondria severity scale (css): german validation and development of a short form the role of health anxiety in online health information search the role of internet usage in the development of adolescents structural equation modeling with amos. basic concepts, applications, and programming internet addiction among chinese adolescents: prevalence and psychological features the intolerance of uncertainty construct in the context of anxiety disorders: theoretical and practical perspectives seeking and sharing health information online: comparing search engines and social media bsi: administration, scoring, and procedures manual-ii cyberchondria: parsing health anxiety from online behavior research methods in applied linguistics: quantitative, qualitative, and mixed methodologies cyberchondria scale (cs): development, validity and reliability study cyberchondria and intolerance of uncertainty: examining when individuals experience health anxiety in response to internet searches for medical information the cyberchondria severity scale (css): an examination of structure and relations with health anxiety in a community sample anxiety sensitivity and intolerance of uncertainty as potential risk factors for cyberchondria: a replication and extension examining dimensions of each construct problematic internet use and internet searches for medical information: the role of health anxiety cyberchondria: examining relations with problematic internet use and metacognitive beliefs examining the specific facets of distress tolerance that are relevant to health anxiety discovering statistics using ibm spss statistics online health search pharmaceutical marketing and the new social media health-related possible selves in young and middle adulthood internet addiction and cyberchondria-their relationship with well-being health anxiety, cyberchondria, and coping in the current covid- pandemic: which factors are related to coronavirus anxiety panik bozukluğunda sağlık kaygısı envanteri (haftalık kısa form) türkçe geçerlilik ve güvenilirlik çalışması anxiety symptomatology: the association with distress tolerance and anxiety sensitivity intolerance of uncertainty as a mediator of the relationship between mindfulness and health anxiety factors associated with internet addiction among adolescents distress tolerance and psychopathological symptoms and disorders: a review of the empirical literature among adults relationship between anxiety, depression, sex, obesity, and internet addiction in chinese adolescents: a short-term longitudinal study commentary: thoughts on cyberchondria-a new challenge for healthcare providers cyberchondria: overlap with health anxiety and unique relations with impairment, quality of life, and service utilization the relationships between health anxiety, online health information seeking, and cyberchondria: systematic review and meta-analysis how internet users find, evaluate, and use online health information: a cross-cultural review cyberchondriasis: fact or fiction? a preliminary examination of the relationship between health anxiety and searching for health information on the internet anxiety sensitivity and intolerance of uncertainty as potential risk factors for cyberchondria kısa semptom envanteri: türk gençleri için uyarlanması the health anxiety inventory: development and validation of scales for the measurement of health anxiety and hypochondriasis cognitivebehavioral treatment for severe and persistent health anxiety (hypochondriasis) sıkıntıya dayanma Ölçeği: geçerlik ve güvenilirlik çalışması mediation in experimental and nonexperimental studies: new procedures and recommendations the distress tolerance scale: development and validation of a self-report measure health-related internet habits and health anxiety in university students from headache to tumour: an examination of health anxiety, health-related internet use and 'query escalation' cyberchondria: challenges of problematic online searches for health-related information cyberchondria: towards a better understanding of excessive health-related internet use prevalence of internet addiction and its association with stressful life events and psychological symptoms among adolescent internet users the curious case of cyberchondria: a longitudinal study on the reciprocal relationship between health anxiety and online health information seeking cyberchondria: studies of the escalation of medical concerns in web search internet addiction: the emergence of a new clinical disorder internet addiction: diagnosis and treatment considerations the role of anxiety sensitivity, intolerance of uncertainty, and obsessive-compulsive symptoms in the prediction of cyberchondria publisher's note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations key: cord- -sqyhzxty authors: song, yuanlin; jiang, jinjun; wang, xun; yang, dawei; bai, chunxue title: prospect and application of internet of things technology for prevention of saris date: - - journal: clinical ehealth doi: . /j.ceh. . . sha: doc_id: cord_uid: sqyhzxty abstract the internet of things (iot) includes three core procedures: full spectrum perception, reliable transmission, and intelligent processing. it can be applied for the prevention and control of sari (severe acute respiratory infection). by combining sensors, information technology, artificial intelligence, and available dynamic networking devices, iot could realize long-distance communication between hospitals, patients, and medical devices, which could ultimately improve current medical conditions. the new coronavirus, which is known as '' -ncov" (ncov), was discovered in cases of viral pneumonia in wuhan and was named by the world health organization on january , . the coronary virus is a large family of viruses known to cause influenza as well as other serious diseases such as middle east respiratory syndrome (mers) and severe acute respiratory syndrome (sars). the virus is a new strain of coronavirus that has not been found previously in humans (fig. ). [ ] [ ] [ ] since december , wuhan has continued to perform influenza and related disease surveillance and has reported a number of cases of viral pneumonia. the incidence rate has increased rapidly; it has exceeded , cases by feb , , resulting in an enormous burden on prevention and control efforts. the common respiratory symptoms of this infection are fever, cough, and breathing difficulties. in more severe cases, pneumonia, sars, kidney failure, and even death can occur. currently, there is no specific treatment against the new coronaviruses. supportive therapy is the primary treatment for this disease and for preventing complications. there have been a considerable number of medical teams supporting wuhan. in addition, numerous cases or virus carriers have been reported in places outside of wuhan. this has compounded the difficulties in preventing and treating ncov pneumonia, such as ( ) efficiently learning the updated interim guideline; suggest rapid learning interim and revised management guideline; ( ) better managing suspected cases; ( ) performing a consultation on difficult diagnose patients to improve the success rate of supportive treatment, and ( ) directing and ensuring quality control for clinical practice. the internet of things (iot) technology can be employed to achieve these goals. based on advanced information technology (it) and electronic medicine, the medical iot (miot) has experienced four major evolutions, including the development of wireless sensing technology, use of internet technology in clinical medicine, use of radio frequency identification (rfid), and artificial intelligence (ai) applications, to realize the iot medical model. wireless sensing technology was first used in fetal electrocardiograms, and the use of wireless sensing technology to monitor the breathing status of patients dates back to the s. pope et al. used a temperature sensor at the head end of a tracheotomy to send a patient's breathing parameters to a signal receiver in real time through an fm transmitter. the new technology was then used in the intensive care unit (icu) to monitor newborns and children, and assist health care workers identify apnea in a timely manner. wireless stethoscopes designed for anesthesiologists to listen to a patient's heart rate and breathing sounds during surgery have also been developed. as heart rate, breathing rate, electrocardiogram, body temperature, and blood pressure. with the development of technology, physiological parameter recorders based on wireless sensing technology are being introduced to the medical field, such as esophagus ph, pulse blood volume, co decompression, and wireless breathing frequency monitors during movement. moreover, with the development of microelectronics and integrated industrial technology, numerous sensing technologies have been developed, such as the ''smart t-shirt". , the concept of instant ''plug and play" modular installation has further accelerated the development of the ''sensor network," which can be easily applied for heart, lung function, and other joint monitoring. in , for the first time, gandsas and montgomery in the united states, using the wireless application protocol (wap), connected to a laptop using the internet to obtain telemedicine data including blood pressure, pulse, breathing frequency, moisture, oxygen saturation, and electrocardiograms, in ''anytime and anywhere." this technology was mainly implemented through the cell digital packet data (cdpd) of digital cell phones to achieve highspeed transmission of data, anytime and anyp lace (up to , bps). in the st century, with the combination of wireless sensors and internet technology, clinicians have been able to apply this convenient and more accurate measurement technology for the observation and prevention of disease. for example, in a highrisk population with congestive heart failure, the use of a wireless cardiopulmonary monitoring system can discover early signs of chen-shi breathing. in the iot industry chain, rfid is one of the core technologies for item identification. it applies a radio frequent spatial coupling signal (among magnetic or electromagnetic fields) to achieve wireless information transmission and for identification purposes. rfid was first used in the field of supply chain management; it is now also used in the medical industry. in , virtua medical installed rfid systems in four hospitals to track and manage thousands of patients and staff, as well as , assets. two years later, a clinical trial was conducted in new jersey, usa, using an implantable rfid microtag. physicians could view chronic disease information from an implanted verichip in the patients' upper arm. advances in wireless sensing and it have also laid the foundation for the use of the miot in respiratory disorders. in the field of the monitoring of respiratory physiology, the application and development of wireless sensing technology is particularly prominent. the types of sensors include: ( ) abdominal band-type: indirect determination of moisture volume ; ( ) inductance plethysmograph method ; ( ) video monitoring system ; and ( ) radio detector: measuring chest motion. these innovations have also led to the application of the miot to sleep medicine. in , the zhongshan hospital fudan university and shanghai respiratory research institute established the first sleep center medical model, the ''cloud plus terminal medical internet of things sleep laboratory," which interacts with five district and community hospitals on the platform. the application of cloud plus terminal medical internet of things facilitates the early diagnosis and management of sleep respiratory disorders. using this new healthcare model, patients can visit a community hospital, and yet receive high-quality healthcare service from the zhongshan hospital. this reduces the problem of overcrowding in large hospitals and assists community physicians to resolve the challenges inherent to the diagnosis and management of difficult cases. it realizes the prospect of ''the experts are linked by the cloud-the public enjoys modern medical treatment." for example, a successful wireless sensing pulmonary function meter has been developed at zhongshan hospital fudan university, by which patients can complete lung function monitoring from home. , physicians can view the patient's lung function changes and recommend treatment based on the data, simultaneously, online, and propose a three-level linkage of miot models. on october , , the chinese alliance against lung cancer established the ai committee for the early diagnosis and treatment of lung cancer. the objective of this new committee is to certify and further train well educated physicians to apply ai in the early diagnosis of lung cancer. potential value of internet of things-assisted prevention of sari as mentioned above, the incidence of saris is increasing rapidly, which imposes an enormous burden on prevention and treatment. there have been numerous cases of saris and instances of individuals carrying the virus exporting it to other places, which makes prevention and treatment more difficult. at present, there are a number of medical teams available to support wuhan. however, it may be possible to significantly improve the efforts of these medical teams by employing iot technology. the iot is an important part of the next generation of it. the core of the iot is a network that extends and expands based on the internet; the user side can be extended, thereby improving information exchange and communication among the ''things." miot is the science of intelligent identification, location, monitoring, and ai services for sari patients using rfid, global positioning systems, different sensing devices, information exchange, and communication according to an agreed arrangement. if we can delve deeper into its function, it can undoubtedly yield unexpected results. in the clinical work of preventing saris, rfid readers can be installed on robots, and uhf stickers can be read when a drug is uploaded to the robot to confirm drug distribution. this clinical application has gained relevant experience. , once a robot activates its rfid reader, it can obtain the appropriate information and quantity of all the drugs in a drawer, and through the precise matching of the drug information, a drug can be sent to the sari medical units correctly. the geisinger medical center in danville, pennsylvania, usa, has adopted embedded rfid robots to ensure that a drug is actually delivered to all units of use, with images transmitted instantly. it is hoped that wuhan's two mountain infectious disease hospitals can also be equipped with similar equipment to achieve the corresponding services. rfid technology can also be used to build a medical waste management system, which can easily monitor and track the entire sari medical waste process including generation, recycling, transportation, and treatment. there are already a number of medical teams to support wuhan. however, it is necessary to elucidate methods of faster learning and providing high-quality implementation interim guidance. in the management of lung nodules, we have gained extensive experience to prove that augmented reality (ar) technology based on the iot a process is a technology that integrates virtual information with the real world. it uses a variety of technical means including multimedia, d modeling, real-time tracking and registration, intelligent interaction, and sensing to simulate computer-generated text, images, d models, music, video, and other virtual information, and apply these to the real world. can play to the current level of different levels of the handicraft workshop-style diagnosis and treatment model upgraded to the national, and even international standards of modern flow ingesting works. , this experience could also be easily applied to assisting sari prevention and control. we could incorporate the key points of the interim guidance into the a process, adding quality control to ensure that they are performed with high quality. will have a concise, easy-to-promote and universal effect. auxiliary consultation to improve successful rescue rate sari is reported to have a high incidence with a staggering number of deaths. causes of death are related to sepsis, hypoxic respiratory failure, acute respiratory distress syndrome (ards), multiple organ failure, and complications and pregnancy. if these complications can be addressed in a timely manner, early support for treatment and monitoring, and immediate oxygenation therapy for patients with respiratory distress, hypoxemia, or shock can undoubtedly reduce mortality. however, the large number of doctors participating in the rescue work, many with no corresponding experience, require the guidance of experienced experts; however, experienced experts are extremely limited. the iot can assist in this task. with ar technology, experts and front-line physicians can perform and support consultations through ar, which can be easily and quickly completed without time, space, and location constraints, while reducing the probability of infection and avoiding the requirement of isolation after consultation. the management of suspected patients is a difficult problem. the reason is that at present, only patients with fever or those who have become ill can be identified. the numerous possible carriers of the virus are not isolated and become the next generation of infection. ask them stay in their own homes or stay in community to reduce personal transmission. the guidelines suggest selfprotection including maintaining basic hand and respiratory hygiene, maintaining safe eating habits, and avoiding close contact with anyone who exhibits symptoms of respiratory diseases (such as coughing and sneezing). however, the effect of the execution of this guideline is not known and there is no quality control. hence, there is an urgent requirement to apply the iot to assist in the management of this group of patients. compared with traditional medicine, miot can monitor the physiological and pathological state of the suspected patients throughout the process, reflecting the management advantages including providing personalized health solutions for different groups of people. through the use of wireless sensor equipment and modern it, patients can enjoy professional services, thereby ensuring the safety of suspected patients and also avoiding the infection of their families. during a period when the sari incidence rate is high, requiring the participation of the co-working hospitals and numerous doctors, it is challenging to ensure the quality of medical care. this is an unavoidable challenge. to complete the diagnosis and treatment of saris efficiently and accurately, while ensuring security, it is necessary to precisely coordinate the division of labor between the hospitals and physicians at all levels of the miot. clinical quality control based on the internet of things lies in the application of the inherent advantages of the three basic processes of the internet of things identified above, while utilizing its ten basic functions, real-time, transparent and efficient quality control of the medical internet of things classification. the original data on the patient side and the medical information processed by the cloud computer in the five-step method can be stored in a seamless link, real-time online form in the vast space of the medical center's cloud server. automatic grading mode with the default setting. in addition to the intelligent management of the classification model of disease risk, it is easy to process and classify large amounts of information. moreover, through high-speed information quality monitoring and a professional epidemiological data statistics model, we can effectively obtain the results of the most current quality control, efficiently monitor and provide an early warning of potential risks, and feedback to hospitals and physicians at all levels, in a timely fashion, the formation of the three-level linkage of corrective programs, ultimately achieving patient and social satisfaction. the current miot is not mature; there are numerous urgent problems to be resolved. however, it is expected to contribute considerably to the prevention and treatment of saris. this technology completely complies with the national control of diseases, the ''prevention and control of the forward, focus on sinking" requirements. the issues related to ''borrowing" the iot technology are: ( ) faster leaning and ensuring a high-quality implementation guideline; ( ) better managing suspected patients; ( ) effectively dealing with difficult patient consultations and improving the success rate of rescue; and ( ) commanding and ensuring effective quality control. moreover, the development and diffusion of the miot must address the following unresolved issues: ( ) improve interoperability: the adoption of open standards must ensure that products from different manufacturers can communicate with each other; ( ) ensure that there is no leakage; and ( ) enhance the distribution network: sensors must communicate with services such as the internet and wireless networks such that information can be relayed to remote health workers. to solve these problems, there is an urgent requirement to develop a modern medical model -miot and to use this to improve the current telemedicine, coupled with mobile phones and other popular user terminals. this should be more conducive to achieving: ''easy to prevent, easy to diagnose, easy to treat diseases, and to easily communicate with experienced doctors." community physicians can be advised in real-time by medical center experts through the cloud-plus terminal miot technology to achieve early diagnosis and optimal co-management of sari patients. shanghai strong main body key disciplines. clinical management of severe acute respiratory infection when novel coronavirus (ncov) infection is suspected laboratory testing for novel coronavirus ( -ncov) in suspected human cases cases of viral pneumonia were detected in wuhan in december, mostly by seafood city operators taped on heart rate and electrocardiogram telemetry transmitters a wireless respiration failure detection system development of a wireless stethoscope for auscultatory monitoring during anaesthesia possibilities of radio recording of biosignals using digital memories ambulatory esophageal ph monitoring using a wireless system noninvasive monitoring of arterial blood volume using photoplethysmography method with wireless bluetooth patient monitoring by transcutaneous pco -measurement and simultaneous control of the relative local perfusion (author's transl) wireless biomedical signal monitoring device on wheelchair using noncontact electro-mechanical film sensor . ghz wireless sensor network for smart electronic shirts wireless hydrotherapy smart-suit network for posture monitoring wireless vital sign telemetry to hand held computers a comparative study of a new wireless continuous cardiorespiratory monitor for the diagnosis and management of patients with congestive heart failure at home major issues include professional liability crisis, charity care, taxation of physician practices the security implications of verichip cloning wearable transducer to monitor respiration in a wireless way monitoring patient's setup and motion with a wireless cctv system in semi beam's eye view biomonitoring with wireless communications who's who at ats, chunxue bai focuses on innovation mini wireless sensoring spirometry, bringing the lung function test to the community hospitals quantitative ct analysis of pulmonary nodules for lung adenocarcinoma risk classification based on an exponential weighted grey scale angular density distribution feature automated classification of pulmonary nodules for lung adenocarcinomas risk evaluation: an effective ct analysis by clustering density distribution algorithm the authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. key: cord- -au sqzxw authors: dores, artemisa r.; geraldo, andreia; carvalho, irene p.; barbosa, fernando title: the use of new digital information and communication technologies in psychological counseling during the covid- pandemic date: - - journal: int j environ res public health doi: . /ijerph sha: doc_id: cord_uid: au sqzxw the use of digital information and communication technologies (icts) has enabled many professionals to continue to provide their services during the covid- pandemic. however, little is known about the adoption of icts by psychologists and the impact of such technologies on their practice. this study aimed to explore psychologists’ practices related with the use of icts before and during the covid- lockdown, to identify the main changes that the pandemic has brought and the impact that such changes have had on their practice with clients, and also identify the factors that potentially have affected such changes. the portuguese psychologists association announced the study, and psychologists responded to an online survey during the mandatory lockdown. the results showed that these professionals continued to provide their services due to having adopted icts. comparing with face-to-face interventions, psychologists recognized that additional precautions/knowledge were needed to use such technologies. despite the challenges identified, they described the experience with the use of icts as positive, meeting clients’ adherence, and yielding positive results. psychologists with the most years of professional experience maintained their services the most, but those with average experience showed the most favorable attitudes toward the use of technologies and web-based interventions. the coronavirus disease (covid- ) pandemic represents an unprecedented global challenge in our era, strongly affecting people's lives, namely, the exercise of various professional activities [ , ] . this unique circumstance, associated with the current availability of several digital tools, has contributed exponentially to the digital revolution that we have witnessed in recent years, with impact on the social, economic, and professional domains of life [ ] . in this scenario, e-health has emerged as one viable solution to allow the continuity of the provision of health services, particularly considering the public health measures that have been taken as a result of the national emergency state, which has limited people's access to in-person services [ ] . e-health is broadly defined as the provision of services related to health supported by a safe and cost-effective use of information and communications technologies (icts) [ ] . the path toward the progressive adoption of icts in the field of psychology had already begun before the covid- pandemic, albeit in varied degrees across different countries. if documentation guiding and/or regulating professional practice is already available in some countries, a legal or normative void still exists in others. this has implications regarding the availability of the psychological services offered, which are still quite scarce in some countries, as is the case in portugal. for example, before the covid- pandemic, few portuguese psychologists adopted guided and unguided psychological internet interventions ( . % and . %, respectively) [ ] , despite the already recognized advantages of this type of intervention [ ] [ ] [ ] [ ] [ ] . many of the guidelines for the online practice of psychology are the result of the work of several national and international associations and bodies (e.g., the american psychological association and european federation of psychologists' associations), in an effort to set forth a consensus about the practices that they aim to guide. in portugal, after a first approval of at-distance interventions by the ordem dos psicólogos portugueses (opp; transl.: portuguese psychologists association) in , the opp issued a document with the guidelines for these types of services very recently, precisely at the peak of the covid- pandemic [ , ] . in its first issue, this association claimed that psychological intervention should always be conducted within the same obligations and responsibilities (i.e., ethical principles and deontological and legal norms), regardless of the format of the intervention, as defined in the code of ethics. although the opp recognized the potential benefits of web-based interventions and use of itcs, it also launched warnings about the need for a better understanding of the effects of the different modalities of remote intervention (e.g., written, audio or audiovisual support) compared to face-to-face intervention. the opp additionally warned of the fact that the specificities of cyber space could elude the means of control available to psychologists, which could put privacy and confidentiality at risk [ ] . in its second document, entitled "opp guidelines for professional practice: provision of psychology services mediated by icts" [ ] , a set of recommendations for the adoption of these technological and digital resources were presented. in addition to the opp, other projects have been carried out in portugal (and elsewhere) with the aim to issue good practices to be adopted for the use of these digital and technological means in the area of health in general, and of psychology in particular. as an example, the european project therapy . -counseling and therapeutic interactions with digital natives, financed by the erasmus + program, sought to issue the appropriate integration of icts in counseling and therapy, especially for younger populations and refugees [ , ] . in several countries, an increasing number of studies have also sought to characterize the attitudes of psychologists toward the inclusion of icts in their professional practice and to gather evidence about the efficacy and effectiveness of psychological interventions mediated by icts (e.g., [ , [ ] [ ] [ ] [ ] [ ] [ ] ). regarding the psychologists' attitudes, the results of the studies are not consistent (e.g., [ , ] ). in portugal, a recent study assessing psychologists' attitudes revealed a slightly negative/neutral position regarding internet interventions and greater acceptability of blended treatment interventions when compared to standalone internet interventions. however, these attitudes seem to depend on different factors, such as knowledge and training [ ] . a study in other countries revealed that, among those who have employed any online means of practicing counseling and therapy, . % had a positive or very positive opinion about the use of these tools. in this study, e-mail was the most widely used online tool, and the computer and smart phone were the most frequently used equipment [ ] . different circumstances and factors seem to contribute to explain the different attitudes and the adoption (or not) of this type of resources [ ] . such factors include the therapist's theoretical model/orientation, geographic area, previous experience with the use of these resources, presence or absence of previous training, perception about the usefulness of these tools, ease with their use, and years of clinical practice [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] . regarding the effectiveness of these online means, in general, studies allow us to conclude that internet interventions may be efficacious and cost-effective [ , , , , , [ ] [ ] [ ] [ ] . they seemed to be at least as effective as face-to-face interventions in a large group of clients receiving treatment for psychological disorders, namely, for generalized anxiety and other types of anxiety disorders [ ] [ ] [ ] [ ] [ ] [ ] [ ] , depression [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] , and stress [ , ] . the recognized advantages of using icts do not refer merely to their power to compensate for the limitations of traditional interventions (e.g., travelling requirements for customers or therapists), nor to their use as complementary means. there are several advantages associated with implementing internet interventions [ ] [ ] [ ] [ ] , , ] . these include easy accessibility, high adaptability, flexibility and convenience, evolution at the client's pace, easy adherence and treatment monitoring, privacy and possibility of anonymity, cultural adaptability, low cost, and high potential for dissemination [ ] . conversely, the main challenges identified in the use of icts include ethical concerns (e.g., security, privacy, confidentiality, and an absence or lack of deontological orientation), clients' ict illiteracy, and negative attitudes toward internet interventions [ ] . others can be added, such as a lack of access to technological and digital tools by some users, technological problems in their use, and changes to the setting and regarding the therapeutic relationship [ , , , , [ ] [ ] [ ] [ ] . because different circumstances and factors seem to contribute to explain psychologists' attitudes and the adoption (or not) of this type of resources, it is important to identify potential changes in the use of icts during the covid- pandemic. given the measures of physical distance and isolation that most governments imposed with the state of emergency (e.g., [ ] ), how have psychologists dealt with the provision of counseling and therapy to their clients? the aims of this work were to (a) analyze how the attitudes of professionals in the field of psychology have changed in relation to the use of icts in the context of psychological monitoring during the lockdown; (b) assess whether the practice of psychological counseling and therapy includes greater use of icts during the lockdown period; (c) identify the factors that potentially have affected such changes; and (d) study the possible adoption of guidelines for at-distance psychological monitoring by psychologists who are using icts during the period of physical distance the sample in this study comprised psychologists who were registered in the opp. most were women ( , or . %). the mean age was . years old (sd = . ; min = , max = ), with ( . %) in the age group between and years, and ( . %) in the group between and years. the number of years of professional experience ranged between one or less and (m = . ; sd = . ). the sample reflects national representation (including continental portugal and islands), with of the portuguese districts participating in the study. the most represented districts were porto (n = ; . %), lisbon (n = ; . %), braga (n = ; . %), and coimbra (n = ; . %). these cities comprehend the national opp delegations with the most psychologists [ ] . most participants held master's degrees (n = ; . %). most were specialists in clinical psychology (n = ; . %), and ( . %) had one or more advanced specialties, including in psychotherapy (n = ; . %), among others. participants worked mainly with adults (n = ; . %), followed by adolescents (n = ; . %), children (n = ; . %), and the elderly (n = ; . %), in the areas of anxiety disorders (n = ; . %), mood disorders (n = ; . %), personality disorders (n = ; . %), neurocognitive disorders (n = ; . %), among others. for more information, see table . a questionnaire was developed for this study and a pilot test was conducted to check its comprehension level and adequacy for the current purposes. the questionnaire included questions divided into three sections: (i) socio-demographic data, with questions; (ii) experience before the covid- pandemic, with questions; and (iii) current experience using icts in psychology sessions (namely web-based interventions during the lockdown period), with questions. the questions were designed based on a previous questionnaire developed for the purpose of studying the use of icts in the provision of therapy and counseling [ ] , as well as on the advantages and challenges identified in the literature about the use of these technologies on psychological counseling. socio-demographic data included gender, age, education level, number of years of professional experience, district where the respondent was practicing, area of specialization, and targeted population in the respondent's practice (including development stages and most frequent disorders). the remaining two sections of the questionnaire focused on information about the use of icts in the respondents' clinical practice. questions included which tools and devices were used, clients' degree of satisfaction with the use of icts, advantages and difficulties identified by the professionals, impact on clients' adherence and on therapeutic results, among others (cf. appendix a). this study was approved by the local ethics committee (approval no.: ce a), and the questionnaire was made available at the web-based survey platform limesurvey [ ] . opp sent this anonymous online self-report questionnaire to its members via e-mail and published it on its webpage. the questionnaire was also sent to the authors' professional institutions via their mailing lists and was made available via professional social media, such as linkedin. this procedure ensured that all psychologists registered at opp (a mandatory requirement for practicing psychology in portugal) were invited to take part in this study. the e-mail containing the questionnaire included the study's description and aims, followed by an informed consent form. if the person agreed to participate, a link gave them access to the questionnaire. the data were collected during april and may , precisely at the peak of the covid- pandemic, and during the lockdown period. data were exported from limesurvey [ ] into ibm spss statistics commuter license [ ] for analysis. descriptive statistics (e.g., frequency distributions) were conducted for the sample characteristics (e.g., age, gender, educational and professional background)) and for the data pertaining both to the period before and during the covid- pandemic. data before the covid- pandemic included aspects such as the use of digital tools in professional practice, professional experience with this type of tool, and adherence of clients to therapeutic activities based on digital technologies. during the covid- lockdown, analyses considered the following aspects: maintenance of psychological support services, percentage of clients who have maintained the use of psychological counseling or therapy, frequency and duration of the therapeutic sessions, therapeutic adherence, therapeutic relationship, feedback from the clients, and results of the at-distance sessions. additionally, a thematic analysis for the open-type questions was performed on the open-ended questions. two independent raters (a.g. and i.p.c.) have proceeded to the classification of the categories in each answer. conflicts were solved by a third rater (a.r.d.). to explore the correlations between professional experience, age and the selected outcome variables, point-biserial and spearman correlations were performed. regarding the use of digital technologies for providing at-distance psychological counseling and therapy before the covid- pandemic, most (n = ; . %) had rarely or never used digital tools in clinical practice before the covid- pandemic ( table ) . the reasons pointed out by portuguese psychologists for never or rarely (n = ) having used digital technologies in psychological counseling and therapy were they considered it very impersonal (n = ; . %), inefficient (n = ; . %), ineffective (n = ; . %), not safe enough (n = ; . %) or ethical (n = ; . %), and for their lack of knowledge on how to apply these technologies in psychological counseling and therapy (n = ; . %). under "other", participants additionally shared that they did not feel the need to use these means of providing counseling and therapy before, that these means were not part of their institutions' policies, and that they preferred in-presence interventions. among the portuguese psychologists who used icts to provide at-distance psychological counseling and therapy (n = ), the most often used tools were video conferences (n = ; . %), telephone calls (n = ; . %), e-mails (n = ; . %), and social networks (n = ; . %). other tools used were audio conferences (n = ; . %), online intervention platforms (n = ; . %), smartphones and tablet apps (n = ; . %), online forums (n = ; . %), chats (n = ; . %), short-message services (n = ; . %), and virtual rooms (n = ; . %). concerning the technological devices used for providing at-distance psychological counseling and therapy, the most frequently used device was a computer (n = ; . %), followed by a telephone/smartphone (n = ; . %) and by tablets (n = ; . %) ( table ). among the psychologists that used digital technologies in psychological counseling and therapy previously to the covid- pandemic (n = ), none considered their experiences with these tools to be negative or very negative. nevertheless, (of these ) psychologists ( . %) considered their experiences to be neither negative nor positive. most of the respondents considered their experience with digital technologies to be either positive (n = ; . %) or very positive (n = ; . %). regarding the involvement of their patients in the therapeutic activities that were delivered through digital technologies, most of the psychologists rated it as moderate (n = ; . %), followed by high (n = ; . %) and low involvement (n = ; . %). only six psychologists ( . %) considered their patients' involvement in this type of activities very high, and seven psychologists ( . %) rated their patients' involvement as very low (table ). with regard to the advantages that portuguese psychologists considered might be experienced or already were experienced through the use of icts in psychological counseling and therapy, geographic flexibility was the most frequently selected advantage (n = ; . %), followed by scheduling flexibility (n = ; . %), the possibility of them reaching new groups of people in need of psychological counseling and therapy (n = ; . %), and by their easier access to some target-groups, such as persons with disability, refugees, among others (n = ; . %). they added the cost-benefit relationship (n = ; . %) and the possibility of obtaining new business areas (n = ; . %). twelve ( . %) of the respondents considered that they have never benefitted, or will never benefit, from any advantage through the use of icts in psychological counseling and therapy. on the contrary, through the analysis of the challenges that psychologists had already faced or were afraid of facing when using new icts in psychological counseling and therapy, the most frequently referred challenge was the difficulty in establishing and/or maintaining the therapeutic relationship (n = ; . %), followed by the lack of non-verbal communication (n = ; . %), reduced therapeutic adherence (n = ; . %), reduced client engagement in the sessions (n = ; . %), and reduced privacy (n = ; . %). other challenges referred by the psychologists were the interruption of the sessions (n = ; . %), ethical concerns (n = ; . %), possible misunderstandings (n = ; . %), difficulties in therapeutically approaching some problems/topics (n = ; . %), the substantial decrease or increase of the sessions' frequency (n = ; . %), lack of security (n = ; . %), establishment of boundaries (n = ; . %), and time management (n = ; . %). under the category "other", they also mentioned technical problems. four ( . %) respondents considered to never have faced or feared to face challenges in the future related to the use of digital technologies in their professional practice (table ) . during the covid- pandemic, and specifically during the lockdown period, only ( . %) of the psychologists discontinued the provision of psychological counseling and therapy to their clients (table ). these psychologists reported that the main reasons for interrupting their professional activities were the suspension of activities on the part of the institution where they worked, activity suspension on the part of their clients for various reasons (e.g., considering themselves to be info-excluded populations or presenting digital illiteracy, financial difficulties, or sensing that the clinical setting is lacking), psychologists' own personal unavailability during this period (e.g., due to new family responsibilities), and considering that digital means were inadequate for the target population (i.e., children) or clinical condition (e.g., attention deficits) that they were treating. all the other psychologists (n = , . %) were able to continue the sessions with their cases due to the use of icts. among the psychologists who continued to provide at-distance psychological services, ( . %) previously read guidelines and other documents that support their at-distance psychological practice. the documents that these psychologists consulted the most were materials made available by the opp (e.g., written material, videos, and webinars), guidelines from apa and from the international psychoanalytical association (ipa), scientific papers, and manuals about online therapeutic interventions (including the therapy . project). regarding additional cautionary procedures implemented by the psychologists for at-distance interventions, respondents referred the careful definition of rules and ethical limits, namely in terms of privacy, confidentiality, security, schedules, forms of contact, session duration and frequency, as well as how to proceed when unforeseen situations occur (e.g., technical failures such as problems with the internet connection, technology problems such as problems with the tools/equipment used, or interruptions). caution about the type of software and the type of technology used were mentioned, also related with non-exposure of personal life, as well as the conditions of the physical space and the psychologist's personal appearance/presentation, and personal well-being. psychologists also referred several precautions and procedures associated with the actual therapeutic process, namely regarding verbal and non-verbal communication (e.g., minimizing the occurrence of overlaps, interruptions, and misunderstandings), greater session structuration and directivity (which involved greater previous preparation for some of them), avoidance of emotional themes that require in-person support, which distance prevents, parent follow-up in sessions with children, and assessment of clients' level of comfort with the new format. when focusing on the technological tools used to provide at-distance psychological counseling and therapy, video conference was the most frequently used (n = ; . %), followed by phone calls (n = ; . %), social networks (n = ; . %), e-mail (n = ; . %), audio conference (n = , . %), smartphones and tablet apps (n = ; . %), online intervention platforms (n = ; . %), chats (n = ; . %), online forums (n = ; . %), and virtual rooms (n = ; . %). computers were the most frequently used technological device to provide psychological services during the covid- pandemic (n = ; . %), followed by telephones/smartphones (n = ; . %) and tablets (n = ; . %) ( table ). most of the respondents (n = ; . %) have continued to provide their services to most of their clients, i.e., twenty-seven ( . %) of the psychologists continued to provide counseling and therapy to between % and % of their clients, and psychologists ( . %) to between % and % of their clients. however, for psychologists ( . %), the number of clients decreased to a range of between % and %, and for another psychologists ( . %) that number diminished to a range of between % and %. these psychologists referred, as main reasons for these reductions, low client adherence, lack of client's necessary privacy, confidentiality and non-interruption conditions at home, the fact that clients preferred in-presence contacts (considering such forms of intervention to be more effective than, or feeling uncomfortable with, the new format), had financial difficulties, had difficulties managing the new routines (including caring for the children at home), and lacked the technological means for at-distance sessions. in some cases, the client's condition was stable, and the therapeutic process had come to an end, or it was requiring no immediate sessions. considering the frequency of the counseling and therapy sessions among the clients who continued to use this service, a small majority of the psychologists (n = ; . %) referred that their clients have maintained the previous frequency, but psychologists ( . %) reported a decrease in the number of sessions, and six ( . %) reported a significant decrease in that number. despite that, seven psychologists ( . %) reported an increase in the number of sessions during the covid- pandemic, and one ( . %) reported a significant increase. the same pattern was found for the duration of the counseling and therapy sessions, with ( . %) psychologists reporting a maintenance of the duration of each session, ( . %) reporting a decrease, and ( . %) reporting a significant decrease in the duration of the sessions. nevertheless, psychologists ( . %) stated that the duration of the counseling and therapy sessions increased during the covid- pandemic. regarding the results of the current therapeutic sessions, when compared to former in-presence sessions, most psychologists (n = ; . %) considered the results to be more of less the same, four ( . %) reported obtaining better results with at-distance sessions, and ( . %) considered that at-distance sessions have yielded worse results than in-presence sessions. similarly, from the points of views that clients shared with their psychologists, at-distance and in-person sessions were more or less the same (n = ; . %). six ( . %) of the respondents reported receiving better feedback (i.e., the clients preferred the online sessions), and one ( . %) received much better feedback. even so, ( . %) psychologists received worse feedback from their clients about this type of intervention. in what concerns therapeutic adherence to the ict sessions, the majority of psychologists considered it to be more or less the same during the covid- pandemic, comparing to the pre-covid sessions (n = ; . %), and psychologists ( . %) reported an improvement. nevertheless, other psychologists reported a decrease (n = ; . %) or a significant decrease (n = ; . %) in the therapeutic adherence of their clients ( table ). the vast majority of respondents considered that the therapeutic relationship between the psychologists and their clients was maintained (n = ; . %), with only three ( . %) psychologists reporting an improvement in those relationships. however, ( . %) psychologists considered that those relationships have worsened during this period. regarding the advantages that portuguese psychologists viewed as associated with their current use of new icts in psychological counseling and therapy, geographic flexibility was the most frequently selected (n = ; . %), followed by scheduling flexibility (n = ; . %) and the possibility of reaching new groups of persons in need of psychological counseling and therapy (n = ; . %). other advantages that they mentioned were the cost-benefit relationship (n = ; . %), the easier access of psychologists to some target groups, such as persons with disability and refugees, among others (n = ; . %), and the possibility of obtaining new business areas (n = ; . %). under the category "other", they further mentioned the possibility of providing secure interventions in the current covid- pandemic context, which ensured the possibility of maintaining the interventions. only two ( . %) respondents considered that at-distance psychological counseling and therapy does not offer any advantages. through the analysis of the challenges that psychologists currently face when they provide at-distance psychological counseling and therapy sessions, the most frequently referred difficulty was lack of non-verbal communication (n = ; . %), followed by reduced privacy (n = ; . %), the difficulty in establishing and/or maintaining the therapeutic relationship (n = ; . %), session interruptions (n = ; . %), reduced therapeutic adherence (n = ; . %), difficulties in approaching some problems/topics therapeutically (n = ; . %), ethical concerns (n = ; . %), and a reduction in patient engagement in the sessions (n = ; . %) ( table ). other challenges that counsellors referred to (under the category "other") were the establishment of boundaries (n = ; . %), the significant decrease or increase in session frequency (n = ; . %), the time management of the sessions (n = ; . %), possible misunderstandings (n = ; . %), and lack of security (n = ; . %). under this category, they additionally mentioned technology problems (e.g., equipment adjustments) and technical failures (e.g., internet connection). five respondents ( . %) did not report any difficulty or challenge in providing at-distance psychological counseling and therapy. in table , the psychologists' practices are presented pre-and post-covid- for easy comparison of the main results described previously. significant point-biserial correlation coefficient were positive between the aspect "continue to provide psychological counseling to customers regularly" and years of professional experience, r pb = . , p = . . significant correlations were negative between "frequency of psychological counseling sessions" and both years of professional experience, r s = − . , p < . , and age, r s = − . , p = . . the aspect, "duration of psychological counseling sessions" also displayed a significant negative correlation with age, r s = − . , p = . . thus, regarding respondents' professional experience, psychologists with more years of experience maintained their professional services during the covid- pandemic more than professionals with less years of experience. nevertheless, the frequency of the sessions decreased for the professionals who had more years of professional experience. regarding age, older psychologists reported a decrease in session frequency and duration. no significant results were found in any of the other variables analyzed. this study aimed to explore psychologists' attitudes and practices related with the use of icts before and during the covid- pandemic lockdown period, for identification of the main changes that have occurred in the provision of counseling and therapy. the impact of age and years of professional experience on the use of icts was also inspected. in this study, psychologists' use of icts in their professional activity before the covid- pandemic is in accordance with the literature, namely in terms of previous experience of their use, tools, devices, professionals' satisfaction with their use, advantages, and perceived challenges. these results reproduce those by mendes-santos ( ), in which only . % of the inquired portuguese psychologists admitted to having used digital technologies in their professional practice. the results of the present study also showed that most portuguese psychologists had never or rarely used digital technologies as a means of delivering psychological counseling and therapy before the covid- pandemic [ ] . there is also a high degree of similarity between the tools most frequently used in our study and the tools used, the resources that psychologists most recommend to their clients (e.g., telephone calls, e-mails, video conferences, social networks, and apps) [ ] , and the most used devices (e.g., computers and smart phones) reported in other research (e.g., [ ] ). additionally, according to previous studies, accessibility/geographic flexibility, convenience/(scheduling) flexibility, and cost-effectiveness/low cost are amongst the most recognized advantages of using icts (e.g., [ , ] ). regarding the disadvantages, the major challenges in this study, pertaining to ethical concerns and to the difficulty in establishing and/or maintaining the therapeutic relationship due to different reasons, were also identified in the literature [ , , , , , [ ] [ ] [ ] [ ] . the analysis of the reasons given for not using icts before the covid- pandemic revealed that lack of knowledge and training about the correct use of icts was particularly relevant, which might explain professionals' concerns about efficiency, effectiveness, and ethical issues. these same factors were associated with more negative attitudes toward the use of technologies in a previous study [ ] . training thus seems to be a necessary step in order to increase the use of icts. however, before the covid- pandemic, available training was scarce, perhaps also because the professionals themselves perceived the use of icts as unnecessary, as they indicated in this study. the emergence of the covid- pandemic brought about relevant changes in the use of icts. barriers to their use by both professionals and clients have been reduced, as the availability of information about their use in various formats has increased. the percentage of psychologists who have adopted icts in their practice during the lockdown period was very high in this study, and the vast majority of respondents were able to maintain their professional activity due to the inclusion of these means in their practice. this shows an enormous capacity of adaptation and flexibility, both on the psychologists' and clients' parts. this phenomenon was observed not only in portugal but in other countries too, such as the united states, where the provision of at-distance psychological services has been raised from . % to . % [ ] . additionally, other health services are also adopting the online modalities, namely in medicine [ , ] , with the professionals reporting positive perceptions regarding the telehealth services. however, the implementation of icts in such a short period of time leads to questions about the conditions under which they were implemented. our results showed that more than half of the psychologists have read about the use of icts, and some had already used these tools, even if not exclusively, in their professional practice before, which is consistent with a previous study [ ] . they additionally identified a number of materials that were informative of at-distance psychological practice. these materials also contained information on additional care that needs to be adopted in at-distance psychological monitoring sessions and that is different from the procedures that these professionals might have adopted in the use of digital technologies in the context of their social relationships. their concerns about web-based session pertained to a diversity of aspects considered to be critical in the e-health literature (e.g., the clear definition of rules and ethical limits) that can be different from face-to-face services [ ] . however, it is noteworthy that most professionals have not offered any input on any additional measures that they might have adopted (e.g., use end-to-end encrypted technology), nor has it been possible, within the scope of this work, to identify how psychologists were capable of responding effectively to the new requirements and specificities that they reported they have adopted. despite the great availability of webinars and the training and specialty documents made available during this period (e.g., [ ] ) by accredited entities, often free of charge, little is known about how such information transfer to the professional contexts. this study provides important information in that regard by confirming the pertinence and usefulness of such materials among the psychologists. in general, the tools and devices used before the lockdown period were the same that were used during the covid- pandemic, although there was an increase in the use of several of them during the covid- pandemic (e.g., video conferences, computers, and telephones/smartphones). the primary use of computers and smartphones in this study is in line with the findings from previous research, although psychologists in our study used mostly video conference and telephone calls, whereas e-mail was the most widely used tool in a previous study [ ] . psychologists who do use icts in their practice tended to report a positive or very positive experience regarding the use of these online technologies in counseling and therapy. similarly, a study focused on the attitudes of psychotherapists towards online therapy during the covid- pandemic have also found a positive attitude of the professionals with regard to this therapy modality [ ] . research has recognized several advantages associated with using icts in such contexts, and participants in our sample identified equivalent advantages [ ] [ ] [ ] [ ] , , ] . their reported advantages were the same before and during the lockdown period, although they added a new advantage during the lockdown period, i.e., the possibility to conduct secure interventions. the number of professionals who did not see any advantage in the use of icts after the pandemic decreased to a practically residual value. however, the implementation of these modalities was not without difficulties. both before and during the pandemic, psychologists identified a set of challenges in the adoption of dicts in professional practice. some decreased during the pandemic, possibly due to the professionals' increased experience (e.g., establishing/maintaining the therapeutic adherence). however, others were particularly worsened during the period of mandatory lockdown (e.g., loss of privacy and risk of interruption). although generally positive, the results regarding psychologists' experiences/results and clients' adherence to therapeutic activities based on digital technologies were variable, as reported in previous studies (e.g., [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] [ ] and [ , ] respectively). it is important in the future to understand which ingredients explain this variability, both in relation to individual and to disorder aspects, and to work with the group of clients who have failed to adhere to the new format. from the limitations identified in this study, some clients might benefit from better advertisement of this type of services and of the scientific evidence of its effects, together with the possibility of receiving a reduction in the price of the services provided. increasing clients' digital literacy will also contribute to their adherence to web-based interventions. this is already happening among the new generations, whose members are already known as digital natives [ , ] , but it is still difficult when working with specific populations, such as the elderly and people that live in rural areas, or when performing some psychological acts, such as psychological assessment and rehabilitation practices [ ] . additionally, other challenges might be more difficult to overcome, such as the sense that an adequate therapeutic setting is lacking. this aspect has been particularly exacerbated by the situation of mandatory lockdown that has brought together all who live in the same physical space, namely risking privacy. regarding the correlations between the professionals' characteristics (i.e., age and years of professional experience) and the use of icts in professional practice, the results showed that were the professionals with more professional experience who presented greater maintenance of psychological support services, but less frequently. however, because it was also the professionals with more professional experience who presented greater maintenance of psychological support services, the decrease in the frequency of the sessions reported by them might have been an intentional procedure to help their own and their clients' adaptation to the new format. the results failed to show a correlation between age and the use of icts, except for frequency and duration of the therapeutic sessions, which was significantly shorter for the oldest than for the youngest psychologists. the decrease in these aspects in the group of older psychologists from the pre-to the during-lockdown period could be possibly explained by the greater discomfort that these professionals experienced with the use of icts. the influence of the personal characteristics of the professionals in their attitude towards online psychological counseling was also reported in another study [ ] , with the psychotherapists who had previous experience with online psychotherapy, who thought that the patients they attend to had positive experiences in this modality, that adopted cognitive behavioral therapy in their practices (in comparison to psychodynamic therapists), and that lives in north america (in comparison to europe) exhibiting a more positive attitude towards online psychotherapy. this work has some limitations, namely the sample size and the use of a questionnaire that has not been previously validated to study the attitudes of psychologists toward icts (namely toward web-based interventions). however, the data collection took place during the period of absolutely unique and exceptional sanitary measures to prevent the pandemic dispersion of the coronavirus sars-cov- , which causes covid- . to understand the impact of these circumstances on professionals' practice is of the utmost relevance, despite the fact that these same circumstances have limited the time to conduct the data collection and the availability of participants in the study. still, the collaboration of the opp in this scenario, advertising the study and making the questionnaire available to all its members, contributed to ensure national representation of the participants. the process of adapting an existing instrument, namely obtaining the respective authorizations, would require an extended period of time that would risk missing this window of opportunity. instead, the questionnaire was adapted from an instrument that was previously used by the authors and that was tested in a pilot-study. future studies could focus on exploring the reasons that seem to be interfering either negatively or positively with clients' adherence to, and satisfaction with, ict sessions, so that personalized healthcare services can be provided and tailored to the specificities of each case. it is widely known that the covid- pandemic and associated restrictive measures of physical contact have significantly changed many professional activities. the current work has contributed to our understanding of that impact in the practice of psychology and psychotherapy, in close relation with the use of icts. awareness of these changes can guide future professional practice by allowing the replication of the best practices and experiences shared by the psychologists during the period of maximum lockdown. it can also help to overcome the main difficulties and limitations experienced, for example, by guiding future training in this area, stimulating the creation of guidelines for ict-based professional practice in different countries, and of measures to promote knowledge of and adherence to these guidelines that are becoming increasingly available. although the completion of the questionnaire is anonymous, some socio-demographic and professional data will be requested, as well as answers to closed-and open-ended questions. these questions focus on the use of digital technologies in psychological counseling. please read each question carefully. it is important that your answers are sincere. if you accept to participate in this study, please click on "next" to proceed to the informed consent. all data will be collected and processed in an anonymous and confidential way, so you will never be asked for your name, professional number, or other personal data that could identify you. the data will be used for research purposes and will never be analysed at the individual level. your participation is completely voluntary, and your contribution is very important for us, to better understand the role of digital technologies in confinement situations like the one we are going through now. although you can quit answering the survey at any time without any consequences, we really appreciate your collaboration. if you want to clarify any question or if you need more information, please contact: andreia geraldo (andreiageraldo.psic@gmail.com). we appreciate your cooperation. andreia geraldo, researcher artemisa r dores, responsible for the project if you intend to participate in this study, select the option below to proceed to the survey. i declare that i have read the information above, have become aware of the research aims, and agree to participate in this study. socio-demographic and professional data . in case you consider that you have adopted some additional precautions in at-distance psychological counselling, when compared to those you adopt in relation to the use of digital technologies in the context of your social relationships, please indicate them. . what are the advantages that you identify in at-distance psychological counselling? geographic flexibility, both for professionals and clients (i.e., they can interact from any location) . what are the difficulties that are reported by your clients in relation to maintaining at-distance psychological counselling? thank you for your collaboration. an early view of the economic impact of the pandemic in charts. international monetary fund imfblog how the economy will look after the coronavirus pandemic: the pandemic will change the economic and financial order forever the long and short of the digital revolution decreto do presidente da república n.º -a/ portuguese psychologists' attitudes towards internet interventions: an exploratory cross-sectional study internet-delivered psychological treatments advantages and limitations of internet-based interventions for common mental disorders internet and psychological treatment internet interventions for depression: new developments effectiveness, mediators, and effect predictors of internet interventions for chronic cancer-related fatigue: the design and an analysis plan of a -armed randomized controlled trial parecer /ceopp/ da ordem dos psicólogos portugueses (opp) linhas de orientação para a prática profissional opp: prestação de serviços de psicologia mediados por tecnologias da informação e da comunicação (tic). [guidelines for professional practice opp: provision of psychology services mediated by information and communication technologies (ict) chegar mais perto dos que estão longe: therapy . [getting closer to those who are far: therapy . online interventions: counteracting the exclusion of young people in counselling and therapy internet-based and other computerized psychological treatments for adult depression: a meta-analysis a comprehensive review and a meta-analysis of the effectiveness of internet-based psychotherapeutic interventions online counseling and therapy formental health problems: a systematic review of individual synchronous interventions using chat therapist-supported internet cognitive behavioural therapy for anxiety disorders in adults computer-based psychological treatments for depression: a systematic review and meta-analysis online counseling: a narrative and critical review of the. literature attitudes towards digital treatment for depression: a european stakeholder survey the advantages and disadvantages of online and blended therapy: survey study amongst licensed psychotherapists in austria psychologist treatment recommendations for internetbased therapeutic interventions in the eyes of the beholder: exploring psychologists' attitudes towards and use of e-therapy in norway attitudes toward psychological telehealth: current and future clinical psychologists' opinions of internet-based interventions attitudes towards the use of computerized cognitive behavior therapy (ccbt) with children and adolescents: a survey among swedish mental health professionals attitudes towards internet interventions among psychotherapists and individuals with mild to moderate depression symptoms efeito da prática profissional baseada na evidência nas atitudes dos psicólogos face à intervenção psicológica eletrónica use or not to use: clinicians' perceptions of telemental health australian health professionals' attitudes toward and frequency of use of internet supported psychological interventions do web-based interventions improve well-being in type diabetes? a systematic review and meta-analysis the effect of technology-based interventions on pain, depression, and quality of life in patients with cancer: a systematic review of randomized controlled trials a systematic review of web-based interventions for patient empowerment and physical activity in chronic diseases: relevance for cancer survivors e-therapy in the treatment and prevention of eating disorders: a systematic review and meta-analysis internet-based psychodynamic versus cognitive behavioral guided self-help for generalized anxiety disorder: a randomized controlled trial effectiveness of internet-based cognitive behaviour therapy for panic disorder in routine psychiatric care predicting long-term outcome of internet-delivered cognitive behavior therapy for social anxiety disorder using fmri and support vector machine learning treatment of social phobia: randomised trial of internet-delivered cognitive-behavioural therapy with telephone support guided internet-de¬livered cognitive behavior therapy for generalized anxiety disorder: a randomized controlled trial individuallytailored, internet-based treatment for anxiety disorders: a randomized controlled trial randomized controlled trial of internet-delivered cognitive behavioral therapy for posttraumatic stress disorder randomised controlled non-inferiority trial with -year follow-up of internet-delivered versus face-to-face group cognitive behavioural therapy for depression internet-based treatment of depression: a randomized controlled trial comparing guided with unguided self-help internet-based behavioral activation and acceptance-based treatment for depression: a randomized controlled trial behavioral activation-based guided self-help treatment administered through a smartphone application: study protocol for a randomized controlled trial a randomized controlled trial evaluating a manualized telecoaching protocol for improving adherence to a web-based intervention for the treatment of depression netmums: a phase ii randomized controlled trial of a guided internet behavioural activation treatment for postpartum depression clinician-assisted internet-based treatment is effective for depression: randomized controlled trial standardized web-based cognitive behavioural therapy of mild to moderate depression: a randomized controlled trial with a long-term follow-up internet administered guided self-help versus individualized e-mail therapy: a randomized trial of two versions of cbt for major depression internet-based versus face-to-face cognitive-behavioral intervention for depression: a randomized controlled non-inferiority trial stress management for middle managers via an acceptance and com¬mitment-based smartphone application: a randomized controlled trial randomized controlled trial of internet-based stress management perceived drivers and barriers to the adoption of emental health by psychologists: the construction of the levels of adoption of emental health model survey of psychologists' telebehavioral health practices: technology use, ethical issues, and training needs south african psychologists' use of the internet in their practices online counseling: an exploratory survey of italian psychologists' attitudes towards new ways of interaction censo dos membros efectivos da opp [association numbers. census of effective members of the opp limesurvey: an open source survey tool the covid- telepsychology revolution: a national study of pandemic-based changes in u.s. mental health delivery the effect of the covid- pandemic on physicians' use and perception of telehealth: the case of lebanon suddenly becoming a "virtual doctor": experiences of psychiatrists transitioning to telemedicine during the covid- pandemic psychotherapists' attitudes toward online therapy during the covid- pandemic key: cord- - s r authors: lu, t.; reis, b. y. title: internet search patterns reveal clinical course of disease progression for covid- and predict pandemic spread in countries date: - - journal: nan doi: . / . . . sha: doc_id: cord_uid: s r effective public health response to covid- relies on accurate and timely surveillance of local pandemic spread, as well as rapid characterization of the clinical course of disease in affected individuals. de novo diagnostic testing methods developed for emergent pandemics are subject to significant development delays and capacity limitations. there is a critical need for complementary surveillance approaches that can function at population-scale to inform public health decisions in real-time. internet search patterns provide a number of important advantages relative to laboratory testing. we conducted a detailed global study of internet search patterns related to covid- symptoms in multiple languages across countries on six continents. we found that internet search patterns reveal a robust temporal pattern of disease progression for covid- : initial symptoms of fever, dry cough, sore throat and chills are followed by shortness of breath an average of . days [ % ci . - . ] after symptom onset, matching the precise clinical course reported in the medical literature. furthermore, we found that increases in covid- -symptom-related searches predict increases in reported covid- cases and deaths . days [ % ci . - . ] and . days [ % ci . - . ] in advance, respectively. this is the first study to show that internet search patterns can be used to reveal the detailed clinical course of a disease. these data can be used to track and predict the local spread of covid- before widespread laboratory testing becomes available in each country, helping to guide the current public health response. accurate real-time surveillance of local disease spread is essential for effective pandemic response, informing key public health measures such as social distancing and closures, as well as the allocation of scarce healthcare resources such as ventilators and hospital beds. during the current covid- pandemic, surveillance has primarily relied on direct testing of individuals using a variety of de novo laboratory testing methods developed specifically for the emergent pandemic. while laboratory testing serves as an important gauge of epidemic spread, it suffers from a number of important limitations: . during the early stages of a novel pandemic, laboratory tests specific to the pandemic do not yet exist and therefore must be developed de novo , leading to significant delays before test-based surveillance can begin. . even once a test has been developed, scaling of manufacturing, distribution and test processing capacity takes a significant amount of time, resulting in limited testing capacity in many countries. it is therefore difficult to achieve population-scale surveillance with laboratory testing in the crucial early stages of an emergent pandemic. . even after tests are widely available, delays in test administration and processing make laboratory testing a lagging indicator relative to disease onset. . laboratory testing often requires individuals to leave home and congregate at testing centers, increasing exposure both for those being tested and for the health workers administering the tests. . laboratory tests are typically invasive, involving blood draws or deep nasal swabs. . laboratory testing can be expensive, especially when testing large numbers of people. alternative surveillance approaches are needed to overcome these limitations and serve as a complement to laboratory testing, especially during the critical early stages of a pandemic. aggregated de-identified internet search patterns have been used to track a wide range of health phenomena, including influenza , mers , measles , abortion and immunization compliance, and are a potential alternative source of information for surveilling pandemic spread. previous uses of these data have yielded valuable lessons in their appropriate use, with an emphasis on avoiding non-specific search terms and avoiding complex models that tend towards overfitting. when harnessed appropriately, internet search patterns possess a number of powerful advantages relative to laboratory testing: these studies relied on a limited number of patients, and were published weeks and months after the initial spread of the pandemic. it would thus be beneficial to pandemic tracking, case diagnosis and treatment if these clinical patterns could be ascertained earlier and at population-scale. we conducted a detailed global study across countries on six continents to determine whether internet search patterns can provide reliable real-time indicators of local covid- spread, and whether these data can reveal the clinical progression of covid- . we selected countries from diverse regions of the world (table ), in which robust search data were available for the search terms of interest. we obtained data on reported covid- cases and deaths for each of these countries from a publicly available dataset maintained by the center for systems science and engineering at johns hopkins university. we collected daily relative search volume (rsv) data on a per-country basis for the period of january , through april , , from google trends using the pytrends api . google has limited availability in china, we accessed search trend data from the search engine weibo. we accessed data for the following common symptoms of covid- : "fever", "cough", "dry cough", "chills", "sore throat", "runny nose" and "shortness of breath", as well as the general terms "coronavirus", "coronavirus symptoms" and "coronavirus test". we also looked at other less common symptoms such as loss of smell and loss of taste, but the search data on those terms were too sparse for many countries. since daily search data are inherently noisy, all search data were smoothed with a -day moving average. and confirmed that robust data were available for these translated search terms on google trends. we conducted temporal correlation studies to characterize the relationships between search data and reported covid- cases and deaths. for each country and search term, we calculated the pearson correlation coefficient between the time series of search volumes for that search term and the time series of covid- cases. we then shifted the search term data by a variable lag, and identified the lag that yielded the highest correlation. we computed the mean and standard deviation of these optimal lags for each search term across all countries. we then repeated these analyses substituting reported covid- deaths for reported covid- cases. we then investigated whether internet search data can be used to characterize the clinical course of covid- symptoms over time. we used "coronavirus symptoms" as the index search term, since it peaked first among all other search terms. for each country, the date of peak search volume for the index search term was defined as the index date, and the dates for all other search data for that country were defined in relation to this date (day , day , day , etc.). with the data from all countries thus aligned, cross-country ensemble average curves were calculated for each search term. (e.g. day values for "fever" searches in each of the countries were averaged together to calculate the day value of the ensemble average "fever" curve. the same was done for day , etc.) the ensemble average curves for all search terms were then overlaid on one plot, providing a search-data-based view of the clinical course of illness. as above, all results were calculated using search data smoothed with a -day moving average. for . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . visualization purposes a -day moving average provided slightly clearer plots; plots based on -day moving average are included in the supplementary materials. we begin by presenting examples from individual countries. figure shows search volumes for the terms "fever" and "dry cough", alongside reported covid- cases and deaths for china, iran, italy, united states and india. even though outbreaks occurred at different times in each country, the temporal relationships between the search terms and reported covid- cases and deaths are similar across countries. figure a shows the lags between search volumes for the term "fever" and covid- -related deaths across countries, along with a histogram showing the distribution of these lags. figure b shows the same information for the term "dry cough". the average lags between searches and reported cases are shorter than those between searches and reported deaths since cases are usually diagnosed and reported before deaths. the cross-country variability of the average lags between searches and reported cases is greater (higher standard deviation and larger confidence intervals) than that for reported deaths, likely since case reporting is more sensitive to local testing capacity and rates which vary greatly between countries. we found that the general term "coronavirus" has greater variability in its lags to reported cases and deaths (largest standard deviation and confidence intervals), compared to other symptom-specific terms, as would be expected for a non-specific search term. we examined the progression of symptom-related search terms over time in order to characterize the clinical course of covid- . figure shows examples from individual . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . we calculated ensemble average curves for each search term across countries. figure a shows the ensemble average search volumes for "fever", "cough", "dry cough" and "shortness of breath", indexed by searches for "coronavirus symptoms", alongside reported covid- cases and deaths. figure b shows this same analysis for additional search terms "sore throat", "runny nose", "chills", and "coronavirus test", also indexed by searches for "coronavirus the clinical progression that emerges from these data presents the following picture: as the pandemic begins to take hold in a country, people search for "coronavirus symptoms" and "coronavirus test", followed by initial symptoms "fever", "cough", "runny nose", "sore throat" and "chills", followed by searches for "shortness of breath" about days after the search for initial symptoms. the medical literature reports the clinical progression of covid- in terms of the number of days between initial symptom onset and shortness of breath (dyspnea). therefore, we examined a range of possible search-term-based definitions for initial symptom onset, based on various combinations of the earliest-peaking search terms "fever", "cough", "coronavirus symptoms" and "coronavirus test". table shows the lags between these different definitions for initial symptom onset and searches for "shortness of breath". the average lag between the searches for "fever" and "shortness of breath" was . days [ % ci . - . ]. for "cough" it was . days [ % ci . - . ]. these lags, and lags deriving from other symptom onset definitions, are all around days, precisely matching the clinical course of the disease reported in the literature. this is the first study to show that internet search patterns can be used to reveal the detailed clinical progression of a disease. during emergent pandemics, this level of detail can provide . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . public health officials not only with predictions of local pandemic spread, but also with a detailed understanding of the stage of illness and the manifestations of the disease in the local environment. recent studies have indicated that the spread and severity of the disease can be affected by local conditions, and search volume data can be a valuable complementary tool in studying potential local variations in disease presentation. we found that increases in symptom-related searches such as "fever" predict increases in reported the use of search data is subject to a number of important limitations. internet infrastructure and digital access levels differ across countries and communities. some countries currently lack sufficient search volume to support robust search-based tracking, though in the long-term, digital access rates are increasing worldwide. search data may be subject to socio-economic, geographic, or other biases inherent in the local digital divide. , , even though selecting specific search terms increases the signal to noise ratio, changes in search volumes for symptom-related terms such as "fever" could result not only from increases in covid- cases, but also general curiosity about the pandemic, other diseases (e.g. influenza, lassa fever ), news coverage, or other factors. due to privacy considerations, search data are provided as aggregated relative search volumes rather than raw counts, so while it is possible to predict an increase in cases, it may be difficult to infer the magnitude of such an increase. while search data have important limitations, laboratory testing is subject to a wide range of limitations listed above, in addition to reliability issues which cause certain types of laboratory tests to need to be repeated in order to increase reliability. recent studies have examined internet search data related to the current covid- pandemic. some studies looked only at the search term "coronavirus", , , , while others looked at additional search terms such as "handwashing", "face masks", "quarantine", "hand disinfection", "sars", "mers", "antiseptic", and "sanitizer". none of these studies . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint examined specific symptom-related search terms. in our analysis, we found that general non-symptom-specific search terms such as "coronavirus" have a greater variability in their relation to reported cases and deaths. this is likely due to the fact that individuals seeking general information on the pandemic may search for "coronavirus" even if they are not experiencing specific symptoms themselves at that time. a few studies have examined internet search data using terms related to covid- symptoms, but these covered only one country or a small number of countries: one study looked at "coronavirus" and "pneumonia" only in china and found that internet searches were correlated with daily incidences of covid- . another looked at "covid", "covid pneumonia", and "covid heart" only in the us, and found that these terms were correlated with covid- daily incidence and deaths. yet another study looked at "loss of smell" in countries, and found these to be correlated with covid- cases . another study developed a model incorporating a large selection of search terms, including some symptom-related terms, in countries; however this latter study focused on optimizing the performance of a predictive model, rather than studying the detailed temporal relationships between patterns of symptoms . the present study is the first to conduct a detailed investigation of multiple covid- symptom-related search terms across a large number of countries. it is also the first to conduct a detailed analysis of the temporal relationships between different symptom-related searches in order to characterize the clinical course of illness for covid- . future work includes training a robust predictive model with various machine learning techniques to provide more granular predictions for increases in covid- cases and deaths. such models can also take into account additional sources of data such as news reports, testing capacity, public health mitigation measures, climatological and air quality variables, among others. the ability of search data to not only predict future increases in cases, but also reveal the clinical course of symptoms in emergent pandemics is significant. given the numerous limitations of laboratory testing, search data are a valuable complementary source for population-scale tracking of pandemics in real time. these data can be used today to guide the public health response to the covid- pandemic in countries worldwide. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint all data used in this study are publicly available through the sources referenced in the methods section. b.y.r. and t.l conceived of the study, supervised its conduct, and oversaw data collection. b.y.r. and t.l. designed the study, conducted statistical analysis, drafted the manuscript and formulated the implications of the results. all authors contributed substantially to the revision of the manuscript and approved the final manuscript as submitted. table . countries included in the study, categorized by region. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint table . average lag in days from search-based symptom onset to searches for "shortness of breath" across countries. different search-term-based definitions for symptom onset were examined by looking at different combinations of early-peaking search terms. . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. the copyright holder for this preprint this version posted may , . . figure . search volumes (purple) for the terms "fever" (left) and "dry cough" (right), alongside reported covid- cases (cyan) and deaths (orange) for china, iran, italy, us and india. even though outbreaks occur at different times in different countries, the relationships between the search terms and reported covid- cases and deaths are similar across countries. to highlight the temporal relationships between the curves, the magnitude of each curve was independently normalized to fit the vertical dimensions of the plot. . (a) search volumes for the terms "fever" , "cough", "dry cough", "shortness of breath" (black), indexed by searches for "coronavirus symptoms", shown alongside covid- cases (dashed line cyan) and deaths (dashed line orange). (b) search volumes for the terms "sore throat", "runny nose", "chills", and "coronavirus test", algonside "shortness of breath" (black), indexed by searches for "coronavirus symptoms", shown alongside covid- cases (dashed line cyan) and deaths (dashed line orange). . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. (which was not certified by peer review) the copyright holder for this preprint this version posted may , . . cc-by-nc-nd . international license it is made available under a is the author/funder, who has granted medrxiv a license to display the preprint in perpetuity. (which was not certified by peer review) the copyright holder for this preprint this version posted may , . . https://doi.org/ . / . . . doi: medrxiv preprint pandemic preparedness and response--lessons from the h n influenza of pandemic (h n ) surveillance for severe illness and response covid- mass testing facilities could end the epidemic rapidly the burning building fast, portable tests come online to curb coronavirus pandemic reappraising the utility of google flu trends high correlation of middle east respiratory syndrome spread with google search and twitter trends in digital epidemiology: assessment of measles infection through google trends mechanism in italy measuring the impact of health policies using internet search patterns: the case of abortion internet activity as a proxy for vaccination compliance google trends: opportunities and limitations in health and health policy research big data. the parable of google flu: traps in big data analysis assessing ebola-related web search behaviour: insights and implications from an analytical study of google trends-based query volumes clinical characteristics of hospitalized patients with novel coronavirus-infected pneumonia in wuhan clinical and epidemiological characteristics of coronavirus disease clinical features of patients infected with novel coronavirus in wuhan interim clinical guidance for management of patients with confirmed coronavirus disease (covid- ) cssegisanddata/covid- high temperature and high humidity reduce the transmission of covid- effects of temperature and humidity on the spread of covid- : a systematic review exposure to air pollution and covid- mortality in the united states profiles of a health information-seeking population and the current digital divide: cross-sectional analysis of the - california health interview survey beyond access: barriers to internet health information seeking among the urban poor mass media reportage of lassa fever in nigeria: a viewpoint development and clinical application of a rapid igm-igg combined antibody test for sars-cov- infection diagnosis covid- : testing times the second worldwide wave of interest in coronavirus since the covid- outbreaks in south korea, italy and iran: a google trends study tracking covid- in europe: infodemiology approach association of the covid- pandemic with internet search volumes: a google trendstm analysis online information search during covid- applications of google search trends for risk communication in infectious disease management: a case study of covid- outbreak in taiwan google searches for the keywords of 'wash hands' predict the speed of national spread of covid- outbreak among countries perception of emergent epidemic of covid- / sars cov- on the polish internet infodemiological study using google trends on coronavirus epidemic in wuhan predicting covid- incidence through analysis of google trends data in iran: data mining and deep learning pilot study retrospective analysis of the possibility of predicting the covid- outbreak from internet searches and social media data, china trends and prediction in daily incidence and deaths of covid- in the united states: a search-interest based model the use of google trends to investigate the loss of smell related searches during covid- outbreak tracking covid- using online search symptom onset search-term-based definition days to fever fever the authors declare no competing interests. key: cord- -d mgma authors: duan, li; shao, xiaojun; wang, yuan; huang, yinglin; miao, junxiao; yang, xueping; zhu, gang title: an investigation of mental health status of children and adolescents in china during the outbreak of covid- date: - - journal: j affect disord doi: . /j.jad. . . sha: doc_id: cord_uid: d mgma objective: : the sudden outbreak of coronavirus disease (covid- ) has had a dramatic effect on the mental health of the public. in the present study, we demonstrated the psychological effects on children and adolescents associated with the epidemic . methods: : by using convenience sampling method, questionnaires, such as spence child anxiety scale, child depression inventory and coping style scale, were distributed to participating children and adolescents online. results: : the anxiety levels of children and adolescents were ( . ± . ) and ( . ± . ), respectively. . % respondents were suffering from depressive symptoms. seven significant factors associated with increased levels of anxiety, including female, resident in urban regions, emotion-focused coping style. nine factors associated with increased levels of depression, such as smartphone addiction (or . , % ci . – . ), internet addiction (or . , % ci . – . ), and resident in hubei province (or . , % ci . – . ). two additional factors associated with decreased levels of depressive symptoms: hours spend on internet per day before the epidemic (or . , % ci . – . ) and tendency to apply problem-focused coping style (or . , % ci . – . ). conclusion: : our findings indicate that the covid- outbreak has had a significant psychosocial impact on children and adolescents. findings of current levels of anxiety and depression not only highlight the need to address emotional distress for children and adolescents during the epidemic but also provide researchers with scientific fundamentals to formulate targeted interventions based on the significant influencing factors. since the outbreak of the covid- at the end of , a series of effective epidemic preventive and control strategies have been developed and conducted by the chinese government to curb the spread of the virus. the general public, especially the high-risk group (e.g., elderly people with chronic underlying diseases, children, and others with lower immunologic function), are suggested to stay indoors in quarantine at home, which is recognized as the best way to eliminate chances of being infected. however, due to the long period of separation from the outside world, the impact of multi-channel negative information, excessive fears of being infected, and even the shame and guilt for the infection, many people across china, particularly residents at the epicenter of the outbreak in wuhan, are experiencing fear, loneliness, panic, anxiety and depression amid the coronavirus outbreak (duan and zhu, ; xiang et al., ) . these negative physical and psychological costs have also been reported in previous studies, such as suicide, substantial anger and sleep disorders (barbisch et al., ; rubin and wessely, ; wang et al., ) . for children and adolescents, due to the lower incidence of infection and mortality than adults, professionals were less focusing on the unique clinical features of covid- and mental health status in children (ma et al., ) . furthermore, evidence has shown that children and adolescents who experienced disasters might suffer from greater stress and trauma because of the lack of development of proper emotional reactions and coping techniques (lazarus pj, ; roussos et al., ) . when emotional support and spiritual guidance from family members, teachers, as well as other significant peers and adults in their lives could not be met due to home confinement, deferred back-to-school and lifestyle changes, it adversely affected the mental health of children and adolescents (sprang and silman, ) . additionally, according to the latest data from the china's national health of commission, the overall situation in the chinese mainland is somewhat coming to a plateau ( national health commission of china, ) . the number of new confirmed cases has dropped dramatically, but authorities are warning that strict prevention measures should stay in place to prepare for a potential comeback of the virus. therefore, in order to effectively cooperate with the government's epidemic prevention work, more than million children and adolescents were confined at home and informed to postpone the start of the spring semester until further notice (wang et al., ) . meanwhile, online studying courses which was delivered by tv broadcasts and the internet have been opened gradually to students to guarantee their needs of learning, according to a guideline released by the china's ministry of education (ministry of education of china, ) . however, due to lack of monitoring and evaluation of students participating in online courses, the inefficiency of the experience of these courses, unstable network signals, and often no networks in remote rural areas, many parents, students, and teachers have complained of the poor effectiveness of online learning. these effects can be compacted by adverse consequences of reduced vision, as well as unconscious smartphone/internet addiction, and may further contribute to mental distress in children and adolescents. here, we assessed the current status of mental health issues among children and adolescents affected by the epidemic and analyzed its influencing factors to provide scientific guidance to psychological professionals and the government in formulating targeted policies. this study was designed as a cross-sectional online questionnaire survey that was administered during the spread of covid- in china. we developed the online questionnaires on the official website of "questionnaire star," which is recognized as a professional online questionnaire survey, evaluation, and voting platform. a set of self-rating questionnaires were distributed to children and adolescents who ranged from grade one in primary school to grade three in high school (aged from to years) in mainland china by employing convenient sampling method. with the help of directors in education bureau, we distribute e-questionnaires to teachers and told them the purpose, significance and announcements of this survey in details. then, they assigned it to school-students and their guardians. before completing questionnaires, all participants and their guardians were debriefed on the study purpose and contents. once consented, participants began filling out the set of questionnaires online. we also included our email addresses and phone number to the first page of questionnaires so that participants could consult and interact with us at any time. the set of questionnaires involved in this study were mainly composed of seven sections. -sociodemographic characteristics were self-designed and included questions regarding sex, age, place of residence, number of siblings, region, family status, education level, and number of electronics owned. -covid-related questions were developed by our research team and assessed the current situation of the respondents' family members involved in anti-epidemic work, the degree of concerns about the epidemic, the implementation of the precaution and control measures, the impact of the epidemic on their course of study and graduation, as well as the electronic products/internet use patterns before and during the epidemic. -the chinese version of spence child anxiety scale (scas) is a -item self-report likert's scale that measures anxiety in children and adolescents (zhao et al., ) . spence (spence, ) first developed it by using community samples, and it has since been used widely in other countries (essau et al., ) . this study evaluated the anxiety status of participants affected by the epidemic on a -factor scale, including items such as separation anxiety, physical injury fear, social phobia, panic disorder, obsessive disorder, and generalized anxiety. -the child depression inventory (cdi) is a -item self-report measure designed to evaluate the severity of depressive symptoms in children and adolescents (kovacs & beck, ) and has demonstrated satisfactory levels of reliability and validity in the chinese population (wang et al., ) . in general, the total score of cdi is or higher can be identified as possessing clinical depressive symptoms, while scores of - indicate subclinical depression, and scores of or lower are considered normal (stewart and sun, ) . in this study, we classified respondents with clinical depressive symptoms with scores or higher. -the short version of smartphone addiction scale (sv-sas) is a -item self-rated developed by kwon et al. (kwon et al., and is recognized as a valid screening tool for the prevalence of smartphone addiction. it uses cut-off points by sex, where and classify "excessive smartphone uses in male and female users," respectively. -the internet addiction scale (ias) consists of items derived from the dsm-iv-tr (cooper, ) [diagnostic and statistical manual of mental disorders ( th edition, text revised)] in order to identify diagnostic criteria of pathological gambling and the degree of preoccupation and compulsiveness to go online. according to young (young, ) , three types of internet-user groups were identified as internet addicts (scores of or higher), possible internet addicts (scores - ), and non-addict ( or lower). in this study, we defined participants with internet addiction with scores or higher. -the coping style scale (css) was developed by chen et al. in based on the theory of social interaction and self-regulation and measures middle school students' competence in coping with stress (chen sl et al., ; folkman et al., ). this -item self-rate scale can be divided into two subscales of problem-focused coping and emotion-focused coping. the former consists of three dimensions, including solving problems, seeking social support, and positive rationalizations, while the later contains four dimensions, including endurance, avoidance, expressing emotions, and fantasy/denial. all data were analyzed using spss (version . , spss inc., chicago). apart from descriptive statistics and frequency analysis of demographic characteristics and covid-related information, t test and analysis of variance (anova) were used to analyze the difference in levels of respondents' anxiety. moreover, multiple linear regression and bivariate logistic regression analyses were used to assess the association between outcome variables (the reported level of anxiety and clinical depressive symptom) and potential predictors (e.g. concerns related to the epidemic, smartphone/internet addiction, and coping style) while adjusting for other identified explanatory variables. in the process of running models, the forward stepwise selection algorithm was used, and variables in the model were screened based on significance levels of the wald inclusion test statistic being less than . . moreover, pearson's correlation analysis was performed to analyze the associations among scas, cdi, smartphone/internet addiction, and coping style, and the statistical significance level was set at . . the sample size of our survey recruited online included , chinese students. as table shows, the study participants comprised , males ( . %) and , females ( . %) collected from provinces in mainland china. however, considering that hubei is the hardest-hit place of this epidemic, we classified the sample sources by residential areas, namely ( . %) in hubei province and , ( . %) from other regions. respondents were mainly from nuclear families ( . %), extended families ( . %), and single-parent families ( . %). the sample comprised children aged - years ( . %) and adolescents aged - years ( . %) and who were in primary school ( . %), secondary school ( . %), or high school ( . %). among them, ( . %) and ( . %) who failed the high school/college entrance examination and had to return to secondary/high school again for resuming their study, respectively. additionally, there were ( . %) respondents who already had electronics before -year-old. furthermore, an in-depth analysis of , respondents who possessed electronic devices showed that the average age at which they started owning electronic devices was . ± . years, and the highest rate of smartphone penetration was . % ( / ), followed by computer . % ( / , ) and tablets . % ( / , ). after analyzing the information related to the epidemic (table ) , ( . %) respondents reported that their family members were involved in the anti-epidemic work, and ( . %) of them were health care professionals (e.g., frontline doctors and nurses). a total of ( . %) of the participants stated that their family members, friends, or other acquaintances had been infected with coronavirus. in all, . % of respondents clearly reported concerns about this epidemic, and almost the same proportion ( . %) reported that they have strictly implemented protective measures (e.g., wear mask, wash hands, or avoid public places and crowds). moreover, , ( . %) and , ( . %) participants reported that the epidemic has affected their learning and graduation, respectively. average time per day spent on internet during the epidemic was also much longer than before, whereas study, as the main reason for internet use, became more significant during the epidemic. after conducting surveys on the willingness to engage in medicine in the future, we found that more people ( . %) were determined to work in healthcare than those ( . %) who were a little uncertain after the outbreak. among all respondents, ( . %) and ( . %) of them had scores above the threshold for clinical depressive symptoms ( or higher) and internet addiction ( or higher). as for sas, due to the gender differences in cutoff values for smartphone addiction, it was found that the prevalence rates of smartphone addiction in male and female respondents were . % and . %, respectively (table ) . additionally, as shown in table , we analyzed the current situation of the respondents' anxiety level and coping style by variables of gender and age, and the results showed that the anxiety levels of adolescents were significantly higher than those of children (f= . , p< . ), and the total scores of scas for females were significantly higher than those of males (f= . , p< . ). the anxiety levels of males and females in each group of children and adolescents are listed in table . in table , results of anova and t test analyses showed that there were variables had significant difference in scas scores (levels of anxiety) (p< . ). additionally, in order to further analyze the significant factors associated with the level of respondent' anxiety, we conducted multiple linear regression analysis and obtained the following factors to construct a multiple linear regression model of anxiety: clinical depression levels, implementation of the precaution and control measures, sex, family member or friend was infected with coronavirus, occupation of the mother involved in the epidemic, region (e.g., rural, urban), and emotion-focused coping style, which accounted for . % of the total variance ( table ) . (table ) . anxiety symptoms are the most common clinical diagnoses in children and adolescents and can act as a significant risk factor for contributing to other psychiatric disorders in adulthood. moreover, it may worsen by facing the increasingly complex social milieu, and being a cause of impairments in various life domains and can increase societal costs for families (bodden et al., ; essau et al., ) . similarly, clinical depressive symptoms in children and adolescents are common and recurrent diseases associated with memory impairments (günther et al., ) , poor interpersonal relationships (lam et al., ) , and even high-risk suicidal behaviors (yorbik et al., ) . with the sudden outbreak and rapid spread of covid- at the end of , the mental health of children and adolescents in china has been put at risk. according to the results of our investigation, we found that levels of anxiety in children and adolescents during the epidemic were much higher than before (zhao et al., ) . as shown in table , the overall scores of five dimensions (including separation anxiety, physical injury fear, social phobia, panic disorder, and generalized anxiety) were higher than before the outbreak, especially the fears of physical injury in children and social phobia in adolescents. this finding may be due to the fact that at the initial stage of the outbreak, protective and therapeutic responses were not yet in place, and the surge in the number of confirmed cases and deaths led children to become excessively concerned about physical damage to themselves and their family caused by exposure to coronavirus. in addition, dong et al. (dong et al., ) reported that chinese adolescents demonstrated higher levels of social-evaluative fears than adolescents in western countries. moreover, negative reports on domestic and foreign social media regarding covid-related discrimination and unfair treatment of chinese people, as well as rumors and misinformation around origins of the infectious disease may be an important reason for worsening their social phobia and other mental disorders (calisher et al., ) .. the findings were in line with previous research indicating that anxiety, depression, and feelings of helplessness can be worsened by sars-related social discrimination in public places (zheng et al., ) . having a family member or friend infected with coronavirus was also significantly associated with increases in anxiety levels. this may be due to the fact that respondents are concerned about the health of the infected and simultaneously afraid of being a suspected or confirmed case, given their level of direct contact. we suggest that the increased awareness of infection control in respondents is consistent with their psychological state of anxiety, thereby reducing the risk of infection. in traditional chinese culture, fathers often assume the role of breadwinners, while mothers tend to perform the bulk of housework and childcare and have more emotional interaction with their children (lee, ) . this appeared to be an important factor affecting the anxiety level of the respondents, who believed that if their mother was not medical staff, this exposed them to insufficient and non-standardized protections, as well as a weaker social support network within their mothers' work environment. given that the outbreak originated in a highly dense urban area, and there have been subsequently many more deaths in this area, it could be the case that anxiety levels of urban residents are higher than those of rural areas. significant sex differences in anxiety levels have been reported in many studies (essau et al., ; li et al., ) , potentially resulting in women being more sensitive to the stressful events of covid- and correspondingly expressing stronger concerns and fears. this study also revealed that the prevalence of clinical depression symptoms in children and adolescents was . % during the covid- outbreak, which was much higher than the generally estimated . % (stewart and sun, ) in china and %- % (phillips et al., ) in western countries. results of regression analysis (table ) and pearson's correlation analysis (table ) showed that levels of clinical depression symptoms were significantly associated with anxiety. evidence has shown that anxiety disorders and depression demonstrate higher overlap in symptomatology, clinical instability, and treatment (lonigan et al., ) . it has been estimated that roughly %- % of depressed youth display comorbid anxiety disorders and about %- % of anxious youth have depression (axelson & birmaher, ) . this suggests that attention should be paid to the proper medication and psychological interventions. due to the significant number of confirmed patients, as well as the relative shortage of medical resources and first-line medical staff, the rate of mortality in hubei province is higher ( . %) than the national mortality rate ( . %) (national health commission of china, ). moreover, a series of effective prevention and control measures (e.g., cities on lockdown, traffic controls, postponed semesters) in wuhan (the capital of hubei province) were implemented more strictly and lasted longer than other regions, leading to the interviewed children and adolescents experiencing greater isolation and for a longer period of time. these factors may explain why living in residential areas was a significant factor affecting depression levels of the respondents. moreover, we also found that students' graduation being affected by the epidemic was significantly associated with increases in depression. chinese students may be particularly academically motivated, particularly in successfully completing entrance examinations (essau et al., ) that will determine the quality of higher education and job prospects. given the current state of the epidemic however, students are taking classes online at home, which may have controversial results due to poor efficiency in managing online courses, visual impairment in students, and instability in internet signals. during the epidemic, . % respondents reported that they spent more than five hours per day online, which could be a potential risk factor for addiction to the internet or smartphones. studies have shown that smartphone/internet overuse may lead to mental or behavioral problems, causing poor performance in study, decreased real-life social interaction, neglecting personal life, relationship disorders, and mood dysfunction (kahraman and demirci, ; soni et al., ) . psychopathological screening studies on internet addicts have shown that anxiety disorders are the most common diagnoses in this group (liu et al., ) , and adverse consequences in sleep disturbances/decreased sleep quality caused by electronic media use at night have been shown to be significantly associated with depressive symptoms (lemola et al., ) . in contrast, we observed that respondents spend much less time online pre-outbreak than they did during the outbreak (table ) , which can also explain the significant correlation between the average time per day spend on internet before the epidemic was significantly associated with the decrease of levels of depressive symptoms. review of the literature has revealed that stress-coping theory posits a broad framework of problem-focused and emotion-focused coping strategies (herman and tetrick, ). problem-focused strategies tend to concentrate efforts towards fixing a stressful problem, while emotion-focused strategies concentrate on minimizing the emotional outcomes of the problem using strategies such as endurance, avoidance, or venting to someone else. it has been reported that greater degrees of ruminative coping instead of actively seeking for solutions were related to high levels of depressive symptoms, while problem-focused and distractive coping style was negatively associated with depression (li et al., ) . we also observed that a problem-focused coping style was closely related to decrease in clinical depression, while an emotion-focused coping style was negatively associated with the increase in depression. this study investigated the mental health status of children and adolescents in mainland china during the epidemic of covid- . specifically, we examined anxiety and depression and investigated the related influencing factors. the findings of this study not only remind researchers and government officials should be more concerned about the mental health of children and adolescents who are often neglected in the wake of the outbreak due to their comparatively lower mortality than elderly adults, but also provide a scientific basis for the formulation of targeted psychological interventions. this study also has several limitations. first, data collection was completed by distributing questionnaires to children and adolescents online. the results and conclusions may be influenced by the degree of understanding and cooperation of the respondents, especially the younger children who need help of their guardians. second, the stylized answer forms of those questionnaires are convenience for statistics, but it limits the response of subjects and let them fail to express their subjective views on relevant issues. third, due to the fact that the epidemic situation in hubei was not fully controlled during the investigation period, it was difficult to collect data and the sample size was relatively small in this region, which limited the applicability and generalization of the results. therefore, in follow-up studies, we plan to increase the sample size (especially in hubei and other hard-hit regions), employ face-to-face surveys, qualitative interview methods, follow-up studies, and collect responses from primary guardians so as to further improve the study design and the scientific nature of the results. the authors have no conflicts of interest to disclose. no conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication. i would like to declare on behalf of my co-authors that the work described was original research that has not been published previously. 'repeated' represents students failed the high school/college entrance examination and had to return to secondary/high school again for resuming their study, while 'fresh' means students have not yet taken the above-mentioned examinations. stepwise selection procedure was employed to select the model from levels of depression, smartphone/internet addiction and subscales of coping style scale, as well as variables listed in table which had statistical difference in levels of anxiety (total scores of scas). table (contents of the covid-related information), as well as levels of anxiety and its six dimensions, smartphone/internet addiction, and subscales of coping style scale. abbreviation: na, not applicable. relation between anxiety and depressive disorders in childhood and adolescence is there a case for quarantine? perspectives from sars to ebola societal burden of clinically anxious youth referred for treatment: a cost-of-illness study the psychological impact of quarantine and how to reduce it: rapid review of the evidence statement in support of the scientists, public health professionals, and medical professionals of china combatting covid- preliminary development of coping style scale for middle school students the notice of the leading party members' group of the ministry of education of the communist party of china on the prevention and control of coronavirus pneumonia and the reform and development of education. ministry of education of china the transcript of the press conference on february , . the national commission of china the smartphone addiction scale: development and validation of a short version for adolescents update on the covid- outbreak as at : on march. china's national health of commission diagnostic and statistical manual of mental disorders fears in chinese children and adolescents and their relations to anxiety and depression psychological interventions for people affected by the covid- epidemic anxiety symptoms in chinese and german adolescents: their relationship with early learning experiences, perfectionism, and learning motivation psychometric properties of the spence child anxiety scale with adolescents from five european countries appraisal, coping, health status, and psychological symptoms verbal memory and aspects of attentional control in children and adolescents with anxiety disorders or depressive disorders problem-focused versus emotion-focused coping strategies and repatriation adjustment internet addiction and attention-deficithyperactivity disorder: effects of anxiety, depression and self-esteem an empirical-clinical approach toward a definition of childhood depression. depression in childhood: diagnosis, treatment, and conceptual models response style, interpersonal difficulties and social functioning in major depressive disorder helping children after a natural disaster: information for parents and teachers gender ideology and the domestic division of labor in middle-class chinese families in hong kong adolescents' electronic media use at night, sleep disturbance, and depressive symptoms in the smartphone age the roles of sex, gender, and coping in adolescent depression anxieties in mainland chinese and singapore chinese adolescents in comparison with the american norm problematic internet use and health in adolescents: data from a high school survey in connecticut a single-center, retrospective study of covid- features in children: a descriptive investigation anxiety and depression in children and adolescents: negative affectivity and the utility of self-reports suicide and social change in china posttraumatic stress and depressive reactions among children and adolescents after the earthquake in ano liosia the psychological effects of quarantining a city prevalence of smart phone addiction, sleep quality and associated behaviour problems in adolescents structure of anxiety symptoms among children: a confirmatory factor-analytic study posttraumatic stress disorder in parents and youth after health-related disasters resilience and depression in children: mental health promotion in primary schools in china ten-year secular trends in sleep/wake patterns in shanghai and hong kong school-aged children: a tale of two cities mitigate the effects of home confinement on children during the covid- outbreak the epidemiological characteristics of depressive symptoms in the left-behind children and adolescents of chongqing in china timely mental health care for the novel coronavirus outbreak is urgently needed clinical characteristics of depressive symptoms in children and adolescents with major depressive disorder caught in the net: how to recognize the signs of internet addiction--and a winning strategy for recovery psychometric properties of the spence children's anxiety scale (scas) in mainland chinese children and adolescents exploratory study on psychosocial impact of the severe acute respiratory syndrome (sars) outbreak on chinese students living in japan key: cord- -xlyi cnl authors: radic, aleksandar; ariza-montes, antonio; hernández-perlines, felipe; giorgi, gabriele title: connected at sea: the influence of the internet and online communication on the well-being and life satisfaction of cruise ship employees date: - - journal: int j environ res public health doi: . /ijerph sha: doc_id: cord_uid: xlyi cnl this study aims to elucidate the idiosyncratic effects of the internet and online communication on the well-being and life satisfaction of cruise ship employees. cross-sectional surveys and covariance-based structural equation modelling tools were used. in addition, univariate variance analysis was used to address the effects of socio-demographic variables (years of service on a cruise ship, working department on a cruise ship, gender, age, educational level and place of residency) on latent variables of the conceptual model. the conceptual model draws on existing theory and previous research and was empirically tested on a sample of cruise ship employee internet users. result show that while being onboard a cruise ship, employees experience strong social pressure to be constantly available and they fear of missing out on important information and life events. thus, relatedness to friends and family needs satisfaction is of paramount importance for cruise ship employees because they are fully aware that they are dispensable and replaceable to cruise ship companies, however to their friends and family, they are indispensable and unique. moreover, employees who engage in other tasks/activities while taking part in online communication with friends and family exhibit reduced performance, which leads to poor interaction and social dissatisfaction. lastly, employees experiencing under-reciprocating exchanges show significant negative effects on their well-being. overall, the results provided several important theoretical and practical implications relevant to cruise tourism and human resource management. cruise tourism growth for predicted by [ ] will not be achieved due to the recent cruise tourism crisis caused by the covid pandemic. on march , all cruise lines suspended their cruise operations for at least days [ ] . however, none of the cruise companies have filed for bankruptcy or cancelled their new builds. therefore, the prediction of the addition of , new crew members and officers each year until remains an achievable possibility [ ] . a recent study by [ ] revealed that cruise ship employees are exposed to long working hours and detachment from friends and family, which leads to poor social interactions and feelings of loneliness. however, advances in information and communication technologies have led to their widespread and increased usage by employees. internet access has become a basic necessity, a more essential element of their mundane ship life with paramount importance on their well-being and life satisfaction. enhanced connectivity is instrumental for bolstering morale and reinforcing job satisfaction, which ultimately strengthens the capability for communication between employees and their significant others back home and hence, reduces the feeling of loneliness [ ] . moreover, maritime labour convention [ ] recommended rational access to the internet with reasonable charges for services. from november to february , only a handful of cruise companies (disney cruise line, holland america, azamara, and princess cruises) have provided free-of-charge specialized cross-platform messaging internet applications for their employees. interestingly, the seafarers happiness index, which covers aspects of job quality, including mental and physical health and relationships at home and onboard, showed significant increases in happiness for cruise ship employees from . to out of [ , ] . given the unique work and life conditions on cruise ships, whereby employees are set apart from their loved ones [ ] , free internet access should be a universal entitlement [ ] because of its ability to enhance seafarer morale, engagement, well-being and life satisfaction [ ] . although significant amounts of research have been done on the positive effects of internet and online communication on social pressure [ ] , fear of missing out [ ] , relatedness to friends and family need satisfaction [ ] , perceived social support [ ] , well-being [ ] and life satisfaction [ ] , these effects were never studied in the peculiar environment of a cruise ship where life and work contexts are so intertwined such that the distinction between one and the other is blurred [ ] . this study aimed to elucidate the idiosyncratic effects of the internet and online communication on the well-being and life satisfaction of cruise ship employees. we reviewed existing theory and previous studies on the effects of the internet and online communication on social pressure, fear of missing out, internet multitasking and relatedness to friends and family need satisfaction. we investigated the influence of social pressure and fear of missing out on relatedness to friends and family need satisfaction and internet multitasking. finally, we proposed relatedness to friends and family need satisfaction and internet multitasking as possible catalyst influencers of perceived social support which, in the end, are the impetus towards the well-being and life satisfaction of employees. the conceptual model draws on existing theory and previous research and was empirically tested on a sample of employee internet users. finally, we addressed the effects of socio-demographic variables (years of service on a cruise ship, working department on a cruise ship, gender, age, educational level and place of residency) on latent variables of the conceptual model. this study is exploratory in nature and presents work addressing a major research gap, given that the effects of internet and online communication on the well-being and life satisfaction of cruise ship employees have never been empirically tested. the results of this study will contribute towards the further development of cruise tourism theory and strengthen existing theories, such as theory of belongingness [ ] , self-determination theory [ ] , uses and gratification theory [ ] , conservation of resources theory [ ] and the paradigm of positive psychology [ ] . while working and living on cruise ships, employees are detached from their family and friends [ ] . primary communication instruments include internet and online communications [ ] , which play significant roles in integrating work and family domains [ ] . these communication instruments provide social capital, information and wider perspectives [ ] . in recent survey conduct by [ ] , the authors concluded that onboard crew members experience strong social pressure for being constantly available to their family and friends. today, almost every cruise ship employee has a mobile device [ ] . although mobile devices allow users to be constantly available, they also create an environment that increases social pressure [ ] . moreover, the social pressure to be constantly available is strongly related to communication load [ ] , with a suppressing effect on well-being via social overload [ ] . based on the social norm of reciprocity in friendship and family ties, psychological tensions and social pressures may arise [ ] . thus, based on the literature review and empirical findings, the following hypothesis is proposed: there is a positive relationship between the internet and online communication and social pressure. in their theory of belongingness, baumeister et al. [ ] argue how human beings have an irresistible need to be a part of a group. the hardest part for cruise ship employees is being away from home and missing so many important life events and quality time with family and friends [ ] . similar results have been reported by [ ] who demonstrated in their comprehensive report how due to being uncontactable at sea, seafarers miss key life events. cruise ship employees are fully aware that work-life on a cruise ship comes with a great burden [ ] . however, they are not willing to tolerate any lack of connectivity [ ] . internet and online communication are closely linked to fear of missing out [ ] . fear of missing out is defined as "a pervasive apprehension that others might be having rewarding experiences from which one is absent" [ ] . thus, we can conclude based on compensatory internet use theory [ ] that individuals who feel that their life needs are not fulfilled and that they are missing important life events and social motivation will experience strong stimulation to use online communication and social networking sites. based on the theoretical background, literature review and empirical findings, the following hypothesis was derived: hypothesis (h ). there is a positive relationship between the internet and online communication and fear of missing out. the internet, social networking sites and online communication have become our liaisons, special amusers, cerberus of our memories and, in times of need, even our counsellors [ ] . based on self-determination theory [ ] , relatedness to friends and family need satisfaction is one of three basic psychological needs that foster healthy self-regulation and promote mental and physical health. opportunities to experience positive feelings of proximity and affection with family and friends at home through the use of online communications benefits seafarers and their friends and family [ ] . interestingly, in the latest survey conducted by [ ] , although provisions of internet access for personal use had positively affected seafarer mental health and morale, home-related anxieties have remained the same, despite speculation that increased communications with family might generate more anxieties. the effects of the internet and online communication and relatedness to friends and family need satisfaction remain unclear [ ] . previous studies have shown that internet and online communication are mainly linked to positive outcomes of relatedness to friends and family need satisfaction [ , ] . few studies have demonstrated the opposite [ , ] . thus, based on literature review, theory and empirical findings, we put forward the following hypothesis: there is a positive relationship between the internet and online communication and relatedness to friends and family need satisfaction. social pressure is conformist behaviour with multiple determinations when an individual or group craves specific social attention; this leads them to behave in certain ways unconcerned of any prestige advantage [ ] . a recent study conducted by [ ] showed how computer-meditated communication is portrayed and influenced by robust rules of conduct, where communication arrangements are under constant social pressure. interestingly, within seafarers, social pressure has shown a strong influence on their work-life at sea due to their social isolation, imbalanced family life, separation from home, family and friends, and lack of free onboard communication facilities [ ] . moreover, relatedness proposes that cruise ship employees need to feel connected with their family and friends at home [ , , ] : when employees feel satisfied with this need, they experience higher levels of work engagement and well-being. most employees own smartphones and experience social pressure to make themselves available to friends and family at home, thereby satisfying needs for relatedness. online communications induced by social pressure and relatedness to friends and family needs satisfaction has been directly linked with significant effects on user life satisfaction [ ] . thus, based on the literature review, theoretical background and empirical findings, we put forward the following hypothesis: there is a positive relationship between social pressure and relatedness to friends and family need satisfaction. the constant occupation with smartphones has created a peculiar mindset in users with specific feelings of being permanently online and connected [ ] . interestingly, online communication and social media are closely linked to fear of missing out [ ] ; % of u.s. social media users suffer from fear of missing out [ ] . thus, tensions related to social relationships may be the prevailing cause for the fear of missing out [ ] . the uses and gratifications theory provides us with pragmatic theoretical lenses instrumental in understanding underlying motives and multitasking behaviours [ ] . the theory proposes that hidden roots of social and psychological needs create certain expectations from various media, which guides users towards specific models of media exposure, culminating in need gratification [ ] . looking at the advantages of online communications to those living and working at sea, fear of missing out has a significant impact; % of crew members need to be connected with the outside world, and % would leave their current company to join some other that would provide better onboard connectivity [ ] . previous research also showed that higher levels of fear of missing out had an impact on a higher tendency to internet multitask [ ] . cruise ship employees have long working hours and rarely go ashore; the tendency to internet multitask leads to some occupational injuries [ ] . consequently, based on theory, literature review and empirical findings, the following hypothesis is derived: there is a positive relationship between fear of missing out and internet multitasking. widespread adoption of the internet, online communication, and social network sites have empowered people across the globe to grow their social network [ ] . however, such a gift comes with responsibility because various social network sites compete for our attention by streaming content based on well-programmed algorithms founded on our likes, fears, and needs [ ] . the digital world is doused with ever-growing social network sites that are changing the online behaviour of digital technology users but also how human beings interact with one another in real life [ ] . moreover, human beings are social animals [ ] in need of relatedness to friends and family, regardless of recent technological advancements. relatedness comes in the form of affective needs, which tend to intensify delightful and affecting experiences, and social needs that tend to bolster existing connections with family and friends [ ] . thus, the need for relatedness is a feeling of satisfaction that comes from being a connected part of a community where individuals manifest a willingness to care about each other [ ] . being away from their homes in an isolated environment [ ] , cruise ship employees are constantly looking for social support [ ] . interestingly, perceived social support, characterized as the tangible or intangible support received from an individuals' social circle, is associated with superior life satisfaction [ ] . consequently, based on theory, literature review and empirical findings, the following hypothesis is derived: hypothesis (h ). there is a positive relationship between relatedness to friends and family need satisfaction and perceived social support. the term internet multitasking refers to "any combination of internet use with other media or non-media activities" [ ] p. . the main reasons for multitasking are social interactions with friends and family and information seeking [ ] . cruise ship employees are most often in different time zones than their friends and family. this leaves them with a limited time frame for online communication and significantly lowers the opportunities for giving and receiving much needed social support. moreover, cruise ship employees are under extreme time pressure due to long working hours [ ] ; this is when individuals perceive that if they engage in internet multitasking, they would be efficient [ ] . however, the effects of internet multitasking on retention of information during online messaging and cognitive load, showed significant retention loss among simultaneous multitasks [ ] . similarly, internet multitasking has been associated with lower gratification and perception memory achievement and sensitivity and moderate standard bias [ ] . moreover, different types of multitasking have robust effects on task performance. task performance significantly decreased when the given task was a secondary task, when a neurological obstruction was high, and when the behavioural reaction was present [ ] . thus, based on literature review and taking in consideration the conflicting empirical results of internet multitasking effect on perceived social support, and bearing in mind the importance of perceived social support for cruise ship employees, the following hypothesis is derived: there is a positive relationship between internet multitasking and perceived social support. the social relationship is flexible and essential for individual vitality because human beings exist within larger social contexts where friends and family play important roles [ ] . moreover, if a persons' social context is supportive of significant relationships, then these individuals encounter elevated feelings towards psychological needs, which can be satisfied through social synergy [ ] . social support has four dominant aspects in the creation of well-being and life satisfaction: main effect (adding particular supplementary function to mental health), mediating effect (intervening in relations between its precursors and health results), indirect effect (preventing disorders by framing mental health) and moderating effect (lowering the risk of any mental health-related components) [ ] . previous studies have shown that the internet and online communication have a positive effect on social capital [ ] where social capital is an antecedent of social support [ ] . internet and online communication are bonding and bridging social capital [ ] , which are of paramount importance for satisfying the social support need of cruise ship employees who tend to use online communication to contact geographically dispersed close friends and family [ ] . thus, social support plays multiple roles in individuals' well-being and life satisfaction [ ] . considering the importance of well-being and life satisfaction of employees, based on literature review and empirical findings, we put forward the following hypotheses: there is a positive relationship between perceived social support and well-being. people should focus on how to be happy, satisfied and filled with positivity [ ] . thus, psychological well-being is an essential part of positive psychology. psychological well-being is related to ones' feelings and evaluations about their life [ ] . moreover, well-being is seen as a psychological well-being that develops based on the eudaimonic dimension of well-being [ ] and as happiness that is built around life satisfaction based on the hedonic dimension of well-being [ ] . in the context of cruise ship employees, well-being is a fusion of eudaimonic (efficiency) and hedonic (thrill) dimensions. interestingly, gibson et al. [ ] argues how due to work-life time constraints, task assignments and job anxiety, employees experience poor well-being. moreover, radić [ ] questions the life satisfaction and well-being of employees who are economic gladiators in pursuit of an unobtainable economic freedom. thus, moore [ ] calls cruise ships "misery machines" where in recent years, as walker [ , ] point out, there has been a substantial increase in suicide rates due to the poor well-being of employees. interestingly, perceived social support from online communication has had a positive effect on well-being [ , ] . social network sites have provided ambient awareness that increases the well-being of its users [ ] . life satisfaction is related to a subjective, comprehensive evaluation of one's quality of life [ ] . moreover, life satisfaction draws from the individual's psychological aspects and is related to one's hedonic satisfaction [ ] , where at the same time, perceived social support from an individual's social networks has the potential to strengthen a person's life satisfaction [ ] . in the q / report, seafarers happiness index [ ] showed how free online communications have a significant impact on employee life satisfaction; as ang et al. [ ] argue, computer-mediated communications can enhance life satisfaction. online communication and social network sites can drive a person towards achieving superior life satisfaction and better quality social relationships [ ] . interestingly, although employees use online communication and social network sites to strengthen their close interpersonal connections and enhance their life satisfaction [ , ] argue that weak ties are also valuable due to their potential to positively influence life satisfaction. figure illustrates the research model and hypotheses of this study. freedom. thus, [ ] calls cruise ships "misery machines" where in recent years, as [ , ] point out, there has been a substantial increase in suicide rates due to the poor well-being of employees. interestingly, perceived social support from online communication has had a positive effect on wellbeing [ , ] . social network sites have provided ambient awareness that increases the well-being of its users [ ] . life satisfaction is related to a subjective, comprehensive evaluation of one's quality of life [ ] . moreover, life satisfaction draws from the individual's psychological aspects and is related to one's hedonic satisfaction [ ] , where at the same time, perceived social support from an individual's social networks has the potential to strengthen a person's life satisfaction [ ] . in the q / report, [ ] showed how free online communications have a significant impact on employee life satisfaction; as [ ] argue, computer-mediated communications can enhance life satisfaction. online communication and social network sites can drive a person towards achieving superior life satisfaction and better quality social relationships [ ] . interestingly, although employees use online communication and social network sites to strengthen their close interpersonal connections and enhance their life satisfaction [ , ] argue that weak ties are also valuable due to their potential to positively influence life satisfaction. figure illustrates the research model and hypotheses of this study. the theoretical framework of this study was based on a literature review; the conceptual model and hypotheses were tested based on a convenience sample. the post-positivistic paradigm was adopted in this study because as [ ] argues, the post-positivistic paradigm takes into consideration the fact that in human behaviour studies, observations are imperfect with potential inaccuracies; thus, all theories could be amended. action research strategy allows the research to use different models of contemporary knowledge in solving genuine industry issues and applying obtained results outside the boundaries of the study [ ] . thus, action research strategy was used. the research model was evaluated using a cross-sectional survey and covariance-based structural equation modelling the theoretical framework of this study was based on a literature review; the conceptual model and hypotheses were tested based on a convenience sample. the post-positivistic paradigm was adopted in this study because as [ ] argues, the post-positivistic paradigm takes into consideration the fact that in human behaviour studies, observations are imperfect with potential inaccuracies; thus, all theories could be amended. action research strategy allows the research to use different models of contemporary knowledge in solving genuine industry issues and applying obtained results outside the boundaries of the study [ ] . thus, action research strategy was used. the research model was evaluated using a cross-sectional survey and covariance-based structural equation modelling (cb-sem). cb-sem allows testing and validation of current theories and comparisons of different theories [ ] . in summary, this study used a deductive approach followed by a cross-sectional time horizon and quantitative techniques for data collection. a comprehensive self-reported online survey in english was designed at surveymonkey®. possible participants were invited to take part in the survey via facebook group "crew center". the main criteria was that participants had to be onboard and employed by a cruise company. the survey was online from august to december , and the final sample consisted of cruise ship employees (see table ). the sample comprised males and females from different geographical areas ( . % from europe, . % from north america, . % from southeast asia, . % from south america, . % from central america, . % from africa, and . % from australia). most respondents were between - years old ( . %) followed by respondents - years old ( . %) and - years old ( . %). among the participants, . % were employed in the hotel department, . % were from the deck and technical department, and . % were from the entertainment department. most ( . %) had worked in the industry for over six years. a large share of respondents had a bachelor's degree ( . %). this extreme unrepresentative value was related to the convenience sampling method. overall, the sample was a very good representation of employee demographics [ ] . internet and online communication were assessed using a five item scale designed to asses internet and online communication usage (all the measures are included in the appendix a). participants indicated on a five point scale from (once per week) to (several times per week) how often they use the internet for communication; from (less than an hour) to (more than hours per day) how many hours per day (on average) they spend on internet communication; from (once a week) to (several times per day) how often they use a) instant messenger, b) social networking sites, and c) chat rooms. internal consistency in the present sample was acceptable (cronbach's α = . ). social pressure to be permanently available was assessed with an adapted perceived norm scale [ ] that had four items (e.g., "people from my private social environment think that it is important that i'm constantly available") and is rated on a five point scale ranging from (does not apply at all) to (fully applies). internal consistency in the present sample was acceptable (cronbach's α = . ). fear of missing out on important life events and information was assessed with a three item scale (e.g., "if i would use the internet less frequently, i would be missing out on important things") developed by [ ] . participants rated the items on a five point scale ranging from (does not apply at all) to (fully applies). internal consistency in the present sample was good (cronbach's α = . ). relatedness to friends and family need satisfaction was assessed with three item scale (e.g., "i feel that my friends and/or family sincerely care about me") developed by [ ] . participants rated the items on a five point scale ranging from (strongly disagree) to (strongly agree). internal consistency in the present sample was good (cronbach's α = . ). internet multitasking was assessed with a five item scale (e.g., "how often do you use the internet while you simultaneously are in a conversation with another person") developed by [ ] . participants rated the items on a five point scale ranging from (never) to (very frequently). internal consistency in the present sample was acceptable (cronbach's α = . ). perceived social support was assessed by the multidimensional scale of perceived social support [ ] , which consisted of six items (e.g., "there is a special person who is around when i am in need") and was rated by participants on a seven point scale ranging from (strongly disagree) to (strongly agree). internal consistency in the present sample was acceptable (cronbach's α = . ). well-being was assessed by the world health organization well-being index [ ] . it comprises five items (e.g., "i have felt cheerful and in good spirits") and was rated by participants on a six point scale ranging from (all the time) to (at no time). internal consistency in the present sample was good (cronbach's α = . ). life satisfaction was assessed by satisfaction with life scale [ ] . it consists of five items (e.g., "in most ways my life is close to my ideal") and it was rated by participants on a seven point scale ranging from (strongly disagree) to (strongly agree). internal consistency in the present sample was good (cronbach's α = . ). structural equation modelling (sem) was computed using the amos software packet (ibm, chicago, illinois), and the maximum likelihood method was used to estimate the parameters from the conceptual model (see figure ). the kolmogorov-smirnov and shapiro-wilks test showed that none of the variables were normally distributed. thus, a maximum likelihood (ml) estimator with enough resistance capabilities to none-extreme deviations from the normal distribution [ ] was used. model fit was tested based on the χ and cmin/df statistics, the comparative fit index (cfi) and the root mean square error of approximation (rmsea) as recommended by [ ] . the univariate analysis of variances (anova) was used in search of differences among employee demographics and conceptual model variables. the model showed an acceptable fit to the data with the following values: χ ( ) = , , p = . ; rmsea = . , lo = . , hi = . ; cmin/df = . and cfi = . . although the general indicator χ was significant, with such a large number of degrees of freedom, χ is not reliable; it is better to rely on other indicators. rmsea was close to the limit that indicates an excellent model ( . ), cmin/df was within the limits that represent a good model, whereas the cfi was close to the lower limit of acceptability of the model [ ] [ ] [ ] . table shows the hypothesized paths of the conceptual model. the zero-order correlations between social pressure and fear of missing out, demonstrate that these two variables are strongly interrelated. social pressure and fear of missing out show very high correlations (r = . , p < . ). this significant relationship is reasonable: social pressure as a concept is closely connected to the concept of fear of missing out as these social processes on the internet and social network sites are synthesized. because social interaction through the internet and online communication are of paramount importance for cruise ship employees [ ] , accomplishing this pursuit is a way towards well-being and life satisfaction. this conception is supported by results from [ , ] , who reported that a recently developed free-of-charge specialized cross-platform messaging internet application (by a handful of cruise companies) had an immediate impact on the happiness index of employees by increasing to % from q / to q / . except for hypotheses , and , all other hypothesized relationships were supported in the final model ( figure ). hypotheses and were supported showing how the internet and online communication had positive effects on social pressure (β = . ) and fear of missing out (β = . ). interestingly, hypothesis was not supported, demonstrating that the internet and online communication did not have a positive effect on relatedness to friends and family need satisfaction (β = . ). as predicted in hypothesis , social pressure had a positive effect on relatedness to friends and family need satisfaction (β = . ), and fear of missing out has a positive effect on internet multitasking (β = . ) as predicted in and hypothesis . hypothesis predicted that relatedness to friends and family need satisfaction had a positive effect on perceived social support; this hypothesis was supported (β = . ). hypothesis predicted that internet multitasking had a positive effect on perceived social support; however, this hypothesis was not supported (β = − . ). hypothesis predicted that perceived social support had a positive effect on well-being; however, this hypothesis was not supported (β = − . ). lastly, hypothesis predicted that perceived social support had a positive effect on life satisfaction; this hypothesis was supported (β = . ). in pursuit of elucidating the peculiar socio-demographics characteristics of cruise ship employees and unrevealing significant differences on sample and research model variables, anova was used. the results disclosed the following pivotal differences. the hypotheses and were supported showing how the internet and online communication had positive effects on social pressure (β = . ) and fear of missing out (β = . ). interestingly, hypothesis was not supported, demonstrating that the internet and online communication did not have a positive effect on relatedness to friends and family need satisfaction (β = . ). as predicted in hypothesis , social pressure had a positive effect on relatedness to friends and family need satisfaction (β = . ), and fear of missing out has a positive effect on internet multitasking (β = . ) as predicted in and hypothesis . hypothesis predicted that relatedness to friends and family need satisfaction had a positive effect on perceived social support; this hypothesis was supported (β = . ). hypothesis predicted that internet multitasking had a positive effect on perceived social support; however, this hypothesis was not supported (β = − . ). hypothesis predicted that perceived social support had a positive effect on well-being; however, this hypothesis was not supported (β = − . ). lastly, hypothesis predicted that perceived social support had a positive effect on life satisfaction; this hypothesis was supported (β = . ). in pursuit of elucidating the peculiar socio-demographics characteristics of cruise ship employees and unrevealing significant differences on sample and research model variables, anova was used. the results disclosed the following pivotal differences. the cruise ship employees come from various countries around the globe. thus, it is important to determine whether any fundamental differences were present based on country of residence and the research model variables. employee place of residence had prominent effects on internet and online communication (f( , ) = . , p = . ), fear of missing out (f( , ) = . , p = . ), social pressure (f( , ) = . , p = . ), relatedness to friends and family need satisfaction (f( , . ) = . , p = . ), internet multitasking (f( , . ) = . , p = . ), perceived social support (f( , . ) = . , p = . ) and well-being (f( , . ) = . , p = . ). differences between place of residence and fear of missing out (η = . ) were low. differences among place of residence and internet and online communication (η = . ), social pressure (η = . ), relatedness to friends and family need satisfaction (η = . ), internet multitasking (η = . ), well-being (η = . ) and perceived social support (η = . ) were moderate. student's t-test revealed that employees from africa showed significant differences the goal of the current study was to investigate and model complex mutual interactions the internet and online communication had on social pressure, fear of missing out, internet multitasking and relatedness to friends and family need satisfaction, perceived social support, well-being and life satisfaction of cruise ship employees. the results revealed that the internet and online communication have positive effects on social pressure and fear of missing out, whereas social pressure and fear of missing out have positive effects on relatedness to friends and family need satisfaction and internet multitasking. moreover, internet multitasking had a positive effect on perceived social support, which in turn had a positive effect on life satisfaction. while onboard a cruise ship, employees are detached from their family, significant others and friends. thus, employees experience strong social pressure to be constantly available and fear of missing out on important information and life events. due to their rigorous schedule, i.e., working to hours every day of the week in an isolated environment, the internet and online communications are wonderful instruments that can meet the demands of social pressure and reduce the fear of missing out experienced by cruise ship employees. the internet and online communication are essential to employees from collectivistic cultures where friend and family ties are strong. thus, providing these employees instruments to maintain close contacts with friends and family at home will reduce their social pressure and fear of missing out, which in turn will create harmony among employee social groups. moreover, the provision of online communication creates an interactive platform for validation through communication acceptance; validation boosts employee sense of belonging and strengthens their relationships with friends and family at home. these results are supported by the theory of belongingness by [ ] and are in line with previous studies conducted by [ , , , ] . free time is the single most precious commodity for cruise ship employees. tight work schedules, ship itineraries, in-port safety duties, and daily job demands leave employees with very narrow time windows for engagement in social networking sites and online communication. thus, social pressure and fear of missing out effects on relatedness to friends and family need satisfaction force employees to engage in internet multitasking. within the minimal and confined space of a cruise ship, employees lose almost all points of reference to the outside world and friends and family at home; thus, social pressure and fear of missing out fuels the need for relatedness to friends and family and internet multitasking. employees spend on average between four and six months onboard in small shared cabins while working long hours every day of their full contract length and are in desperate need of the support of family and friends. thus, if cruise ship companies provide ad libitum internet access to online social networks and communication, employees will enjoy a strong network of supportive family and friends that can help them enhance life satisfaction. interestingly, employees with bachelor's degrees exhibited high levels of internet multitasking, which correlated positively with their working memory and ability to divert their attention among different tasks. these results are supported by self-determination theory of [ ] as well as uses and gratification theory [ ] . moreover, these findings are in line with previous studies conducted by [ , , , ] . cruise ship employees understand they are dispensable and replaceable to cruise ship companies [ ] . thus, relatedness to friends and family needs satisfaction is of paramount importance for employees because they know that to their friends and family, they are indispensable and unique. moreover, living and working onboard a cruise ship is a lifestyle; while this particular lifestyle may seem to disregard certain employee rights, it does not dehumanize the employee. when employees feel their needs towards relatedness are satisfied, they experience elevated social support through strengthened connections with friends and family. although cruise ship employees work and live on the high seas, they do not thrive alone: donne [ ] pointed out that each man relies on others. however, there are slight differences between employees and their relatedness to friends and family who need satisfaction. experienced employees with + years exhibited stronger needs towards relatedness because they feel lonely and isolated for a significant period of their life. employees working in the entertainment department and employees coming from south america expressed greater needs because they have a wide circle of close ties at home. lastly, the youngest cruise ship employees between - years exhibited greater needs towards relatedness because at their age, friendships are highly complex and offer significant self-disclosure and support. these results are also supported by self-determination theory of [ ] and are in line with previous studies conducted by [ , ] . onboard cruise ship operations are in constant flux, and employees come and go frequently. for many employees these crew changes become the only point of reference. during long contracts, tiredness of employees builds up as physical pain, exhaustion and psychic fatigue. however, cruise ship employees are required to continuously work until the completion of their contracts. experiences like these build highly intense relationships where employees need social support. in such an environment, employees lean on their friends and family at home for support, which comes in many forms most often as empathy, compassion and providing care. social support is the foundation of healthy relationships that improves employee life satisfaction. employees who spent + years living and working on cruise ships are exposed to prolonged periods of loneliness and isolation, which affects them in ways that would require social support to achieve happiness and life satisfaction. male employees look for social support to feel happy and satisfied, whereas female employees engage in social support to pursue increased well-being. employees with master/doctoral degrees enjoy social support from their close ties because they understand the benefits of happiness that comes from such relationships. finally, employees from north america consider deep relationships to have significance in enhancing social support because larger social networks improve life satisfaction. these results are supported by the positive psychology paradigm by [ ] . moreover, these findings are in line with previous studies conducted by [ , , ] . although keeping in touch with family and friends at home is essential for cruise ship employees, the majority of cruise companies charge significant prices for internet and online communication services. thus, due to limitations of internet and online communication use because of high service prices, poor coverage and slow data connection, employees are prevented from satisfying their needs through their friends and family at home. moreover, due to in-port manning duties, employees cannot use free internet services off ship, which leads to chronic emotional distress, frustration, anger, despair, and anxiety. this finding is in line with [ ] who argued that people heavily dependent on social networking sites to satisfy their needs towards relatedness to friends and family may experience a lack of social capital outcomes; this can trigger detrimental impacts on their well-being. employment on cruise ships carries many occupational safety hazards [ ] . to compensate, employees engage in internet multitasking; however, doing so exposes employees to added distractions and prevents them from safely or effectively completing their tasks. moreover, employees who engage in other tasks/activities while taking part in online communication with friends and family exhibit reduced performance, which leads to poor interaction and social dissatisfaction. this finding was in line with [ ] who argued that multitasking is related to reduced enjoyment in messages and reduced recognition memory performance. similarly, Örün et al. [ ] argue that retention of communication content during online messaging is significantly worse while multitasking. lastly, the perceived social support from internet and online communications and social networking sites with family and friends at home is dependent on reciprocity. employees experiencing under-reciprocating exchanges show significant negative effects on their well-being. these results are supported by the conservation of resources theory [ ] and the reciprocity norm [ ] . perceived social support is a multi-dimensional construct highly dependent on personality traits. personality can affect perceived social support relationship with well-being to the point of being non-significant [ ] . overall, our results provided several important theoretical and practical implications relevant to cruise tourism and human resource management. this study contributes to academic literature in several ways. first, working and living on a cruise ship, carries a heavy burden where cruise ship employees are detached from their family and friends at home [ ] and their main communication instruments are the internet, social networking sites and online communications [ ] . thus, although this study is of an exploratory and pioneering nature regarding the effects of internet communication on employee well-being and life satisfaction, this study was founded on well-known theories. overall, the results are consistent with the theory of belongingness [ ] , the self-determination theory [ ] , the uses and gratification theory [ ] and the conservation of resources theory [ ] and confirmed previous studies that reported positive effects of the internet and online communication on social pressure on fear of missing out [ , ] ; relatedness to friends and family needs satisfaction [ ] ; fear of missing out on internet multitasking [ ] ; relatedness to friends and family needs satisfaction and perceived social support [ ] ; and life satisfaction [ ] . intriguingly, the results did not confirm positive effects of internet and online communication on relatedness to friends and family needs satisfaction, internet multitasking on perceived social support, and perceived social support on well-being. however, these findings are supported by previous studies, such as [ , , , ] . a second contribution to the literature is the final model's broad scope and applicability towards achieving life satisfaction of employees who work and live in a specific workplace, as described by [ ] . comprehensive measurement scales (based on previously confirmed scales by [ , [ ] [ ] [ ] [ ] for measuring internet and online communication effects on well-being and life satisfaction of cruise ship employees showed strong reliability and validity. these tools can be used in future studies as instruments for measuring internet and online communication effects on the well-being and life satisfaction of various employees. this study offers valuable practical recommendations for cruise ship companies. unhappy, detached and dissatisfied employees can become unproductive and disengaged, which can lead to high employee turnover, absenteeism, and increased expenses due to health care costs and insurance premium fees [ ] ; this can harm the profitability of cruise ship companies. companies that provide free-of-charge internet and access to social networking sites and online communication, will satisfy employee needs for belongingness. this will allow employees to enjoy a strong network of family and friends to achieve and maintain life satisfaction. moreover, companies who understand and appreciate the value of employees who flourish in life satisfaction should reinforce their core values by setting their "true north" towards providing employees instruments to maintain close contacts with their friends and family at home. such provisions by companies would suppress employee social pressure and fear of missing out. this would create harmony among their close ties and social groups, ultimately leading to life satisfaction of cruise ship employees. this study has several limitations. first, this study utilized a cross-sectional time horizon; there is space for potential causality and reciprocal relationships among components [ ] . future studies should use a longitudinal time horizon to investigate the effects of the internet and online communication on well-being and life satisfaction during various stages of cruise ship employee contracts. second, common method bias is expected in this study due to self-reported answers collected from employees who agreed to participate in the survey. to lessen this challenge, a cautiously composed and validated survey was used following the suggestion of [ ] . as such, participant anxiety related to giving right or wrong answers was at least reduced to its lowest possible level, if not completely avoided. nevertheless, components that were used in this study could only be measured by particular, authentic impressions of employees. third, participants in this study were recruited via the facebook group "crew center". thus, the sample is not representative of the general population of all cruise ship employees who use the internet and online communications because many cruise ship employees are not members of the aforementioned group and as such are underrepresented. the fourth limitation is related to the sampling method, i.e., the convenience sample method. this method could have limited the generalizability of the overall findings. the fifth limitation is related to the research model. even though components showed satisfactory levels of validity and reliability, component constituents should be tested in future work on wider populations of employees. the sixth limitation is the quantitative analyses that were used to evaluate research data. future studies could mix qualitative and quantitative techniques to obtain comprehensive knowledge about the effects of internet and online communication on well-being and life satisfaction of cruise ship employees. taking into consideration these limitations, the results of this exploratory study cannot be generalized and should be judged with limited and careful interpretation. cruise line international association. state of the cruise industry outlook major cruise lines have suspended operations for days working at sea occupational and health safety on cruise ships: dimensions of injuries among crew members an investigation into connectivity at sea. available online maritime labour convention seafarers happiness index seafarers happiness index what's eating cruise ship employees: exploring dimensions of engagement the human right to free internet access new survey from ics and ecsa paints positive picture for seafarer internet access basic psychological needs satisfaction and fear of missing out: friend support moderated the mediating effect of individual relative deprivation digital stress over the life span: the effects of communication load and internet multitasking on perceived stress and psychological health impairments in a german probability sample daily technology interruptions and emotional and relational well-being facebook use and its association with subjective happiness and loneliness mobile-mediated multimodal communications, relationship quality and subjective well-being: an analysis of smartphone use from a life course perspective tell me a story: socio-emotional functioning, well-being and problematic smartphone use in adolescents with specific learning disabilities istraživanje kvaliteta uslužnog process u kruzing turizmu the need to belong: desire for interpersonal attachment as a fundamental human motivation self-determination theory and the facilitation of intrinsic motivation, social development, and well-being social and psychological resources and adaptation positive psychology: an introduction online communication with strong ties and subjective well-being in japan a network analysis of facebook use and well-being in relation to key psychological variables: replication and extension connecting activities on social network sites and life satisfaction: a comparison of older and younger users producers and directors work-family role integration and personal well-being: the moderating effect of attitudes towards personal web usage online-specific fear of missing out and internet-use expectancies contribute to symptoms of internet-communication disorder why social network site use fails to promote well-being? the roles of social overload and fear of missing out evaluation of hotel websites: progress and future developments (invited paper for 'luminaries' special issue of international journal of hospitality management) social media and the fear of missing out: scale development and assessment motivational, emotional, and behavioral correlates of fear of missing out a conceptual and methodological critique of internet addiction research: towards a model of compensatory internet use family functioning and information and communication technologies: how do they relate? a literature review fear of missing out and its link with social media and problematic internet use among filipinos social media elements, ecologies, and effects the contribution of mobile social media to social capital and psychological well-being: examining the role of communicative use, friending and self-disclosure the sense of should: a biologically-based framework for modeling social pressure. phys. life rev theotokas, i. the role of communication to the retention of seafarers in the profession understanding computer-mediated communication attributes and life satisfaction from the perspectives of uses and gratifications and self-determination permanently online and permanently connected: development and validation of the online vigilance scale report: % of social media users suffer from fomo social and emotional correlates of the fear of missing out why do media users multitask?: motives for general, medium-specific, and content-specific types of multitasking utilization of mass communication by the individual network environments and well-being: an examination of personal network structure, social capital, and perceived social support young people, social media and connective action: from organisational maintenance to everyday political talk the politics; translated by h. rackham on the use of the mass media for important things self-determination theory and the role of basic psychological needs in personality and the organization of behavior identity construction in transient spaces: hospitality work on-board cruise ships new kids on the ship': organisational socialisation and sensemaking of new entrants to cruise ship employment what happens on facebook stays on facebook? the implications of facebook interaction for perceived, receiving, and giving social support effect of multitasking, physical environment and electroencephalography use on cognitive load and retention boring is bad: effects of emotional content and multitasking on enjoyment and memory multitasking and task performance: roles of task hierarchy, sensory interference, and behavioral response why social relationships are important for physical health: a systems approach to understanding and modifying risk and protection the sage handbook of social network analysis surfing alone? the internet and social capital: evidence from an unforeseeable technological mistake toward an economy of well-being the effect of internet use on well-being: meta-analysis effects of nature-based solutions (nbs) on eco-friendly hotel guests' mental health perceptions, satisfaction, switching barriers, and revisit intentions talent management and the cruise industry will giant cruise ships destroy the wonders their passengers claim to love misery machines and crew member suicides suicide attempts by filipino crew members demonstrate need for shipboard psychologists i see you, i know you, it feels good" -qualitative and quantitative analyses of ambient awareness as a potential mediator of social networking sites usage and well-being subjective well-being: the science of happiness and a proposal for a national index social interactions and well-being research design: qualitative, quantitative, and mixed methods approaches doing action research in your own organization an updated and expanded assessment of pls-sem in information systems research the world cruise industry: a profile of the global labour market. seafarer's international research centre (sirc) how can perceived autonomy support influence enrollment in elective physical education? a prospective study the multidimensional scale of perceived social support world health organization. world health organization info package: mastering depression in primary care. fredriksborg: who, regional office of europe, psychiatric research unit the satisfaction with life scale confirmatory factor analysis for applied research bentler department of psychology university of california los angeles cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives principles and practice of structural equation modeling dimensionalizing cultures: the hofstede model in context. online readings psychol devotions upon emergent occasions, and severall steps in my sicknes the norm of reciprocity: a preliminary statement the relationship between social support and subjective well-being across age an investigation into hospitality cruise ship work through the exploration of metaphors games researchers play -extreme-groups analysis and mediation analysis in longitudinal occupational health research common method biases in behavioral research: a critical review of the literature and the recommended remedies key: cord- -b rucgmg authors: di carlo, francesco; sociali, antonella; picutti, elena; pettorruso, mauro; vellante, federica; verrastro, valeria; martinotti, giovanni; di giannantonio, massimo title: telepsychiatry and other cutting edge technologies in covid‐ pandemic: bridging the distance in mental health assistance date: - - journal: int j clin pract doi: . /ijcp. sha: doc_id: cord_uid: b rucgmg at the end of a novel coronavirus (covid‐ ) was identified in china. the high potential of human to human transmission led to subsequent covid‐ global pandemic. public health strategies including reduced social contact and lockdown have been adopted in many countries. nonetheless, social distancing and isolation could also represent risk factors for mental disorders, resulting in loneliness, reduced social support and under‐detection of mental health needs. along with this, social distancing determines a relevant obstacle for direct access to psychiatric care services. the pandemic generates the urgent need for integrating technology into innovative models of mental healthcare. in this paper we discuss the potential role of telepsychiatry and other cutting‐edge technologies in the management of mental health assistance. we narratively review the literature to examine advantages and risks related to the extensive application of these new therapeutic settings, along with the possible limitations and ethical concerns. telemental health services may be particularly feasible and appropriate for the support of patients, family members and health‐care providers during this covid‐ pandemic. the integration of telepsychiatry with other technological innovations (e.g., mobile apps, virtual reality, big data and artificial intelligence) opens up interesting future perspectives for the improvement of mental health assistance. telepsychiatry is a promising and growing way to deliver mental health services but is still underused. the covid‐ pandemic may serve as an opportunity to introduce and promote, among numerous mental health professionals, the knowledge of the possibilities offered by the digital era. initially identified in december in china, the coronavirus disease is now affecting more than countries and territories around the world. given the current unavailability of an effective cure or vaccine for covid- , a public health strategy of reduced social contact and distancing has been adopted worldwide to slow the spread of the virus. this strategy may on the one hand achieve its goals, but on the other hand, it may undermine normal social support systems, yielding to loneliness and reduced support from family and friends. these conditions may be risk factors for the development of anxiety, stress, depressive symptoms [ ] [ ] [ ] [ ] in the general population, they may worsen previously diagnosed diseases [ ] , as well as they may favor the onset of new problematic behaviors (e.g., internet misuse [ , ] ). during the pandemic ni y. m. et al. ( ) found that almost one-fifth of respondents to their online survey, targeting chinese adults, reported probable anxiety and depression. another survey, carried out in the same time among italian people, registered high/very high levels of depression, anxiety and stress in a large part of the sample [ ] . because of the contagiousness of covid- [ ] , physical access to the mental health care system could increase the risk of infection. fear of contracting the disease in a healthcare setting, transportation restrictions and isolation at home have become important barriers to treatment for many people [ ] . zohu et al ( ) reported that the decrease of hospital visits led to a reduction of routine psychiatric care for many patients with mental disorders in china, during the covid- outbreak. consequently, . % of their sample self-reduced drugs dosages and . % stopped taking their psychiatric drugs because they could not have a prescription from their doctor. they also found out that . % of new patients with anxiety disorders, insomnia, psychosis or depression, could not receive timely diagnoses and treatment during this period. these data globally underline the urgency of quickly adapting to current situation, in order to guarantee continuity of care for at-risk populations, such as that of psychiatric patients. in this respect, the global concern about the psychosocial consequences of covid- pandemic has led governments to call for suggestions to deal with these effects. the use of digital health technology has been proposed as an important strategy to reinforce heath systems [ ] . this article is protected by copyright. all rights reserved according to the world health organization, telemedicine is "the delivery of health care services, where distance is a critical factor, by all health care professionals, using information and communication technologies" [ ] . when applied in the field of psychiatry these methods are known as telepsychiatry (tp). telemental health services can be particularly useful and appropriate for the support of both patients and health care workers during this pandemic, allowing providing assistance and care to those who need it by reducing the risk of infection. illustrative of that is the case of a californian psychiatric clinic in which all patients with existing appointments received care via tp, allowing to provide uninterrupted care while minimizing potential covid- exposure to patients and clinicians [ ] . technology-based tools offer telemedicine visits for patients, as well as support for institutions, by facilitating the provision of information and data sharing, while creating virtual spaces for meetings and clinical briefings [ ] . it is interesting to underline that psychiatric care is relatively unique: with only rare exceptions, psychiatrists are not expected to physically touch their patients to provide the best services. verbal and visual assessments, which can easily occur via telemedicine, are all that is needed to engage with patients appropriately obtain full diagnostic accuracy and provide most types of treatments. this makes using a -way tele-videoconferencing for clinical assessments and treatment delivery a reasonable method for psychiatric treatment [ ] . the aim of this narrative review is to evaluate tp utility and feasibility, updating and summarizing the most important findings about its use in the modern psychiatric practice, in order to inform the potential use of tp during the pandemic. this article is protected by copyright. all rights reserved pandemic. only english-written papers were considered for inclusion. resources were filtered using abstracts (where available), before evaluating full texts, and the included studies were selected based on their relevance to the aim of this review. the reference section of each examined article was also reviewed to identify other potential studies. a total of records was found from bibliographic search. of these, record were excluded after abstract evaluation (n= because were not english written, n= because were other study types, n= because were not relevant to the aim of this review). of for patients, tp improves access to care, reduces appointments' waiting time and decreases travel time and costs [ ] . some patients report feeling more comfortable, and can be more open and honest, when discussing difficult subjects from their devices, because the virtual space of the session instills a feeling of "protection". another important advantage of tp in terms of rapport building is the possibility for patients (especially immigrants, refugees, and asylum seekers) to receive psychiatric assistance in their native language without the assistance of an interpreter [ ] . in their review cowan et al. [ ] reported several advantages to tp, both from the patient's and the clinician's points of view. by working in different settings (schools, prisons, homes, and hospitals) and with different populations (prisoners, students, employees, hospital patients and outpatients) clinicians can have flexibility in scheduling appointments and an increased diversity of practice. they may also feel safer when this article is protected by copyright. all rights reserved evaluating patients without any risk of physical aggression. furthermore, the use of tp allows clinicians to remotely consult other colleagues. given a reasonable initial apprehension, discomfort and fear, patients generally report an increase in comfort and satisfaction after having used tp a few times. comparably to patients, after a few trials, clinicians reveal an improved attitude toward tp as well, suggesting that increased exposure for clinicians may be important to alleviate their concerns about rapport [ ] . with regards to efficacy, in a recent narrative review by hilty et al. [ ] , tp was found to be effective in terms of patient acceptance, ability to increase access to care and provision of good educational outcomes. moreover, tp was found to be valid and reliable when compared to in-person services. telemental health effectiveness compared to faceto-face treatment has been proven in various randomized controlled trials (rct). this has been proved to be valid also for psychotherapy. [ ] [ ] [ ] . both users and, particularly, nonusers of telemedicine reported disliking the loss of personal contact with patients. the decreased ability to detect nonverbal cues during video conferencing may limit the building of the relationship, since clinicians report some difficulties in picking up nuances and emotions [ ] . some clinicians may feel and look clumsy and uncomfortable or may have a hard time engaging with patients; eye contact can feel artificial across the screen as well. clinicians have also expressed discomfort in being unable to take physical actions to reassure or assist their patients [ ] . when randomly assigned to evaluate in-person vs videoconference therapy sessions psychiatrists reported lower therapeutic alliance in telemental health settings; they also persisted on being hesitant to use video conference for their therapy sessions because they believed that the therapeutic alliance was at risk [ ] . concerns have been voiced about certain patients possibly being non-suitable for tp, including individuals with psychotic symptoms, acute crisis or at risk of self-harm [ ] . patients with auditive, visual or cognitive impairments appear not to be eligible for telemental health services [ ] . furthermore, it has been proposed to carefully consider the role of substance abuse and past episodes of self-injurious or violent behavior before selecting a patient for a tp intervention [ ] . technical difficulties such as transmission unable to start, spontaneous this article is protected by copyright. all rights reserved disconnections, poor audio/visual quality and audio/visual delay may seriously affect the tp session. sound quality may affect more substantially the interaction; although audio quality may be prioritized, poor visual transmission, resulting in decreased ability to detect nonverbal cues, remains a concern [ ] . clinicians are often reluctant to try tp because it is not something they are accustomed to [ ] . new approaches to professional techniques require deliberate conscious efforts. there are deep-rooted cultural traditions in medicine that impede the diffusion of tp. physical proximity has an important part in clinical practice, with an enormous cultural significance. over the course of two thousand years, a doctor's physical presence has been regarded as necessary for clinical practice and patients' healing. telemental health is not equally distributed among professional categories other than psychiatrists. for example, more than two-thirds of psychologists in the us have never used videoconferencing to deliver their therapeutic services in [ ] . finally, despite easier and cheaper access to the internet the so called "digital divide" remains often reported in the literature and can impede tp diffusion [ ] . these reports still reveal that the elderly, racial/ethnic minorities, and individuals from lower socioeconomic classes continue to trail behind in access to the internet and thus in taking advantage of tp opportunities [ ] . to address this limitation the lancet commission on global mental health recommended adoption of digital interventions alongside traditional in-person treatments, rather than as replacements [ ] . a relevant concern about using tp is privacy. it is important to ensure conversations' confidentiality and to protect health information and sensitive data. another confidentiality issue is of technological nature, concerning network security and encryption [ ] ; in fact, videoconferencing occurs through the internet connection, it is therefore essential to guarantee that the meeting will not be easily hacked into, or viewed by others [ ] . clinicians and organizations who are supplying telemental health services must adapt their connection pathways and equipment to the health information protection laws, such as the usa's health insurance portability and accountability act (hipaa). in recent years a variety of platforms (like zoom, bluejeans, simplepractice and others) have been developed with opportune security and confidentiality [ ] , but privacy cannot be this article is protected by copyright. all rights reserved guaranteed in its entirety. in fact, the fbi warned about the weakness of such platforms after some unidentified users invaded several school-sessions, a phenomenon called "zoombombing". because of the covid- pandemic, some governments have issued an emergency waiver suspending the requirement to comply with information protection laws, as it happened in the usa for hippa, in order to facilitate access to telemedicine services [ ] . however, the privacy issue remains a compelling one. similar with in-person interventions, clinicians are legally obligated to seek informed consent and review possible risks and benefits of the treatment [ ] . nittari et al. [ ] recently reviewed legal issues related to informed consent, data protection, confidentiality, physician's malpractice, and liability in telemedicine. the main issue reported by the authors is the global uncertainty of the legislation in the field. it has been suggested that clinicians need to update their knowledge and adapt to the newest regulations related to online interventions [ ] . tp has been proven to be valid and reliable in a variety of psychiatric disorders. depression is one of the most studied, and the evidence supporting the treatment of depressive disorders via tp is robust [ ] . a large number of systematic reviews and meta-analyses have examined technology-based interventions for depression, but only few of those have attempted to synthesize the data with the purpose of determining the admissibility and influencing factors of the implementation of tp [ ] . there are different forms of internet-delivered treatments for depression. one approach, based on bibliotherapy and the therapist's guidance via email or phone, uses primarily text-based materials. there are also short treatments that are not generally delivered with guidance and do not usually target people with clinical depression; although they can reach more people at a lower cost, they do so with a presumably diminished effect [ ] . globally, it seems that computerized treatments, administered with the support of a therapist, are much more effective than the unsupported ones, and that is confirmed in many open studies [ ] . patients with depressive disorders have been shown to benefit from tp, as reported by several studies in which patients' symptoms improved more in the telemental health group than in the traditional setting ones. a study evidenced how patients taking this article is protected by copyright. all rights reserved antidepressant in a high-intensity telemedicine-based collaborative care model reported less drug-related side effects than the low-intensity practice-based collaborative care patients [ ] . evidence from several studies has underlined that the use of tp for delivering mental healthcare services can improve symptoms of depression among older adults. the effects are the reduction of depression symptoms and the improvement of comorbid health conditions. a review analyzed the combination of tp with mobile apps targeting depression, anxiety disorders and substance abuse, reporting similar results [ ] . stanmore et al., in a meta-analytic review, showed that tp improves not only the patient's global cognition score, but also his/her cognitive functioning, which is often affected in depressive disorders [ ] . in pruitt and colleagues assessed patients' satisfaction in a group of military personnel with depression , who were included in a randomized clinical trial that utilized a type of psychotherapy known as behavioral activation therapy for depression (batd); this was delivered in-office and at home by videoconferencing [ ] . they discovered that patients' satisfaction was very high, with no significant difference between the two modalities, which is indicative of a comparable quality of care, and a satisfying level of contrary to popular belief, higher levels of end-of-treatment satisfaction for in-person care were detected to be mostly associated with younger age patients. conversely, the satisfaction for in-home care was more frequently associated with older and symptomatic individuals. plausible explanations for such results include stigma, symptom severity and convenience. younger patients, or those with less debilitating symptoms, may be more satisfied with traditional care treatments, while the elderly, or more symptomatic patients, may prefer a home-based care setting [ ] . it has been proposed that telemental health strategies could be a useful resource in the treatment of patients with obsessive compulsive disorder (ocd). with this kind of this article is protected by copyright. all rights reserved patients several barriers to treatment, such as logistic and financial barriers, as well as shame and fear of stigmatization or discrimination [ ] , could lead to the customary long delay from the onset of the disorder to the first treatment delivery [ ] ; telemental health therapies have the potential to overcome these barriers [ ] . in their extensive review videoconference-based cbt. all these approaches were based on the cbt principles. therapists' level of involvement ranged from direct contact through video or phone, to completely automated computer programs with no direct involvement. all telemental health interventions resulted in a decrease in ocd symptoms for active participants. they observed that the most efficient studies included contact with the therapist (higher effect size and lower dropout rate), although approaches that did not include visual or the use of tp has been tested in patients with schizophrenia as well. in their review, kasckow et al. [ ] summarized many studies involving telephone-based, internet-based and videoconference-based interventions and concluded that tp approaches were feasible with this kind of patients and that they could improve treatment outcomes. tp may help to engage these patients, to improve their medication adherence rates, to detect the exacerbation of psychotic symptoms and to prevent hospitalizations; nonetheless, only a strong patient-clinician engagement seems to guarantee an improvement in clinical and social functioning and an adequate relapse prevention [ ] . patients with schizophrenia exhibited high satisfaction levels when using telepsychiatry approaches; in fact, many patients reported feeling more comfortable using tp methods to connect with their psychiatrist and, additionally, anonymous web-based interventions may help them reduce stigmatization feelings. this article is protected by copyright. all rights reserved prompts related to follow up appointments, hygiene, physical exercise, symptom management and supporting messages [ ] . psychotherapy is now the front line, gold standard treatment for common mental health disorders in evidence-based medicine [ ] . yet, due to barriers including access, cost and stigma, estimates suggest that up to two-thirds of individuals do not access or receive evidence-based care [ ] . tp could help overcome those barriers. herbst et al. [ ] showed how, in a population of patients, an internet-based, therapist-guided cbt with exposure and response prevention led to significant reductions in the main symptoms of ocd; they also observed that this improvement remained stable throughout the -month follow-up period after treatment. the i-cbt did not inhibit a solid patient-therapist working alliance; in fact, a strong and stable patienttherapist relation was established [ ] . matsumoto et al. [ ] examined the feasibility of videoconference-delivered cbt in adult patients with mild to severe ocd, panic disorder (pd) and social anxiety disorder (sad). they determined that there was a significant reduction in symptoms for each group, with a remission rate of % for ocd, % for pd and % for sad. the therapeutic alliance significantly improved throughout the treatment and its scores were comparable to those of in-person cbt. the dropout rate was very low ( %), since % of patients completed the videoconference-delivered cbt treatment. interestingly, % of the participants affirmed to prefer videoconferencedelivered cbt to face-to-face cbt. additionally to tele-health services, in recent years digital technologies have found other secondly, the applications of virtual reality (vr) are very promising for the mental health field. vr is defined as a computer-generated simulation, a set of computer generated images and sounds, representing a place or situation with which the person can interact utilizing special electronic equipment [ ] . immersion in vr allows the patient to savor emotions he has already experienced in the real world, yet also to face dysfunctional behaviors and traumatic situations in a safe setting. it allows each individual to act as the protagonist and to regain control of his actions, to feel as the active builder of his own experience and, over time, of the changes he brings into his own life [ ] . the best applications of vr are found in ptsd, specific phobias, body-image disorders, such as nervous anorexia, and autism spectrum disorders. finally, today psychiatrists have the unprecedented opportunity to benefit from the use of artificial intelligence (ai) and, more specifically, of machine learning. this technique allows the clinicians to analyze huge amounts of data with the aim of producing a biologically founded re-classification of major psychiatric disorders. increasing evidence suggests that the sub-classifications obtained from machine learning data analysis have better predictive power of treatment outcome than the traditional dsm/icd models. in a new era of evidence-based psychiatry these objectively measurable transdiagnostic endophenotypes will allow the clinicians to make an early diagnosis, to individualize treatment and to make therapeutic adjustments to reduce disease relapses that are tailored for each individual. this has been called computational psychiatry [ ] . covid- pandemic brought to reduced social contact and impeded face-to-face interactions, also in delivering psychiatric assistance. the purpose of this paper was to review our knowledge about tp to inform its use during this pandemic, still lacking original studies addressing this topic. as shown, many are the advantages of tp, such as this article is protected by copyright. all rights reserved an improved access to care, the possibility for patients to receive psychiatric assistance in their native language, a reduction of travel time and costs and a comparable effectiveness to in-person care [ ] . however, is also important to consider some limitations of telehealth. first of all, more studies are needed to ensure a broader and more reliable validation of tp, in terms of diagnostic reliability and therapeutic efficiency in the short and long term. secondly, it is necessary to consider that tp is not suitable for all patients. moreover, some concerns remain about tp use during acute crisis: important cues such as facial expressions, tone and posture can be missed in the tp consultation. for this reason, there must be regular screenings for adverse events and a procedure for crisis intervention must be readily available [ ] . an important recommendation is to connect with on-site staff who may intervene if necessary [ ] . the privacy issue is also not to be underestimated: patients should be educated about personal privacy problems rising from using tp services outside the traditional clinical setting, like when speaking in rooms where family members can hear the conversation, or in public places where anyone might listen [ ] . the clinician himself must maintain privacy rules and ensure that no one will interrupt the session [ ] . despite the advantages and the technological progress tp is globally underused: in , only about % of psychiatrists in the united states had used tp [ ] . a recent article examining a large population of subjects from to concluded that, although telemedicine care had substantially increased during that time, its use was still not widespread in [ ] . in addition, in , the same clinicians accounted for more than half of all tele-mental health visits that year [ ] , suggesting that tp practice had been undertaken by very few, selected clinicians. during this pandemic period several authors all over the world underlined the need to promote online mental health care services and encourage their use [ , , ] . in fact, tp can capably respond to the mental health needs of people in isolation or quarantine, reducing infection risk. tp observes social distancing, avoids care interruptions and increases public health outcomes [ ] . considering that psychiatric population is more vulnerable to stress than general population, a disruption in care is concerning as it may increase the risk of symptom exacerbation and relapse. tp is precious to ensure continuity of care for these populations at higher risk of decompensation [ ] . in this regard, recently the us centers for medicare and medicaid services waived restrictions this article is protected by copyright. all rights reserved on originating sites for telehealth [ ] and the ability of healthcare professionals to prescribe remotely has been expanded to cover controlled substances [ ] . in china, in addition, artificial intelligence programs have been used as interventions for psychological crises during the pandemic. furthermore, online psychological counselling services have been widely established -h services on all days of the week [ ] . considering the experiences from past serious virus pandemics, improvement of mental health care are needed [ ] . the increasing demand for psychiatric services in the overburdened mental health care system presents a risk of creating a global public mental health crisis throughout the world [ ] . despite initial difficulties to adapt the system to tp requirements, this crisis can be an opportunity to improve the healthcare system and to expand its accessibility for patients also for the future [ ] . the review here presented has some limitations. mainly, it has been conducted in a narrative way and using only one electronic database to search for articles. this decision has been undertaken in order to rapidly review the literature to inform the unfolding pandemic situation, but possibly limited the number of references obtained. there is the need for future research to investigate the topic in a systematic way, in order to rigorously assess the contribution of tp in covid- pandemic. focused and continuous trainings of clinicians are required to increase the awareness of the benefits and risks of tp. in relation to this, and considering the present pandemic, formally teaching tp in medical schools, residency trainings and continuing medical education programs should be encouraged. among other benefits, this may be a way to help cultural acceptance of tp both among patients and medical staff. in conclusion, tp is destined to expand in the future and mental health professionals have a strategic role to help ensure that these technologies respect the therapeutic relationship and remain rigorous in their scientific foundation. accepted article mental health, risk factors, and social media use during the covid- epidemic and cordon sanitaire among the community and health professionals in wuhan, china: cross-sectional survey mental health consequences during the initial stage of the coronavirus pandemic (covid- ) in spain epidemic of covid- in china and associated psychological problems patients with mental health disorders in the covid- epidemic preventing problematic internet use during the covid- pandemic: consensus guidance problematic internet use (piu), personality profiles and emotion dysregulation in a cohort of young adults: trajectories from risky behaviors to addiction substantial undocumented infection facilitates the rapid dissemination of novel coronavirus (sars-cov- ) mental health response to the covid- outbreak in china global mental health and covid- . the lancet psychiatry a health telematics policy in support of who's health-for-all strategy for global health development: report of the who group consultation on health telematics the psychiatrist-patient relationship of the future: anytime, accepted article this article is protected by copyright. all rights reserved anywhere? a viewpoint from paris on the covid- pandemic: a necessary turn to telepsychiatry telemedicine for child collaborative or integrated care current directions in videoconferencing tele-mental health research therapeutic alliance in videoconferencing psychotherapy : a review barriers to use of telepsychiatry: clinicians as gatekeepers international review of psychiatry telepsychiatry and other technologies for integrated care : evidence base , best practice models and competencies psychotherapy for depression in older veterans via telemedicine: a randomised, open-label, non-inferiority trial is telepsychiatry equivalent to face-to-face psychiatry? results from a randomized controlled equivalence trial randomized clinical trial of telepsychiatry through videoconference versus face-to-face conventional psychiatric treatment resisting and promoting new technologies in clinical practice : the case of telepsychiatry practical issues in delivery of clinicianto-patient telemental health in an academic medical center original article ~ an evaluation of the use of interactive television in an acute psychiatric service specialist physicians' knowledge and beliefs about telemedicine: a comparison of users and nonusers of the technology. telemed e-health survey of psychologists ' telebehavioral health practices : technology use , ethical issues , and training needs the digital divide among low-income homebound older adults : internet use patterns , ehealth literacy , and attitudes toward computer / internet use navigating the digital divide : a systematic review of ehealth literacy in underserved populations in the united states the lancet commission on global mental health and sustainable development ethical standards for telemental health must be maintained during the covid- pandemic telemedicine practice: review of the current ethical and legal challenges internet-based and other computerized psychological treatments for adult depression: a meta-analysis the use of technology for mental healthcare accepted article this article is protected by copyright. all rights reserved delivery among older adults with depressive symptoms : a systematic literature review internet-based cognitive behaviour therapy for symptoms of depression and anxiety: a meta-analysis free range users and one hit wonders : community users of an internet-based cognitive behaviour therapy program smartphones for smarter delivery of mental health programs: a systematic review the effect of active video games on cognitive functioning in clinical and non-clinical populations : a meta-analysis of randomized controlled trials predicting post treatment client satisfaction between behavioural activation for depression delivered either inperson or via home-based telehealth barriers to treatment and service utilization in an internet sample of individuals with obsessive-compulsive symptoms the brown longitudinal obsessive compulsive study: clinical features and symptoms of the sample at intake no talking, just writing! efficacy of an internet-based cognitive behavioral therapy with exposure and response prevention in obsessive compulsive disorder the potential of telemental health accepted article self-help with minimal therapist contact for obsessivecompulsive disorder: a review stepped care versus standard cognitivebehavioral therapy for obsessive-compulsive disorder: a preliminary study of efficacy and costs telepsychology for posttraumatic stress disorder: a systematic review bdnf concentration and impulsiveness level in post-traumatic stress disorder telepsychiatry-based care for the treatment follow-up of iranian war veterans with post-traumatic stress disorder: a randomized controlled trial telepsychiatry in the assessment and treatment of schizophrenia treatment compliance in the long-term paranoid schizophrenia telemedicine study mhealth based interventions for the assessment and treatment of psychotic disorders : a systematic review depression in adults: treatment and management urgent need for improved mental health care and a more collaborative model of care working alliance in internet-based cognitive-behavioral therapy for obsessive-compulsive disorder internet-based cognitive behavioral therapy accepted article this article is protected by copyright. all rights reserved with real-time therapist support via videoconference for patients with obsessivecompulsive disorder, panic disorder, and social anxiety disorder: pilot single-arm trial using science to sell apps : evaluation of mental health app store quality claims a literature overview of virtual reality ( vr ) in treatment of psychiatric disorders : recent advances and limitations transforming experience : the potential of augmented reality and virtual reality for enhancing personal and clinical change review machine learning for precision psychiatry : opportunities and challenges telemedicine use in a large commercially insured population rapid growth in mental health telemedicine use among rural medicare beneficiaries, wide variation across states. health aff mobilization of telepsychiatry in response to covid- -moving toward st century access to care the role of telehealth in reducing the mental health burden from covid- telemental health in the context of a pandemic: the covid- experience available from: accepted article this article is protected by copyright dea okays telehealth to prescribe opioids amid covid- emergency online mental health services in china during the covid- outbreak. the lancet psychiatry timely mental health care for the novel coronavirus outbreak is urgently needed. the lancet psychiatry impact of human disasters and covid- pandemic on mental health: potential of digital psychiatry ijcp_ _f .pdf