Web-based Multimedia Mapping for Spatial Analysis and Visualization in the Digital Humanities: a Case Study of Language Documentation in Nepal
Web-based Multimedia Mapping for Spatial Analysis
and Visualization in the Digital Humanities: a Case Study of Language
Documentation in Nepal
Shunfu Hu1 & Brajesh Karna2 & Kristine Hildebrandt3
Published online: 17 January 2018
# Springer International Publishing AG, part of Springer Nature 2018
Abstract
There has been a growing interest in utilizing geographic information systems (GIS) in the digital humanities and social sciences (DH).
GIS-based DH projects usually emphasize spatial analysis and cartographic capability (e.g., displaying the locations of people, events,
or movements), however, GIS alone cannot easily integrate multimedia components (e.g., descriptive text, photographs, digital audio,
and video) of DH projects. Multimedia mapping provides a unique approach to integrating geospatial information in digital map format
and multimedia information, which is useful for DH integration into spatial analysis and visualization. As cartographic mapping and
GIS evolve from a traditional desktop platform to the World Wide Web, it is of significance to design and develop a Web-based
multimedia mapping approach that could carry out spatial analysis and incorporate multimedia components, which is greatly beneficial
to the DH applications. Our objectives of the language documentation research project in Nepal were to (1) use geo-tagging equipment
to collect audio and visual recordings of three types of socio-linguistic data: language attitudes and practices interviews, free-form
narratives, and elicited vocabulary and grammatical paradigm sets, from representative speakers of the four endangered languages in
twenty-six Manang villages; (2) design and develop a Web-based, interactive multimedia atlas that can display data points correspond-
ing to the speakers, links to the three types of data gathered in multimedia format, provides friendly user interface for the manipulation
and spatial analysis of all the data. It is anticipated that the Web-based, interactive, and multimedia language atlas can bring all local and
international stakeholders, such as the speech communities, linguists, local government agencies, and the public, together to raise
awareness of language structures, language practices, language endangerment, and opportunities for preservation, all through this easy-
to-use means that enhance the geo-spatial representation in engaging visual and sensory (multimedia) formats. Google Maps API and
JavaScript are employed to develop this online, interactive, and multimedia language atlas.
Keywords Multimedia mapping . Spatial analysis . Visualization . Digital humanities . Language documentation . WWW
Introduction
There has been a recent and noticeable increase in connections
between humanities and geography (including geographic
information system (GIS)), particularly in visualizations and in
projects that spatially represent historical, narrative, and textual
descriptions. This movement has been termed by Harris et al.
(2011) as the BGeohumanities.^ By Bhumanities,^ we mean spa-
tial considerations of disciplines concerned with the human con-
dition, and involving largely qualitative, introspective and spec-
ulative methods of inquiry (e.g., literature, anthropology, philos-
ophy, history, communication studies, and languages/linguistics).
The linkages between social sciences and GIS geography have
been substantive and productive, particularly in the rapidly
expanding realm of the digital humanities (DH) (cf. Goodchild
and Janelle 2004).
In such projects, mapping is used in the spatial visualiza-
tion of multimedia information, including digital still images,
sound, and video. One discipline that is poised to greatly
Electronic supplementary material The online version of this article
(https://doi.org/10.1007/s41651-017-0012-4) contains supplementary
material, which is available to authorized users.
* Shunfu Hu
shu@siue.edu
1 Department of Geography, Southern Illinois University
Edwardsville, Edwardsville, IL, USA
2 Department of Computer Sciences, Southern Illinois University
Edwardsville, Edwardsville, IL, USA
3 Department of English Language and Literature, Southern Illinois
University Edwardsville, Edwardsville, IL, USA
Journal of Geovisualization and Spatial Analysis (2018) 2: 3
https://doi.org/10.1007/s41651-017-0012-4
http://crossmark.crossref.org/dialog/?doi=10.1007/s41651-017-0012-4&domain=pdf
https://doi.org/10.1007/s41651-017-0012-4
mailto:shu@siue.edu
benefit from a deeper collaboration with GIS is that of linguis-
tics, particularly applied dimensions such as typology, socio-
linguistics, historical linguistics, and language documentation
and description. Examples of such notable interdisciplinary
collaboration include the AUTOTYP linguistic typology pro-
ject (http://www.spw.uzh.ch/autotyp/), Der sprechende
Sprachatlas BThe Speaking Language Atlas^ (http://
sprachatlas.bayerische-landesbibliothek-online.de), and The
Linguistic Atlas of the Middle and Atlantic States (http://us.
english.uga.edu/lamsas/) (Kirk and Kretzschmar 1992).
Recent geolinguistics publications also reflect this shift in col-
laborative momentum (cf. Auer and Schmidt (2009), Lameli
et al. (2011), and Gawne and Ring (2016)).
At the same time, in a recent paper on sampling in dialec-
tology research, Buchstaller and Alvanides lament that until
recently, BThe majority of sociolinguistic work [could] be de-
scribed as spatially naïve, using geographical space merely as
a canvas…on to which the results of linguistic analysis [could]
be mapped.^ (2013: 96). This need for inclusion and testing of
different types of spatial factors alongside social ones is in-
creasingly being addressed in regions like the USA and Great
Britain (Trudgill 1974; Auer and Schmidt 2009; Lameli et al.
2010; Buchstaller and Corrigan 2011; Cheshire et al. 1989,
1993; Labov et al. 2006; Kretzschmar 1996; Kretzschmar
et al. 2014; Britain 2009 and also the rise of Bgeohumanities^
Dear et al. 2011), but is still in its infancy in other regions of
the world (but cf. Stanford 2009 for a report on adjusted
spatial factors on language practices and tonal patterns of
Sui communities in China).
Compounding this general gap, precious few quantitative
studies have investigated language attitudes and practices in
linguistically diverse areas. This gap is unfortunate because it
is often these attitudes and reports of language practices that
can shed light on shifting ideologies as precursors to endan-
germent in areas where languages compete among each other
and with prestige varieties (Giles et al. 1977; Coupland et al.
2006). In the Nepal scenario, two of the languages in our
larger documentation project (Manange and Gurung) are
threatened but viable, while the other two (Nar-Phu and
Gyalsumdo) are highly endangered (technically moribund),
with very few active younger speakers. Our overall aim, there-
fore, is to build an online, interactive atlas that contributes
towards what Britain (2009: 142) terms Bsocially rich
spatiality,^ taking into account speaker practices and networks
as they intersect with geo-physical location.
The potential for mutual benefit in the GIS-linguistics and
GIS-language documentation collaborative contexts cannot
be overstated and is the focus of this paper. Increasingly, lan-
guage documentation (particularly of vulnerable or threatened
speech communities) relies on an awareness and understand-
ing of the spatial-temporal interplay of language practices,
structural variation, and contact dynamics, all working togeth-
er to form a more comprehensive profile of the contributing
variables to the survival and threat scenarios of these lan-
guages. Spatial visualization of documentation, through maps
and atlases, for example, also benefits grammatical descrip-
tion in itself as a product or output, as grammars vary widely
in their coding and conceptualization of space-time continua
(e.g., Slobin 1996; Bickel 1997; Harrison 2008). This varia-
tion can be more deeply appreciated in tandem with GIS ap-
plications as relevant to the language communities.
GIS is defined as a computer program for the capture, stor-
age, manipulation, visualization, and spatial analysis of
geospatial features (e.g., points, lines, or polygons) and their
attributes (Chang 2015). The attribute data of the geospatial
features in GIS are commonly alphanumeric values stored in
an attribute table, thus termed as structured data. Therefore,
the spatial analysis of GIS is often conducted using a struc-
tured query language (e.g., BCountry_Name^ = BNepal^). As
a result, GIS has traditionally lacked the ability to integrate
non-structured data, such as digital photographs, digital audio
and digital video clips (i.e., multimedia components). In the
past two decades, a new trend of developing multimedia map-
ping systems has been seen in the literature. Multimedia map-
ping refers to the integration of computer-assisted mapping
systems and multimedia technologies that allow one to incor-
porate not only geospatial information in digital map format,
but also multimedia information. The development of multi-
media mapping techniques has gone through several stages,
including the emergence of interactive maps and electronic
atlases (Openshaw and Mounsey 1987; Rhind et al. 1988;
Shepherd 1991), the development of Bhypermaps^ in the
1990s (Wallin 1990; Laurini and Milleret-Raffort 1990;
Cotton and Oliver 1994; Cartwright 1999), the integration of
hypermedia GIS systems (which features hypertext, hyper-
links and multimedia) and GIS in the late 1990s and early
2000s (Shiffer 1998; Bill 1998; Hu 1999; Soomro et al.
1999; Chong 1999; Hu et al. 2003; Yagoub 2003;
Goryachko and Chernyshev 2004; Belsis et al. 2004). After
having compared the various multimedia mapping techniques,
ranging from desktop-based multimedia mapping to Web-
based hypermedia GIS, Hu (2012) pointed out that the former
relies heavily on computer programming languages (e.g.,
Visual Basic), and digital mapping software (e.g., ArcView,
ArcGIS, or MapObjects) with local data storage, local access,
and single user. The media format is often in Microsoft
Windows with large file sizes, such as .tiff for images, .avi
for digital video, and .wav for digital sound. The latter relies
on both computer programming languages (e.g., Visual Basic)
and markup languages (e.g., HTML), and Internet map server
(IMS) (e.g., MapObjects IMS, ArcGIS IMS) with remote data
storage, network access, and Internet users. The media format
is often Web-based with small file sizes, such as jpeg for
images, .mov for digital video, and .wav for digital sound. In
both cases, the multimedia map application developers must
invest in dedicated computer hardware (e.g., Web server, data
3 Page 2 of 14 J geovis spat anal (2018) 2: 3
http://www.spw.uzh.ch/autotyp
http://sprachatlas.bayerische-landesbibliothek-online.de
http://sprachatlas.bayerische-landesbibliothek-online.de
http://us.english.uga.edu/lamsas
http://us.english.uga.edu/lamsas
server) and computer software (e.g., map server), plus IT per-
sonnel. The developer often faces a steep learning curve to
become knowledgeable about the coding in native language of
the map server. Now, as cartographic mapping system and
GIS evolve from traditional desktop platform to Web-based
online platform, there is a need and an opportunity, to develop
a Web-based multimedia mapping approach to integrating
geospatial information in digital map formats and multimedia
information, which is of significance for DH-centered visual-
ization and spatial analysis.
Our objectives in our language documentation research in
Nepal were to (1) use geo-tagging equipment to collect audio
and visual recordings of three types of socio-linguistic data:
language attitudes and practices interviews, free-form narra-
tives, and elicited vocabulary and grammatical paradigm sets,
from representative speakers of the four endangered lan-
guages in 26 Manang villages; (2) design and develop a
Web-based, interactive, and multimedia atlas that can display
data points corresponding to the speakers, links to the three
types of data gathered in multimedia format, provides friendly
user interface for the manipulation and spatial analysis of all
the data. It is anticipated that the online atlas can bring all local
and international stakeholders, such as the speech communi-
ties, linguists, local government agencies, and the public, to-
gether to raise awareness of language structures, language
practices, language endangerment, and opportunities for pres-
ervation, all through this easy-to-use means that enhance the
geo-spatial representation in engaging visual and sensory
(multimedia) formats. The following section describes our
methods and the atlas design and functionality.
Methodology
Study Area
As alluded to, Nepal has a high degree of linguistic diversity,
with approximately 100 languages attested (CBS 2012;
Kansakar 2006). Most are Btribal^ languages (indigenous, lo-
cally bounded, and strongly connected to community cultural
identification and practices), concentrated in a couple of vil-
lages over a small area. As an example, the Manang District is
both culturally and linguistically heterogeneous and can be
divided into four ethno-linguistic areas across 26 villages:
Manang Gurung and Gyalsumdo to the south (where
Manang Gurung and Gyalsumdo speakers live), the Nar val-
ley to the north (where Nar-Phu speakers live), and the upper
Manang valley in the west (where Nyeshangte speakers live)
(Fig. 1). All languages in this area are Tibeto-Burman.
The Manang District is appropriate for a case study of Web-
based multimedia mapping as it intersects with geo-linguistics
and language documentation and it has undergone rapid envi-
ronmental, economic, and infrastructure development and
changes over the past 15 years, including the ongoing con-
struction of its first motorable road and the population shifts
associated with this (see Laurance 2014 for commentary on
road-building impacts). Some Manang communities have also
witnessed population movements associated with both the rise
of boarding schools in the capital Kathmandu, and also the
rise of migrant worker opportunities where young adults relo-
cate to Gulf States like Saudi Arabia, Bahrain, and United
Arab Emirates for long-term employment (Hildebrandt et al.
2015). These changes have mixed impacts. On the one hand,
they can benefit rural communities by connecting them to
business and other opportunities available only to more cen-
trally located marketplaces. On the other hand, these changes
can trigger language shift as local residents (particularly youn-
ger ones) may emigrate away from their areas of traditional
language practice for education and job opportunities. These
changes introduce new, complex variables behind language
contact and language endangerment beyond just social vari-
ables, and further motivates a spatial perspective of language
practices and patterns in this area.
Data Source and Data Set
The primary data source was the sociolinguistic interviews
conducted in the 26 villages of the Manang District across
the four language groups.1 There was a total of 87 interviews
conducted between 2012 and 2014. Each interviewee was
asked a total of 61 questions (see Appendix for the full set
of questions). The data sets for this project included the de-
scriptive text from the interviews gathered from locally orig-
inating and residing speakers of the four languages, geograph-
ic coordinates in longitude (x) and latitude (y) of the residences
where each interview took place, digital photographs, and
digital video clips. Digital photographs were taken using
Cannon SLR 40-D digital camera and saved in JPEG format.
Digital video clips were acquired using a Sony Handycam
HDR-XR550 digital video camcorder and stored in MPEG
format. As part of the project agreement with the funding
agency, the multimedia content from this project were ar-
chived with the SHANTI (Sciences, Humanities, and Arts
Network of Technological Initiatives) Collection (http://
shanti.virginia.edu/wordpress/?page_id=414). SHANTI is
housed in the University of Virginia as a publisher of
websites and other digital content on languages and cultures
of the Tibetan Plateau and greater Himalayan region. In
addition, Google Maps was employed as the base map for
the integration or Bmashup^ of the multimedia information
1 All interviews began with an oral consent process (originally composed in
English and given in Nepali, the regional contact language), which was based
on a script approved by SIUE’s Institutional Review Board (IRB) for informed
consent in research involving human subjects. This consent process included
respondent awareness that his/her information would be made available for
public access, through audio-visual and through still (photograph) images.
J geovis spat anal (2018) 2: 3 Page 3 of 14 3
http://shanti.virginia.edu/wordpress/?page_id=414
http://shanti.virginia.edu/wordpress/?page_id=414
collected during the sociolinguistic interviews. The next
sections describe the process of data preparation (BUse of
XML to Store the Data^ section), data loading and display
on Google Maps (BUse of jQuery JavaScript Library to
Load the XML File onto the Google Maps^ and BUse of
Google Maps JavaScript API V3 to Display the Data^
sections , respectively), and how the user interface for spatial
analysis and visualization was developed (BUse of JavaScript,
XHTML, and CSS to Design the User Interface and Functions
for Spatial Analysis and Visualization^ section).
Use of XML to Store the Data
There are various ways to store and prepare the spatial data for
display on Google Maps. Among them, XML is the simplest
and easiest method to use due to its free and open source.
XML (Extensible Markup Language, stored with the exten-
sion .xml in plain text format) is similar to HTML but does not
have any predefined tags and is platform independent. The
authors defined the tags based upon project-specific require-
ments. Below is an example of the XML file (e.g.,
Languages_pts_2016.xml) that contains the information for
one sociolinguistic interview, including the speaker ID, the
language name, the village name, the coded interview ques-
tions and responses (e.g., Aindex, Bindex), interviewee age
group, longitude (x), latitude (y), picture ID, video link, and
other information. For each interview, all relevant information
was placed in a pair of and tags, each representing
the point location of the interview. There are currently 87 pairs
of such tags in the entire XML file.
Notepad++, a free source code editor which supports sev-
eral programming languages running under the Microsoft
Windows environment (http://notepad-plus-plus.org/), was
used to prepare the XML data file. It was also used to edit
all of the JavaScript code for the project. For those using
different platforms, such as macOS or Linux, there are
numerous open-source editors, such as Sublime Text
and Atom.
Fig. 1 Locations of 26 villages in
the Manang District of Nepal
3 Page 4 of 14 J geovis spat anal (2018) 2: 3
http://notepad-plus-plus.org
Use of jQuery JavaScript Library to Load the XML File
onto the Google Maps
jQuery is a cross-browser JavaScript library designed to sim-
plify the client-side scripting of HTML. It is a free, open-
source software designed to create dynamic Web pages.
Therefore, jQuery (version 2.1.4) is selected in our applica-
tion. There are two ways to integrate jQuery library to the
application. One way is to download the jQuery library and
place it at the same place where the main Web page (e.g.,
index.html) is located, which is illustrated with the following
code in HTML.
Another way is simply to point to the URL where the
jQuery library is located, as shown below.
The code below shows how the XML content (i.e.,
Languages_pts_2016.xml) is uploaded to the Google
Maps using JavaScript jQuery.get function ($.get in short)
at the initialization of the Google Maps (i.e., function
initMap()) and jQuery(data).find(Bpt^).each(function(){ });
is used to retrieve the information for each data point
(notice the and tags in the xml file mentioned
above). A variable, xmldoc, is declared to withhold the
information (e.g., language_name, village_name, picid,
and video) for each point. Another variable, latlng, is de-
clared to withhold the locational information (i.e., longi-
tude (x) and latitude (y)).
Fig. 2 Conceptual framework for
the integration of the Google
Maps API, XML, and HTML for
Web-based multimedia mapping
J geovis spat anal (2018) 2: 3 Page 5 of 14 3
Use of Google Maps JavaScript API V3 to Display
the Data
The launch of Google Maps in 2005 has revolutionized
Web mapping service applications on the Internet.
Based on Asynchronous JavaScript and XML (AJAX),
a new type of client/server interaction was introduced in
Google Maps to maintain a continuous connection be-
tween the client and the server for immediate
downloading of additional map information (Peterson
2008). In addition to implementing a better client/
server interaction, Google also provides programmers
free access to its code in the form of an Application
Programming Interface (API). In other words, the API
consists of a set of routines or functions that can be
called by a programmer using JavaScript. Linking the
Web page with the Google Maps API is straightforward
in version 3, using a single HTML, shown below:
This is a standard HTML directive to include an external
JavaScript file, served by maps.google.com. This element is
added to the …
section of the Web page where
the map is loaded.
The next key step is to initialize the API and load the map
onto the Web page as follows. First, we created a initMap()
function to launch the Google Maps. In the initMap() func-
tion, we defined the center of the map to be displayed on the
Fig. 3 Initial launch of the Web-based multimedia atlas, which displays the three tabs, the Google Maps, the locations of all 87 sociolinguistic interviews
across four languages in Manang, Nepal, and the map legend
3 Page 6 of 14 J geovis spat anal (2018) 2: 3
http://google.com
Web page at 28.576671° N, 84.257245° E. Another function,
pushMapData(), was developed to load all data points from
the xml to the map. A third function, AddMapLegend, was
developed to load the map legend.
Fig. 4 Search result from Languages by Village tab: in this case, the user selected Chame village. There were six sociolinguistic interviews conducted in
Chame village. The marker icons on the map also indicate Gyalsumdo and Gurung as the mother tongue languages as reported in the interviews
J geovis spat anal (2018) 2: 3 Page 7 of 14 3
Use of JavaScript, XHTML, and CSS to Design the User
Interface and Functions for Spatial Analysis
and Visualization
In the design of the Web-based multimedia digital atlas,
the layout design was adopted. There are three rows. The
first row contains three tabs, the second row is the map
container and the third row contains map legends. The
three tabs serve as a user interface for spatial analysis
and visualization. The first tab is named Languages by
Village. With this tab, the user can search for the lan-
guage identified by interviewees as their mother tongue.
To do so, the user can first select a village from a
dropdown list of all the villages where linguistic inter-
views were conducted, and click the Search button. The
Google map is zoomed into that selected village and
customized marker icons that represent locations where
the sociolinguistic interviews were conducted in that vil-
lage are shown on the map. In this case, there are four
customized marker icons for four different languages—
the green balloon for BNyeshangte,^ the yellow balloon
for BManang Gurung,^ the purple balloon for BNar-Phu^
and the orange balloon for BGyalsumdo.^ In addition,
tooltips (e.g., language name) to the markers are provid-
ed. In addition, these marker icons are clickable. When
the user clicks on an icon, it will launch a Google Maps
API standard Info-window in which the speaker ID, age,
mother tongue, residential picture, and a link to the video
clip of associated recordings will be displayed.
The second tab is named Villages by Language. With
this tab, the user can search for all the villages in
Manang District where a selected language is spoken.
To do so, the user can first select a language from the
dropdown list of the four languages, and click the Search
button. The search result will be displayed on the map as
customized marker icons. Similarly, each marker icon is
clickable. The third and last tab is Question & Response.
With this tab, the user can select an interview question
from the question dropdown list (see Appendix for the
entire list of the sociolinguistic survey questions), and
click the Search button. The search results will be
displayed on the map as different marker icons, each
representing an answer to that question. This function
provides the user a tool to further examine the spatial
distribution of language practices and attitudes in
Manang languages.
Fig. 5 Search result from Villages by Language tab: in this case, the user selected Nyeshangte language, all locations of interviews with this language are
displayed on the map
3 Page 8 of 14 J geovis spat anal (2018) 2: 3
Inside second row is the map container where the
Google Maps is displayed. A few standard Google Maps
controls, such as Pan and Zoom controls, Map Scale con-
trol, and Map Type control–Roadmap and Satellite, are
available for the user to interact with the map or satellite
imagery. The last row is where the map legend is placed. If
the user selects the first tab or the second tab, the map
legend contains four customized marker icons that repre-
sent four different languages; if the user selects the third
tab, the map legend contains customized marker icons that
represent different responses for the interview questions.
Finally, all the pieces were assembled. JavaScript is
the native language of Google Maps. In addition,
Google Maps is built from HTML and formatted with
CSS (Cascading Style Sheet). Therefore, JavaScript,
HTML, and CSS are used to develop the functionali-
ties. These include creation of a user interface in the
form of tabs using bootstrap and Ajax, uploading XML
data file via jQuery, displaying points of locations for
the sociolinguistic interviews with customized marker
icons via Google Maps API, and providing the spatial
analysis functions via Google Geocoder. Figure 2 illus-
trates a conceptual framework for the integration of the
Google Maps API and other JavaScript libraries in the
World Wide Web environment.
In Fig. 2, the workflow begins with a user-initiated request
for the URL of our project website: https://mananglanguages.
isg.siue.edu/atlas/. With the request, the server finds the page
from the requested location and loads the default page, in this
case, Index.html. The server then sends the response back to
the user with the content of the project page. We are using
JavaScript, which is client-sided in its scripting, so all of the
code execution takes places in the client browser and reduces
the load on the server. This makes our server fast enough to
process additional incoming requests. On the client side, the
server loads the JQuery Library, including the Google Map
API. It then loads a custom written JavaScript, which sends
request to the server to retrieve and load the data from our
XML into the background. This makes the script dynamic
and more responsive to the user. When all processes are exe-
cuted completely, the user can see the map, with various mark-
er icons plotted on the map with respective legends.
Fig. 6 Search result from the Question & Response tab. Here, the user selected the question BDo you think there will be child learners of your mother
tongue in future?^ The different marker icons on the map indicate different answers from the interviewees
J geovis spat anal (2018) 2: 3 Page 9 of 14 3
https://mananglanguages.isg.siue.edu/atlas/
https://mananglanguages.isg.siue.edu/atlas/
Results
The use of Google Maps API V3 provides an efficient and
familiar mechanism to deliver digital cartographic information
to Internet users with a user-friendly interaction. With Google
Maps Standard Map Type control, the user is able to choose
one of the two map types: Roadmap and satellite imagery.
Figure 3 shows an outlook of the online map service for our
language documentation project in Google Chrome. At the
initial launch of the Web page, locations of the sociolinguistic
interviews are displayed within the map container. Notice also
that in Fig. 3, the three tabs are visible: Languages by Village,
Villages by Language, and Questions & Responses. Figure 4
demonstrates the result when the user selects a village, for
instance, Chame village, from the drop-down list and then
clicks the Search button to show all the sociolinguistic inter-
views conducted in that village, six in this example. The dif-
ferent icons on the map also indicate the type of Tibeto-
Burman languages, Gyalsumdo and Gurung identified as
mother tongues by the interviewees in Chame. Figure 5 dem-
onstrates the result when the user searches for all the villages
where one, two, three, or four selected languages is spoken.
Figure 6 demonstrates the result when the user chooses the
Questions & Responses tab, selects a question from the drop-
down list of all the questions, and clicks the Search button.
Figure 7 illustrates the Info-window when the user clicks on a
marker icon on the Google Map. The Info-window displays
multimedia information related to the interview conducted at
the marker location, including the speaker id, age, mother
tongue, village name, a link to download the transcript of
the questions and responses, and a hyperlink to the video clip
of the interview. Figure 8 illustrates a sample video clip that
may be played, and which is housed in the SHANTI
Collection at the University of Virginia.
Discussion
BMethodology^ and BResults^ sections detailed the workflow
and elements of the BDocumenting the Languages of Manang,
Nepal^ atlas. However, this multimedia atlas is not the only
example of ongoing efforts use geospatial technology in the
digital humanities, particularly in geolinguistics and online,
interactive language mapping. We summarize here examples
Fig. 7 When the user clicks on a marker icon on the Google Map, an
Info-window is launched to display all relevant information, including the
speaker id, age, mother tongue, village name, a link to download the
transcript of the questions and responses, and a hyperlink to the video
clip of the interview (see Fig. 8)
3 Page 10 of 14 J geovis spat anal (2018) 2: 3
of three projects with parallels to the Manang Languages
Atlas, their similarities, and their differences.
First, Saint Mary’s University uses online multimedia map-
ping for the Mi’kmaw Place Names Digital Atlas (URL: http://
sparc.smu.ca/mpnmap/). This project makes use of ArcGIS
maps and not Google Maps. Additionally, Adobe Flash is
required to load the content of the map in the browser.
ArcGIS is not open source and can be expensive for some
programs. Similarly, Flash is well known for causing spikes
in CPU usage and comes with security holes. Our project
uses JavaScript and HTML, and the data are encoded on an
open-source XML that uses fewer computer resources.
Fig. 8 Top: A video clip of a
Gurung speaker that is linked
from the Info-window on the atlas
(Fig. 7) in which the speaker was
describing his apple orchard using
Gurung language; Bottom:
Language transcript. The first line
is the local language (Gurung),
transcribed in the IPA
(International Phonetic
Alphabet). The 2nd line is the
Nepali free translation. The 3rd
line is the free translation in
English. The audio was tran-
scribed in ELAN (download is
free at https://tla.mpi.nl/tools/tla-
tools/elan/) and the language
transcript is synchronized with the
video, which was done by the
THL (Tibetan Himalayan
Library) using a Drupal platform
J geovis spat anal (2018) 2: 3 Page 11 of 14 3
http://sparc.smu.ca/mpnmap/
http://sparc.smu.ca/mpnmap/
https://tla.mpi.nl/tools/tla-tools/elan/
https://tla.mpi.nl/tools/tla-tools/elan/
Second, Language Landscape is a nonprofit British orga-
nization that uses maps to plot languages as they are used
around the world (URL: http://languagelandscape.org). The
project directors encourage people to get involved in the
project by adding their own audio clips to the project,
which are then displayed visually and interactively. This
project uses the Google Map API for the map, JavaScript,
and HTML. This map includes pins which display
information in audio format which can be played in real
time without page redirects. Our project similarly displays
map pins related to different languages and on click, and
also provides information like speaker name, age, village.
Like Language Landscape, our project also includes links to
other resources.
Third, Atlas of Pidgin and Creole Language Structures
project spatially represents pidgin and creole languages, and
their structures, from around the world (URL: http://apics-
online.info/contributions#2/30.3/10.0). The map is
developed using JavaScript and Leaflet JavaScript library
for mobile friendly interactive maps. More information
about the Leaflet JavaScript library can be found at (http://
leafletjs.com/index.html). The library is open source and
lightweight in terms of CPU usage. Pins are plotted to the
map with various colors and also display an Info-window
with hyperlinks for additional information on a given lan-
guage. This map lacks legends, which results in the need to
click on each pin to learn more about language types.
Conclusion
This paper has demonstrated a new and relatively easy ap-
proach called Web-based multimedia mapping that could in-
tegrate not only geospatial data but also multimedia data in
the form of digital photographs, digital sound, and digital
video clips—which is very useful for digital humanities and
social sciences. The use of Google Maps API and JavaScript
libraries allow us to employ the digital maps and satellite
imagery from Google Maps with only a few JavaScript codes.
The apparent benefit of using existing Maps APIs may be of
value to those who do not have the resources to invest in a
dedicated computer for map servers and data servers. We
used only a Web host account provided by the home institu-
tion to upload the atlas Web page (index.html, 48 kilobytes)
and the xml file (languages_pts.xml, 30 kilobytes). The latter
stored only the Bpointers^ to the digital multimedia files that
are related to the social linguistic interviews and narratives
collected for the project (archived in SHANTI). SHANTI
offers the state-of-the-art technology to deliver the multime-
dia content with the highest quality at a faster speed (e.g.,
video streaming) and is associated free of charge to this pro-
ject. It is worth mentioning that for those who do not have
their own video streaming service available, free video
streaming professional services are provided by third parties,
such as YouTube. In addition to the Google Maps API,
Yahoo! Maps APIs and Bing Maps APIs can also be used
for the same purposes. It is also beneficial for college students
to start off with multimedia mapping technology and for or-
ganizations, such as the university in this case study, to see
the benefits of the new technology.
The Web-based, interactive, and multimedia language atlas
can be accessed on the Internet; therefore, it can bring all
local and international stakeholders, such as the speech com-
munities, linguists, local government agencies, and the pub-
lic, together to raise awareness of language structures, lan-
guage practices, language endangerment, and opportunities
for preservation, all through this easy-to-use means that en-
hance the geo-spatial representation in engaging visual and
sensory (multimedia) formats.
One example of this potential contribution may be found
in Hildebrandt and Hu (2017) which, through quantitative
analysis of spatial distributions of respondent answer types,
demonstrates that non-structural (language attitude and use)
variables reveal different degrees of vitality vs. endangerment
in Nepal. The Web-based multimedia mapping approach of-
fers a unique tool for a spatial analysis and visualization of
variations in self-reported attitudes and practices across the
four Manang languages, with adjusted spatial factors (e.g.,
location of communities to a newly built motor road, location
of communities to the district headquarters, location within
closely clustered communities) alongside traditional social
factors (e.g., gender, age, education, occupation, and so on).
As a result, this research project contributes to new under-
standings of the relationship between space and language
practices in Nepal.
Acknowledgements This work is supported by the National Science
Foundation’s Division of Behavioral and Cognitive Sciences -
Documenting Endangered Languages (funding no. 1149639):
BDocumenting the Languages of Manang^ and by an equipment support
grant from SIU Edwardsville. We are grateful to members of the Gurung,
Gyalsumdo, Manange and Nar-Phu-speaking communities in Manang,
Nepal, for their help in gathering these data. We are grateful to Dubi
Nanda Dhakal, Oliver Bond, Sangdo Lama, and Ritar Lhakpa Lama for
assistance with interviews. We are grateful to Saita Gurung and Manisha
Chaudhary for assistance with atlas construction and development. All
errors are the responsibility of the authors.
Funding This research was supported by the US National Science
Foundation’s Division of Behavioral and Cognitive Sciences-
Documenting Endangered Languages (funding no. 1149639):
BDocumenting the Languages of Manang^ and by an equipment support
grant from SIU Edwardsville.
Compliance with ethical standards The authors of this paper
will agree, accept, and comply with all the ethical standards set by the
journal.
Conflict of Interest The authors declare that they have no conflict of
interest.
3 Page 12 of 14 J geovis spat anal (2018) 2: 3
http://languagelandscape.org/
http://apics-online.info/contributions#2/30.3/10.0
http://apics-online.info/contributions#2/30.3/10.0
http://leafletjs.com/index.html)
http://leafletjs.com/index.html)
Ethical Approval The field data collection (i.e., field interviews) was
approved by the SIUE’s Institute Research Board. The authors used an
approved oral informed consent script for data collection.
Informed Consent The authors have given the informed consent to pub-
lish this article in the Journal of Geovisualization and Spatial Analysis if
accepted.
References
Auer P, Schmidt J (eds) (2009) Language and space: an international
handbook of linguistic variation. Mouton de Gruyter, Berlin.
https://doi.org/10.1515/9783110220278
Belsis P, Gritzalis S, Malatras A, Skourlas C, Chalaris I (2004).
Enhancing knowledge management through the use of GIS and
multimedia. Lecture Notes in artificial intelligence (Subseries of
Lecture notes in computer science), vol 3336. pp 319–329
Bickel B (1997) Spatial operations in deixis, cognition, and culture:
where to orient oneself in Belhare. In: Nuyts J, Pederson E (eds)
Language and conceptualization. Cambridge University Press,
Cambridge, pp 46–83. https://doi.org/10.1017/CBO97811390
86677.003
Bill R (1998) Multimedia-GIS concepts and applications. Geo-Inf Syst
11:21–24
Britain D (2009) Language and space: The variationist approach. In: Auer
P and Schmidt JE (eds) Language and space: An international hand-
book of linguistic variation, vol 1. Mouton de Gruyter, Berlin, pp.
142–162
Buchstaller I, Corrigan KP (2011) How to make intuitions succeed:
Testing methods for analyzing syntactic microvariation. In:
Maguire W and McMahon A (eds) Analyzing variation in English:
what we know, what we don’t, and why it matters. Cambridge
University Press, Cambridge, pp. 30–48
Buchstatller I, Alvanides S (2013) Employing geographical principles for
sampling in state of the art dialectological projects. J Linguist Geogr
1:96–114
Cartwright W (1999) Development of multimedia. In: Peterson MP,
Gartner G (eds) Multimedia cartography. Springer, New York, pp
11–30. https://doi.org/10.1007/978-3-662-03784-3_2
Central Bureau of Statistics (2012) National population and housing cen-
sus 2011. Government of Nepal National Planning Commission
Secretariat
Chang KT (ed) (2015) Introduction to geographic information systems.
McGraw-Hill Education, New York
Cheshire J, Viv E, Henrik M, Bert W (eds) (1989) Dialect and education.
Multilingual Matters, Clevedon/Philadelphia
Cheshire J, Viv E, Pamela W (1993) Non-standard English and dialect
levelling. In: Milroy J, Milroy L (eds) The grammar of English
dialects in the British Isles. Longman, London, pp 53–96
Chong AK (1999) Orthoimage mapping bases for hybrid, multimedia,
and virtual environment GIS. Cartography 28(1):33–41. https://doi.
org/10.1080/00690805.1999.9714298
Cotton B, Oliver R (1994) The Cyberspace Lexicon—an illustrated dic-
tionary of terms from multimedia to virtual reality. Phaidon Press
Ltd., London
Coupland N, Hywel B, Betsy E, Garrett P (2006) Imagining Wales and
the Welsh language. J Lang Soc Psychol 25(4):351–376. https://doi.
org/10.1177/0261927X06292803
Dear M, Ketchum J, Luria S, Richardson D (eds) (2011) Geohumanities:
art, history, text at the edge of place, 226–240. Routledge, London
Gawne L, Ring H (2016) Mapmaking for language documentation and
description. Lang Doc Conserv 10:188–242. http://nflrc.hawaii.edu/
ldc/. Last Accessed 28 Nov 2016
Giles H, Taylor D, Bourhis R (1977) Dimensions of Welsh identity. Eur J
Soc Psychol 7:29–39
Goodchild MF, Janelle DG (eds) (2004) Spatially integrated social sci-
ence. Oxford University Press, New York
Goryachko VV, Chernyshev AV (2004) Multimedia and GIS-
technologies in atlas mapping. Vestnik Moskovskogo Universiteta
5(2):16–20
Harris TM, Bergeron S, Rouse LJ (2011) Humanities GIS: place, spatial
storytelling and immersive visualization in the humanities. In: Dear
M, Ketchum J, Luria S, Richardson D (eds) Geohumanities: art,
history, text at the edge of place. Routledge, London, pp 226–240
Harrison KD (2008) When languages die: the extinction of the world’s
languages and the erosion of human knowledge. Oxford University
Press, New York
Hildebrandt K, Dhakal DN, Bond O, Vallejo M & Fyffe A (2015) A
sociolinguistic survey of the languages of Manang, Nepal: co-
existence and endangerment. NFDIN J 14.6:104–122. http://www.
nfdin.gov.np/securi/. Last Accessed 28 Nov 2016
Hildebrandt KA, Hu S (2017) Areal analysis of language attitudes and
practices: A case study from Nepal. Language Documentation and
Conservation. Special Publication 13: 152-179, http://scholarspace.
manoa.hawaii.edu/handle/10125/24753
Hu S (1999) Integrated multimedia approach to the utilization of an
Everglades vegetation database. Photogramm Eng Remote Sens
65(2):193–198
Hu S (2012) Multimedia Mapping on the Internet Using Commercial
APIs. In Peterson MP (ed.) Online maps with APIs and
Mapservices, 61–71, Springer
Hu S, Gabriel AO, Bodensteiner LR (2003) Inventory and characteristics
of wetland habitat on the Winnebago Upper Pool Lakes, Wisconsin,
USA: an integrated multimedia-GIS approach. Wetland 23(1):82–
94. https://doi.org/10.1672/0277-5212(2003)023%5B0082:
IACOWH%5D2.0.CO;2
Kansakar TR (2006) Research on the typology of Nepal’s languages.
Nepalese linguist 22:106–128. http://www.digitalhimalaya.com/
collections/journals/nepling/. Last Accessed 28 Nov 2016
Kirk JM, Kretzschmar WA (1992) Interactive linguistic mapping of dia-
lect features. Lit Linguist Comput 7(3):168–175. https://doi.org/10.
1093/llc/7.3.168
Kretzschmar WA (1996) Quantitative areal analysis of dialect features.
Lang Var Chang 8(01):13–39. https://doi.org/10.1017/S09543945
00001058
Kretzschmar W, Juuso I, Bailey CT (2014) Computer simulation of dia-
lect feature diffusion. J Linguist Geogr 2(01):41–57. https://doi.org/
10.1017/jlg.2014.2
Labov W, Ash S, Boberg C (2006) Atlas of North American English:
phonology and phonetics. Mouton de Gruyter, Berlin. https://doi.
org/10.1515/9783110167467
Lameli A, Kehrein R, Rabanus S (eds) (2010) The handbook of language
mapping. Mouton de Gruyter, Berlin
Laurance WF (2014) Roads benefit people but can have massive envi-
ronmental costs. National Geographic (online). https://blog.
nationalgeographic.org/2014/10/19/roads-benefit-people-but-can-
have-massive-environmental-costs/. Accessed 5 January 2018
Laurini R, Milleret-Raffort F (1990) Principles of geomatic hypermaps.
In: Proceedings of the 4th International Symposium on Spatial Data
Handling, Zurich, Switzerland, 2, 642–655
Openshaw S, Mounsey H (1987) Geographic information systems and
the BBC’s Domesday interactive videodisk. Int J Geogr Inf Syst
1(2):173–179. https://doi.org/10.1080/02693798708927802
Peterson MP (2008). International Perspectives on Maps and the Internet:
An Introduction. In Peterson MP (ed.), International Perspectives on
Maps and the Internet (pp. 3–10), Springer
Rhind DP, Armstrong P, Openshaw S (1988) The Domesday machine: a
nationwide geographical information system. Geogr J 154(1):56–
58. https://doi.org/10.2307/633476
J geovis spat anal (2018) 2: 3 Page 13 of 14 3
https://doi.org/10.1515/9783110220278
https://doi.org/10.1017/CBO9781139086677.003
https://doi.org/10.1017/CBO9781139086677.003
https://doi.org/10.1007/978-3-662-03784-3_2
https://doi.org/10.1080/00690805.1999.9714298
https://doi.org/10.1080/00690805.1999.9714298
https://doi.org/10.1177/0261927X06292803
https://doi.org/10.1177/0261927X06292803
http://nflrc.hawaii.edu/ldc
http://nflrc.hawaii.edu/ldc
http://www.nfdin.gov.np/securi
http://www.nfdin.gov.np/securi
http://scholarspace.manoa.hawaii.edu/handle/10125/24753
http://scholarspace.manoa.hawaii.edu/handle/10125/24753
https://doi.org/10.1672/0277-5212(2003)023%5B0082:IACOWH%5D2.0.CO;2
https://doi.org/10.1672/0277-5212(2003)023%5B0082:IACOWH%5D2.0.CO;2
http://www.digitalhimalaya.com/collections/journals/nepling
http://www.digitalhimalaya.com/collections/journals/nepling
https://doi.org/10.1093/llc/7.3.168
https://doi.org/10.1093/llc/7.3.168
https://doi.org/10.1017/S0954394500001058
https://doi.org/10.1017/S0954394500001058
https://doi.org/10.1017/jlg.2014.2
https://doi.org/10.1017/jlg.2014.2
https://doi.org/10.1515/9783110167467
https://doi.org/10.1515/9783110167467
https://blog.nationalgeographic.org/2014/10/19/roads-benefit-people-but-can-have-massive-environmental-costs/
https://blog.nationalgeographic.org/2014/10/19/roads-benefit-people-but-can-have-massive-environmental-costs/
https://blog.nationalgeographic.org/2014/10/19/roads-benefit-people-but-can-have-massive-environmental-costs/
https://doi.org/10.1080/02693798708927802
https://doi.org/10.2307/633476
Shepherd ID (1991) Information integration and GIS. In: Maguire DJ,
Goodchild MF, Rhind DW (eds) Geographical information systems:
principles and applications, vol 1. Longman Scientific and Technical
Publications, Essex, pp 337–357
Shiffer MJ (1998) Multimedia GIS for planning support and public dis-
course. Cartogr Geogr Inf Syst 25(2):89–94. https://doi.org/10.
1559/152304098782594562
Slobin D (1996) Two ways to travel: verbs of motion in English and
Spanish. In: Shibatani M, Thompson SA (eds) Grammatical con-
structions: their form and meaning. Oxford University Press,
Oxford, pp 195–217
Soomro TR, Zheng K, Turay S, Pan Y (1999) Capabilities of multimedia
GIS. Chin Geogr Sci 9(2):159–165
Stanford JN (2009) One size fits all? Dialectometry in a small clan-based
indigenous society. Lang Var Chang 24:247–278
Trudgill P (1974) Linguistic change and diffusion: description and expla-
nation in sociolinguistic dialect geography. Lang Soc 3(02):215–
246. https://doi.org/10.1017/S0047404500004358
Wallin E (1990) The map as hypertext—on knowledge support systems
for the territorial concern. In: Proceedings of the first European
conference on geographical information system, EGIS’ 90I. EGIS
Foundation, Munich, p 1125–1134
Yagoub MM (2003) Building an historical remote sensing atlas and mul-
timedia GIS for Al Ain. GEO: Connexion 2. 7:54–55
3 Page 14 of 14 J geovis spat anal (2018) 2: 3
https://doi.org/10.1559/152304098782594562
https://doi.org/10.1559/152304098782594562
https://doi.org/10.1017/S0047404500004358
Web-based...
Abstract
Introduction
Methodology
Study Area
Data Source and Data Set
Use of XML to Store the Data
Use of jQuery JavaScript Library to Load the XML File onto the Google Maps
Use of Google Maps JavaScript API V3 to Display the Data
Use of JavaScript, XHTML, and CSS to Design the User Interface and Functions for Spatial Analysis and Visualization
Results
Discussion
Conclusion
References