Issues in Science and Technology Librarianship	Winter 1998

DOI:10.5062/F4KK98S8

URLs in this document have been updated. Links enclosed in {curly brackets} have been changed. If a replacement link was located, the new URL was added and the link is active; if a new site could not be identified, the broken link was removed.

Designing the Next-Generation Chemistry Journal: The Internet Journal of Chemistry

Steven M. Bachrach, Darin C. Burleigh and Anatoli Krassivine
Department of Chemistry and Biochemistry
Northern Illinois University
DeKalb, IL 60115
smb@smb.chem.niu.edu

Abstract

The development of the Internet as a means for mass communication affords scientists an opportunity to re-evaluate the way they communicate scientific results. The computer is no longer just a way for preparing a conventional print article but is an electronic publication medium in its own right. In this article, we will discuss how our new journal, the Internet Journal of Chemistry, is designed to take advantage of the new technologies available in this electronic medium. We will discuss the development of our concept of an electronic journal, our decision on the scope and coverage of the journal, the financial logistics, and how the journal will be implemented. Woven within will be our perspectives on how the technologies may be utilized and how this new medium compares with traditional print publication.

Introduction

The world of scientific communication changed in early 1994 with the advent of the first graphical web browser. This browser, Mosaic, developed by NCSA and distributed for free, demonstrated to many scientists how electronic articles could be delivered to the community. Prior to Mosaic, the Internet provided inferior tools for scientific discourse. File transmittal via ftp was difficult due to an unwieldy user interface. Gopher solved this interface problem, but delivered flat text files or separated graphics files and no means for joining them within one window. The world wide web had been operational but the only available browser could just display text with hyperlinks. Graphic files had to be displayed using an external program. Mosaic offered the ability to commingle text, graphics, and hyperlinks. For scientists, and chemists in particular, articles must be allowed to contain text intertwined with graphics.

We recognized the power of this medium and decided to test its applicability to chemistry by hosting the First Electronic Computational Chemistry Conference (ECCC-1) in November 1994. Full details on the conference can be found elsewhere (Bachrach 1995), and we will not discuss here some of the greater ramifications of this event with regards to electronic conferences. Rather, we just note here that a number of important communications tools were already utilized at this conference. The articles were delivered as hypertext documents written by each author with no translation or creation services offered by the conference organizers. Realize that at this time there were no full-blown HTML editors available, though some translation converters were already in place. Nevertheless, over 60 articles were submitted. Most contained color images; some contained animation files as attached files. A number of them had hyperlinks attached to regions of an image. Perhaps most significantly, six or so articles contained hypermolecules.

Rzepa, Whitaker, and Winter had earlier in 1994 defined the chemical MIME types which defined the appropriate identifiers for files that contain chemical information. (Rzepa 1994) These include specifications to indicate that the files contained 3-dimensional structures of molecules. Using the chemical MIME type, when these files were selected by a conference participant, instead of receiving just a static image of the molecule, the 3-dimensional structure was delivered and sent as input into the user-selected molecular viewing program. The participant could then manipulate the molecule (therefore called a "hypermolecule"), rotate it, zoom in on a particular region, change the color of the atoms, etc. This was the first large-scale example of interactive manipulation of chemical data within scientific discourse, and provides a key reason for our interest in electronic journals.

ECCC-1 and subsequent electronic chemistry conferences {(Rzepa 1995)} had convincingly shown that chemists could author articles in hypertext, include figures, and embed a variety of chemical data. The web clearly provided a medium for scientific discourse and we awaited the announcement and appearance of chemistry journals which would incorporate this technology. What follows is a historical account of how we decided to launch an independent new chemistry journal, called the Internet Journal of Chemistry.

Waiting for Godot - Publishers Hesitate to Act

The period from 1995 through early 1996 was marked by a number of important technological developments on the web. (Bachrach 1996) Some of these had peripheral effect on the state of electronic journals. The Netscape Corporation was formed and released its browser, called Navigator. AltaVista was developed and launched to great success, since there was now a very fast search engine providing access to millions of web pages. Microsoft announced that it would provide a web browser (which it did late in the summer of 1996). The Microsoft announcement put a stamp of approval on the Internet and helped spur on the hype and intensified use of the Internet. HTML specifications grew to include frames, subscript and superscripts, and tables. VRML was announced with serviceable, but slow, browsers to manipulate the virtual reality environments.

Also released during this period were two significant new technologies which had direct impact on publication. With the release of Navigator 2, Netscape defined the protocol for creating plug-in modules. Plug-in modules are extensions to the browser that provide the browser with additional functionality, usually the ability to image new types of information within the browser window. For example, one might write a plug-in that will play audio, so a sound file can be directly embedded within a web page. This technology now allows users to add features that will customize their web presentations to handle specialized data, especially as interactive applications.

The {Chime} plug-in by MDL was an important release for chemists. Chime is a molecular visualization tool built around the {RasMol program} developed by Sayles. RasMol is a standalone application to render 3-dimensional molecules and allow the user to manipulate the structure. With the Chime plug-in, this ability is now placed seamlessly within the browser window. A single web page can therefore contain text, static color graphics, and an active molecule which can be manipulated by the user.

The display above is an example of Chime. The molecule can be rotated, scaled, and the display characteristics altered. If you do not see a molecule here, then you do not have the Chime plug-in loaded on your computer. Proceed to the {MDL Chime} site to download and install this free utility. We have written a {tutorial on using Chime and creating web pages with Chime objects}.

The other important development was Java, a computer programming language that is cross-platform compliant. In principle, Java allows for a single program to run on any computer. Netscape incorporated the Java virtual machine within its browser. Therefore, a web author can embed a Java program, called an applet, within a web page. This applet can display information (such as run a movie), or allow the user to manipulate information (such as playing a game). The applet appears within the web page providing, in a sense, a further extension of the browser capability.

Early Java applets in chemistry included simple programs to draw chemical structures which then can be submitted as search queries into a chemical database and to draw and manipulate spectra. (A nice example of this technique is available from the NIST WebBook. This link will take you directly to the {IR spectrum of benzene}.) The {ChemSymphony suite}, developed in late 1996, provides Java code to render and manipulate 3-dimension structures, much as Chime does, but with greater flexibility in representations.

While web technology was improving, the chemistry publishers generally sat on the sidelines. A handful of electronic journals were launched by the established publishing houses. Publication of the e-journal of this time involved at least one of the following steps which continued to tie them to paper: articles could (or had to be) submitted on paper, reviewers were sent hard copy, reviews were returned by regular post, the journal was simultaneously printed on paper and electronically, or a year-end compendium of all published articles appeared on paper.

A number of major publishers either committed to or began to release electronic versions of their published journals. These versions were (and still remain) electronic delivery of print. They did not contain any added features to the publication that e-publishing affords, such as color images or animations. Many journals provided just PDF versions of their journals, which truly maintains the print ideal. The main feature that these publishers were stressing was the improved speed of publishing, that articles could be obtained weeks if not months before the print versions were available.

By Spring 1996, we recognized what a great opportunity lay before us to truly advance chemical publication using the Internet and web technology. The established publication houses seemed unable to move to take advantage of this opportunity. We therefore gathered a group of chemists with electronic publishing experience to develop a new, fully electronic journal and act as the editorial board.

The Birth of the Internet Journal of Chemistry

We were greatly disappointed in the nature of e-journals as of Spring 1996. Further, we feared that the poor use of the Internet as a means for publication might lead to disenchantment with the whole concept of an e-journal. What if the chemical community missed this real opportunity to change the way we communicate? Therefore, we decided to create a new journal that would have one guiding principle: a total commitment to the electronic medium. All aspects of the journal would be carried out across the Internet. Articles would be submitted electronically using ftp or some related procedure. Referees would be contacted by e-mail and they would submit their reviews by e-mail or through a web form. Accepted articles would appear in electronic form only. Only with this commitment did we feel that the true power of the Internet could be applied to chemical publications.

The next decision was to define the scope of the journal. The editorial board discussed this issue intensely. The discussions quickly boiled down to one of two options. The first was to create a specialized journal, one that focused on a particular sub-discipline of chemistry. The obvious first choice was Internet chemistry, but we discarded this option since the potential audience is too small to really make an impact and sustain a journal. Other possibilities included computational chemistry and organic chemistry. Computational chemistry already had a couple of electronic journals, one of which, the {Journal of Molecular Modeling}, while getting off to a slow start, was beginning to develop a firm subscription base and beginning to encourage the use of Internet publication features.

The second option was to launch a journal of chemistry, one that covers the entire discipline. This approach has two advantages. We would have the largest potential chemistry audience and no other e-journal was attempting to reach such a broad spectrum of chemists. The disadvantage of this approach is a potential difficulty in growing a base of subscribers and authors. A broad range approach may diffuse the audience too much and discourage a critical mass of users.

We finally decided to launch a broad range journal. We fully believed that all areas of chemistry can be better communicated through electronic media than on paper. This approach provides the best opportunity to grow a subscription base and have a large impact on the field of chemistry.

The concept of the Internet Journal of Chemistry (IJC) was thus born.

Scope and Goals of IJC

The Internet Journal of Chemistry aims to publish high-quality chemistry first and foremost. The journal will accept original articles in all areas of chemistry. All submitted articles will be peer-reviewed for scientific validity and novelty. The articles will be abstracted by Chemical Abstracts and therefore contained within the premier research database of chemistry. The journal will also publish reviews of chemistry web sites. Subscribers will have a means for asking questions or making comments on the accepted articles, providing a "living document" whereby a discussion of the article can be maintained and preserved.

The journal is firmly committed to the electronic medium. Towards that end we will encourage authors to take advantage of the publication opportunities afforded by the Internet. Some examples of this "enhanced chemical publication" {(Bachrach et al. 1996)} are:

Molecules cannot be effectively rendered in two-dimensions, such as on the printed page. Authors will be able to publish the full 3-D structure of the molecule, which, through the use of chemical MIME types, can be directed into a molecular viewer. The reader can then manipulate this structure.
Color is often useful in representing a third or fourth dimension in a variety of plots, such as electron density maps, multi-dimensional NMR, and protein structures. The cost for producing and distributing color images on the Internet is the same as for a black-and-white image. Therefore, authors are free to include color images within their electronic articles.
Dynamic processes are best represented in a movie or animation, which cannot be distributed in print. However, movies can be easily attached to an electronic document. A simple hyperlink will direct the movie to the user's video imaging program for visualization and analysis.
Spectra are oftentimes not included in conventional publications, due to page limitations or a perceived notion that they have limited value. Electronic publication offers the opportunity to publish spectra, and not just as static images. They could be presented within, for example, a Java applet which allows the reader to manipulate the spectra - resize them, zoom in on particular features, obtain the actual (x,y) data points, etc.
Publication of large data sets, even raw data, is prohibitively expensive in traditional print. However, since large data sets can be cheaply stored and made available on request over the web, authors can deposit these materials with the article. Tools for analyzing the data can also be included within the article, such as a spreadsheet or plotting utility. Further, since the data sets are delivered in electronic form as created by the authors, with no reformatting or retyping by the publisher, there is no loss in accuracy of the contents.

We also believe that electronic publishing can reduce the costs of making materials available to the chemistry community. The real point of scientific publication is to share scientific insight with as wide an audience as possible. The recent economic trends in conventional publication have reduced subscriptions and thereby limited the readership. We are firmly committed to provide the journal at very reasonable rates. This now leads us to discuss some of the financial aspects of the journal.

Financing IJC

The last major process before launching the journal was arriving at the appropriate financial backing. Towards that end we had numerous discussions with a variety of potential publishers. We finally settled on joining forces with a small venture capital firm which was already involved in a number of electronic publishing concerns. This arrangement provides a great deal of flexibility for the journal organization to remain nimble. Technology is changing so rapidly that any Internet venture has to be able to quickly adopt new protocols and adapt to new environments. Private venture capital financing provides us with the funds to purchase the server and network connectivity, support system programmers and SGML conversions, maintain archival storage, and market the journal. We believe that the success of the journal is dependent upon a level of professionalism and stability. Private funding allows us to provide these services to the authors and subscribers.

The journal is part of a for-profit operation. Therefore, at some point the journal will charge a subscription fee. For all of 1998, IJC will be available to all at no cost. Subscription fees have not yet been set, but our aim is to keep fees reasonable to reach as wide an audience as possible. Subscribing to the journal will grant access to the current year's articles plus all previous publications. In a sense, the journal is really a database, and subscription provides access to the full database for the time-period licensed.

Implementing IJC

Our main inspiration for starting the journal lay in the belief that chemists should be able to include so much more information in their publications than what can be done on paper, what we have called {enhanced chemical publication}. {(Bachrach et al. 1996)} As we started to design the web site for the journal, we realized that we had been focusing on only half of the publication process. We had been stressing the empowerment of the author due to Internet publication.

The other side is the reader! We recognized that the web can provide a mechanism for empowering the reader as well. Some primitive aspects of this have always been present within the web browsers - users can select the font and point size, set the screen width, define how emphasized text should appear. Realize that this is a significant advance over the printed page, where the publisher determines these factors and the reader cannot change them.

This concept, however, goes much further than just layout. A concrete example should suffice to explain reader empowerment. All measurements have units attached to them, such as "feet" for a measurement of length and "gram" for mass. Chemists use a variety of different units, and standardization comes slowly. The standard unit for energy is joules. However, many practitioners, particularly in the US, consider calories to be the more convenient unit. Which unit should be used within the journal? For the print journal, this decision must be made and remains fixed. A reader has to abide with the choice even if another unit would be more suitable to that individual.

The web may give an impression to users that the pages they view are static and predefined. This is in fact often not true. Rather, web servers often provide dynamic web pages, created upon the request of the user. An example is the many web sites that provide current stock prices, which clearly are in flux.

How does this dynamic aspect apply to the chemistry journal? Continuing with our units example, a reader can select the option to have all energy numbers provided in units of joules. The server then converts the units appropriately wherever encountered and delivers the selected units embedded right in the document.

The beauty of this system is that the author writes the article using whatever units she desires, and each reader selects the units to be delivered. The web server provides the conduit for passing the materials in the appropriate fashion.

Other examples in chemistry include how 3-dimensional structures and spectra should be passed to the user, how references should be formatted, and which navigational tools should be made available. Readers can thereby customize the journals to fit their own needs.

Putting this concept into place required creating a customized web server and parser. The parser converts the authors' manuscript, written in HTML, into a metafile that contains new tags that indicate the presence of data that must be manipulated by the server itself. The server then reads these metafiles and, taking into account the users specification, converts them into standard HTML, which it delivers to the user's browser. The metafile is analogous with the Chemical Markup Language (CML) proposal of Murray-Rust. (Murray-Rust 1997) It differs in that the web server interprets the CML tags while in true CML, the browser would handle the interpretation. The server was implemented and tested in the Fourth Electronic Computational Chemistry Conference (ECCC-4) recently held on the Internet. Further details of this customized web server and delivery procedure will be presented in another article.

On thing that should be made clear is that the IJC web server provides the appropriate files and tools as requested by the user in the format also requested by the user. Therefore, a user is not restricted from access to any of the information within the journal just because they lack a particular plug-in or are using an older version of a browser.

Conclusions

The Internet Journal of Chemistry aims to deliver enhanced chemistry publication in all areas of chemistry. Utilizing the technologies of the Internet, such as customized web servers, browser plug-ins, and Java applets, authors will be able to include significantly more information (some of it within interactive tools) to readers. Readers will be able to customize the journal to best serve their special needs. Our hope is that chemistry publications will be substantially improved - that better communication between author and the audience will spur on greater understanding and more research. The journal is available at {http://www.ijc.com/}.

Acknowledgment is made to the editorial board of IJC for their tremendous help and encouragement with this project: Henry Rzepa, Stephen Heller, Thomas Pierce, Toni Kazic, Peter Murray-Rust and Jan Labanowski. The Camille and Henry Dreyfus Foundation Informatics Award provided support for the development of the journal concept. IJC is supported by InfoTrust, Ltd. and we thank Dr. Roger Bilboul for his involvement.

References

Bachrach, S. M. 1995. Electronic Conferencing on the Internet: The First Electronic Computational Chemistry Conference. Journal of Chemical Information and Computer Science 35: 431-441.

Bachrach, S. M. 1996. The Internet: A Guide for Chemists. American Chemical Society, Washington, DC.

Bachrach, S. M. et. al. 1996. Publishing Chemistry on the Internet. Network Science, 2(3) URL: {http://www.netsci.org/Science/Special/feature07.html}

Murray-Rust, P. et. al. 1997. The World Wide Web as a Chemical Information Tool. Chemical Society Reviews, 26(1): 1-10.

Rzepa, H. S. et. al. 1994. Chemical Applications of the World-Wide Web. Journal of the Chemical Society, Chemical Communications, 1907.

Rzepa, H. S. 1995. Electronic Chemistry Conferences. Trends in Analytical Chemistry 14: 240 URL: {http://www.elsevier.nl/inca/homepage/saa/trac/rzepa.htm}