Kansas Geological Survey, Open File Report 96-37

Future of Scientific Communication, part 4 of 6

Electronic Publication in the Sciences

The number of electronic scholarly publications has grown significantly in the last five years. At the end of 1995, a survey of full-text, peer-reviewed journals in the areas of science, technology, and medicine discovered over 100 on-line (Hitchcock and others, 1996). The University of Waterloo Scholarly Societies Project provides links to over 150 full-text archives of serial publications published by scholarly societies (University of Waterloo, 1996, Taubes, 1996). These electronic publications have followed a number of models ranging from the traditional journal model to radically different forms of scientific communication. It should be emphasized that scholarly scientific publication is where authors are writing not for direct financial remuneration in the form of royalties, but rather primarily to communicate information (for the advancement of knowledge, with attendant benefits to their careers and professional reputations).

The advantages of electronic information technology are striking. Global networks effectively eliminate geographic considerations from the framework of research. The developing global networks provide instant access to remote resources, facilitate real-time collaboration, and make possible dynamic documents that incorporate by link sound, video and external documents. Another advantage is that search and retrieval technology for electronic databases is more comprehensive and more efficient in dealing with the ever-growing body of scientific information. Print turned information into a material commodity that could record and archive scientific results taking us beyond the oral traditions of a preliterate culture. Books and serials have a relatively high cost for production of the first copy and a relatively low cost for subsequent copies. However, technological advances in scientific communication can now begin to separate access to scientific information from ownership, assuring access to research and data with questions of the physical location of the materials becoming secondary. The economics are favorable, with the costs of digital storage and network connections plummeting. Electronic communication systems appear to be indispensable in providing real-time access to researchers in less developed countries. It has been pointed out that it is easier and less expensive to get a computer connected to the Internet than to build, stock, and maintain conventional libraries (Ginsparg, 1996).

The limitations and costs of the printed page have worked against thorough presentation of data and limited the reproducibility of research results. Paper publications contain a fixed text that provides a predefined linear access and very limited interaction. Non-linear access, if provided, is limited to tables of contents and indexes (Okerson, 1992). In electronic media the possibility emerges of reader interaction and substantially more complete publication of all the data, methods, and theory on which research is based. The next step is to share ongoing research products, intermediate results, and data. Electronic publication improves reproducibility of the research and better still permits continued manipulation and enhancement to address user-defined questions or site-specific application of technology. As emphasis shifts from ownership to access, models of information provision and reproducibility permit, in principle, a degree of resource sharing among institutions and individual researchers far greater than that allowed by traditional scientific communication.

Selected Examples of Electronic Publication

Electronic publication in the sciences is a rapidly evolving area with a number of ongoing experiments by established societies, publishers, libraries and research scientists. A number of researchers are providing on-line access to manuscripts ("preprints") and in some cases complete collections of their personal and research groups' works (Denning and Rous, 1995; Schwab, and others, 1996). Commercial publishers have announced major plans to provide electronic access to 100's of scientific journals (Taubes, 1996). The following is a brief survey of the experiments in electronic publishing that are underway in the sciences.

The American Chemical Society (http://pubs.acs.org) has followed an aggressive but traditional journal model for electronic publication. The ACS provides its subscribers electronic search and retrieval access to refereed and edited articles that have appeared in print in any of 23 American Chemical Society journals. The articles can be downloaded complete with tables and images or reprints delivered by mail or facsimile. The ACS site also provides public access to a broad but limited selection of articles and news that were created as electronic publications. The ACS views electronic publication as a possible replacement for ink on paper, but not of the traditional journal. Peer-review and editing are viewed as critical to maintaining a reliable science archive. Uploading nonpeer-reviewed research on a network is viewed as a possible risk to the integrity of the archival record and the advance of science (American Chemical Society, 1996).

A slightly more radical approach has been advanced by the Society for Industrial and Applied Mathematics (SIAM) (http://www.siam.org). With the 1997 issues SIAM will provide electronic access to the full text of all its journals. Subscribers will have access to a searchable database including the full text of SIAM journal articles, available in various formats (PostScript, Adobe Acrobat PDF, and DVI). Articles can viewed on screen and/or printed. Presently, subscribers to the Journal on Scientific Computing have access to PostScript versions of unedited but accepted papers up to a year before they are scheduled to appear in the printed journal. Access is also provided to published articles in current and archived issues of the journal dating back to January 1994. Access to the preprint files on this server is limited to current individual subscribers.

The Association for Computing Machinery (ACM) (http://www.acm.org/) provides full-text access to over 42,000 citations. The association has committed to providing full on-line access to all its publications for members and subscribers. Currently two ACM journals are available on-line and a number of other electronic publishing experiments are underway. In addition ACM is a leader in developing electronic publishing plans and policies that are rapidly moving the association away from print journals and toward a digital library that can deliver individual articles on demand (Denning and Rous, 1995).

SCIENCE magazine (http://science-mag.aaas.org/science/) combines a conservative approach with experiments in different forms of electronic communication. SCIENCE provides general table of contents access, search and retrieval capabilities, and document delivery via mail or fax. Also accessible are a number of on-line experiments in electronic communication. These include the full texts that are available in a form specifically designed for the world-wide-web (http://science-mag.aaas.org/science/content/current/e-perspectives/). These enhanced texts contain hypernotes linking the text to figures and other scientific information available on-line. Experiments in on-line forums cover selected topics in science policy and foster on-line interaction and collaborative authoring (http://sci.aaas.org/nextwave/public.html).

One of the first and most radical experiments in on-line communication of research information is a set of automated archives that have been operational in physics and related disciplines (Ginsparg, 1996; http://xxx.lanl.gov/). Starting in 1991, the physics archives are meant as an on-line replacement of printed scientific journals and the traditional publication process. At present the physics archives serve over 35,000 users worldwide from over 70 countries, and processes more than 70,000 electronic transactions per day (Ginsparg, 1996). In some fields of physics, the on-line archives have supplanted traditional research journals as the source of topical and archival research information (Ginsparg, 1996).

The Genome Database (http://gdbwww.gdb.org/) is another example of on-line publication that increases collaboration across institutions and nations to create a new type of scientific communication. The Genome Database (GDB) is the official central repository for genomic mapping data resulting from the Human Genome Initiative. The GDB stores and curates genomic mapping data and results submitted by researchers worldwide and provides this information electronically to the scientific community. The content and structure are dynamic and distributive in that published results are evolving through time and represent the contributions of numerous researchers. Documents are searchable using textural and graphical query mechanisms and incorporate other links to other documents. The GDB allows on-line public curation and third party annotation that has as its goal the collaborative effort to analyze the structure of human DNA and determine the location of the estimated 100,000 human genes.

These experiments in on-line publication and communication highlight the rapidly evolving changes in the scientific publishing tradition. Overall these experiments appear to be moving at various speeds and to various degrees away from the traditional bound serial as the unit of published output. In its place the individual article or research product has increased in importance and has become part of a stream flowing into structured databases. In place of journal subscription, organizations are moving to provide access to these published works through providing access rights to the database (Taubes, 1996). Scientific communication is still primarily static text and graphics that use digital and network technologies to facilitate access and delivery. However, the real revolution is occurring when the primary research artifact is itself electronic. Electronic publications, such as the Genome Data Base, are radically changing the relationship between interpretative result and the underlying data on which they are based. New forms of publication are arising that can only be displayed in an electronic environment, using sophisticated "hypertext"; functions to provide an interactive research product that can be customized to fit user specified requirements. Functions, such as video and audio, keyword searches, links to related material, and automatic notification, are revoltionary changes in scientific communication resulting from electronic publication (Taubes, 1996).