DISCIPLINING ELECTRONIC SCIENTIFIC PUBLISHING?
Rob Kling and Geoffrey W. McKim
Center for Social Informatics
School of Library & Information Science
Indiana University

Even though established scientific journals are printed on paper for distribution to their readers, it is commonplace for their articles to be written with computerized  word processors and also to be typeset with computer compositors. Electronic publishing refers specifically to the distribution of materials, such as journal articles, through electronic media (including computer networks and CDs). Since many people print electronic articles in order to read them carefully, electronic publishing is but one intermediate state in a complex processing chain between authors and their readers.

Electronic publishing opportunities  have captured the imagination of many scientists, manifested in an array  of electronic journals, working paper servers, electronic discussion lists, and other Internet-based compendia. The  formats of these forums differ in many important details, including their relationship to paper publications, the forms of editing and reviewing, the extent to which they act as 'fast paper distribution systems" or include datasets and pictures that would not be readily included in paper publications, etc.  These opportunities have also provided many analysts and scientists with the opportunity to question norms about the character of legitimate publishing.

North American scientific societies such as the American Psychological Association (APA), the American Chemical Society (ACS) and the Association for Computing Machinery (ACM) have created quite different policies about the role of various forms of electronic posting in advance of paper journal publication for their respective fields. Unfortunately, much of the literature about scientific e-publishing homogenizes the character of publishing and  gives conflicting advice to journal editors, scientific societies, and scientists.  Electronic publishing in various forms is alterately presented as: (1) inevitable, rendering such debates about the role of electronic publishing in scientific communication moot; (2) a necessary component of publishing reform, in which scientists no longer give their articles away to journals and then buy them back through subscriptions; and (3) a threat to effective scientific communication, that allows unreviewed writings the same publication status as peer-reviewed, high-quality research.

In order to sort out and analyzes the roles electronic publishing may and does play in scientific communication, we provide a model of scientific publishing that that has the following characteristics:

1. It focuses on scientific publishing as a set of communicative practices (rather than  the economics of production or the symbolic dimensions of publishing venues), embedded within particular scientific communities.
2. It characterizes effective publication of a work along three dimensions: publicity, access (both long and short-term),
   and trustworthiness.
3. It is media-independent, and treats various forms of electronic publishing as equivalent to paper-based publishing forums with similar communicative properties.
4. It helps us to explicitly examine disciplinary differences in publishing practices.

Electronic publishing forums  vary in their structure and roles from one field (or closely related set of fields)  to another. These differences seem to reflect variations in paper-publishing practices across the disciplines.

The contrast between some branches of  physics, molecular biology, and information systems is instructive.  For example, the electronic working paper ("e-print") server (http://xxx.lanl.gov) at Los Alamos, which is used to distribute working papers in several fields of  physics, has become central to the communications system of the field (Odlyzko 1996).

While the E-Print server at Los Alamos National Labs is the best-known of these electronic working paper servers in the U.S., there are about 11 others in current use.  Biologists, on the other hand, circulate working papers after acceptance for publication in a journal within small invisible colleges and broader access depends upon publication in archival journals. Many biological fields, however, use digital databases to share important data sets, such as Protein Data Bank,  Genbank, and FlyBase; they serve as repositories for genomic sequences that have been published in refereed journals.  Many of these shared databases, such as FlyBase (http://flybase.bio.indiana.edu) , used by fruit fly researchers, contain much more than gene sequences, including published articles, researcher directory information, and fly stock lists.  ISWORLD, an extensive Web-based collection of links, papers, course syllabi, tools, and resources for information systems researchers, and maintained in a distributed fashion  (http://www.isworld.org) by various IS researchers, represents yet a different form of e-media. While the top-level Web site itself is sponsored by MIS Quarterly, Information Systems researchers act as section editors for the many sub-pages of the site.

These disciplinary differences both in the role and character of electronic publishing are socially shaped by the practices and structure of the field. A critical set of shaping processes are those by which trust is constructed.  First, scientific communications forums are shaped by the extent to which scientists to trust the quality and value of articles that they see in the forum.  Formal peer review is only one process out of many by which trust is constructed. Others are tied to collegial social networks (Chubin and Hackett 1990).  This form of trust is crucial to the construction of a readership for a forum. Another important shaping process is that by which scientists trust that they will become adequately visible and credited for their work.  We call this set of shaping processes the dialectics of trust.

The dialectics of trust are not at all limited to electronic media; scientists are eager to publish reports in journals such as  Science and MIS Quarterly, because such forums provide higher visibility and rewards.  Conversely, scientists who are willing to share materials (data, working papers, research reports, etc.) must have enough confidence that the sharing will not hurt  their own career advancement or future access to resources.  If a scientist publishes a paper in a low-status journal (paper or electronic), will this be considered a "wasted publication" regarding their career-advancement?  If a scientist posts a working paper in advance of publication (or acceptance for journal publication), is she taking a significant risk that someone else will either (a) plagiarize their work, or (b) take their work further, more quickly, and produce a higher-impact report?

Different fields construct trust in different ways.  Our preliminary research suggests that the construction of trust in a field may depend on such factors as relative centralization of control and funding in a field, degree of industrial integration, the role of oral culture, and degree of concentration of communications channels in the field.

These observations suggest that publishing practices are part of a more complex social system that varies from one field to another. In identifying some publishing practice of a field as a "best practice' it is important to understand how it works within the fields social arrangements for assessing trustworthy communications, allocating credit, and controlling violations of important norms about these matters.  It is easy to identify some important  publishing practices, such as using disciplinary working-paper server or having papers publicly posted for comment before journal review.  Other fields may not be able to effectively adopt such practices unless they also transform their social arrangements for assessing trustworthy communications, allocating credit, and controlling violations of important norms about these matters.  However, some practices, such as citation styles may be much easier to translate across the disciplines.
 
References:

1. Chubin, Daryl and Hackett, Edward. 1990. Peerless Science: Peer Review and U.S. Science Policy. Albany, NY: State University of New York Press.
2. Kling, Rob and Lisa Covi. 1995. "Electronic Journals and Legitimate Media  in the Systems of Scholarly Communication."  The Information Society. 11(4):261-271.
3. Kling, Rob and Geoffrey McKim. 1998a. The Shaping of Electronic Media in Supporting Scientific Communication: The Contribution of Social Informatics. Presented at "European Science and Technology Forum: Electronic Communication and Research in Europe" Darmstadt/Seeheim, 15 to 17 April 1998. <http://academia.darmstadt.gmd.d/>http://academia.darmstadt.gmd.d
   e/seeheim/. Sponsored by Academia Europea.
4. Kling, Rob and Geoffrey McKim. 1998b.  "Scholarly Communication and the Continuum of Electronic Publishing" Working draft. Center for Social Informatics. Indiana University.
5. Read & contribute to the Social Informatics Home Page http://www.slis.indiana.edu/SI a resource about research, teaching, conferences & journals