The Impact of Electronic Publication on Scholarly Journals Copyright The American Mathematical Society. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage. To disseminate otherwise, or to republish, contact the author or The American Mathematical Society. Reprinted from "The Impact of Electronic Publication on Scholarly Journals" by John Franks, _Notices of the American Mathematical Society_, Volume 40, Number 9, November 1993, pp. 1200-1202, by permission of the American Mathematical Society. Published in Volume 17, Number 1, of CAUSE/EFFECT, a quarterly magazine published by CAUSE, the association for managing the using information technology in higher education. For further information, contact Julia Rudy at CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301 USA; 303-939-0308; e-mail: jrudy@CAUSE.colorado.edu THE IMPACT OF ELECTRONIC PUBLICATION ON SCHOLARLY JOURNALS by John Franks What will electronic journals be like? When most academics consider this question they naturally interpret it to mean what will electronic journals be like for them--how will their daily use of research journals be different when these journals are routinely accessed on their desktop computers? But another interpretation of the question may be more important--it is certainly more controversial. What will the organization and economics of electronic journals be like? What does electronic publication mean for the publisher and the librarian? Will the editorial or peer-review process be affected? Will the way journals are funded and marketed change? And how will all this affect the academic? Declining and shifting costs Electronic publishing will come to scholarly research journals before it comes to other kinds of publications. There are several reasons that we academics will be the first. The content of our journals is relatively homogeneous and generally amenable to being put in an electronic form. Also the readership is much more highly connected to the Internet than the general population and we are generally more favorably inclined to dealing with documents in electronic format. Finally, at the present time most libraries are under extreme financial pressure. What librarians refer to as the "serials crisis" has been brought about by dramatic increases in the cost of scholarly journals (an average of 13.5 percent annually for more than a decade) combined with financial constraints facing most academic institutions today. The result is a strong economic motive to find less expensive ways for scholars to communicate their work. We don't know what the economics of electronic publishing will turn out to be, but several of its aspects are becoming clear. The relationship of electronic publishing to traditional publishing is analogous to the relationship of personal computing to mainframe computing. Electronic publishing is likely to be much more decentralized. It is much easier to do and the costs can be much less than traditional publishing. This is especially true for an organization or department which must have a computing infrastructure for other purposes. In such a setting the marginal cost of an electronic publication can be quite low. Moreover the costs which remain are being redistributed. For example, the emergence of TeX as a near standard for mathematics moves much of the cost of composition and typesetting to the author or the author's institution.[1] This trend will accelerate--there will be increasing pressure on authors to provide manuscripts in a standard form which needs little massaging by the publisher. Since transferring articles from an electronic source to paper will normally be done by the reader, the printing costs will be shifted to the user or to the user's library. Distribution via the Internet is not free, contrary to popular opinion, but the costs are largely borne by the reader's institution and perhaps to a smaller extent by the taxpayer. Since the reader's institution usually must have Internet access for other reasons, the marginal cost is low. The effect on scholarly journals of all of these shifts is to decrease the value added by the publisher and to increase the value added by the institutions of the author and reader. It is worth examining who contributes to the value of a journal and which "value added" items contribute to the cost. The contribution of the author of a scholarly article (presumably the primary value) is almost never a direct contributing factor to the cost of the journal. The next most important value added is the certification achieved by the editorial and peer-review process. While the publisher plays an important organizational role in this process, the work of editor and referee is usually done by unpaid volunteers, at least for primary journals. What the publisher of a traditional journal does provide is printing, distribution, and "production editing." Production editing is a relatively costly item which improves and standardizes the style of articles. It is what makes the difference between a "camera-ready-copy" journal and a typeset one. The cost shifts which will come with electronic publishing, together with the dramatic increases in the cost of commercial scholarly journals, have led some in the library community to suggest that it may be time for a dramatic restructuring of the process of research publication. Patricia Battin, then University Librarian and Vice President for Information Systems at Columbia, urged that universities take a much greater role in the publishing enterprise. The advent of electronic capabilities provides the university with the potential for becoming the primary publisher in the scholarly communication process. At the present time, we are in the untenable position of generating knowledge, giving it away to the commercial publisher, and then buying it back for our scholars at increasingly prohibitive prices. The electronic revolution provides the potential for developing university controlled publishing enterprises through scholarly networks supported either by individual institutions or consortia[2] Battin might have added that we are also giving away our efforts as editors and peer reviewers and then buying them back, at rapidly increasing prices. Access versus ownership Not surprisingly, others, especially publishers, have a very different vision of how electronic delivery might change the nature of scholarly publication. To quote Ann Okerson of the Association of Research Libraries: We have lived for many generations with a world in which the technology of publication meant that access required ownership ... New electronic technologies allow the possibility of uncoupling ownership from access, the material object from its intellectual content. This possibility is revolutionary, perhaps dramatically so[3] I don't know what Okerson has in mind as a revolutionary possibility, but I think I know what commercial publishers mean by uncoupling ownership from access. It is a code phrase for a possibility I find disturbing. If I subscribe to a mathematics journal and receive an issue, what I have acquired, aside from a certain quantity of paper, is a limited right to use the content of the articles printed on the paper. I have the right to read the issue and to make limited photocopies for my personal use. I do not have the right to make large-scale reproductions or to sell reproductions. But my limited rights to the contents last as long as the paper on which they are printed. Unfortunately, what publishers seem to have in mind when they speak of access without ownership is a model in which a one-year subscription entitles the subscriber to one year of access. When the subscription ends, so does the access. This has little to do with ownership--no matter what the form of the journal, the subscriber always owns the medium and never owns the message. It is a question of limited rights in perpetuity versus the same rights with a fixed time limit which makes it necessary to repeatedly repurchase access. Such an arrangement is very advantageous to the publisher. It becomes much harder for a library to cancel a long- running subscription, if, for example, the quality of a journal declines or the price increases dramatically. The big loser in such a model is the librarian, who is supplanted as archiver and probably cataloger by the publisher. In fact, with this model the librarian becomes little more than a purchasing agent for scholarly journals. Even this role could be threatened by publishers who have hopes of marketing their journals directlyto individuals rather than to libraries. A new form of publishing In light of these very different visions of electronic publishing, it is interesting to see what new developments are actually taking place. Since the time when photocopying machines became widely available, the informal distribution of "preprints" of scholarly articles has been an important component of scholarly communication. Some would argue that it is now the most important component because formal publication is such a time-consuming process. A major criticism of this "scholarship by preprint" system is that scholars who have not managed to break into the right distribution circles may not have timely access to scholarly work in their field. It is natural that with the arrival of electronic mail, this process is tending to move from paper to an electronic format. The ease and economy with which articles can be widely distributed electronically have led a number of volunteers to set up article databases for their subdiscipline or their organization. For example, one of the best of these is run by the Institute for Mathematical Sciences at Stony Brook. It provides access to the articles in their preprint series in both TeX and PostScript format. As scholars have gained experience with this kind of publishing they have learned that e-mail is not a very good way to do electronic document distribution. Its sole advantage is widespread availability, but it is by far the most cumbersome method for the user. Anonymous ftp is a substantial improvement, but still better are distributed electronic document browsers like Gopher and Mosaic, which were explicitly designed for this purpose. These article collections certainly constitute a form of electronic publication, but are they journals? The main thing missing, of course, is the peer-review process. Also, at present, it is normally assumed that articles in such a collection are preprints and will be formally published elsewhere in a traditional journal. This raises several questions, as yet unanswered. Commercial publishers would like to have preprints removed from electronic databases when an article appears in a traditional journal. This may seem reasonable at first, but, in fact, tends to conflict with the interests of the author. When I publish an article, I receive a certain number of reprints which I am free to distribute to interested scholars. If that supply is exhausted I routinely photocopy more and continue to distribute them. This practice is surely widespread and, I believe, perfectly consistent with copyright transfer agreements of most journals. It would be both easier and cheaper for me to respond to these reprint requests electronically. I would be disinclined to publish in a journal which did not permit me to do so. And if requests for electronic reprints become routine, I will certainly want to automate the process of sending them to such an extent that no intervention on my part is required. At that point I am running my own electronic database in conflict with the publisher's wishes. Indeed, with modern electronic document delivery systems using client programs like Gopher and Mosaic, it is quite possible for my electronic reprints to reside physically on my desktop computer but appear to others to be in a discipline-oriented database. This is an issue whose resolution is not clear, but current practice seems to be to leave articles available electronically after publication and I know of no instances of publishers attempting to enforce their removal. Not surprisingly, it has occurred to some that the main thing required to turn a preprint database into a true journal is a volunteer editor. In fact, there now exist electronic journals, edited and fully refereed, and available to scholars electronically without cost. The costs of running such a journal are comparable to those of an article database and are typically borne by the editor's institution or research grant. What happens next? In some form or other, free electronic access to scholarly articles seems to be here to stay. There are simply too many advantages for the author, the librarian, and the institution as a whole. The author has a strong incentive to have his or her work easily accessible to other scholars and he or she has the final say on where it is published. How great the competition free journals will provide for commercial journals remains to be seen. Certainly a younger scholar trying to establish research credentials for promotion or tenure may prefer a traditional publication, even if it offers less accessibility to his or her work. A great deal depends on the reaction of librarians to these innovative publications. A serious journal needs to be archived by libraries or consortia of libraries. Librarians have two strong incentives to support this new model of publication. First, it preserves their traditional role of archiver and cataloger--a role which at least some publishers seem to covet. Second, library budgets can surely better support the cost of archiving than the "prohibitive prices" that Patricia Battin complains about. Faced with increases in the prices of traditional journals, librarians have begun to take action by campaigning to persuade faculty not to submit papers to the most expensive journals or to serve on their editorial boards. Whether they will equally vigorously support inexpensive or free electronic publications by archiving them or forming archive consortia remains to be seen. The rapid change in the nature of publication is just beginning to impact the librarian, but that impact will be profound. For the most part librarians are not well prepared for the changes being forced upon them by the pace of technological developments. It will be a difficult time for them. But on the plus side, being a librarian is rapidly becoming a glamorous high-tech position. How will it all turn out? No one knows for sure, but likely we will see a mixture of models. I doubt that primary research journals can be successfully marketed in a way that requires an annual repurchase of access rights, but specialized databases which have greater value added by the publisher perhaps can. We will see more free subscription journals, but I doubt they will supplant traditional journals. However, as traditional journals gradually move to an electronic format they will be under great pressure to cut costs and pass on the savings to subscribers. Printing and distribution account for about 30 percent of the cost of a traditional research journal and hence represent substantial potential savings. Journals can also save by making greater efforts to shift composition costs to the author. This will take the form of stricter requirements for the format of submissions (or real page charges if those requirements are not met). This is a desirable thing. The scholar, too, must contribute to ameliorating the serials crisis. Most likely, traditional journals will sell electronic subscriptions much as they currently do subscriptions to paper versions. Their customers will continue for the most part to be libraries. With a journal subscription a library will receive the right to redistribute the contents to its institutional patrons and the responsibility of assuring that it distributes only to those patrons. The library also will acquire the right and responsibility to archive those journal contents, perhaps as part of a consortium. ======================================================================== Footnotes: 1 TeX is a text formatting system designed by Donald Knuth of Stanford which is especially useful for scientific and symbolic text. It is widely used in the scientific community for personal printing and is used by many commercial publishers for journals and texts. Documents written with TeX markup can be automatically processed to produce high quality typeset output. 2 Patricia Battin, "The Library: Center of the Restructured University," College and Research Libraries, 1984, pp. 170- 176. 3 Ann Okerson, "Synopsis," in University Libraries and Scholarly Communication, A Study Prepared for The Andrew W. Mellon Foundation (Washington, D.C.: Association of Research Libraries, 1992). Available via Gopher server at: orion.lib.virginia.edu, port 70. ======================================================================== John Franks is a professor of mathematics at Northwestern University. His primary research interest is dynamical systems. Currently he is the chair of the Committee on Electronic Products and Services of the American Mathematical Society. Date Revised: 08/19/94 (jar) The Impact of Electronic Publication on Scholarly Journals