FEATURE ARTICLE
Free Internet Access to Traditional Journals
Can scientists find ways to share published research without high cost? The experiences of one society suggest it can be done cheaply, even profitably
Thomas Walker
The Serials Crisis, Web Style
Every year libraries must cancel journal subscriptions to continue receiving other journals that are increasingly costly. (And publishers often raise subscription prices as their subscriber base shrinks, creating a feedback loop.) Unfortunately, rather than ending the libraries' serials crisis, the beginning of Web access to traditional journals may have intensified it. Publishers now offer licenses to electronic versions as add-ons to regular subscriptions. The Web versions enable library patrons to access and search the journals without leaving their office or laboratory computers, and many of them are enhanced with extra material and sophisticated indexing and search capabilities. But there is of course a cost for this service. For example, the American Chemical Society offers libraries site licenses for the Web versions of its journals for 25 percent more than for paper subscriptions alone. Ironically, then, in these early stages of the Web's evolution some libraries are paying more for journals because they are paying for two versions and for the enhanced access expected as technology allows it. Although indeed they are providing more service and convenience, this is not the world of "free" digital information envisioned by some prophets of the Internet.
Within some fields of science, meanwhile, alternative and economical do-it-ourselves models have emerged. Most notably, physicists and mathematicians now routinely submit their manuscripts as "e-prints" to an Internet server at the same time they submit them to traditional journals. If they rewrite a manuscript, they submit the new version, but the old one stays in the on-line archive, helping to resolve questions of priority in a more objective fashion than if only anonymous referees have access. The famous e-print server called "xxx" (its Internet address is xxx.lanl.gov) at Los Alamos National Laboratory currently handles 25,000 submissions annually at a cost of no more than $15 per paper, including overhead (Odlyzko 1998). Although most of the papers are eventually published, their e-prints are available more easily and are probably consulted more frequently. The founder of the e-print archive and its current overseer is physicist Paul Ginsparg, who started it by himself because it seemed a good use of the developing technology (Taubes 1993, Ginsparg 1996).
Investigators in many areas of physics and mathematics, where (in contrast to the life sciences) there is a tradition of widely circulating information in preprint form, thus have open electronic access to much of the recent literature in their fields prior to its traditional publication. But can free access be granted for the final, refereed articles in these and other fields? There are barriers to implementing such a model, but these barriers hardly mean there are not exciting prospects for new kinds of publishing, or for saving money doing it. Strangely, a small entomological society's electronic-publication project may demonstrate to larger scientific societies that they can grant such access and reap rewards from it—both in improved service to their members and financially. To understand this, you must first know the technology that makes free access affordable.
» Post Comment