|
From the Canadian Science Writers' Association Life In the Fast Lane: Dan Hogan
Now, with growing interest in a novel breed of online publications, science writers have a tantalizing new opportunity to get a behind-the-scenes look at research in progress. These Web-based publications, called electronic preprint or "eprint" servers, are not journals in the traditional sense. They are more like dispatches from the frontiers of science, fired off by researchers straight from the lab, often before peers have had a chance to review their work. To many, the philosophy of these servers is appealing: take the business of communicating research results out of the hands of the established "dead-tree" print-based journals, blamed for being too slow for the free flow of scientific knowledge in today's fast-paced world of the Internet; instead, let scientists share their results as they make them, via grass-roots Web sites that allow researchers to post both their own findings and comments on their colleagues' work in near-real time. To others, however, eprint servers represent a dangerous, heretical trend that threatens the very credibility of science. Such servers, critics warn, will open the floodgates to a deluge of half-baked, erroneous "discoveries" which would normally be weeded out by the high academic standards that the peer-review process demands. At the very least, eprint servers could seriously weaken the many organizations whose livelihoods depend on their journals' subscription and advertising revenues. How did all this come about? Blame it on the physicists. First came the World Wide Web - a handy new communications tool invented in 1989 by Tim Berners-Lee at CERN, the European Particle Physics Laboratory, to let scientists share their research over computer networks through clever codes called hyperlinks. Then came the original Los Alamos E-Print Archive (Physics Preprints) - a Web server set up at the U.S. Department of Energy's Los Alamos National Laboratory in 1991 by Paul Ginsparg, who was frustrated with the slow pace that traditional print journals were adapting to the physics community's brave new networked world. Now the biomedical community is jumping on the physicists' e-bandwagon with PubMed Central - the brainchild of Harold Varmus, former director of the U.S. National Institutes of Health. Launched in February 2000, the PubMed Central Web site is designed ostensibly to provide researchers with a free archive containing the full texts of papers in the life sciences. As of press time, PubMed Central was also to have an adjunct site called PubMed Express, which would provide unreviewed papers as well. Later this year, it is expected to be joined by two European services: E-Biosci, an initiative of the European Molecular Biology Organization, will likely require unpublished drafts to be peer-reviewed before they make it online; on the other hand, a private venture called BioMed Central reportedly has plans to offer access to unreviewed drafts. Meanwhile, the American Association for the Advancement of Science (AAAS) has unveiled a collaboration among a dozen leading scientific publishers who have been reluctant to grant free access to online versions of their printed articles. The joint effort, which includes Harcourt's Academic Press and John Wiley & Sons, Inc., is not so much of an eprint server as it is a reference-linking service. Based on a special tagging language that can be thought of as an extension to the Web's HTML coding, the new service is designed to let users click on a linked reference in one online journal and view the contents of that article, even if it is located on another journal's Web server. Each publisher will have the option of determining whether their linked articles will be available as abstracts or full text and how much the user has to pay to view the contents. Other online preprint services have also sprung up in various fields. For example, the United Kingdom-based CogPrints (Cognitive Sciences Eprint Archive) currently receives 25,000 new papers a year. And the American Mathematical Society's MathSciNet has an archive of 60 years' worth of Mathematical Reviews and Current Mathematical Publications. Just what is and isn't reviewed varies from service to service, so you should check out each one's stated editorial policies before using or citing any article. Why should science writers care about electronic preprint servers? For starters, they can be a treasure trove of scoops. Although the highly specialized jargon of eprint articles can be difficult to wade through, you often stumble on potential nuggets of interesting news stories. A casual glance at the Los Alamos E-Print Archive, for instance, revealed such enticing article titles as "Can we predict the fate of the universe?" and "Water Ice in 2060 Chiron and its Implications for Centaurs and Kuiper Belt Objects" among the dozens of "new" postings as of early February. However, several caveats should be stressed here, before anyone goes off half-cocked and starts breaking headlines like these. First, make sure that these are legitimate findings made by real scientists from reputable institutions. Although most eprint servers at least initially screen submissions, some rather strange-sounding articles can make it through the cracks. On the American Physical Society's eprint server, for example, an article entitled "A Mass-Metric Scalar Theory of Gravity," written by a retiree from Colorado, supposedly describes a theory that "denies the basis for the existence of black holes." Second, you might make some scientific enemies pretty quickly if you don't bother to contact the source of an eprint article and find out if it's OK to write up a news story before the article has been officially published. If you jump the gun, and the story gets on the wires, it can, at the very least, be an embarrassment to the scientist and his or her colleagues, not to mention the journal in which the article was to be published. Some journals do not have such rigid embargo rules as Science or Nature do, but you should find out in advance. Third, be prepared to download and install some tricky software in order to read the full text of eprint articles. In the case of the Physics Preprints, your computer has to be able to decipher files written in the PostScript language. (Fortunately, programs that can be used to view and print these files exist and can be downloaded free of charge.) Even if you come up empty-handed in your scouring of an eprint server, with no big scoop to show for your online labours, you'll at least have had an exposure to science at the cutting (if not bleeding) edge. Eprint browsing is a great way of learning about the current trends in a certain field and to find out what's hot. Just be careful not to get burned!
For those interested in finding out more about eprint servers, visit the following Web sites: Los Alamos E-Print Archive (Physics Preprints): http://xxx.lanl.gov (See also the Los Alamos press release entitled "Physicist fantasizes electronic knowledge network" at http://w10.lanl.gov/external/news/releases/archive/99-013.html) PubMed Central (general overview): http://www.nih.gov/welcome/director/pubmedcentral/pubmedcentral.htm (see also the National Library of Medicine - http://www.nlm.nih.gov - for announcements) Cognitive Sciences Eprint Archive (CogPrints): http://cogprints.soton.ac.uk/ American Mathematical Society's MathSciNet: http://www.ams.org/mathscinet/ Also, the University of Virginia's Science and Engineering Libraries
web site maintains a list of electronic preprint servers and databases
in various fields: http://www.lib.virginia.edu/science/guides/s-preprn.htm
Published in Sources, Number 45, Summer 2000.
Sources |