True, but if your department and dean recommend tenure based on published work not in open access journals, then [what you are] saying is that I should deny tenure completely. Its not either/or in terms of publications. And what about if the field is a book field? Then do you want web based publications only?
Deborah A. Freund Vice Chancellor and Provost Professor of Public Administration Office of Academic Affairs 304 Tolley Administration Building Syracuse University Syracuse, NY 13244-1100 Phone: 315-443-2494 Fax: 315-443-1839 Email: [email protected] >>> [email protected] 03/14/04 16:56 PM >>> On Sun, 14 Mar 2004, Deborah Freund (Provost, Syracuse University) wrote: > I am a provost and I fear that is unrealistic. Mandating something, > at least in the American system, is a sure way not to get what you > want. I would guess strong suggestions and incentives might be better. Perhaps "mandate" is too controversial a word! I meant whatever the verb is that already describes existing American "publish-or-perish" policy. I suppose that's not exactly "mandating" publishing either, in the sense of "publish or you're fired!" Publish-or-perish policy is, as you put it, a matter of strong suggestions and incentives: The carrot is that publishing will get you hired, promoted, tenured, funded, honored, etc., and the stick is that if you don't do research and publish, you will have to do other kinds of duties. But even this is no longer the up-to-date description of publish-or-perish policy either, for of course research *impact* has already entered the equation too: It is not enough to simply publish. If what you publish is not found to be useful by your peer research community, it is less valued and rewarded than a publication that is found to be useful by your research community. And citations are a strong measure of the degree to which your contribution has been found to be useful by your research community. So far there is nothing new in this. Here comes the new part: The degree to which your articles can be used and cited by your research community is *strongly* dependent on the degree to which they can be *accessed*. And the fact today for every single one of the 2,500,000 articles that appear annually in the planet's 24,000 peer-reviewed journals is that most -- not some, or many, but most -- of its potential users worldwide cannot access (hence use or cite) it, because their institutions cannot afford to pay the access-tolls for the journal in which it happens to appear. (Nor is this is just a matter of the Harvards vs. the Have-Nots, because no university can afford access to anywhere near all of the 24,000 journals, and most universities can only afford access to a smaller and smaller minority of them. Hence although the few Harvards can indeed afford more access to the research output of the many Have-Nots, they still suffer the impact-loss for their own research output, because the journals in which it appears are unaffordable to their would-be users at the far more numerous Have-Not universities!) http://www.eprints.org/self-faq/#29.Sitting This impact-loss is an invariant fact for all those articles *except* the small percentage of them that have been made open-access by their authors -- either by publishing them in an Open Access (OA) Journal (5% of journals are OA) or by publishing them in a conventional Toll Access (TA) journal (95%), but also self-archiving them (in a central or institutional OA Eprint Archive), as shown by the data below: http://www.ecs.soton.ac.uk/~harnad/Temp/tim1.ppt http://opcit.eprints.org/feb19oa/kurtz.pdf http://www.nature.com/nature/debates/e-access/Articles/lawrence.html http://www.ecs.soton.ac.uk/~harnad/Temp/self-archiving_files/Slide0025.gif Now one might respond: "But our existing publish-or-perish policies already implicitly have that covered! We *already* reward publication in proportion to impact, rather than merely a bean-count: Why does OA provision need a special policy of its own?" The answer is very simple: The research community does not yet *realize* how much OA enhances research usage and impact, nor do they yet realize that it is possible to provide OA for all their articles, nor how to go about doing it. The institutional policy that is needed is one that informs researchers about the benefits of OA, provides them with the means of providing OA (institutional OA Eprint Archives), and encourages and helps them to use them (until OA provision becomes as natural a part of academic culture as publication itself). To this end, some examples of the kind of OA provision policy that universities and research funders might (if not mandate then) strongly encourage with incentives would be the following: http://software.eprints.org/handbook/universities.php http://software.eprints.org/handbook/departments.php http://software.eprints.org/handbook/funders.php http://paracite.eprints.org/cgi-bin/rae_front.cgi Stevan Harnad
