Stevan's email has helped to sharpen the debate. He is right that we will not know whether self-archiving will really have uptake (or optimal uptake) until it is readily available to researchers. Nothing in the concerns I raise about incentives in any way should be viewed as slowing down the effort to make open archives readily available. So, Stevan, when do you expect this threshhold of ready availability will be reached?
At that point, my hypothesis is that we will see sub-optimal use of self- archiving in medicine. I believe this will result from the incentive barriers i describe. Whether my hypothesis is true will become evident once the threshhold of ready availability of open archives is reached. The preliminary evidence supporting the hypothesis is that the BMJ Netprints preprint server has not caught on the way the Los Alamos physics one did a decade or so ago. I think this is becasue of the incentives in medicine to publish in brand name journals that disallow prepublication. Presumably the physics journals do not have restrictive prepublication policies, yes (ie no "Ingelfinger-like rule")? Stevan is also correct in pointing out that i do not have a sharply defined, demonstrated, quality assessment method for research articles that would provide an alternative to the journal brand name system. I have sketched a conceptual framework for the alternative in http://www.press.umich.edu/jep/06-02/singer.html . Developing this further will require research and development. It is precisely initiatives such as open archives or biomedcentral (because of its peer review light feature) that will serve as the laboratories for such R & D. The R & D will obviously not be required for the purpose of freeing the literature if my hypothesis proves false or the remaining name-brand journals drop their objections to pre-publication as Stevan predicts they will. However, if my hypothesis proves true (which i think it will), and the remaining name-brand journals persist with restrictive pre-publication policies (and frankly i do not see the NEJM backing away from the Ingelfinger rule any time soon), then realigning incentives by producing brand-name independent quality measures and incorporating them into the incentive system in medicine will become the crucial step to free that literature. I am not sure we can go much further with this debate in the abstract. We need more data now which we should have as the open archives initiative unfolds. However, i think i have shown why the issue of incentives should stay within the scope of a discussion on open archives. xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx Peter A. Singer, MD, MPH, FRCPC Sun Life Chair in Bioethics and Director, University of Toronto Joint Centre for Bioethics Professor of Medicine, University of Toronto Canadian Institutes of Health Research Investigator Associate Editor, Canadian Medical Association Journal e-mail: [email protected] website: http://www.utoronto.ca/jcb fax: 416-978-1911 phone: 416-978-4756 mail: 88 College St., Toronto ON Canada M5G-1L4
