Stevan has, I suspect, misunderstood Sally's point. Sally (if I may speak for her) was simply raising the question: what evidence was there for Stevan's speculation? Stevan now tells us that this was based on anecdotal evidence. Charles
Professor Charles Oppenheim Head Department of Information Science Loughborough University Loughborough Leics LE11 3TU Tel 01509-223065 Fax 01509 223053 e mail [email protected] ____________________________________________________________________________ From: American Scientist Open Access Forum [mailto:[email protected]] On Behalf Of Stevan Harnad Sent: 05 January 2009 11:48 To: [email protected] Subject: Re: Comparing Physicists' Central and Institutional Self-archiving Practices at Southampton On Mon, Jan 5, 2009 at 4:21 AM, Sally Morris (Morris Associates) <[email protected]> wrote: Did the researchers ask physicists why they did or didn't use one or the other repository? If they did not (and the abstract only talks about correlation), then Stevan's explanation is pure speculation, surely? The (Indiana University) study was done on University of Southampton physicists. The (Indiana University) author did not ask the Southampton physicists any questions; he just analyzed the Southampton IR's (and Arxiv's) metadata logs. But we at Southampton did of course did ask our physicists, and they did of course reply that they preferred to continue depositing in Arxiv and Arxiv only, as they had already been depositing for many years, rather than depositing instead -- or in addition (double-depositing) -- in the University's own IR: . Since the answer was more or less obvious, I would be interested to hear what other hypothesis Sally might be speculating was the real reason our physicists preferred to continue depositing in Arxiv rather than their own institution's IR. (I might add that the latest RAE outcome for physics at Southampton might possibly now increase Southampton physicists' inclination to double-deposit -- but even that is mooted by the SWORD protocol, which can and will do the imports automatically.) (I will be taking up a perhaps more interesting question in my next posting, which will be about physicists' belief that a central repository like Arxiv can provide functionality that distributed IRs and their central harvesters cannot provide. The belief is in fact incorrect, but it has furnished the basis for a quite natural and timely research project that Les Carr and I will be submitting as a proposal to JISC: designing a functional demonstration that a central harvester of distributed IR OAI-PMH metadata can duplicate all of Arxiv's current functionality, plus a lot more, built on Southampton's Celestial OAI-PMH harvester . Stay tuned!) Stevan Harnad
