Thomas, what you actually wrote is > Show me an archive, and a university, who will vouch that for a > certain period, all that is in the IR with free full-text > is a equivalent to the university's authors' total research > papers in the same period. Does such a university exist?
Such a university can never and will never exist if you insist on every term in the statement. Mainly because no university authority can ever know all of the university authors' research output with absolute certainty, unless its staff size is very small (say less than 50). Maybe the head of a small research institute can be that sure, but a senior executive simply can't for even a small size university. Insistence on a free full-text is also impossible given current publisher requirements, though deposit of a full-text is achievable. Exactly the same is true of discipline specific repositories, with the proviso that the repository manager must be even more unsure. I assumed that you meant the question seriously and would accept 'close to all'. To be reasonably sure that you are capturing close to all research output requires some audit capability - for example that there is independently collected data on the university's research output to compare with the repository. As it happens, such a situation existed in Australia in 2007 as you probably know. The HERDC data collection for Government provides such an independent estimate. The HERDC is spot-audited by Government to prevent over-claiming. Queensland University of Technology I assert that QUT achieves an acceptable closeness to collecting all research output in its repository. Indeed Paula Callan is in a good position to cross-check the two collections against each other, and does so. The QUT policy statement is widely known within Australia and outside, and you can read the current version approved by the Academic Senate at http://www.mopp.qut.edu.au/F/F_01_03.jsp. The QUT eprints site is certainly up now because I checked. University of Queensland As to UQ, I need not wait until 2009 to know that they will collect all research output for 2008 by March 2009. They are simply implementing the usual Australian Government HERDC report through their repository. In other words the HERDC report will be generated from the repository contents. That guarantees that they will collect the same data that the HERDC requires or suffer financially for it by losing funds from the research block grant. As I wrote, I need no evidence to know this (nor does any other Australian repository manager), though it will be worth confirming in 2009. This policy is weaker than QUT's because it is not necessarily Immediate Deposit (ID), but it is also stronger since it guarantees much closer to 100%. There is a financial penalty for losing publications, often down to the department. Of course there may always be a small number of missing publications in any system. This may be because of laziness on the part of the authors, mislaid documentation, illness, or other reasons. Charles Sturt University and others BTW, Charles Sturt University has exactly the same intention. Probably about ten or more other Australian universities are actively considering the same step as UQ, because it eliminates duplication of work. Disagreement You write > But I hope that > we can agree that, from today's perspective, filling IRs > until we achieve 100% open access will be a very very long > process. Sorry, we can't agree. Filling IRs is happening now. The rate varies by country and situation, of course. I have hopes that IRs in all or most ~40 Australian universities will be capturing substantially all their research output by say two years. It may not all be open access, but it will be deposited. And by filling, I don't mean retrospectivity but that current output is captured and continues to be captured into the future. I could agree with you that filling discipline-specific repositories and covering all disciplines and inter-disciplinary fields will be a very long process, if that will help. Arthur Sale Professor of Computer Science University of Tasmania > -----Original Message----- > From: American Scientist Open Access Forum > [mailto:AMERICAN-SCIENTIST-OPEN-ACCESS-FORUM@LISTSERVER.SIGMAX > I.ORG] On Behalf Of Thomas Krichel > Sent: Sunday, 17 February 2008 3:10 PM > To: american-scientist-open-access-fo...@listserver.sigmaxi.org > Subject: Re: [AMERICAN-SCIENTIST-OPEN-ACCESS-FORUM] How to > Compare IRs and CRs > > Arthur Sale writes > > > In response to Tom's request for one university that will guarantee > > that they collect all their research output, here are two: > > > > Queensland Institute of Technology, Australia, since 2004. > University > > mandate since 2004. http://eprints.qut.edu.au/ Now in its 5th year! > > The site can not be reached on Februrary 17 at 09:41:21 NOVT 2008. > http://qut.edu.au can be, but I don't find such a statement there. > > > University of Queensland, Australia, since beginning of 2008. > > That is for just 1 and a half months? > > > Now achieving annual government research reporting through > their IR. > > This implies 100% coverage of course. > http://espace.library.uq.edu.au/ > > I did not ask you to tell me about them, I asked if there would > be an official from an institutions warrant us that they have > achieved it. I happen to know a bit about the Queensland Institute > of Technology, situation, I hold a QUT staff card and know the > repository manager there. But I don't think that it is worth > discussing the situation in one particular institution here. > > I am not saying that IRs are not a potentially good development > and I am not saying that they will never work. But I hope that > we can agree that, from today's perspective, filling IRs > until we achieve 100% open access will be a very very long > process. > > With cheers from Novosibirsk (sunny, -13C), > > > Thomas Krichel http://openlib.org/home/krichel > RePEc:per:1965-06-05:thomas_krichel > phone: +7 383 330 6813 skype: thomaskrichel >