OA expert Richard Poynder has a new PDF paper on his blog “Open Access, Almost-OA, OA Policies, and Institutional Repositories”. In it he looks at how many fulltext papers are in various repositories, and explores the trend toward the dark side that involves records that state of the PDF that ‘this item is embargoed until…’ .
Poynder’s article also has details and very extensive analysis of the “Button”, sometimes seen on repository record pages, which allows one to request a fulltext copy of an embargoed repository item.
As an aside, he notes…
a suspicion I have long had that repository managers are depositing a lot of historical data.”
Yup, I can confirm that feeling. Not all, of course, but a few do have a lot of historical material jumbled in. I guess that may be because they only have funds and staff to run one repository, which then has to hold everything. Only a few large universities sensibly split their repositories into separate servers/URLs, thus:
* a slimline one for public access to theses and masters dissertations.
* one to capture the flow of all the public-access scholarly OA items, sometimes even with filters that can knock out preprints, conference papers, or which can focus only on papers from individual journal titles.
* plus a more conventionally rambling repository to hold the digitisation of pre-1960s content, image collections, university ephemera, and the ‘local interest’ collections such as newspapers and trade magazines. Sometimes this has a slick public-friendly ‘showcase’ front-end, sometimes it’s just a list browse.
* big U.S. law schools increasingly have their own separate repositories, and their own OJS server for their journals.
* and running alongside all those, an OJS installation to run the university’s current journals (some universities even split their mainstream academic journals from the graduate school / undergraduate / creative writing / alumni magazine titles, having the latter on a second OJS installation).
It’s my feeling that even smaller universities may soon have to adopt such splitting strategies, given the tidal wave of OA content that’s looming on the horizon.
When servers do get split up like this there’s often no public interlinking between them, even in terms of using the front page of each as a platform to publicise the existence of the others.