A lot of people don’t seem to be very happy with OCLCs decision on OAIster. There’s a general problem with centralized services like OAIster. Too much can happen to that single service, and if it’s down it’s down. That’s why I thought about a distributed OA search for some time now and the OAIster decision gave reason to me to ask the YaCy  developer Michael Christen if YaCy could be used for such a search. To make it short: he’s quite optimistic.
Right now YaCy is not able to import data via OAI-PMH but that should be possible to realize. Of course there’s still some work to be done, for example the proper implementation of dublin core. On page 2 of http://yacy.net/material/YaCy_LT2009_Flyer_216x303.pdf (in German) you can find an example of the current possibilities of YaCy with dublin core. BTW: It’s SRU compliant.
We’d like to set up a Yacy instance as a proof of concept. In future it should be possible to cluster some of these instances to a large network of open access search engines. Just to mention the name of this mailing list: kind of a “next generation catalog for libraries”, I guess.
Opinions, anyone? Do you think a distributed OA search could be useful?
Kritik, Vorschläge und andere Äußerungen dazu bitte in die Mailingliste oder hier in die Kommentare. Gerne auch direkt per Mail. Hätte solch ein Projekt Zukunft? Was wären die Anforderungen?