Okay, so OAISter-type harvested content will still be available from
worldcat.org -- at least to the extent that all the various institutions
are willing to sign OCLC's permissions statement. I suspect that many
won't be, just because as far as I can tell many IR's that OAISter is
harvesting have essentially been administratively abandoned by their
institutions, there might not even be anyone who cares enough to pay
attention to OCLC's email.
But it's good that whatever OCLC does have will be available for free
through the worldcat.org interface.
It's still unclear if the worldcat.org interface will support limiting
your search to OAISter content. It's also unclear if worldcat API's will
be provided to the public for free that duplicate the functionality of
the existing OAISter machine-accessible interfaces. The OCLC
announcement was really confusing, every time I read it it seems to mean
something different to me.
But Thomas Krichel makes a really good point about the limitations of
the DC metadata that is the OAI-PMH lowest common denominator in the
first place. I have to admit that this so severely limited the utility
of OAISter in the first place, such that I was not actually able to use
it in my software. In particular, the fact that there's no way to tell
if a record in OAISter actually does point to a digital full text copy,
and if it does, if that copy is publically available or behind a pay
gate. No way to tell from the metadata.
I had some hope that OAISter would try to address this problem. If
_anyone_ has the influence to get OAI-PMH data provider software
developers to add support for better standard metadata formats, and get
institutions to upgrade to such software, it was an entity with the name
recognition, neutrality, and draw of OAISter. (I think some people
probably thought of OAI-PMH provision as useful solely because their
records would then end up in OAISter!). But of course it was clear that
umich lacked the resources/interest to pursue this. I was hoping that
OAISter would go to someone who did. We'll see if OCLC is that someone,
but I don't see much reason to think so.
Jonathan
Received on Mon Sep 21 2009 - 11:26:15 EDT