Hi all,
We periodically create bibliographies to support faculty projects and are at
the point where we either want to roll our own re-usable software or adopt
something from the wild.  We need something that can understand library
metadata on the way in and on the way out.  We also need something that can
be available as a data store to different webapps, e.g. a subject guide, a
faculty publications webapp, etc.  In other words, we need the datastore to
be queryable from a script.
I'm looking at Zotero, which seems to be hot on the harvesting end, but
seems to be conceived of as a tool for a single researcher at their browser.
 I would need to flash Zotero's SQLite instance to a db table on my
webserver so the data would be available to scripts.  Ideally, my harvesting
team could interact with the web application directly so there would be no
harvesting/export/import latency.
Any ideas if this is something we've already tackled in libraryland?
Thanks,
Michael Lindsey
Law Library, UC Berkeley