There was a Federated Search Symposium at the University of Calgary sponsored by The Alberta Library.
Info via Library Boy.
Distant Librarian has a good Federated Search Symposium wrap-up.
Also incredibly useful, the CDL posts their evaluations and assessments online http://www.cdlib.org/inside/assess/evaluation_activities/
The symposium website also has the other presentations, and since I know y'all want to see what the Google Scholar people had to say, here's the PowerPoint from Cathy Gordon, Director of Business Development - Google Scholar: Providing Visibility to Scholarly Literature.
The general tone I get is: metasearch = megapain.
I have a maybe naive question: why do we need to search, in the exploratory sense? That is, within the specific context of library journal article holdings. Don't we already know all of the journal issues we hold? Isn't that the function of the whole licensing department? Don't we have (or at least, have the ability to buy) the article metadata for all of our holdings? What is there to be discovered?
Can't you just buy a service as follows:
1) I tell it what journals I have, ideally by having it talk to the publishers directly (e.g. machine to machine holdings API)
2) Since the service holds the article-level metadata (author, title, abstract) for every article, it just carves out that data into a search subset
3) I do a search on that subset. No federation, since all of the data is in one place.
4) It points me to the fulltext - e.g. using my OpenURL resolver, which knows what stuff is held in what "databases" (websites)