> Could you elaborate a bit? In my mind, the only "semantic web
> technology" of any note is "linked data". How that fits into
> "library search" is anyone's guess, and I'm wondering what,
> specifically, you're referring to when you say that Talis is
> active in this area.
I'm sure there are some Talis people out there who could give a fuller run down of what the company are doing in this area, but as examples:
Developing and supporting RDF Triple store technology (aka 'The Platform')
Exposing "reading list" information as RDF via their Aspire product
Contributing to relevant Ontologies - including discussions of/contributions to BIBO, resourcelist and AIISO ontologies
The http://semanticlibrary.org/ demonstrator of use of RDF/Semantic tech/Linked data for library data
Roy has already noted the excellent work happening at the LoC (the Chronicling America as Linked data http://chroniclingamerica.loc.gov/about/api/ as well as the LCSH mentioned), Libris (Swedish Union cat) and OCLC. The work being done to develop an RDA ontology (http://metadataregistry.org/schema/show/id/1.html) is probably worth a mention, as is the work being done on a Zotero->Bibo mapping https://www.zotero.org/trac/wiki/BiboMapping
The BRII project at Oxford http://brii-oxford.blogspot.com/ (not quite library specific, but relevant) might also be of interest.
So, there is quite a bit going on at the moment in this area. The question of 'library search' I think can be seen at several levels - since we have library data expressed as RDF and examples of this being searched, I guess we can say that examples of library search based on semantic web technologies exist. What I think we are missing is anything 'web scale' (I'd be very happy to be corrected on this if there are good examples!). In a recent blog post Alex describes how linked data might be used in a library context (http://shelter.nu/blog/2009/10/on-identity.html), including a nice (I think) idea of what cataloguing might look like. This type of approach would allow us to take advantage of the data being exposed by VIAF/id.loc.gov etc. and start to create links between our systems - which I where I see the real interesting possibilities starting to emerge.
Finally, probably also worth mentioning that both Yahoo and Google are (I believe) using RDFa embedded in HTML pages - which also points to some interesting possibilities for search - and libraries if we expose our data in this format.
The Open University is incorporated by Royal Charter (RC 000391), an exempt charity in England & Wales and a charity registered in Scotland (SC 038302).