DefiningImageAccess/Project/Talis
From ImageWeb
Talis and Library 2.0
A commercial organization who, according to their Web page, provide library solutions with a heavy emphasis on "Web 2.0".
In particular, they have coined the term "Library 2.0", which is dicussed:
- in a D-Lib opinion piece: http://www.dlib.org/dlib/april06/miller/04miller.html
- in a Talis white paper (November 2005): http://www.talis.com/downloads/white_papers/DoLibrariesMatter.pdf
- and another Talis white paper (February 2006): http://www.talis.com/resources/documents/447_Library_2_prf1.pdf
A strand in their opinion piece is that libraries can and should embrace Wev technologies to "deliver library services outside the walls of the building" rather than "trying, repeatedly, to become the destination site for users". This philosophy resonates very strongly with what we are attempting to do with data webs. Talis are also using RDF as part of their technical solution, and have recruited one of the early Semantic Web activists (Ian Davis).
We are trying to achieve some very similar goals to what Talis do with library data, applied to institutional repositories. We are also looking to locate the individual images as well as complete publications, and aim to use metadata containing some domain-specific information to find related images and information.
November 2007 Update
I recently heard a talk by Ian Davis, describing Talis' Semantic Web Application Platform (http://www.talis.com/platform/index.shtml) at a UK SWIG meeting at HP Labs, and was prompted to revisit Talis' web site. Talis' focus seems to have broadened somewhat from library systems to "managing semantically rich metadata and in delivering software and services for information management".
Judging from Ian's talk, the Talis platform (http://www.talis.com/platform/index.shtml) seems very impressive, and very much the sort of facility that fits nicely with our data web ideas, with one caveat. The platform provides an TDF data store, which is indexed for IR functions. This could be at odds with out philosophy of linking existing web data, rather than creating new databases. But even if that is true, there are certainly elements of our vision for which the platform could be very well suited.
My other reservation about Talis' platform is potential "lock-in": the system is offered as an application service, using which semantic web applications may be built. Talis have ensured that any data used remains open and easily recoverable in the event of problems with the platform service, but if no-one else is providing a compatible service platform that still potentially represents an unmitigated risk for systems that depend on that platform. With good will, which I'm sure does exist, this is a problem which can be overcome.
