I have just done an entry talking about how discovery of library materials may be carried out in a different environment to the library system (in Google Scholar, WorldCat, or a local discovery system like the NCSU catalog, for example).
One issue that arises in this context is how to connect the 'discovery' experience with location or availability information for materials in the library or elsewhere. I have suggested that we would benefit from an ILS service layer which provides a consistent way for this type of machine to machine interaction to occur between discovery systems and the library system.
Here is an example of the type of service one would find in such a layer, the Catalog availability service from NCSU:
The Catalog Availability Service returns current item availability information for a given ISBN in the NCSU Libraries catalog. If the ISBN is unknown, or is not currently available for checkout or use, the service uses OCLC's xISBN service to look for a related catalog item that is currently available.
This service is designed to be consumed by global catalogs such as WorldCat.org, as a catalog localization service to help NC State users locate available copies of library resources of interest. The model we propose here could be adopted and deployed at other institutions capable of implementing a similar lightweight web services interface to their library catalog.[Description: Catalog Availability Web Service: NCSU Libraries]Of course, as we begin to compose services in this way it may be possible to add value by incorporating other services also to enhance functionality. NCSU uses xISBN as described here; another example is the use of our audience level web service by John Blyberg in PatREST.