discovery services at the network level

Peter Brantley mentions that someone who went to RLG's D2D conference made this comment:

The conference covered a wide range of issues, and was very interesting. I think the consensus opinion at the close was that discovery has moved to the network layer and libraries should stop allocating their time and money trying to build better end-user UI, and concentrate instead on delivery, and their niche or customized services such as digitizing special collections, providing innovative end-user tools for managing information, and so forth.

Interesting observation. I'm thinking about how this comment relates to lots of dispersed efforts such as Scriblio to remake the library OPAC using technologies like SOLR. These are important pioneering efforts but will they stick around long?

More specifically, I'm thinking about it in terms of the options available to replace the interface to our OPAC at L&C, and ultimately the Summit Union catalog that we share with other Orbis Cascade Alliance members. The Alliance, through the Summit Catalog Committee, is considering lots of different options, including III's Encore, Endeca, OCLC's forthcoming WorldCat Local, as well as local development options. The statement above, in my mind, would be an argument for WorldCat Local.

WorldCat Local (not sure if this is the correct product name), from what I've heard, is a version of the WorldCat.org catalog that is scoped down to your own holdings and optionally, those of your union catalog. It allows you to use your local system for delivery. By using a discovery service like WorldCat local, we would be tapped into OCLC's web-scale bibliographic database and could, theoretically, benefit from "network effects" only available on that platform. One network effect readily available would be the most up-to-date version of any bibliographic record. But there could also be other network effects, for example, Web 2.0 ish stuff like tagging and comments that really only become useful on a wide scale. And maybe they will offer stuff like FRBRization and relevance ranking that takes advantage of the intelligence available in that size of a database.

OCLC, in my mind, has always been sort of a slow moving behemoth. Using many of their services, such as LDR updating, is often painfully slow and cumbersome. But they appear to have turned a corner on their OpenWorldCat program, especially some of the APIs that they have released recently. I'm listening to them with an open mind.