"Betwixt and between: integrated MARC data with museum object records"
Presenters from University of Southern California, Union Catalog of Art Images (based at UCSD), Los Angeles County Museum of Art, and Research Libraries Group; moderated by Lynda Bunting, MOCA, Los Angeles

Barbara Shepard described USC's digital projects and her PowerPoints are at http://www.usc.edu/arc/arlis The goal of the digital library there is to connect the right user to the right resource at the right time. That user is diverse and dispersed. The resources are growing exponentially and many are born digital. Users want the resource, not just a citation. The right time is any time. Their metadata is qualified DC in a METS wrapper. Their platform is Oracle and XML. They are using a variety of controlled vocabularies, including AAT, ULAN, ADL, LACBD, TGN, LCSH. Documentum is their chosen digital asset management software and it is used also for content management and for the user interface, along with WebChoir.

Stephen Toney talked about the use of MWeb for LACMA's museum system which includes library data. The data is not integrated (yet) with museum information presented first. For example, if you search on an artist's name, you'd be presented with museum hits before library hits. cf http://mweb.lacma.org and http://www.systemsplanning.com/arlis.htm

Brad Westbrook and Trish Rose presented "Searching for a lingua franca for visual resources." The UCAI project combines data from UCSD, Harvard and the Cleveland Museum. A prototype database is being developed with about 750,000 metadata records, some with thumbnails. Extended VRA Core is being used and the problems with integration were discussed. Harvard uses a hierarchical model while Cleveland and UCSD have flat models, with the latter in MARC, some from a previous incarnation in a database. Both models are being tested in the integration. Other issues include synchronization of data, import/export, and maintenance (snapshot vs evolving). cf http://gort.ucsd.edu/ucai

Ricky Erway closed the panel with a presentation on the RLG Cultural Materials Initiative. She asserted that RLG is becoming an XML shop. Data has come into CMI in all sorts of formats and RLG has taken the responsibility of converting the data into a shareable format. After receiving the data, RLG converts it to XML using the schema and style sheets developed for the initiative. The data is then evaluated, tested, and loaded. Access is based on the CIDOC Cultural Reference Model which is event-based. The approach has been to build the database quickly and refine it as it is used. cf http://www.rlg.org

... go to other ARLIS information ...