EAD Harvesting for the National Gallery of the Spoken Word

Nancy Fleck
Head of Technical Services
Michigan State University

Michael Seadle
Head of Digital and Multimedia Center
Michigan State University

The library portion ofthe Digital Library Initiative (Phase 2) National Gallery of the Spoken Word project has used XML-based Encoded Archival Description (EAD) to organize spoken word recordings ranging from 30-second Edison cylinders to multi-hour oral histories. While EAD is well suited to the organizational task, it is too complex to serve as the primary discovery tool. To facilitate discovery, portions of the EAD record are harvested into MARC (for loading into the online catalog) or into Dublin Core (for Open Archives Initiative Metadata Harvesting). The challenge is not to find a single correct mapping, but to define dynamic relationships that meet evolving needs.