| Abstract: | In fall 2004, the UVa Library made a first version Digital Library Repository based on the Fedora available to its community. The Repository contains TEI and image collections produced locally since mid-2003, selected licensed image collections, and UVa Library Special Collections EAD finding aids.
This presentation focuses on the UVa Library implementation of Fedora. Fedora is the underlying architecture for the Repository, but not the complete management, indexing, discovery, and delivery application. The UVa Library's implementation required a large-scale effort to define a local architectural and service overlays specific to UVa's collections and functional requirements. The development process encompassed:
· the creation, documentation, and adoption of new holistic standards for production;
· a detailed analysis of the formats and configuration of media files and metadata for legacy collections for migration to current standards;
· functional requirements for discovery and delivery services;
· specification and development of content models and disseminators based on all of the above; and
· the application of a unified interface design on top of the architecture.
The Repository also includes the first release of the "Collector Tool" that allows users to create personal portfolios of objects. The current tool includes the ability to collect images into personal portfolios and generate slide shows or electronic reserve web sites that include pointers to the images and metadata in the Repository. Later releases will be generalized to support the collection of other object types, a sort of combination shopping cart and basic authoring tool for the entire Repository.
The next steps for the Repository infrastructure development process are an evaluation of the production workflows, usability testing of the interface with groups of faculty and students, and migration to Fedora 2.1. The next steps for Repository content development are a review of additional legacy image and text collections for migration to current production standards and ingestion, and planning for additional content types, including data sets and digital video.
|