What is a Digital Collections Repository (aka Central Digital Repository)?
A digital library management and delivery system that can serve a research
library with broad, comprehensive electronic collections. The UVa Repository
that will contain texts, images, and data from Library collections, licensed
resources, and faculty projects, where the resources are available for discovery
and use, and new uses of the collections can also be collected into the
repository.
What is Fedora™?
The Flexible and Extensible Digital Object
and Repository Architecture (Fedora™)
system, was proposed by the Cornell
Digital Library Research Group in a 1999
D-Lib article. The University of Virginia Library' s Digital Library
Research and Development Group (DLR&D) is collaborating with Cornell
to develop Fedora™ under a $1,000,000 Andrew
W. Mellon Foundation grant, which has also funded a $1,400,000 Fedora
Phase 2 grant. Fedora is the underlying "plumbing" for the Digital Collections Repository.
Why does the UVa Library need a Repository?
- To store "digital objects" and the metadata that describe them.
- To manage access, including rights management.
- To support indexing and retrieval.
- To coordinate delivery of the digital objects.
What is a "digital object"?
Objects can range from a single digital image, to a series of images related
to a single architectural site, to chapters of an electronic book that include
digital page images, to virtual collections that contain multiple objects
themselves.
How do objects get created?
Objects are created manually (TEI electronic text files, EAD finding aids,
and the locally-defined GDMS format for complex, structured objects) or
programmatically (extracted from MARC records, from the IRIS system, or
from the User Collection Tool). When these objects are loaded into a Fedora-based
repository system, METS encoded XML files are created describing the objects
at an administrative and technical level, as well as Dublin Core/OAI compliant
metadata extracted from the full metadata for the objects.
How do objects get into the Repository?
Objects are reviewed for conformity to standards and batch-loaded by Digitization and Publishing Services.
How will users discover the collections?
The collections and metadata files are indexed using the a search engine
that supports XML. Users can search the collections via the web.
What will users find?
Approximately 16,000 images (to date) from five image collections (Jefferson Country
Architecture, the Smithsonian American Art Museum Catlin Collection, selections
from the Fowler Museum of Cultural History at UCLA, and images from the UVa Fine Arts Library and McIntire Department of Art Visual
Resource collections); all new electronic texts created by DLPS to date; and migrated image and text collections from many areas of the Library.
What can users do with what they find?
Descriptive metadata and images can be viewed, and images can be downloaded,
if allowed by licensing; texts can be read online.
When will it be available?
It's available! The Repository launched for the entire UVA community in 2007.
Three prototype digital collection search and delivery interfaces for searching
and browsing image and electronic text collections were presented to the
Library staff for review of the functionality and design in summer 2003.
The services were developed as a collaborative effort between a number of
Library units, and with the input from several committees that addressed
questions of functionality and user requirements. Almost 150 individual
comments were received, which were distilled into a series of categories
for prioritization by staff from many areas of the Library. The first phase UVa production repository
launched its alpha version for evaluation during the 2004-2005 academic year. Our first and second phase production repositories based
on the Fedora architecture were tested with the UVa community in 2004-2005 and 2005-2006.
What is planned for the future?
A number of services and tools are in the planning stages, including new formats such as video and audio and datasets, and data manipulation tools
(annotation, text analysis, etc.) in addition to the existing image and text presentation tools. See the prototype content models currently being tested.
|
