Library of Congress plans new film, video portal

The Library of Congress will in the fall launch the first centralized database portal for motion pictures and video images.

The portal will be 'a gateway to the world's moving-image collections,' said Gregory Lukow, assistant chief of the Library's Motion Picture, Broadcasting and Recorded Sound Division.

The University of Washington, Rutgers Universities Libraries and Georgia Institute of Technology's Interactive Media Technology Center will construct the portal under a $900,000 grant from the National Science Foundation. The Association of Moving Image Archivists originally commissioned the project with funding from the Library of Congress' National Film Preservation Board.

The Moving Image Collections will serve as a metadatabase of 'what's out there,' from old videotapes to digital motion pictures, said James DeRoest, assistant director of computing and communications at the University of Washington.

Until now, there has been no central source of information about the content and format of TV programs, movies and news footage. They are stored in innumerable collections at institutions that control both the media and access to it, Lukow said.

Access and lending policies vary greatly among these repositories, Lukow said. So, instead of storing the Moving Image Collection at a central place, the portal will redirect visitors to the content providers, which will tell them how to download video or order tapes.

The portal will consist of two IBM eServer p630 and two eServer p610 systems, all running SuSE Linux Enterprise Server 8 from SuSE Linux AG of Germany.

The researchers chose Linux and open-source applications to keep costs down, DeRoest said. They had considered Open LDAP for building the directory with Lightweight Directory Access Protocol but ended up using IBM's Directory Server, which is based on the Open LDAP standard.

Going open source

For the database engine behind the portal, Rutgers researchers selected the open-source PostgreSQL, said Grace Agnew, associate university librarian for digital library systems at Rutgers. She described the software as 'one step up from MySQL,' a popular open-source database manager.

Rutgers is designing the Extensible Markup Language schema for mapping metadata from individual collections to the main catalog. For example, an archive of TV news videos will have data descriptions vastly different from those of a collection of old films.

The Rutgers researchers wrote the XML schema for the Union Catalog Core Registry, which defines the primary metadata that portal users need to find what they seek, Agnew said. The Moving Images Collection will use XSL Transformations to make extensible style sheets for displaying diverse records.

The Georgia Tech researchers are evaluating open-source search engines for the moving-image portal, DeRoest said.

Once completed, the portal will be turned over to the Library of Congress as a production service. It will have multiple interfaces to serve the needs of archivists, historians, educators and the public, DeRoest said. For example, the archivist interface might present extensive details about media formats and storage conditions, whereas the educators' interface would give details about ordering copies of films and videos.

During its development phase, the project's Web site is gondolin.rutgers.edu/MIC.

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above