The General Libraries of the University of Texas at Austin received an award in the 1996/97 round of the Library of Congress / Ameritech National Digital Library Competition to support the digitization of this collection of Robert Runyon's photographs. As encouraged by the guidelines, the image files and most of the web pages describing the source collection and providing context are mounted and maintained at the University of Texas. Bibiographic records delivered to the Library of Congress allow intellectual access to the collection to be integrated into American Memory's search capabilities. Each record includes links to the corresponding image files.
The links below provide more detail on different aspects of building this digital collection.
This collection of Robert Runyon's photographs is housed at the Center for American History, a unit of the General Libraries of the University of Texas at Austin. An award in the 1996/97 round of the Library of Congress / Ameritech National Digital Library Competition supported its digitization. The competition guidelines allowed institutions flexibility in conversion methodology and allowed different approaches to describing or cataloging the materials. Applicants could opt to mount and maintain the digital reproductions at their own institution (and make them accessible over the Internet) or send all files to the Library of Congress. The University of Texas chose to mount the files on their campus facilities. Integration of this collection into American Memory uses a simple approach to interoperability.
Bibliographic records were delivered to the Library of Congress
The LC/Ameritech competition guidelines required that awardee institutions provide intellectual access to their materials through descriptive records or finding aids that can be delivered to the Library of Congress and incorporated into American Memory. The University of Texas delivered a set of item-level records, one for each photograph; the records were delivered in the form of tagged text, with each field (such as title, creator, and subject) delimited by start and end tags. At the Library of Congress, this set of descriptive records is treated in the same way as records for a collection digitized by the Library. After any necessary automatic transformation (for example, to add fields needed to distinguish records from different sources), the records are indexed by the InQuery search engine to allow users to explore the specific collection or search across the entire American Memory resource. A search returns a list of item titles; when a title is selected, a bibliographic display that links to the digital reproductions of a photograph is generated dynamically from the descriptive record by the American Memory system.
Each record at the Library of Congress includes links to files for the digital reproductions (at the University of Texas)
When a user's web browser presents the bibliographic display generated by the Library of Congress, it retrieves and embeds an image file from Austin for the thumbnail. Clicking on the thumbnail links to a service image, also retrieved from Austin. The competition guidelines allow awardee institutions to deliver digital reproductions to the Library of Congress or to mount them locally. The digital reproductions of the Runyon photographs are mounted at the University of Texas. Each bibliographic record delivered to the Library includes any necessary links as Uniform Resource Locators (URLs) that can be used over the Internet. A URL, the standard link type on the Web, functions exactly like an address or a call number in the sense that it consists entirely of information about the precise location of an "object." Every file has a URL associated with it. One problem with URLs as links is that they usually refer to specific computers and specific files. The links in these records might fail when the computer systems at the University of Texas must be upgraded or re-organized. To avoid this problem, the URLs provided by the University of Texas are based on an implementation of Uniform Resource Names (URNs), a framework for persistent naming proposed by a working group of the Internet Engineering Task Force. The specific implementation developed at the University of Texas is based on the Lightweight Directory Access Protocol (LDAP), Trivial HTTP (THTTP), and a URN resolution scheme written in PERL.
Interoperability is through a "union catalog" at the Library of Congress to distributed content
The simple level of interoperability implemented for the LC/Ameritech competition integrates remotely stored digital reproductions through a centrally stored repository of descriptive records. The model is therefore of a "union catalog" pointing to distributed content. Another potential level of interoperability would have been distributed searching of remotely stored records, perhaps using the Z39.50 protocol for search and retrieval. Requiring participants to support access via this protocol, which was developed in the library community, however, would have excluded small libraries, archives, and historical societies. The Library of Congress and the Ameritech Corporation were anxious to extend the competition to such institutions. The simple approach to interoperability also permits awardees to integrate the materials into their own digital library systems without unreasonable duplication of effort.