![]() sponsored by the Library of Congress Cataloging Directorate ![]() Martin Dillon Former Executive Director of the OCLC Institute Adjunct Faculty, OCLC Institute Metadata for Web Resources: How Metadata Works on the Web About the presenter: From 1970 to 1985, Martin Dillon served on the faculty of the School of Information and Library Science at the University of North Carolina at Chapel Hill, where his research and teaching focused on topics in library automation and information retrieval. He came to OCLC as Visiting Distinguished Scholar in 1985. In 1986, he assumed the position of Director of the Office of Research, where he guided a staff of 30 in research supporting OCLC's mission of improving access to information. From June1993 until he became executive director of the OCLC Institute in January 1997, he served as director of OCLC's Library Resources Management Division, which is responsible for managing OCLC's Cataloging and Resource Sharing services. As the inaugural director of the OCLC Institute, he led the Institute in forging new ways to facilitate the evolution of libraries through advanced educational opportunities. Full text of paper is availableSummary: This paper begins by discussing the various meanings of metadata both on and off the Web, and the various uses to which metadata has been put. The body of the paper focuses on the Web and the roles that metadata has in that environment. More specifically, the primary concern here is for metadata used in resource discovery, broadly considered. Metadata for resource discovery is on an evolutionary path with bibliographic description as an immediate predecessor. Its chief exemplar is the Dublin Core and its origins, nature and current status will be briefly discussed. From this starting point, the paper then considers the uses of such metadata in the Web context, both currently and those that are planned for. The critical issues that need addressing are its weaknesses for achieving its purposes and alternatives. Finally, the role of libraries in creating systems for resource discovery is considered, from the perspective of the gains made to date with the Dublin Core, the difficulties of merging this effort with traditional bibliographic description (aka MARC and AACRII), and what can be done about the gap between the two. |
|
|
|
![]() |
Library
of Congress |