The Death of Metadata

ABSTRACT In the mid-1990s, information professionals and computer scientists began to pay much attention to the concept of metadata. The rise of the World Wide Web highlighted the need to enable discovery, or finding desired information, on the Internet. Metadata was the hot topic at library and information science conferences and in professional literature, and many librarians and others expended enormous efforts creating and implementing new metadata schemes. One of these schemes, the Dublin Core, emerged in 1995 and quickly became the metadata standard of choice for digital objects. Although abundant resources have been put into the development and implementation of Dublin Core, the schema has largely failed. In addition, many other metadata schemes and profiles have also been developed, most of them specific to a particular community. The result is a Tower of Babel of metadata schemes, and sharing metadata among professional communities is becoming increasingly difficult. Multiple schemes also make the ideal of successful federated searching all but impossible. Full-text searching, offered as a solution by some, is also a failure for most serious information-seeking needs. On the other hand, the implementation of the MARC format in libraries has been the most successful metadata implementation in history. MARC, along with various content standards, has worked successfully for decades as a metadata schema. The library community's implementation of the MARC metadata standard should be more widely adopted or used as a model for metadata schema design. A single, proven, comprehensive metadata standard will better enable discovery and control of information than a proliferation of minor schemes ever will.