Project tackles metadata mess

Metadata is an essential tool for cataloguing in libraries, but publishers and libraries use different systems to create metadata. This makes making communication between the two difficult and means that a great deal of duplication can occur.

A pilot project run by OCLC, whose members include more than 60,000 libraries in 112 countries, hopes to change this. This library service and research organisation is exploring the viability and efficiency of capturing metadata from publishers and vendors upstream and enhancing that metadata in its online library catalogue WorldCat.

OCLC plans to work with up to five major publishers who will provide title information in ONIX format. OCLC will convert this data to the MARC format, which is used by libraries. This will be added to WorldCat and, where possible, automatically enriched through data mining and data mapping. Enriched metadata will then be returned to publishers and vendors in ONIX format for evaluation of the OCLC enhancements. Library pilot partners will evaluate the resulting MARC records for use in library technical services processes.

'It is crucial to the future of cataloguing to find collaborative ways to take advantage of publisher ONIX metadata, and we must find efficient and centralised ways to store, enhance and normalise the metadata for the benefit of both library and publishing communities,' said Renee Register, OCLC global product manager for OCLC cataloguing and metadata services. 'Librarians can and should participate in raising the quality of metadata in the marketplace.'

OCLC hopes the pilot will result in ongoing processes for the early addition of new title metadata to WorldCat. 'For many types of materials, most libraries wait to have a book in-hand before creating the metadata,' explained Register. 'We want to be able to get metadata before then and make the process more efficient. For example, if a library knows a hardback book is coming out in paperback, creation of this metadata should be automatic, whereas today it is manual.'

OCLC also intends this pilot to help break down barriers between publishers and libraries and enable libraries to have a common platform for transforming metadata rather than each library developing its own method.

'By working with the major publishers, we hope to develop a system that everyone can use,' said Register. 'We hope to have some early results at the London Book Fair in April and will need the community to help us decide the way forward.'

The ability to use upstream publisher data effectively is central to the recommendations included in the 'Future of Bibliographic Control' draft report issued in November 2007 and sponsored by the US Library of Congress. 'Development and plans for pilot launch were well underway at OCLC prior to the report and OCLC is pleased that this strategy is in line with the Working Group recommendations,' said Register. OCLC believes that the establishment of data flows and practices that allow the interoperability of metadata and the remix and reuse of existing metadata will be crucial for both libraries and publishers.