CORES: A Forum on Shared Metadata Vocabularies
by Tom Baker
Can different standards communities agree on principles to help translate among a diversity of metadata languages?
The Semantic Web vision of merging information from a diversity of Internet sources presupposes methods for processing diverse types of descriptive metadata. The past decade has seen numerous initiatives to standardise metadata semantics for particular application communities, from multimedia and commerce to scholarly publishing. In practice, there is some overlap between these communities of practice, and implementors often draw pragmatically on available standards in designing metadata for particular applications, 'mix-and-matching' as needed.
Processing metadata scalably will require infrastructures to interpret, translate, and convert among various metadata languages automatically. The construction of such infrastructures will however require the availability of metadata dictionaries in forms usable by software agents, middleware applications, and indeed human beings. Such dictionaries should ideally be based on conventions, shared across metadata communities, for declaring metadata vocabularies machine-understandably.
The objective of the CORES Project, an Accompanying Measure under the Semantic Web action line of the EU's Fifth Framework Programme, is to facilitate the sharing of metadata semantics on several levels. At the level of standardisation initiatives, the project is bringing together key figures from major standardisation activities in a Standards Interoperability Forum to discuss principles and practicalities of 'playing together' in the Web environment. The participants include representatives from initiatives as diverse as DCMI, OASIS, DOI, CERIF, MPEG-7, IEEE/LOM, GILS, ONIX, and the W3C Semantic Web Activity (with apologies for the unexpanded acronyms). While undertaken with reference to Semantic Web technologies, the Forum includes communities that are pursuing different strategies for interoperability.
At the level of infrastructure, CORES is working with organisations that are designing or building value-added services for metadata-using applications. Building on the prior SCHEMAS Project (2000-2002), CORES is expanding a registry of application profiles and liaising with related activities on developing good practice for interoperability between specialised registries. This effort will include a hands-on 'profile-writing' workshop with implementors.
One goal is to test the SCHEMAS model for declaring application profiles as simple sets of RDF assertions - a form which in principle lends itself to merging into 'semantic landscapes' of information providers.
Project partners are Pricewaterhouse-Coopers (Luxembourg), UKOLN, University of Bath (UK), Fraunhofer-Gesellschaft, and SZTAKI.
Thomas Baker, Fraunhofer-Gesellschaft
Tel: +49 30 8109 9027