background image
The Dublin Core Metadata Initiative
The Dublin Core Metadata Initiative is an open forum engaged in the development of
interoperable online metadata standards that support a broad range of purposes and business
models.The initiative developed the Dublin Core Metadata Element Set in 1995/96, with
the goal of supporting simple resource discovery for digital collections in different domains.
In 2000, the basic element set has been extended with the Dublin Core Qualifiers for use
within local applications or specific domains. Despite the criticism that Dublin Core is a
rather crude standard for describing cultural heritage objects, it has been officially adopted
in the meantime by various country governments, including Denmark, UK and Australia.
Besides, the CIMI Consortium has extensively tested the Dublin Core Metadata Standard
within the museum domain and published the results.
<http://dublincore.org/>
Further reference: CIMI Guide to Best Practice: Dublin Core <http://www.cimi.org/public_docs/meta_bestprac_v1_1_210400.pdf>
At present, the first model of previously harvesting data to be stored in a central repo-
sitory that can then be searched is the more viable approach and has already attracted many
followers. On the other hand, the second model that proposes to search at the institutional
level currently faces many difficulties as the technological infrastructure is not yet robust
and fast enough to handle user queries.
Both models, however, still face problems with regard to the quality of search results.
The current Dublin Core Metadata Standard, is a good compromise to enable cross sector
resource discovery today, yet it has some shortcomings that should be addressed in further
developments. One such shortcoming, for example, is the loss of context. As Debbie
Campell, Project Manager of Picture Australia, puts it:"A collection of images may have a
collective title such as `Images of Paul Revere'. But the image title may be reduced to `On a
Horse'. So the loss of context becomes a discovery issue." (Campell in Tennant, 2001) In
addition, there are also problems with differing vocabularies that are used within the several
cultural heritage sectors, which can create a major impediment to accessibility across
sectors.
To minimise information loss in the future, a medium range strategy would be to further
develop Dublin Core into a metadata standard that is still compatible with current Dublin
Core, but that offers more granularity. Although a "Dublin Core II"-standard would not
allow to conduct machine inferences, it would address some of the current shortcomings.
A long-term strategy towards cross-sectoral standards for resource discovery, however,
should try and explore new grounds in the form of knowledge-based modelling with high
expressivity. Knowledge-based modelling would deliver high granularity of metadata and
therefore better search quality, without loosing interoperability.
Authority control
The introduction of automated information retrieval and the demand to exchange data
not only between institutions within one sector but also between sectors and across national
borders brought a dramatic increase in the demand for vocabulary control.Today, the
problem of providing seamless access to digital cultural information resources is not so
much the technology but, terminology, i.e. the fact that people describe the same things
differently. Authority files help to standardise object descriptions across sectors, and would
support persistent identification of objects in networked environments.
198
IX TECHNOLOGY
ON THE RAD
AR