A simpler format for OLAC vocabularies and schemes

Steven Bird sb at UNAGI.CIS.UPENN.EDU
Tue Sep 24 22:25:11 UTC 2002


Thanks for the positive feedback.  While we await more reactions let me
jump in and say that Gary and I are working on a revised version of
the proposal to bring it into line with new developments in the Dublin Core
Metadata Initiative (DCMI).  We'll preserve the new extensibility that
people seem to appreciate, but also make syntactic changes to maximize
interoperability with the wider digital libraries community.

In the past we've basically gone it alone in working out how to represent
our own DC qualifications in XML.  However, the timing of these
recommendations and our forthcoming workshop present us with a new
opportunity to standardize our implementation.

If you'd like to learn more about what's happening in DCMI with qualifiers
and XML, please see the following article and the material it cites:

  Recommendations for XML Schema for Qualified Dublin Core
  Proposal to DC Architecture Working Group
  http://www.ukoln.ac.uk/metadata/dcmi/xmlschema/

Next week we'll circulate a proposal for how OLAC can conform with this.

Note that this is only about XML implementation and not OLAC content.
For those who only care about disseminating metadata, conformance with the
DCMI recommendations will ensure maximal interoperability with the wider
digital libraries community, so that your metadata pops up all over
cyberspace.


Back on the subject of extensibility...  The key innovation in our recent
proposal, that we'd still like more feedback on, is for the OLAC
vocabularies to be changed from being centrally enforced standards to
recommended practices.  Under this model, any archive will be able to adopt
and promulgate its favorite ontologies, while the OLAC Process is still
used to identify community-agreed best practices that everyone should
follow.

For instance, consider the sourcecode vocabulary, which is only relevant to
the software archives and which may need constant updates.  Under the
proposed model, the vocabulary wouldn't actually need to reside on the OLAC
site; it could live wherever it could be easily maintained.  However, the
OLAC site would host the details of any associated working group, so that
others could discover the group and contribute to the revision of the
vocabulary.  It would also host any associated OLAC recommendation, so that
everyone would know that the OLAC community had adopted a certain
vocabulary as best practice.

-Steven



More information about the Olac-implementers mailing list