In translating OLAC metadata into the XML representation advocated in
"Recommendations for XML Schema for Qualified Dublin Core"
[http://www.ukoln.ac.uk/metadata/dcmi/xmlschema/], I've run into a
problem with refinements.
We have a vocabulary, OLAC-Role = [author, editor, transcriber, sponsor,
informant], which serves as a refinement for BOTH dc.creator and
dc.contributor. As long as this refinement was represented as an attribute
there was no problem, e.g.:
<dc.creator refine="editor">...</dc.creator>
<dc.contributor refine="editor">...</dc.contributor>
However, if refinements are represented as full elements we are forced to
give them a unique affiliation with an unqualified element, via the
substitutionGroup attribute.
I've thought of two kludges:
(i) partition our refinements between creator and contributor
(ii) drop the creator/contributor distinction since its partially
redundant given our refinement.
In both cases, idiosyncracies of the XML implementation of DCQ and of
the expressive shortcomings of XML schema are in the driving seat.
It would be nice if there was a transparent way to implement this
simple idea.
Thanks for any advice,
-Steven
--
Steven Bird Email: <[log in to unmask]> Web: http://www.cs.mu.oz.au/~sb/
A/Prof, Dept of Computer Science, University of Melbourne, Vic 3010, AUSTRALIA
Senior Research Assoc, Linguistic Data Consortium, University of Pennsylvania
|