On Wed, 3 Oct 2001, Roland Schwaenzl wrote:
>
> In can't see, why a registry requires to change content of the actual DC namespaces.
> I agree with Eric about this.
>
> Their content has to be fixed. They will be needed and used by other applications.
> They form the machine understandable reference. As such they should become a part of the recommendations.
I agree with this, we need a canonical RDF schema expression of DC
vocabulary. What I was not aware of was that we had already fixed the
content of the schemas.
>
> Why a namespace aware RDF application would need to retrieve all
> registry information?
The intention is that the Registry 'adds value' to simple resolution of an
RDF schema, so that one can 'structure' the data in the schema, retrieve
information from a number of schemas etc I see it as a matter of judgement
how much 'structure' one builds into an RDF schema expression of DC terms,
and how much structure one would expect to derive from the added value of
the registry.
The current schema(s) has very little structure, so for example there is
no single 'class' for encoding schemes. There is no way for the Registry
to retrieve a list of all encoding schemes unless one uses a 'canned
search' relying on known semantics e.g. searching for all members of
SubjectScheme and DateScheme and CoverageScheme classes etc etc (as far as
I can see.)
Rachel
>
>
> Additional information needed by the registry can reside wherever you want.
>
>
> cheers
> rs
>
---------------------------------------------------------------------------
Rachel Heery
UKOLN
University of Bath tel: +44 (0)1225 826724
Bath, BA2 7AY, UK fax: +44 (0)1225 826838
http://www.ukoln.ac.uk/
|