> On Mon, 4 Feb 2002, Wagner,Harry wrote:
>
> > > From: Andy Powell
> > > ... DCMI needs to provide help and guidance to those
> > > people who are choosing not to use RDF.
> >
> > Absolutely. We should provide guidance on encoding DC in all practical
> > encoding methods, including XML, RDF and HTML. Personally, I don't care how
> > people choose to encode DC, so long as it is DC they are encoding!
>
> How would you know it was DC?
Obviously, if I use something within my project, I may taylor a XML DTD to
fit my datamodel. I may, or may not use RDF. I may, or I may not, use the
whole triple machinery for traversing graphs when users search. It is none
of you're business what I'm using there.
Outside the internals of my software, encodings start to matter. What my
project present to others, in terms of search results or data exports, is
important. And in that context there are basically two options for XML
encoding of DC metadata, and they are the RDF/XML ones from DCMI and the
XML one from OAI. Both are there, and I can live with them. But I don't
think the world will be any better by increasing the number of encodings
beyond what we have.
I still don't understand why people who don't use RDF (that don't use
graphs and triples) cannot use RDF/XML for metdata exchange. Rather
than reiterating this "we should provide guidance on encoding DC in
all practical encoding methods, including XML, RDF and HTML", tell me
what is lacking, given that we have
http://www.openarchives.org/OAI_protocol/openarchivesprotocol.html,
http://www.dublincore.org/documents/2001/11/28/dcmes-xml/,
http://www.nlc-bnc.ca/bath/bp-app-d.htm,
and
http://www.dublincore.org/documents/2001/11/30/dcq-rdf-xml/
In my view, that would possibly be to pollute the OAI with some
dcterms elements and qualifiers. Otherwise I cannot see that there is
much lacking.
Sigge
|