Stefan-
Thanks for the clarification. Follow-up comments noted below.
----- Original Message -----
From: "Stefan Kokkelink" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Tuesday, May 22, 2001 4:39 AM
Subject: Re: Expressing encoding of dcq refinements
> Tim Cole wrote:
> > ...
> > For instance, consider dc:format. An example from "Expressing
Qualified
> > Dublin Core in RDF" is:
> >
> > <?xml version="1.0"?>
> > <rdf:RDF xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"
> > xmlns:rdfs="http://www.w3.org/2000/01/rdf-schema#"
> > xmlns:dc="http://purl.org/dc/elements/1.1/"
> > xmlns:dcq="http://dublincore.org/2000/03/13/dcq#">
> > <rdf:Description>
> > <dc:format>
> > <dcq:IMT>
> > <rdf:value>text/html</rdf:value>
> > <rdfs:label>HTML</rdfs:label>
> > </dcq:IMT>
> > </dc:format>
> > </rdf:Description>
> > </rdf:RDF>
> >
> > Yet, my understanding of dcq semantic meanings is that IMT implies
> > dcq:medium refinement of dc:format. An IMT encoding does not make
sense
> > for the dcq:extent refinement of dc:format.
>
> You are right, we should have used dcq:medium instead of dc:format.
> We will change it to
>
> <dcq:medium>
> <dcq:IMT>
> <rdf:value>text/html</rdf:value>
> <rdfs:label>HTML</rdfs:label>
> </dcq:IMT>
> </dcq:medium>
>
> (Nevertheless: the original example is sematically correct. The
> information given there is implied in the dcq:medium version.
>
yes both ways imply that the given rdf:value and rdf:label are values for
the dc:format property, but only the second version makes clear that they
are also values for the dcq:medium property.
> > ...
> >
> > Following similar logic, I believe that the rdf schema for dcq
> > (http://dublincore.org/2000/03/13/dcq) really shouldn't include a range
> > declaration for dc:format.
>
> Right, we changed this in our proposal
> (appendix) at
>
> http://www.mathematik.uni-osnabrueck.de/projects/dcqual/qual21.3.1/
>
> Note : Our proposal is a modified version of the orininal
> schema at http://dublincore.org/2000/03/13/dcq. They are not
> in sync (yet).
>
Thanks. Sorry I missed that.
> > Shouldn't the range restriction go with dcq:medium instead?
>
> I don't think so. We might want to use other schemes too.
>
okay, makes good sense, but then why were range restrictions on dcq:spatial
and dcq:temporal (refinements of dc:coverage) retained?
Tim Cole
University Library
University of Illinois at Urbana-Champaign
|