|
|
On Tue, 13 Feb 2001, Andy Powell wrote:
...
> DC 1.1 (the element set) is fine
> DCQ 2000-07-11 (the qualifiers) is fine
> DCQ 1.0 does *NOT* exist - there is no such thing!
>
> Therefore, putting a version number in the namespace URI for DCQ seems a
> little odd?
It does indeed. Now, here lies problem. All properties in DCQ yyyy.mm.dd
arises as subproperties of DC x.y, that is in the metadata for DCQ there
has to be something like
<rdf:description about=".../dcq-yyyy.mm.dd">
...
<dcq:requires resource=".../dcmes-x.y.">
...
</rdf:description>
I rember a conversation between DanBri, Stu and myself back in Ottawa
about merging everything which the dc-usage committee likes together in a
single namespace. This would have the virtue of giving us finite number of
namespaces ;)
Two questions in relation to this:
- Is there hooks for this in EOR toolkit merging and splitting schemas in
the EOR and/or SWAG
- Is there anything in RDF(S) saying that an application should use the
<rdfs:isDefinedBy rdf:resource = ...>
construct for forming the name space URIs? That would indeed help increase
the stability of the vocabularies
Sigge
|
|
|
|