On Sun, 2004-01-25 at 21:56, Thomas Baker wrote:
>
> Thank you for raising this, Pete. This is essentially what
> I was trying to get at in an earlier posting:
>
> On October 17, Thomas Baker wrote:
> > A careful reader might wonder, with regard to to Figure 6 in
> > Appendix B, whether the following is intended:
> >
> > [resource URI] --dc:identifier--> [value URI] --rdfs:label--> "[resource URI]"
> >
> > ...where the URIs at both ends are identical.
This is a very good analysis, and I think you are perfectly correct.
I think the fundamental point is that *you need to know* if the encoding
of value of a DC property (say, in XHTML) is
* a value URI, or
* a value string
because
<resource> <dc:identifier> "aURI"
and
<resource> <dc:identifier> <aURI>
are *fundamentally* different things. The first says that "aURI" is an
identifier of the resource. The second says "aURI" identifies an
identifier.
I think in the previous discussion, there were examples of giving a URI
to a container of identifiers, and using that container as value of the
dc:identifier property.
>
> After chewing on it a bit, my tentative and more simple-minded
> conclusion at the time was that the use of dc:identifier to
> hold a URI could simply be considered harmlessly redundant
> in an architecture that supported the identification of the
> subject resource with a URI. But now I'm not so sure...!
Well, there might be multiple identifiers. Say, a URL and a URN.
> I'm also not sure whether this is fundamentally a question
> of usage with respect to dc:identifier or a question of
> data model. I can think of reasons to discuss this issue in
> the Abstract Model; or in the Appendixes evaluating the RDF,
> XML, and XHTML encodings; or in the original RDF, XML, and
> XHTML specifications evaluated in the Appendixes. (But not
> all three!)
It is a question for the abstract model. The AM *must* be able to
distinguish between the use of a URI as a value string and the the use
of a URI as a value URI. It does so, currently. But I'm not sure all
encodings do...
/Mikael
--
Plus ça change, plus c'est la même chose
The more things change, the more they stay the same
|