> -----Original Message-----
> From: ext Jon Hanna [mailto:[log in to unmask]]
> Sent: 14 January, 2003 15:31
> To: [log in to unmask]
> Subject: Re: rdfs:isDefinedBy revisited
>
>
> > "All DCMI namespace URIs will resolve to a machine-processable DCMI
> > term declaration for all terms within that namespace."
> >
> >
> > Please place your arguments relative to that - or
> > argue why you think this recommendation is in error (!) and
> needs (!)
> > to be changed.
>
> I don't think there is much debate on this. The debate seems
> to be exactly
> what this means - those claiming that the document
> "machine-processable DCMI
> term declaration" is what is identified by the URI and those
> (ahem, well me
> anyway) claiming that the document is a representation of
> what is identified
> by the URI.
>
> It's a bit angles-on-pinheads, but the latter opinion allows
> for different
> documents to be retrieved according to HTTP headers so there
> is at least
> some practical point to all this.
I think that this latter view, of representations of the namespace,
is the most productive, as well as better aligned with the present
concensus view of the REST architecture.
A namespace is, well, a set of names. And a representation of that
set of names would involve some enumeration of the names. A
machine-processable DCMI term declaration document seems a reasonable
realization of such an enumeration and hence of a representation
of the namespace.
Of course, one can argue that such a DCMI representation will likely
include information that is not, technically, part of the namespace
proper (the simple set of names) and as such may not be a fully valid
representation, by saying too much, but that is a separate issue entirely.
And, as Jon points out, DCMI term declarations in different forms
could all be equally considered valid alternate representations of the
same namespace.
Thus,
1. The namespace URI denotes the namespace
2. Dereferencing the namespace URI may provide a representation
of that namespace
3. A DCMI term declaration is a representation of a namespace
That said, there is the residual nagging question of whether a
namespace *defines* a term, and as such, can validly occur as
the object of rdfs:isDefinedBy, and if a particular schema
(XML, RDF, RELAXNG, whatever) or even a URI denoting the
organization as the defining authority might better be identified
as the defining resource, with of course a URIref distinct from
the namespace URI.
Something more to think about...
(not that there isn't enough already ;-)
Cheers,
Patrick
|