Andy,
> OK, I'll have a go at re-wording. My personal view is that
> it is worth noting in the para that XML-based encodings do
> use a combination of namespace URI reference and term name to
> encode the term URI reference.
OK, but I'd like to see this clearly separated out from the issue of
creating/assigning URI references. And I'd say that _some_ XML encodings
use a combination of namespace URI reference and term name to encode the
term URI reference in _some_ contexts.
> > 3. An XML Namespace is _not_ a collection of URI references.
>
> I don't think I say that anywhere do I? If I do, then I
> didn't mean it!
> :-)
No, you're right, you don't say it ;-) But the use of the expression
"URI references within the same XML namespace" had me worried enough to
think the question needed clarification.
> > The "Namespaces in XML" spec [1] says (this is reproduced
> in section 1
> > of the current doc, and I'm suggesting above that it could
> be removed
> > ;-)):
> >
> > "An XML namespace is a collection of names, identified by a URI
> > reference [RFC2396], which are used in XML documents as
> element types
> > and attribute names."
> >
> > This says that the XML namespace itself, the collection of
> names, is
> > identified by a URI reference. It does _not_ say that the
> individual
> > names within that collection are URI references.
>
> I know... and it is correct... I'm unlcear why this sentence
> should be removed.
OK, maybe it isn't necessary to remove it, but I don't think it is
appropriate to have it where it is in the document currently, right in
the middle of introducing the idea of identification of DCMI terms by
URI references, because it conveys the impression that XML Namespaces
are integral to this identification and I think we're agreeing that they
aren't.
Would it be possible to structure the document so that
(a) the introduction divides into two main paragraphs
- the first paragraph talking about identifying terms using URI
references (without ever mentioning namespaces of any form)
- the second paragraph talking about representing URI references using
namespaces/qualified names, one mechanism for which is XML Namespaces
(b) section 2 provides the general guidelines and introduces the idea
that terms are grouped into "vocabularies" (or whatever we call the
groups! I don't really mind as long as it isn't "XML Namespace"!) i.e.
some combination of the current text and the amendments I suggested last
time?
(c) section 3 provides examples for choosing/constructing/assigning URI
references for terms, discussion about persistence, pros and cons of
including version numbers, different URI schemes etc - not a million
miles away from the present stuff, but without the emphasis XML
Namespaces!
(c) there is a new section 4 which talks about the fact that URI
references may be represented as qualified names - and the XML
Namespaces stuff gets introduced here. (However, I think we have to be
very careful not to suggest that the reverse is true i.e. that all XML
QNames in XML documents equate/map simplistically to URI references.)
> > FWIW, I think the "DCMI Namespaces" are examples of what the RDF
> > Primer calls "vocabularies".
>
> Yes :-(
The DCMI Namespace Policy doc suffers much more from the "XML Namespace
problem", I think. The intro to that document is much more problematic
because it seems to highlight the XML Namespace over the URI:
===
An XML namespace [XML-NAMES] is a collection of names, identified by a
URI reference [RFC2396], that are used in XML documents as element types
and attribute names. The use of XML namespaces to uniquely identify
metadata terms allows those terms to be unambiguously used across
applications, promoting the possibility of shared semantics. DCMI adopts
this mechanism for the identification of all DCMI terms.
===
I'm not sure of the procedure for addressing this. Is the Namespace
Policy document "owned" by the DC Architecture WG? If so, would there be
any chance of scheduling a work item for next year to review the
Namespace Policy document and perhaps to align it more closely with the
terminology used in the recent RDF recommendations?
Pete
|