Re: URIs quack like a duck

David Carlisle wrote:

> > In the case of the namespace spec, the obligation to support any properties
> > of namespace names beyond the fact that they are syntactic URIs seems
> > questionable.
>
> Some of the proposals have involved comparing the bytes returned by
> dereferencing a namespace name using an http uri scheme.
>
> that would change the answer to the question
>
> is
>
> <x:stylesheet xmlns:x="http://www.dcarlisle.demon.co.uk/foo.html">
> ...
>
> an XSL stylesheet
>
> from the current answer of "No" to a proposed answer of
> "yes" if foo.html is a copy of the html page returned by
> http://www.w3.org/1999/XSL/Transform, the document is an XSL
> stylesheet (until such time as anyone edits the text in either file.)
>
> This would not be an acceptable change.

I agree with you.  But I'd add that creating a new attribute would relieve much
of the pressure for making that change in the first place by solving the
underlying problem in a different way.  (If anyone asks, the underlying problem
is that people want somehow to attach descriptive information -- I'm
deliberately using a very general term here -- to namespaces.)

Paul Abrahams

Received on Thursday, 1 June 2000 11:13:18 UTC