Re: URIs for "xml" and "xmlns" prefixes?

David Brownell wrote:
> 
> The namespace spec states that there are two "implicitly declared"
> namespace prefixes, "xml" and "xmlns".  But it doesn't give their
> associated URIs, as needed for uniformity within implementations.

I don't see a need for a URI for "xmlns".  Names with the "xmlns"
prefix  are never seen by applications.

We know we need a URI for "xml".

> It should state what the associated URIs are ... I'd propose that the
> URI for the "xml" namespace be that of the XML spec itself, and that
> for the "xmlns" namespace it be the namespace draft.
> 
> Some attention should also be paid to these and related issues:
> 
>     - Assume a prefix declared to use the same URI as "xmlns";
>       for discussion, say "declare-namespace".  Shouldn't
>       "declare-namespace:foo='some-uri'" attributes declare a
>       prefix "foo"?  And shouldn't "declare-namespace='some-uri'"
>       define the default namespace in the current scope?
> 
>     - Same assumption; shouldn't "declare-namespace:xmlns='some-uri'"
>       redefine (within scope) what the "xmlns" prefix represents?
>       Similarly with "declare-namespace:xml='some-uri'", applying
>       to "xml:space", "xml:lang", and so forth.
> 
> One way of looking at those is whether "xml" and "xmlns" are to be
> treated as keywords (ideally, not) or prefixes which, like any other,
> can be used as appropriate for the task at hand (preferable).
> 
> The current draft is inconsistent on such issues; it's written as
> if those are keywords, but says that those prefixes are "declared",
> implying the more consistent and general treatment.

The intention on "xmlns" is that's it's a keyword.

James