Re: The Kesselman/Connoly proposal (was Re: Re Deprecate/Undefined )

At 11:15 AM 6/25/00 -0400, Michael Champion wrote:
>So, this proposal is:
> - the original intent of most of the Namespace WG (right?)
> - acceptable to those who believe that namespaces must be "real" URIs (?)
> - consistent with existing XSLT practice (if not the letter of the spec)
> - consistent with the DOM Level 2 CR

I'm not completely settled down to liking this proposal, but I'm very happy
with this exchange:
>Q: OK, then, what's the namespace name of the root element?
>A: ../foo , per the namespaces spec as written.
>
>        But be careful about calling it anything else,
>        like "namespace URI" -- that terminology suggests
>        that you're talking about the absolute form
>        of ../foo w.r.t. the relevant base.

Establishing a strong separation between 'namespace names' (roughly,
literal) and 'namespace URIs' (roughly, absolutize) and then determining
the relationship between them seems like a good way to move forward.  That
way, those of us who are only concerned with namespace names can get our
work done, while those more excited about namespace URIs can also get their
work done.

It doesn't seem like supporting both those views of a document should be
that difficult, given the information (base URIs) already provided by the
Infoset.

>As Dan C. says, "this is the best we can do".  Let's do it!

It's a promising path, anyway.  Laying out rules for namespace names -
which should be very simple - and rules for namespace URIs - which will be
more complex - will still take some work, but at least it seems feasible.

(I'm not real happy with Dan's "absolutize for XML 2.0" suggestion, but
moving forward with this separation might create an environment in which we
can get some experience of how this really works.  If we find
absolutization is great in practice, then fine, if it proves unwieldy, then
fine.  I'd like to see practical experience with this issue fed into the
2.0 spec.)

Simon St.Laurent
XML Elements of Style / XML: A Primer, 2nd Ed.
http://www.simonstl.com - XML essays and books

Received on Sunday, 25 June 2000 14:41:38 UTC