Re: [URIEquivalence-15] Namespaces in XML -- URI, IRIs and equivalence

Might I respectfully ask that the Tag solicit and consider the experiences 
of implementors before finalizing any decisions regarding the 
interpretation of namespace names?  I am not trying to signal that any one 
proposal or another is necessarily problematic.  I am signalling that, as 
I mentioned in an earlier note, seemingly simple features of namespaces 
have already proven to be extremely challenging in building high 
performance XML implementations--in this case, I speak from first hand 
experience. 

Of course, there may be good architectural reasons for making decisions 
that compromise performance, and we don't yet know that there are problems 
with any of the proposals.    I'm merely suggesting that as you converge 
on a set of workable options, you check with implementors and consider 
their reports in making any final decision.  Thank you!

------------------------------------------------------------------
Noah Mendelsohn                              Voice: 1-617-693-4036
IBM Corporation                                Fax: 1-617-693-8676
One Rogers Street
Cambridge, MA 02142
------------------------------------------------------------------







Tim Bray <tbray@textuality.com>
Sent by: www-tag-request@w3.org
06/07/02 02:02 PM

 
        To:     www-tag@w3.org
        cc:     (bcc: Noah Mendelsohn/Cambridge/IBM)
        Subject:        Re: [URIEquivalence-15] Namespaces in XML -- URI, IRIs and equivalence


Larry Masinter wrote:

>>I think we can coherently assert and (probably should)
>>that character-by-character means that %6a and %6A 
>>and 'j' are the same character. -Tim
> 
> It's imperative that you don't treat %2f and '/' as the
> same character, so I'm not sure how coherent it would be.

I assumed we'd specifically except reserved characters, as enumerated in 
the RFC.  Or is there another gotcha? -Tim

Received on Friday, 7 June 2002 15:10:02 UTC