Re: PROPOSAL: Errata text to deal with the issue of predeclared 'xml' and 'xmlns' prefixes

Jeni Tennison wrote:
>
> On 25 Sep 2009, at 12:35, Julian Reschke wrote:
>> Clarifying... this is *really* academic, right? Or are you expecting 
>> people to actually use things which are in the "xml" or "xmlns" 
>> namespaces?
>
>
> Yes, this is *really* academic! I think the spec could easily be 
> changed without affecting any authors.
I concur.  However, I also agree with Philip that making changes like 
this at this time, even via errata, would be a conformance requirement 
change, and that would be bad from a W3C process perspective.  Philip is 
also correct that indicating these prefixes are pre-declared would 
pierce the carefully constructed veil between RDFa and Namespaces in 
XML.  I don't want to do that, really.  I would much rather just require 
that people who want to use xml: as a prefix declare that prefix.  It's 
such an edge case, it won't effect any of our real constituents.

Another approach would be to put in the errata that we recognize that 
the current restriction makes it impossible to create an XSLT-based 
implementation that conforms to this one edge case, encourage authors to 
avoid that edge case, and indicate that in a future version we intend to 
expand the rules so that an XSLT-based implementation is possible (by 
expressly permitting the pre-definition of the 'xml' prefix mapping).

If it sounds like I am waffling on this, it's because I am.  I don't 
have a personal stake in this, I just know it is an issue some people 
care about and I hoped that by putting out a specific errata proposal it 
would spur discussion.  Let's get to consensus on the best way forward 
so we can close this one out.  There are, of course, more!

-- 
Shane P. McCarron                          Phone: +1 763 786-8160 x120
Managing Director                            Fax: +1 763 786-8180
ApTest Minnesota                            Inet: shane@aptest.com

Received on Friday, 25 September 2009 14:21:58 UTC