RE: No error for missing namespace attribute

Thanks for the clarification, Olivier.  I guess we can all put it on our
to-do lists. ;-)

Best wishes,

Melinda 

> -----Original Message-----
> From: olivier Thereaux [mailto:olivier.thereaux@gmail.com] On 
> Behalf Of olivier Thereaux
> Sent: Wednesday, January 03, 2007 7:38 AM
> To: Jukka K.Korpela
> Cc: Grant, Melinda; www-validator@w3.org
> Subject: Re: No error for missing namespace attribute
> 
> Hello,
> 
> On Jan 2, 2007, at 19:02 , Jukka K. Korpela wrote:
> > On Tue, 2 Jan 2007, Grant, Melinda wrote:
> >
> >> The document at http://www.savagecreek.net/CSS/t1.htm passes 
> >> validation although it is missing the mandatory html namespace 
> >> attribute.
> >
> > The reason is that the requirement on explicit use of the xmlns 
> > attribute has been expressed in prose only, not in the DTDs. The 
> > reason to this is that it cannot be expressed in a DTD, due to the 
> > limitations of the formalism. The attribute is declared 
> #FIXED, which 
> > means by definition that the attribute, if present, must have the 
> > exact value given, but it does not specify that it is required.
> 
> Correct. See also http://lists.w3.org/Archives/Public/www-html-editor/
> 2000OctDec/0045 for the rationale behind the HTML Working 
> Group's choice to declare it as #FIXED rather than #REQUIRED. 
> Regrettable from the perspective of validation, but the 
> culprit here is the DTD, not HTML.
> 
> There is an entry in the validator's bugzilla[1] on this 
> issue, if anyone is interested in either adding this check to 
> the validator or making another small checker that could 
> later be integrated with the validator through the Unicorn[2] 
> framework, feel free.
> 
> [1] http://www.w3.org/Bugs/Public/show_bug.cgi?id=68
> [2] http://www.w3.org/QA/2006/obs_framework/
> --
> olivier
> 
>

Received on Thursday, 4 January 2007 00:22:58 UTC