Re: validator does NOT validate XHTML 1.0

scripsit Devon Y.:
> -- Brant Langer Gurganus wrote:
> >Actually, I believe it does validate because it is automatic via
> >the DTD file:
> ><!ATTLIST html
> > %i18n;
> > id          ID             #IMPLIED
> > xmlns       %URI;          #FIXED ''
> > >
> >As you can see, the DTD tells that it is a fixed attribute.  If
> >the system reference was not in the DOCTYPE declaration, it would
> >be invalid.
> Thank you for answering that. Now I'm confused though. Since the  
> Recommendation clearly says the document must have the namespace in it -  
> if a document doesn't, then it is not a valid document, right? Or does 
> including it in the DTD as a #FIXED attribute, equal to it actually being 
> in the document?

I'm not really an XML/SGML guru ('though such folks are known to hang
out here), but it seems to me that the contents of the DTD are
considered to be part of the document.  In generic XML, the sort of
things that in (X)HTML go in the DTD actually appear in the document
itself (like setting #FIXED attributes, defining entities [&amp; etc.],

> I wouldn't think so, because leaving the namespace out of the doc directly, 
> makes it unusable in Opera 6.04 and Mozilla 1.1a (appears like all inline 
> style-less text & no images). I would think that if a #FIXED attribute in a 
> DTD was equal to the attribute being in the doc itself, that Opera & 
> Mozilla would still display the page as any XHTML 1.0 doc w/ namespace 
> directly in it, would be displayed.

That's _bizarre_, and very broken behaviour if the browsers do that.
I've never seen such a thing, though.  Can you send one or more URI that
exhibit this?  

If, however, the effect is what you suggest, then bugs need to be filed
with both Opera and Mozilla.

Thanasis Kinias
Graduate Student, Department of History
Arizona State University
Tempe, Arizona, U.S.A.

Ash nazg durbatulūk, ash nazg gimbatul,
Ash nazg thrakatulūk agh burzum-ishi krimpatul

Received on Monday, 19 August 2002 06:09:29 UTC