Re: xmlns in HTML5 (was: Telecon Agenda- Thursday 1500 UTC)

On Thu, 16 Jul 2009, Toby Inkster wrote:
> On Thu, 2009-07-16 at 09:20 +0000, Ian Hickson wrote:
> > On Thu, 16 Jul 2009, Toby Inkster wrote:
> > > Ian Hickson wrote:
> > > 
> > > > Authors must not use elements, attributes, and attribute values that
> > > > are not permitted by this specification or other applicable
> > > > specifications.
> > > >  -- http://www.whatwg.org/specs/web-apps/current-work/#semantics-0
> > > 
> > > Which "other specifications" are "applicable"?
> > 
> > Pretty much any that claim to be and that the people affected agree
> > are applicable.
> 
> So assuming that people deem the HTML 4.01 Recommendation to be 
> "applicable", the much debated @summary, @longdesc and @profile may all 
> be used in HTML5?

If someone intends to follow HTML4 rather than HTML5, then HTML5 isn't 
relevant. HTML4 doesn't claim to add to HTML5, so it's not applicable to 
the quoted statement above.


> A document using @summary is conformant if and only if the person
> assessing the document's conformance deems HTML 4.01 to be applicable?

HTML4 is a poor example because it is replaced wholesale by HTML5, they 
don't really work together. However, yes, whether you decide to evaluate a 
document's conformance against HTML4, HTML5, or, say, DOCBOOK, depends 
entirely on your own desires. If one person looks at a document and 
considers HTML5 relevant and says "this is invalid because it has the 
wrong DOCTYPE", they are no more right or wrong than another who looks at 
it and decides that HTML4 is more relevant and says "this is invalid 
because it uses a <video> element".

(BTW, summary="" is actually permitted, though marked obsolete, in HTML5.)

-- 
Ian Hickson               U+1047E                )\._.,--....,'``.    fL
http://ln.hixie.ch/       U+263A                /,   _.. \   _\  ;`._ ,.
Things that are impossible just take longer.   `._.-(,_..'--(,_..'`-.;.'

Received on Thursday, 16 July 2009 10:59:44 UTC