W3C home > Mailing lists > Public > www-html@w3.org > January 2000

Re: tricky XHTML 1.0 namespace question

From: Henry S. Thompson <ht@cogsci.ed.ac.uk>
Date: Mon, 17 Jan 2000 17:07:49 -0500 (EST)
To: XML-Dev Mailing list <xml-dev@ic.ac.uk>, www-html@w3.org
Message-ID: <f5bogakz70w.fsf@cogsci.ed.ac.uk>
David Megginson <david@megginson.com> writes:

> ht@cogsci.ed.ac.uk (Henry S. Thompson) writes:
> 
> [on prefixed and unprefixed attributes]
> 
> > Why should they state it explicitly, when the Namespace REC explicitly
> > says that is NOT the case?  The fact that people on this list have
> > been confused about this does not mean the HTML WG is confused or is
> > responsible for sorting out their confusion.  See my message to Dave
> > about this [1] for a succinct statement of why this is a time-wasting
> > red herring which IS perfectly clearly specified in the Namespace
> > REC.
> 
> The language in the Namespaces REC means that the two *can* be
> distinguished, not that they must be, and every Namespaces-based spec
> should include a explicit statement of its usage.  Based on the
> examples in the spec, RDF, for example, doesn't distinguish
> 
>   <rdf:Description about="http://www.foo.com/">
> 
> from
> 
>   <rdf:Description rdf:about="http://www.foo.com/">

I guess my problem here is that asking XHTML to say they _don't_ do
this falls into the category of what the technical writer who taught
me technical writing called the Spurious Negative.  In general, he
said, avoid statements like "Relay 12 does not energise at this point."

RDF arguably made a mistake.  Not a standards error, just a tactical
mistake.  I don't think that means that every XML aplication from here 
to eternity needs to say "Oh, by the way, we didn't make the confusing 
mistake that RDF did."

It is perfectly true that _applications_ can do anything they bl***y
like, including treating the attributes 'html:head' and 'head' as if
they were the same, just as they can treat 'tete' and 'head' as if
they were the same, or 'HEAD' and 'head' or 'foo' and 'baz'.  XML lets 
them do any of those things.  And when they do any of them, they
should document that fact.  But documenting when they _don't_ do any
of those things is crazy.

But _XML itself_ (nor the Namespace REC) does not distinguish these
cases, or make one pair of names easier for applications to identify
than the others.  And the Namespace REC _does_ make clear that a
conforming _processor_ (not application) _must_ distinguish them, when 
it says that

   <!-- http://www.w3.org is bound to n1 and is the default -->
  <x xmlns:n1="http://www.w3.org" 
     xmlns="http://www.w3.org" >
    <good a="1"     b="2" />
    <good a="1"     n1:a="2" />
  </x>

is valid.

ht
-- 
  Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
     2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
	    Fax: (44) 131 650-4587, e-mail: ht@cogsci.ed.ac.uk
		     URL: http://www.ltg.ed.ac.uk/~ht/
Received on Monday, 17 January 2000 17:12:14 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:40 GMT