W3C home > Mailing lists > Public > www-tag@w3.org > April 2008

Re: Namespaces and XML/XHTML

From: Anne van Kesteren <annevk@opera.com>
Date: Fri, 18 Apr 2008 10:32:55 +0200
To: "Henry S. Thompson" <ht@inf.ed.ac.uk>, "Henri Sivonen" <hsivonen@iki.fi>
Cc: www-tag@w3.org
Message-ID: <op.t9sso5yr64w2qv@annevk-t60.oslo.opera.com>

On Fri, 18 Apr 2008 10:17:53 +0200, Henry S. Thompson <ht@inf.ed.ac.uk>  
wrote:
> Henri Sivonen writes:
>>> Exactly -- having aria-role and aria:role as aliases is bad design.
>> Having only aria-foo and not having aria:foo solves the problem.
>
> That proposal is not, as far as I know, on the table -- I would
> certainly argue against it as fundamentally breaking XML design
> principles.

Why? We decide to extend some set of languages with a set of attributes  
that are shared accross those languages. These are attributes that work  
well in HTML and XML, can be used for styling in an identical way in both.  
Can be manipulated through the DOM in an identical way in both and don't  
clash with any of the languages we want to use them in.


Your proposal on the other hand gives different attributes for HTML and  
XML. (In HTML you get "aria:x" not in a namespace and in XML you get "x"  
in the ARIA namespace.) Styling does not work in an identical way. In HTML  
you need to use [aria\:x] and in XML you need to bring in @namespace  
complexity. So you need to write more complex style sheets. The DOM also  
does not work in an identical way as you need to use setAttributeNS /  
getAttributeNS rather than setAttribute / getAttribute. Finally, the  
syntax is also not identical because in XML you would be required to have  
an xmlns declaration in scope.

That you can't see that your proposal is far worse for everyone involved  
is just beyond me.


-- 
Anne van Kesteren
<http://annevankesteren.nl/>
<http://www.opera.com/>
Received on Friday, 18 April 2008 08:33:52 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 26 April 2012 12:47:56 GMT