Re: [Role Module] Namespace for 'role' Attribute

Doug Schepers wrote:
> Hi, David-
> 
> David Woolley wrote (on 10/30/2007 4:17 AM):
>>

> 
> I don't understand your point.  Deterrent to whom?  Embrace and extend 
> what?

Firstly I'm reading this in the context of languages other than XHTML, 
and sort of assuming that SVG is a prime case.

Whom is the developers of browsers and authoring tools.

Embrace and extend refers to the tendency of these people to retarget 
features, often missing their original point.  An example I've come 
across recently is how the desirable feature of CSS, that it doesn't 
introduce executable content, has been lost as the result of vendor 
extensions.

> mind), would still be done in updates to the Role Attribute spec (rather 
> than in host languages), so it would apply equally across conforming 
> host languages.

The problems arise when developers try to be creative, and extend or 
otherwise improvise on the specification without fully understanding it 
or trying to get it updated first - tag soup happened that way.

By retaining the namespace prefix, one is at least putting out a warning 
that the attribute needs to continue to work properly in all its host 
languages, whereas, without a prefix, developers are likely to ignore 
the impact on other languages, with it there will be some psychological 
pressure not to do so.


-- 
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.

Received on Tuesday, 30 October 2007 22:09:14 UTC