Re: Next steps for the ARIA syntax discussion

On Fri, 30 May 2008 16:00:23 +0200, Elliotte Harold  
<elharo@metalab.unc.edu> wrote:

> Al Gilman wrote:
>
>> WAI-ARIA could have ridden on namespaces, and *would have* if
>> namespaces were ready for prime time.  But they're not.

I disagree strongly with this characterisation of the issue.

In SVG, namespaces are a critical part of the ecosystem. In HTML, they are  
simply not there. In XHTML it seems that people have actually slightly  
misunderstood the namespaces spec and the many interoperable  
implementations of that spec.

Anyway, the aria- approach works as implemented with both  
namespace-unaware HTML, and namespace-reliant SVG. That's the strength of  
the proposal. The draft is, I hope, going to be changed to clarify that  
ARIA attributes should always be "in no namespace" to use the 'technically  
correct' (but misleading) terminology of the day.

What it really means is that the appropriate namespace for these  
attributes is the empty string - xmlns:aria="" if you are going to use  
aria:aria-[something] in a namespace aware environment, but  
aria-[something] will work perfectly correctly in both a namespace-reliant  
environment, and in a namespace-unaware environment, because of a careful  
and thoughtful design decision in the namespaces specification that allows  
for the easiest possible transition between the two kinds of environment.

> OK. Seems you are rejecting namespaces in toto because you don't like  
> them.

In no way. I will ahppily charaacterise a lot of the people who have  
supported the "aria-" syntax in this discussion as people who happen not  
to like namespaces, but as someone who is a strong supporter of  
namespaces, I think that believing this is about the pros and cons of  
namespaces per se has got hold of the wrong issue.

> The decision, therefore, comes down to this: how much does following the  
> web architecture matter?

I don't think so. The mistake was suggesting that there should be a  
namespace other than the null string for aria attributes. ("In no  
namespace" is somehow consdered a more useful phrase despite the  
misunderstandings it has caused, but I think it is an idiotic piece of  
terminology to continue with). I hope and believe that the PF group are  
about to correct that error, and therefore have a way of doing ARIA that  
is consistent with actual implementations and the HTML and XML Namespaces  
specifications and their discernible futures.

As Anne has pointed out, aria- actually works happily in any spec that can  
see its way to saying "attributes whose names start with 'aria-' and whose  
namespace is the empty string mean exactly what they mean in the ARIA  
specifications". In practice, as implementors, by assuming that key specs  
(like SVG or MathML) will say that we get very simple and very effective  
implementation of very important accessibility improvements. This seems to  
me an ideal result, and a good demonstration that W3C's technologies are  
actually well-designed and valuable for real-world things.

cheers

Chaals

-- 
Charles McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg lærer norsk
http://my.opera.com/chaals   Try Opera 9.5: http://snapshot.opera.com

Received on Sunday, 1 June 2008 01:03:13 UTC