Re: Drop longdesc, get aria-describedat?

On Mon, 12 Mar 2012 21:56:26 +0100, Silvia Pfeiffer  
<> wrote
> On 13/03/2012, at 5:53 AM, John Foliot <> wrote:
>> Quoting Charles McCathieNevile <>:
>>> On Wed, 07 Mar 2012 20:04:16 +0100, Leif Halvard Silli

>>> The specification can be done in a tiny amount of time if that is  
>>> needed. But without implementor commitment, it just isn't needed very  
>>> urgently. Understanding the basics can be done just as well by  
>>> implementing longdesc...
>> Let's be crystal clear: without further support from the tool vendors  
>> (and I sidestep the fact that the browsers are a significant, but not  
>> exclusive member of that group) @longdesc will languish under-used,  
>> cheating users from functionality they require. But rushing to dump it  
>> and insert something "new" with even less support is stupid, and I will  
>> go so far as to suggest that anyone who fails to understand *THAT* also  
>> deserves the same title.

Sure. What I meant by the below though, is that if it is stupid, but  
*works*, then it isn't so stupid.

>>> If aria-describedat will get implemented, that is pretty much trumps  
>>> for me. But if an ongoing discussion about it is an excuse to do  
>>> nothing for a few extra weeks, I'd rather talk about something more  
>>> productive.
> We're spinning in circles. If implementers would rather implement a new  
> attribute the same across all browsers and for more elements than just  
> img, we should enable them to do so. Refusing to produce a spec because  
> isn't helping.

I absolutely agree. If there is a will to implement, I don't think  
producing a spec is the issue. If that were the roadblock, I could find  
resources to make sure a spec is produced in a timely manner.

I was trying to point out that *without* implementation commitment, having  
a spec is not that useful, since it requires some days of real work and  
the result isn't so different from where we are already. Certainly I think  
it should be obvious to a person of reasonable intelligence and  
familiarity with the state of the art roughly what such a spec might say.



Charles 'chaals' McCathieNevile  Opera Software, Standards Group
     je parle français -- hablo español -- jeg kan litt norsk       Try Opera:

Received on Tuesday, 13 March 2012 00:14:57 UTC