Re: 48-Hour Consensus Call: ARIA-DescribedAT & Longdesc

On Thu, Mar 29, 2012 at 5:45 PM, Janina Sajka <janina@rednote.net> wrote:
> Colleagues:
>
> On 29 March last the HTML-A11Y Task Force teleconference meeting
> reached consensus as follows:
>
> RESOLUTION: The HTML-A11Y Task Force confirms that ARIA-DescribedAT will
> not be ready for HTML 5 in HTML 5's currently published timeframe, and
> therefore reaffirms its support of Laura's authored CP to reinstate
> longdesc (Issue-30).
>
> The TF resolution, together with minutes of the discussion leading up to
> it, is logged at:
> http://www.w3.org/2012/03/29-html-a11y-minutes.html#item03
>
> As usual, if there is objection to this consensus position, please
> respond by replying to this message no later than close of business,
> Boston Time, on Monday 2 April.

I object to this position.

>From what I've seen so far, it is not realistic to expect a future
attribute with semantics like @longdesc to address its use cases so
much better that it justifies deprecating @longdesc. Therefore the
HTML WG should exclude such expectations from its consideration of
whether to make @longdesc conforming. The consensus position suggests
a different timeline for such expectations rather than expressing
their fundamental unrealism. As not everyone who thinks we should meet
@longdesc's use cases necessarily agrees we must meet them within
HTML5's timeframe, the consensus position would continue to damage the
fair consideration of our proposal for Issue 30.

I believe assessments of the likelihood of user agents implementing
@longdesc should (and will) enter into HTML WG's consideration of
whether to recommend @longdesc to authors. I think Opera's inclusion
of a long description context menu item during HTML5's development
period was a positive step here, but it seems unlikely we will see
additional implementations before the WG tries to decide Issue 30. The
response from other user agent vendors has been lukewarm at best.
Recommending without caveat that authors produce long descriptions
that lots of users cannot easily access could be a dereliction of
duty. So if we want to push to make @longdesc conforming, we should
arguably also be pushing to:

   - Suggest in authoring guidance that authors make long descriptions
discoverable using visible webpage elements.

   - Surface test-based implementation status warnings (for @longdesc
*and* other features like <details>) in HTML authoring guidance and in
conformance checkers.

It is possible that open advocacy of these positions might improve the
chances of our Issue 30 proposal.

--
Benjamin Hawkes-Lewis

Received on Saturday, 31 March 2012 09:39:33 UTC