- From: Silvia Pfeiffer <silviapfeiffer1@gmail.com>
- Date: Mon, 10 Dec 2012 16:35:12 +1100
- To: Matthew Turvey <mcturvey@gmail.com>
- Cc: "public-html-a11y@w3.org" <public-html-a11y@w3.org>
- Message-ID: <CAHp8n2mzCyi_YNhvUqQPa+gcLUUwCU9budo1GiPWd5kxRywPPg@mail.gmail.com>
I agree with this suggestion of making @longdesc "obsolete but conforming". I would pair it with a proposal of a new attribute that provides links to longer content descriptions for more than just the <img> element. In addition, I would like to see a rationale document for the @longdesc extension spec that addresses the often-heard objections in a succinct manner. In particular I'd like to see an explanation of how the different browsers different in their implementation and interpretation of the value of the @longdesc attribute (some versions of IE mapping it to a description rather than a link) and how AT deal and fix this situation. I believe this is crucial for people to understand - on top of the author misuse of @longdesc, which is far less harmful. This rationale document can be just in the wiki, since it's additional information for the decision making rather than part of the actual specification. HTH. Regards, Silvia. On Fri, Dec 7, 2012 at 5:38 AM, Matthew Turvey <mcturvey@gmail.com> wrote: > Another potential compromise solution for consensus would be to spec > longdesc as "obsolete but conforming", i.e. effectively "deprecated". This option has previously received some support in the TF and HTMLWG: > > In the HTML-A11Y-TF's original poll: > http://lists.w3.org/Archives/Public/public-html-a11y/2010Apr/0180.html > > Richard and Judy: > http://www.w3.org/2011/04/18-text-minutes.html > > Janina: > http://lists.w3.org/Archives/Public/wai-xtech/2012Mar/0014.html > > Steve: > http://lists.w3.org/Archives/Public/wai-xtech/2012Mar/0031.html > > Cynthia > http://lists.w3.org/Archives/Public/public-html-a11y/2012Sep/0289.html > > James: > http://lists.w3.org/Archives/Public/public-html-a11y/2012Nov/0151.html > > Me: > http://lists.w3.org/Archives/Public/public-html/2011Jun/0254.html > > Since the HTML5 spec already requires UAs to expose longdesc [1] this > option would just result in validators issuing a warning instead of an > error. I think this approach would provide better advice to authors > and is more likely to gain consensus in the HTMLWG. > > [1] http://dev.w3.org/html5/spec/single-page.html#dom-img-longdesc > > -Matt > >
Received on Monday, 10 December 2012 05:36:01 UTC