- From: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
- Date: Thu, 30 Aug 2012 15:56:21 +0200
- To: Peter Winnberg <peter.winnberg@gmail.com>
- Cc: Mathew Marquis <mat@matmarquis.com>, HTML WG <public-html@w3.org>, public-respimg@w3.org
Peter Winnberg, Thu, 30 Aug 2012 15:22:31 +0200: > 2012/8/30 Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>: >> (1) To repeat oneself is an antipattern. In my view, it should be >> considered valid if the <img> does not contain any @alt as long as it >> instead points, via aria-labelledby="", to the picture element: >> >> <picture alt="Alernative text" id="pict" > >> <img src=file aria-labelledby="pict" /> >> </picture> > > This is insufficent because older browsers / AT that have no support > for the picture element Well, they would be told that it exists. > or ARIA would be left with an img element without alt text. I don't understand what you say here. > So the alt attribute is always needed on the img element. So would the opposite be sufficient? I.e. the alt on the <img> but no alt on the <picture> *but* aria-labelledby on the <picture> instead: <picture aria-labelledby="pict" > <img src=file id="pict" alt="Alernative text" /> </picture> Originally, this was what I wanted to propose. However, one issue with is that it seemingly makes the conformance criteria of the <picture> element more intricate. Another, and in theory simpler variant that I considered, was to DROP the alt attribute on the picture element and instead say that any child img element(s) with non-empty alt serve(s), by default, as the alternative text of a picture element. -- leif halvard silli
Received on Thursday, 30 August 2012 13:56:57 UTC