Re: why are we pursuing this idea? (was: Implementation Details request on Issue 204 Decision)

Steve Faulkner, Tue, 21 Aug 2012 14:49:39 +0100:
> hi leif
> 
>> where the AT would - when presenting the link as a long description
>> link - ignore the textual content of the anchor and only present its
>> URL to the user as a long description link. That is what Firefox does
>> now, already.
> 
> not entirely correct
> 
> firefox exposes the text content of the link as the images accessible
> description (not accessible name) this description may not be
> announced by some AT on images.

Thanks for that. So the text of the 'longdesc linked' link is not lost 
- it is just treated as part of the image. An important correction of 
what I said, I agree.[1] 

However, it then seems to me that I was correct in saying that the link 
gets presented as a longdesc link. That is: without its normal link 
text - and instead with a generic longdesc announcement. (I heard that 
announcement when testing with JAWS.)

[1] In addition it gets read twice,  I suppose, when AT continues with 
the element after the <img> element.
-- 
leif halvard silli