[whatwg] Question regarding accessibility for img

[Pentasis]

From: "Benjamin Hawkes-Lewis" <bhawkeslewis at googlemail.com>
To: "Geoffrey Sneddon" <foolistbar at googlemail.com>
Cc: "Pentasis" <pentasis at lavabit.com>; <whatwg at lists.whatwg.org>
Sent: Sunday, November 30, 2008 9:56 PM
Subject: Re: [whatwg] Question regarding accessibility for img


> Geoffrey Sneddon wrote:
>>
>> On 30 Nov 2008, at 16:40, Pentasis wrote:
>>
>>> I notice that it says in the spec under the img-section:
>>>
>>> "There has been some suggestion that the longdesc attribute from HTML4, 
>>> or some other mechanism that is more powerful than alt="", should be 
>>> included. This has not yet been considered."
>>>
>>> May I ask why it has not been considered (yet)?
>>
>> Because there's an issues list of several thousand issues, and as such 
>> not all issues have been considered. If we could do everything at once 
>> we'd have a spec instantly. :)
>
> Perhaps also worth noting that there's already been a quite epic amount of 
> discussion of LONGDESC, if you care to search the archives. I suppose the 
> text might be more accurate if it said "yet been decided".
>
> A rough summary of the currently dominant view in WHATWG would be that 
> visible descriptions are more useful than invisible descriptions and that 
> in any case LONGDESC is poisoned by real-world abuse ( 
> http://blog.whatwg.org/the-longdesc-lottery ).
>
> --
> Benjamin Hawkes-Lewis

Just a random thought (not a major discussion point afaic):
As I understand it, "best-practice" would now dictate that the image is 
simply explained in the actual content. I agree with this on the most part, 
but I can image the explanation and the image being seperated in "distance" 
from each other. Would it be helpfull for screenreaders to include a 
anchor-point on the image that points towards the explanatory text in such 
cases (which COULD be done with the longdesc), or would you think that be 
overkill?

Bert