admin followup Re: longdesc requirements Re: alt tags for graphs

On Mon, 01 Apr 2013 17:00:05 +0300, Jonathan Avila  
<jon.avila@ssbbartgroup.com> wrote:

> [Chaals wrote] (do you have a pointer to that mail? I can't find it :( )
>
> Here is the link to the post that I approved but it evidently was never
> posted to the public HTML a11y list.
> http://www.w3.org/Mail/review?id=ca4a184393bfadaa22d7  This is not the  
> first time this has happened.

Yes, and I have confirmed that it happens to other people. I'm following  
up with W3C.

> [Chaals wrote] I'm not sure I understand. The requirements on user agents
> are:
>
>> User agents should make the link available to all users through the
>> regular user interface.
>> User agents should expose the link to relevant APIs, especially
>> accessibility-oriented APIs.
>> User agents should enable users to discover when images in a page  
>> contain a long description.
>
>> Are you simply proposing that the "should" requirements be strengthened  
>> to MUST? And that there should be a matching requirement to make them
>> keyboard accessible?
>
> Yes, that is what I was hoping.
>
>> I would certainly be happy if the should listed become musts. If you  
>> would like to raise a bug, please do, or I can do it if you prefer.

OK, bug filed. https://www.w3.org/Bugs/Public/show_bug.cgi?id=21566

>> I don't think that the normative requirement to be keyboard-accessible
>> belongs in this spec - it is a general  accessibility requirement for  
>> User Agents that run on devices where keyboards can be used, so is in
>> UAAG. I would propose instead adding an informative note and a  
>> reference,
>> if you think it is necessary to make this particular point.
>
> I would be happy with that.
>
> [Chaals wrote]
>> It would surprise me, as the person who wrote the text of the longdesc
>> spec. I would think the HTML spec is wrong to insist on that, and would
>> not want to follow. But I just read
>> http://www.w3.org/TR/html5/embedded-content-0.html#the-img-element (the
>> Candidate Recommendation draft) all the way through, and as far as I can
>> tell it doesn't have that requirement.
>
> Past discussions about the topic of displaying alt text for images tend  
> to go back to the requirements I pulled from the HTML 5.1 nightly draft  
> below (which historically has been the same in earlier specifications).
>
> http://www.w3.org/html/wg/drafts/html/master/embedded-content-0.html#alt
>
> "If the image is available and the user agent is configured to display  
> that image, then the element represents the element's image data."
>
> "Otherwise, the element represents the text given by the alt attribute.  
> User agents may provide the user with a notification that an image is
> present but has been omitted from the rendering."
>
> "The alt attribute does not represent advisory information. User agents  
> must not present the contents of the alt attribute in the same way as
> content of the title attribute."
>
> So, strictly speaking user agents could provide access to the alt text
> somehow -- but not through a tooltip and not on-screen where the image
> should go.  So it doesn't leave too many options and doesn't provide any
> discussion of recommendations on providing this information to users.

OK, I understand what you are getting at.

Actually, I suspect the bug should be filed on the HTML spec. In practice  
user agents *do* present title and alt in the same way, if they are using  
a title as a fallback where there isn't an alt available, and I am not  
sure that there is anything wrong with that.

>> I agree that user agents that don't make longdesc available to all users
>> all the time are not doing as well as they should. Putting requirements  
>> in specifications doesn't, by itself, change implementations. But I
>> welcome input on strengthening the requirements...
>
> Access to longdesc information in my opinion is more critical than alt  
> text because deciphering a descriptive meaning out of an image requires
> more cognition and visual skill than abstracting the meaning from a
> simple image.

Yeah, I certainly agree about that. The key question is "what changes to  
the spec will actually help achieve that result?" Beyond the spec there is  
plenty of work to do. One of the things that I think would be useful is an  
informative document that talks about best practices for using longdesc.  
And that quite possibly just means doing work on HTML techniques.

cheers

Chaals

-- 
Charles McCathie Nevile - Consultant (web standards) CTO Office, Yandex
       chaals@yandex-team.ru         Find more at http://yandex.com

Received on Wednesday, 3 April 2013 11:49:39 UTC