W3C home > Mailing lists > Public > public-html-a11y@w3.org > September 2012

Re: How to obsolete @longdesc

From: Laura Carlson <laura.lee.carlson@gmail.com>
Date: Wed, 19 Sep 2012 14:44:26 -0500
Message-ID: <CAOavpve2NMoHfZvU1_J=VxyZhZUd6yTUvznLqM8Ev7BsMD_5nQ@mail.gmail.com>
To: Leif Halvard Silli <xn--mlform-iua@xn--mlform-iua.no>
Cc: Sam Ruby <rubys@intertwingly.net>, public-html-a11y@w3.org
Hi Leif,

>From all of the prior longdesc discussions, you are probably aware
that I believe that there is nothing inherently wrong with longdesc.
And it is improved in the new spec text. So in that respect, it is
likely that I could not/will not live with:

*  "obsolete" or "obsolete but conforming". (No throwing longdesc
under the bus for ARIA or another feature.)
* A forced visual encumbrance on sighted users of an on-page indicator
or the description itself. (Receiving a description should be a user
opt-in choice.)

It is likely that the consequence of those would result in a formal objection.

The question I posed to the browser vendors was for a FUTURE attribute
based on longdesc. I suspect the way things work around here, is if
that was agreeable to them and it was *expedited* now in this working
group, it would take ten to twenty years. (Don't hold your breath
you'll suffocate.)

We need longdesc in the spec now.

Best Regards,
Laura
--
Laura L. Carlson

On Wed, Sep 19, 2012 at 10:47 AM, Leif Halvard Silli
<xn--mlform-iua@xn--mlform-iua.no> wrote:
> Sam Ruby, Wed, 19 Sep 2012 11:40:23 -0400:
>> On 09/19/2012 11:32 AM, Laura Carlson wrote:
>>> Hi Leif,
>>>
>>> Obsoleting longdesc in any way is not under consideration.
>>
>> That is not a factual statement.  It may not be something that you
>> personally would be willing to consider.  It may not be something
>> that actually happens.  But it is a factual statement that some
>> people are not only considering that, but actively proposing to
>> obsolete longdesc.
>
> What I would like say is that I think David is right when he suggested
> that it should not be a an ARIA feature but a HTML feature. I would be
> more comfortable with such a solution. When we start to mix in ARIA -
> such as in the debates about 'hidden but revealable on such and such
> conditions', then the waters become unclear. Plus that the argument
> could be made that ARIA would be getting that feature, anyhow, at some
> point. And so, it has the smack of appear to be getting something but
> not getting anything anyhow.
>
> Leif H Silli
>
>>> Best Regards,
>>> Laura
>>
>> - Sam Ruby
>>
>>> On Wed, Sep 19, 2012 at 10:29 AM, Leif Halvard Silli
>>> <xn--mlform-iua@xn--mlform-iua.no> wrote:
>>>> Laura Carlson, Wed, 19 Sep 2012 10:19:53 -0500:
>>>>> Hi Leif,
>>>>>
>>>>>> Thank you for pointing to David's message. Clearly, name change might
>>>>>> be a better idea than we have admitted.
>>>>>>
>>>>>> In that case, a logical 'deal' to consider
>>>>>
>>>>> No 'deal' is in consideration. This is an inquiry only.
>>>>
>>>> Sorry, I don't want to disturb the inquiry so I change the topic so you
>>>> don't feel you need to stand for my words.
>>>>
>>>> The argument has been mad, in this recent discussion, that HTML5 has no
>>>> means for deprecation of features. In the first longdesc poll, there
>>>> were no alternatives to replace it with and thus, true deprecation was
>>>> not possible. But if an alternative emerged, then HTML5 has some
>>>> mechanisms for making features obsolete but conforming, as pointed out
>>>> with in my message with the unlucky word 'deal'.[1]
>>>>
>>>> [1]
>>>> http://www.w3.org/mid/20120919165642623450.594a22b3@xn--mlform-iua.no
>>>>
>>>> Leif Halvard Silli

-- 
Laura L. Carlson
Received on Wednesday, 19 September 2012 19:44:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 19 September 2012 19:44:54 GMT