Re: CP, ISSUE-30: Link longdesc to role of img [Was: hypothetical question on longdesc]

On 3/21/2012 6:44 AM, Charles McCathieNevile wrote:
> On Wed, 21 Mar 2012 11:23:09 +0100, Benjamin Hawkes-Lewis 
> <bhawkeslewis@googlemail.com> wrote:
>
>> On Wed, Mar 21, 2012 at 10:16 AM, Steve Faulkner
>> <faulkner.steve@gmail.com> wrote:
>>> "ARIA is being implemented as an AT-only access mechanism."
>>>
>>> This does not have to be the case, it's not a requirement in ARIA or 
>>> HTML5 and for example there has been positive discussions with the 
>>> Firefox acc team in regards to exposing landmarks to keyboard users.
>
> Although there is definitely no link, and "nothing to see here" 
> implemented (Hmm. For full disclosure, there are easter eggs such as 
> ARIA checkboxes), we have also taken the approach at Opera that we 
> should expose ARIA more generally.

Let me try to clear that one up: I'm happy to see ARIA consumed by 
browsers provided that consumption is handled in a manner that can be 
turned on and off. ChromeVox is a good example of that.

Browsers already make enough trouble. If they start handling ARIA 
without giving users an option, that will overstep well defined boundaries.
CSS selectors do an excellent job of allowing authors to expose ARIA to 
the primary UA.


>
>> Interesting. Do you have a link?
>>
>> Anyhow there's no current spec for linking @aria-describedby to the
>> @longdesc on some image. I don't think it's a particularly elegant
>> approach.
>
> Nor do I.

Well, I guess the inelegant approaches are the best ones!


-Charles

Received on Wednesday, 21 March 2012 17:23:53 UTC