Re: name computation for role="checkbox"

Ok, I think I don't mind about either way. Just all controls have to be
synchronized: if checkbox allows name from subtree then slider should do
same and vise versa.

On Fri, Feb 27, 2015 at 3:07 PM, Bryan Garaventa <
bryan.garaventa@ssbbartgroup.com> wrote:

>  I’ve tested many of these out in the past, and they appear to hold true
> using both element types, those that do and don’t support child elements.
>
>
> http://whatsock.com/tsg/Coding%20Arena/ARIA%20Toggles,%20Checkboxes,%20Links,%20and%20Buttons/ARIA%20Checkboxes/demo.htm
>
>
>
> *Though I think there may be some event firing issues in IE11, either that
> or JAWS16 isn’t reflecting the fired state properly such as on the IMG
> example.*
>
>
>
> *From:* Alexander Surkov [mailto:surkov.alexander@gmail.com]
> *Sent:* Friday, February 27, 2015 6:03 AM
> *To:* Dominic Mazzoni
> *Cc:* W3C WAI Protocols & Formats
> *Subject:* Re: name computation for role="checkbox"
>
>
>
> I think the issue is that certain roles are not expected by the assistive
> technologies to have children. So technically you can do a checkbox that
> way but it might be confusing for AT. On the other hand a slider role [1]
> doesn't allow name from subtree, however it doesn't look much different
> from checkbox in terms of leafability.
>
> [1] http://www.w3.org/WAI/PF/aria/roles#slider
>
>
>
> On Thu, Feb 26, 2015 at 12:30 PM, Dominic Mazzoni <dmazzoni@google.com>
> wrote:
>
>  What's wrong with something like this, assuming it has css to make it
> look like a checkbox and js to make it behave like one?
>
>
>
> <div tabIndex=0 role="checkbox" aria-checked="false">I agree to the
> license terms</div>
>
>
>
>
>
>
>
> On Thu, Feb 26, 2015 at 4:48 AM, Alexander Surkov <
> surkov.alexander@gmail.com> wrote:
>
>  Hi, the spec says [1] that role="checkbox" name is computed from
> "contents". Is there an use case for ARIA checkbox name computed from
> subtree?
>
>
>
> Thanks.
> Alexander.
>
>
>
> [1] http://www.w3.org/WAI/PF/aria/roles#checkbox
>
>
>
>
>