Re: elements with presentational content

I have another example outside of buttons and other activatable content:

Let's say there's this markup that represents current time:

<p>10:55<p>AM</p></p>

In the ideal use case the user should be able to navigate to the outer 
paragraph and hear 10:55 AM. There is no need to step in into the inner 
paragraph since the name of the outer one would've included the sub-tree.

yura


On 2014-05-29, 4:02 PM, Alexander Surkov wrote:
> In case of Firefox we prune the tree in trivial cases like if button 
> contains plain text only but if button has complicated subtree then 
> it's just exposed. So the original issue was about the case when the 
> button has complicated presentational subtree. I believe Firefox could 
> do a good guess about presentability of the content but it'd be nice 
> if ARIA had something to let the author to control that explicitly. Of 
> course it all doesn't make sense if ARIA doesn't allow complicated 
> trees under buttons and doesn't want to. Btw, which browser did you 
> test, was that also Firefox?
>
>
> On Thu, May 29, 2014 at 3:50 PM, Joseph Scheuhammer 
> <clown@alum.mit.edu <mailto:clown@alum.mit.edu>> wrote:
>
>     FWIW, I checked what is exposed in AT-SPI and AXAPI for a simple
>     test case in the test harness [1].  The markup is:
>
>     <div role="button" id="test">
>       Placeholder content
>     </div>
>
>     In both cases, the a11y tree shows a "button" leaf, whose
>     accessible name is "Placeholder content".  Note that they both
>     expose zero children.
>
>     AXAPI:
>     ...
>     AXButton:
>     - AXRole:  AXButton
>     - AXChildren: 0 items
>     - AXTitle: Placeholder content
>
>     AT-SPI:
>     pushbutton:
>     - name:  Placeholder content
>     - Child count 0
>
>     Find attached two pngs that show more information.
>
>     [1]
>     https://dvcs.w3.org/hg/pfwg/raw-file/default/ARIA/1.0/tests/test-files/roles-plain-concrete/roles-plain-concrete-button.html
>
>
>
>     -- 
>     ;;;;joseph.
>
>
>     'A: After all, it isn't rocket science.'
>     'K: Right. It's merely computer science.'
>                  - J. D. Klaun -
>
>