Re: Expectations for presentational element with title child?

The way I wrote up the inclusion/exclusion rules[1], and as I understood it
based on discussion today [2], it would be excluded because the role takes
precedence over native language features.  That's also consistent with
Fred's desire that role=presentation have a decisive effect whenever
possible.

I'm open to hearing arguments that <title> should be treated the same as
aria-label and other global aria attributes on the element, and should
therefore override presentation.  But then we'd want to explicitly mention
"host language alternative text elements and attributes" (as something that
can override a presentation/none role) in the relevant sections of ARIA and
Core-AAM that Rich and Joseph are editing based on today's AAPI call.

~Amelia

[1]: https://lists.w3.org/Archives/Public/public-aria/2016Mar/0158.html
[2]: https://www.w3.org/2016/03/29-aapi-minutes.html#item02

On 29 March 2016 at 15:35, Joanmarie Diggs <jdiggs@igalia.com> wrote:

> Hey all.
>
> I'm doing the golden-hammer-ectomy for WebKit at the moment. I would
> like the task force's expectations for one of my/WebKit's test cases,
> please and thank you. :)
>
> <svg>
>   <g id="test" role="presentation">
>     <title>Venn Diagram</title>
>     <circle cx="50" cy="50" r="20" aria-label="set A"/>
>     <circle cx="50" cy="75" r="20" aria-label="set B"/>
>   </g>
> </svg>
>
> Should the "g" element with id of "test" be:
> a. Excluded from the tree because of the presentational role
> b. Included in the tree because of the "title" element
> c. To be discussed further before consensus can be reached
>
> Thanks!
> --joanie
>
>