Re: Question about a rowgroup note

Joseph: Thanks for your response!

Everyone else: Input please. :)

When I sent this message a couple of days ago [1], I was operating under
the impression that we'd have new tbody, thead, and tfoot ARIA roles --
as I alluded to in that message.

Having explicitly asked about tbody and friends [2], I am starting to
suspect that this might not come to pass after all. :) As a result, the
first part of the current note about rowgroup would still stand / not be
obsoleted. The second part of the note still confuses me. And as you'll
see in Joseph's reply below, I'm apparently not the only one. Since I'm
top-posting, I'll repeat the note and my question here.

<quote>
The rowgroup role exists, in part, to support role symmetry in HTML, and
allows for the propagation of presentation inheritance on HTML table
elements with an explicit presentation role applied.
</quote>

Could someone please elaborate on the use case -- and for that matter
meaning -- of "allows for the propagation of presentation inheritance on
HTML table elements with an explicit presentation role applied." Bonus
points for an example that we might be able to include in the spec so
that readers know exactly what the note is stating.

Thanks!
--joanie

[1] https://lists.w3.org/Archives/Public/public-pfwg/2015Mar/0240.html
[2] https://lists.w3.org/Archives/Public/public-pfwg/2015Apr/0005.html

On 03/31/2015 12:40 PM, Joseph Scheuhammer wrote:
> Hi Joanie,
> 
> You wrote, quoting the note in the current ARIA 1.1 draft (which is also
> part of the REC spec):
>> <quote>
>> The rowgroup role exists, in part, to support role symmetry in HTML, and
>> allows for the propagation of presentation inheritance on HTML table
>> elements with an explicit presentation role applied.
>> </quote>
> 
> The latter part of this note is bizarre -- where rowgroup inherits the
> presentation role.  The reason is that rowgroup has a required context,
> namely a grid or treegrid role.  Thus, authors can do something like:
> 
> <div role="grid">
>   <div role="rowgroup"> ... </div>
> </div>
> 
> And, they might also be able to do:
> 
> <table role="grid">
>   <tbody role="rowgroup"> ... </tbody>
> </table>
> 
> But, author's cannot put a rowgroup inside a presentational table, since
> that kind of table is definitely not a grid nor a treegrid. Technically,
> since it violates the required context role for rowgroup, it's illegal
> to do:
> 
> <table role="none">
>   <tbody role="rowgroup"> ... </tbody>
> </table>
> 
> My best guess is that this part of the note is saying that, since
> rowgroup is structurally equivalent to thead, tbody, and tfoot, it
> inherits the presentation/none role from the ancestor context role. But,
> that structure isn't allowed in the first place.
>

Received on Wednesday, 1 April 2015 13:47:02 UTC