- From: <james.nurthen@oracle.com>
- Date: Sat, 26 Jul 2014 13:20:44 -0700
- To: Steve Faulkner <faulkner.steve@gmail.com>
- Cc: "public-html@w3.org" <public-html@w3.org>
> On Jul 26, 2014, at 12:34, Steve Faulkner <faulkner.steve@gmail.com> wrote: > > Hi James > > Sent from my Chorizone > >> On 26 Jul 2014, at 20:38, "james.nurthen@oracle.com" <james.nurthen@oracle.com> wrote: >> >> >> >>> On Jul 26, 2014, at 10:43, Steve Faulkner <faulkner.steve@gmail.com> wrote: >>> >>> Hi James >>> >>> Sent from my iPhone >>> >>>> On 26 Jul 2014, at 18:59, James Nurthen <james.nurthen@oracle.com> wrote: >>>> >>>> If these are focusable by default (as is currently the case in FF) then 2 additional things need to happen >>>> 1) it MUST be possible for a web developer to reverse this behaviour. The current FF behaviour is causing us issues in some combobox implementations and needs to be resolved. I would suggest allowing tabindex=-1 to ensure the area is never focusable. This does not work in the current FF implementation >>> >>> Does using tabindex=-1 resolve the issue in FF? >> >> No. FF still focuses it. > > Suggest that is a bug, best to file one against FF I agree. Unfortunately I'm on vacation for 3 weeks and have no laptop with me. I'll try to remember to do it when I get back. As you can imagine getting an extra focusable area in a widget plays havoc with widgets which use either roaming tabindex or aria-activedescendent to manage their focus. > >>> >>>> 2) An accessible name and role needs to be provided for this region so AT can determine what to speak when it gets focus. >>> >>> Being scrollable, I would suggest, is a property not a role. >> >> What role would you give it then? > > If the element already has a role there is no need. > <section> for example has a default role of region. > If not would depend on what the element content was, I think. Most of the time I imagine the element would be a div. I seem to recall the mapping for this differ vastly depending on the platform. > >> >>> >>> >>>> >>>> Regards, >>>> James >>>>
Received on Saturday, 26 July 2014 20:21:16 UTC