Yes, that was mentioned at the face to face.
Rich Schwerdtfeger
From: Joseph Scheuhammer <clown@alum.mit.edu>
To: James Craig <jcraig@apple.com>, Steve Faulkner
<faulkner.steve@gmail.com>, Matthew King/Fishkill/IBM@IBMUS
Cc: W3C WAI Protocols & Formats <public-pfwg@w3.org>
Date: 04/01/2014 04:04 PM
Subject: Re: ISSUE-636 CTION-1398 Provide spec. text for
aria-roledescription
> On Apr 1, 2014, at 7:32 AM, Steve Faulkner<faulkner.steve@gmail.com>
wrote:
>
>> On 1 April 2014 15:23, Matthew King<mattking@us.ibm.com> wrote:
>>> My understanding is that <div role="region"
roledescription="chapter" .... would still expose the role of the div to
the AT as region. None of the role information is lost.
>> what do AT currently do, on Mac for example, does VoiceOver announce
both role an role description?
> VoiceOver announces either the AXRoleDescription or nothing, never the
AXRole. AXRole is a non-localized token value that determines
functionality. The string that a VoiceOver user hears is a localized,
human-understandable string… the "role description." VoiceOver does not
speak the role value literally. ...
FWIW, ATK has a localized role name accessor function [1], and it has
been available for many years. Mind you, it requires a valid ATK role
as input to generate the localized name, so it wouldn't currently work
with the proposed role description attribute. But, it's a possible
route for ATK/AT-SPI to consider.
[1]
https://developer.gnome.org/atk/stable/AtkObject.html#atk-role-get-localized-name
--
;;;;joseph.
'A: After all, it isn't rocket science.'
'K: Right. It's merely computer science.'
- J. D. Klaun -
