- From: Richard Schwerdtfeger <schwer@us.ibm.com>
- Date: Tue, 1 Apr 2014 10:14:24 -0500
- To: Joseph Scheuhammer <clown@alum.mit.edu>
- Cc: W3C WAI Protocols & Formats <public-pfwg@w3.org>
- Message-ID: <OFF7551833.B6BF534B-ON86257CAD.005341A7-86257CAD.0053B704@us.ibm.com>
Joseph,
Yes, we agreed to limit the use of this attribute to ONLY group and region.
If you look at the supported text sections I mentioned for group or region
I stated that the property would not be inherited by subclasses:
The following spec. text goes into role group and role region under
supported states and properties:
<li><a class="property-reference-" href="
states_and_properties#aria-roledesc"><code>aria-roledesc</code></a>
(noninhertitable by subclass roles)</li>
Rich Schwerdtfeger
From: Joseph Scheuhammer <clown@alum.mit.edu>
To: Richard Schwerdtfeger/Austin/IBM@IBMUS, W3C WAI Protocols &
Formats <public-pfwg@w3.org>, Richard
Schwerdtfeger/Austin/IBM@IBMUS
Date: 04/01/2014 08:40 AM
Subject: Re: Action : 1398 Provide spec. text for aria-roledescription
Hi Rich,
Thanks for getting the ball rolling. I have a a point of
clarification. Your spec text states (my emphasis) that the role MUST
have a computed value of *group* or *region*. What's not clear from
this is whether the role resolves specifically to group or region, or if
the computed role value can be a subclass of group or region.
For example, in the taxonomy, the "row" role is a subclass of group, but
we are not allowing aria-roledesc on "row" at this point, correct?
My recollection from the telecon is that a role description has more
value for generic roles since it provides a degree specificity for that
role. So, we want to be clear that aria-roledesc applies specifically
to group or region, and nothing subclassed from them.
--
;;;;joseph.
'A: After all, it isn't rocket science.'
'K: Right. It's merely computer science.'
- J. D. Klaun -
Attachments
- image/gif attachment: graycol.gif
Received on Tuesday, 1 April 2014 15:14:59 UTC