I agree. It would be a hack to map the error message to a description. It must be a different mapping.
Error messages must be separated and what is more they need to be visible unlike descriptions as everyone needs to have access to them. Furthermore, the error message may have links to help information in them. Often these are live region or alerts that pop up in the middle of the page. Also, there may be hidden help information in forms on the form elements. You don't want to mix that with error messages. It is also important that designers check to make sure they have mapped the actual error messages to the invalid form elements. This can be verified by having it in a separate relationship.

Rich Schwerdtfeger
----- Original message -----
From: James Teh <ja...@nvaccess.org>
Sent by: accessibility-ia2-boun...@lists.linuxfoundation.org
To: Joanmarie Diggs <jdi...@igalia.com>
Cc: IA2 List <accessibility-...@lists.linux-foundation.org>, ARIA Working Group <public-a...@w3.org>
Subject: Re: [Accessibility-ia2] Mapping of aria-errormessage for ATK/AT-SPI2 and IA2
Date: Tue, Mar 1, 2016 5:33 PM
On 2/03/2016 12:13 AM, Joanmarie Diggs wrote:
>> If you really do want errormessage to be treated as an entirely
>> separate thing, then I have to change my position on this: we need a
>> new relation and new events.
> Really? Why can't we map it via described-by/description-for and then
> use an object attribute to check if the descriptive element is an object
> attribute?
Everyone else is saying that errormessage is inherently different to
description. If that really is how it's meant to be according to the
ARIA spec (I disagree, but that's not relevant here), then mapping it to
description (even with an attribute) is a big hack. You can't have it
both ways; it's either just a specialisation of description or it's not.
Several people now say it's not the former, so it must be the latter.
And if it *is* the former, then it should be included as part of the
description text as well.

FWIW, I don't like this whle thing at all, but it seems wrong to me that
the ARIA spec says it's not at all a description and the a11y specs say
it is. Personally, I think the ARIA spec *should* treat it as just an
extension of description, but I've already made that point and it
doesn't seem popular.


James Teh
Executive Director, NV Access Limited
Ph +61 7 3149 3306
Facebook: http://www.facebook.com/NVAccess
Twitter: @NVAccess
SIP: ja...@nvaccess.org

Accessibility-ia2 mailing list


Accessibility-ia2 mailing list

Reply via email to