RE: Are there really no (normative) constraints on use and exposure of aria-atomic and aria-relevant ?

I agree, these attributes only make sense in combination with valid roles or with aria-live in the creation of live regions.

There is something that has been bothering me for years regarding aria-relevant, and nobody appears to have addressed it when I've brought it up in the past, why is the following in the spec?

"aria-relevant is an optional attribute of live regions. This is a suggestion to assistive technologies, but assistive technologies are not required to present changes of all the relevant types."
http://www.w3.org/TR/wai-aria-1.1/#h-aria-relevant

Why does this even have to be in the spec at all? Personally I think this should be deleted, because it basically gives all AT venders a free pass to ignore making live regions work consistently across the board.

If an AT supports live regions then it should do so consistently, otherwise it can ignore all of these attributes if it does not support live regions at all. There should not be any grey areas like this for something that can cause such wildly varying results during implementation.


Bryan Garaventa
Accessibility Fellow
Level Access, Inc.
Bryan.Garaventa@LevelAccess.com
415.624.2709 (o)
www.LevelAccess.com

-----Original Message-----
From: Schnabel, Stefan [mailto:stefan.schnabel@sap.com] 
Sent: Tuesday, September 12, 2017 9:45 PM
To: Joanmarie Diggs <jdiggs@igalia.com>
Cc: ARIA Working Group <public-aria@w3.org>
Subject: Re: Are there really no (normative) constraints on use and exposure of aria-atomic and aria-relevant ?

Hi Joanie,

i think this means they should be used always *together* with aria-live on elements on the base markup and not used "standalone".  Otherwise, what would a button with aria-atomic=true declared only mean? Makes no sense to me. 

The "use" part of the spec could be more clear here.

- Stefan

Sent from my iPad

> On 12. Sep 2017, at 22:44, Joanmarie Diggs <jdiggs@igalia.com> wrote:
> 
> Hey all.
> 
> According to the ARIA spec, aria-atomic and aria-relevant are each 
> supported on "all elements of the base markup." The ARIA spec does 
> state that aria-atomic and aria-relevant are associated with live regions.
> However, I don't see any normative statement in the ARIA spec 
> indicating that authors should/must not use aria-atomic or 
> aria-relevant if the element is not a live region and not inside of a 
> live region. I similarly do not see a statement that user agents 
> should/must not expose these properties if an author does use them 
> outside of a live region. Is this by design?
> 
> The above, combined with the mappings in the Core AAM for 
> aria-relevant and aria-atomic, seem to suggest that if these 
> properties are present, they should be exposed to ATs even if the 
> properties are used outside of a live region. Is that desired?
> 
> --joanie (who hopes she's just missing something obvious)
>

Received on Wednesday, 13 September 2017 16:23:32 UTC