RE: Safari lacks support for aria-activedescendent: questions

Hi,

I agree both are very different, however aria-activedescendant has no impact on iOS so there is nothing to add support for there.

 

The purpose of aria-activedescendant is to simulate a focus shift in the accessibility tree so that the AT user perceives that focus has moved into another widget, even when actual keyboard focus remains on the originating control. For a combobox, this is on the input element while aria-activedescendant references the child option roles within a listbox.

 

This occurs by tracking specific keyboard keys such as the up/down/left/right arrow keys, and then changing the value of aria-activedescendant to reference the id of the child role in the controlled widget.

 

On iOS, there are no arrow keys in the virtual keyboard to retain focus on a form field and process these key events. Also, attaching a Bluetooth keyboard won’t work to do this either, because the external arrow keys don’t perform the same actions as they do on a Windows keyboard, so there are no arrow key events to process the attribute change.

 

A while back, I tried to change the keyboard settings in an attempt to force this behavior, but found this was impossible to achieve.

 

An ARIA combobox does already work accessibly on iOS with VoiceOver, though it does need some tweaking to ensure the dynamic listbox doesn’t disappear when focus moves away from the input field and a live region announcement mechanism needs to be added to update the user when typing into the virtual keyboard.

 

All the best,

Bryan

 

 

From: Steve Green <steve.green@testpartners.co.uk> 
Sent: Friday, August 30, 2024 6:26 AM
To: Taliesin Smith <talilief@gmail.com>; WAI Interest Group discussion list <w3c-wai-ig@w3.org>
Cc: Karen Lewellen <klewellen@shellworld.net>; Suman Damera <suman.damera@phenom.com>
Subject: RE: Safari lacks support for aria-activedescendent: questions

 

That’s welcome news, but it doesn’t necessarily mean the bug will be fixed on iOS, which has massively more usage than macOS.

 

Steve Green

Managing Director

Test Partners Ltd

 

From: Taliesin Smith <talilief@gmail.com <mailto:talilief@gmail.com> > 
Sent: Thursday, August 29, 2024 10:14 PM
To: WAI Interest Group discussion list <w3c-wai-ig@w3.org <mailto:w3c-wai-ig@w3.org> >
Cc: Karen Lewellen <klewellen@shellworld.net <mailto:klewellen@shellworld.net> >; Steve Green <steve.green@testpartners.co.uk <mailto:steve.green@testpartners.co.uk> >; Suman Damera <suman.damera@phenom.com <mailto:suman.damera@phenom.com> >
Subject: Re: Safari lacks support for aria-activedescendent: questions

 

Hello again Interest group,

 

Thanks for everyone’s work-around ideas and thoughts on deployability.

 

I just got a message from Apple’s Web Accessibility team that they have fixed the aria-activedescendent bug in Sequoia.

 

I have not yet verified this, but I thought I would share the news yo wrap-up this thread.

 

Please test away on Sequoia!

 

Have a wonderful day,

Taliesin

 

 

~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
Taliesin L. Smith
talilief@gmail.com <mailto:talilief@gmail.com>  

taliesin.smith@colorado.edu <mailto:taliesin.smith@colorado.edu>  

Inclusive Design Research Specialist
PhET Interactive Simulations
http://phet.colorado.edu/
Department of Physics
University of Colorado, Boulder

 

 

 

 

 

On Aug 29, 2024, at 03:24, Suman Damera <suman.damera@phenom.com <mailto:suman.damera@phenom.com> > wrote:

 

Hi,

The workaround is to use aria-live. While navigating from one option to another, we need to trigger aria-live with the name of the option so that it announces the option names while traversing them in VoiceOver on Safari. However, this approach might cause the options to be announced twice in other combinations, such as NVDA with Chrome, JAWS with Chrome, and so on. Just some thoughts here.

Thank you.
---

This version improves the flow and clarity of the content while maintaining the original intent. Let me know if you need any more adjustments!

 

On Thu, Aug 29, 2024 at 5:45 AM Karen Lewellen <klewellen@shellworld.net <mailto:klewellen@shellworld.net> > wrote:

I would add that Voiceover works with many  browsers.  a safari bug is 
not  necessarily a voiceover one.  I vote for  a design that broadens 
usability for more people..but hey I am progressive enhancement  kind of 
cheerleader whenever possible.
Kare



On Wed, 28 Aug 2024, Steve Green wrote:

> I don’t know of a workaround for this Safari issue, and I just tell clients not to use design patterns that rely on . aria-activedescendant.
>
> When making recommendations for our clients, we usually go for a third option, which is to design the feature differently, such that all the functionality is retained, but there may be a minor reduction in usability. It’s not ideal, but I think it’s a better trade-off than having good usability for some people and no accessibility for others.
>
> Waiting for browser and AT vendors to fix issues is a hopeless strategy because most don’t publish a roadmap at all, and those that do only publish it a few months ahead. There are countless bugs that should be easy to fix but have been present for years, so you can’t assume that easy bugs will be fixed soon.
>
> Steve Green
> Managing Director
> Test Partners Ltd
>
>
> From: Taliesin Smith <talilief@gmail.com <mailto:talilief@gmail.com> >
> Sent: Wednesday, August 28, 2024 8:39 PM
> To: WAI Interest Group discussion list <w3c-wai-ig@w3.org <mailto:w3c-wai-ig@w3.org> >
> Subject: Safari lacks support for aria-activedescendent: questions
>
>
> Dear WAI ARIA-IG,
>
>
> It seems that it is a well known issue that “Voiceover on the Mac does not properly support aria-activedescendant to ensure accessibility for the blind.” (Webkit bug list<https://bugs.webkit.org/show_bug.cgi?id=269026>).
>
>
> This bug makes it so that the selected option is not always read out - mostly it is not read out, but there are times when it gets read out. The behavior is quite buggy. This bug affects the recommended Design Pattern documented on the ARIA Practices website for grid-combo<https://www.w3.org/WAI/ARIA/apg/patterns/combobox/examples/grid-combo/> and the auto-complete combo<https://www.w3.org/WAI/ARIA/apg/patterns/combobox/examples/combobox-autocomplete-list/>.
>
>
> I have 2 questions:
>
>  1.  Does anyone on this list know of a good work around for Safari?
>
>
> My second question is more nuanced as our team tries to assess the inclusion of this Design Pattern on our website. Pros: the pattern offers important search functionality for long lists of options, including mid-string searches which are important for us. The pattern works when using the Chrome browser.
>
>
>
>  1.  From a WCAG perspective, how do people generally address browser-based bugs that are out of their hands?
>
>     *   Do you include the pattern in your website knowing that some people will need to change browsers to access certain aspects of your website, and wait for the vendor to address it? OR
>
>     *   Do you design an alternative that offers less functionality for everyone?
>
>
> I would appreciate thoughts on both questions 1 and 2. If question 2 is not appropriate for this list, I can re-post that question on the WCAG-IG list.
>
>
> Taliesin
> ~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~.~
> Taliesin L. Smith
> talilief@gmail.com <mailto:talilief@gmail.com> <mailto:talilief@gmail.com <mailto:talilief@gmail.com> >
> taliesin.smith@colorado.edu <mailto:taliesin.smith@colorado.edu> <mailto:taliesin.smith@colorado.edu <mailto:taliesin.smith@colorado.edu> >
>
> Inclusive Design Research Specialist
> PhET Interactive Simulations
> http://phet.colorado.edu/
> Department of Physics
> University of Colorado, Boulder
>
>
>
>
>
>

 

  <https://get.paubox.com/hbfs/Email%20folder/green_lock_v2.png> Secured by Paubox <https://www.paubox.com/>  - HITRUST certified