E-mail List Archives
Thread: Seeking public feedback for W3C ARIA role=form landmark mapping
Number of posts in this thread: 12 (In chronological order)
From: Bryan Garaventa
Date: Tue, Apr 19 2016 1:53PM
Subject: Seeking public feedback for W3C ARIA role=form landmark mapping
No previous message | Next message →
Hello,
Currently the W3C ARIA Working Group is looking into a mapping change for the 'form' role as documented at
http://www.w3.org/TR/2016/WD-wai-aria-1.1-20160317/#form
At present, the form role is not mapped in the same manner as other landmark roles within the accessibility APIs, which is why certain landmark behaviors such as the ability to jump between landmark regions and other such features are not automatically available with forms.
The current proposal is to remap form in the same manner as other landmarks to enable this functionality in the future, which will automatically apply to HTML form elements through their implicit role mappings to the ARIA role=form API mapping.
So the question for the public is, do people want forms to act in the same manner as other landmarks?
This feedback will be returned to the W3C ARIA WG for further consideration of this proposal.
Thanks,
Bryan
Bryan Garaventa
Accessibility Fellow
SSB BART Group, Inc.
= EMAIL ADDRESS REMOVED =
415.624.2709 (o)
www.SSBBartGroup.com
From: Schalk Neethling
Date: Tue, Apr 19 2016 1:57PM
Subject: Re: Seeking public feedback for W3C ARIA role=form landmark mapping
← Previous message | Next message →
How are forms currently mapped? or are they not mapped at all?
On Tue, Apr 19, 2016 at 9:53 PM, Bryan Garaventa <
= EMAIL ADDRESS REMOVED = > wrote:
> Hello,
> Currently the W3C ARIA Working Group is looking into a mapping change for
> the 'form' role as documented at
> http://www.w3.org/TR/2016/WD-wai-aria-1.1-20160317/#form
>
> At present, the form role is not mapped in the same manner as other
> landmark roles within the accessibility APIs, which is why certain landmark
> behaviors such as the ability to jump between landmark regions and other
> such features are not automatically available with forms.
>
> The current proposal is to remap form in the same manner as other
> landmarks to enable this functionality in the future, which will
> automatically apply to HTML form elements through their implicit role
> mappings to the ARIA role=form API mapping.
>
> So the question for the public is, do people want forms to act in the same
> manner as other landmarks?
>
> This feedback will be returned to the W3C ARIA WG for further
> consideration of this proposal.
>
> Thanks,
> Bryan
>
>
>
> Bryan Garaventa
> Accessibility Fellow
> SSB BART Group, Inc.
> = EMAIL ADDRESS REMOVED =
> 415.624.2709 (o)
> www.SSBBartGroup.com
>
>
> > > > >
--
Kind Regards,
Schalk Neethling
Senior Front-End Engineer
Mozilla ::-::
From: Moore,Michael (Accessibility) (HHSC)
Date: Tue, Apr 19 2016 2:03PM
Subject: Re: Seeking public feedback for W3C ARIA role=form landmarkmapping
← Previous message | Next message →
I can see some advantages in making it a part of the region navigation - particularly if the form is not near the beginning of the main region. Of course if there are multiple forms on a page then they would need to be identified with aria-label or aria-labelledby.
Mike Moore
Accessibility Coordinator
Texas Health and Human Services Commission
Civil Rights Office
(512) 438-3431 (Office)
From: Bryan Garaventa
Date: Tue, Apr 19 2016 2:10PM
Subject: Re: Seeking public feedback for W3C ARIA role=formlandmark mapping
← Previous message | Next message →
Currently form is already mapped as a subclass of landmark, this would simply bump them up in the accessibility API chain.
Functionality changes would be noticed most readily using NVDA and Orca, which is why we are seeking public feedback as to whether this change is desirable.
Bryan Garaventa
Accessibility Fellow
SSB BART Group, Inc.
= EMAIL ADDRESS REMOVED =
415.624.2709 (o)
www.SSBBartGroup.com
From: Birkir R. Gunnarsson
Date: Wed, Apr 20 2016 5:57PM
Subject: Re: Seeking public feedback for W3C ARIA role=form landmark mapping
← Previous message | Next message →
I'd propose to treat the form role the same way the region role is
treated, i.e. to expose it as a landmark only if it has an accessible
name.
On 4/19/16, Bryan Garaventa < = EMAIL ADDRESS REMOVED = > wrote:
> Currently form is already mapped as a subclass of landmark, this would
> simply bump them up in the accessibility API chain.
>
> Functionality changes would be noticed most readily using NVDA and Orca,
> which is why we are seeking public feedback as to whether this change is
> desirable.
>
>
>
> Bryan Garaventa
> Accessibility Fellow
> SSB BART Group, Inc.
> = EMAIL ADDRESS REMOVED =
> 415.624.2709 (o)
> www.SSBBartGroup.com
>
>
>
From: Léonie Watson
Date: Thu, Apr 21 2016 2:47AM
Subject: Re: Seeking public feedback for W3C ARIA role=formlandmark mapping
← Previous message | Next message →
> From: WebAIM-Forum on Behalf Of Birkir R. Gunnarsson
> Sent: 21 April 2016 00:58
>
I'd propose to treat the form role the same way the region role is treated, i.e.
> to expose it as a landmark only if it has an accessible name.
>
+1
Léonie.
--
@LeonieWatson tink.uk Carpe diem
From: Steve Faulkner
Date: Thu, Apr 21 2016 3:30AM
Subject: Re: Seeking public feedback for W3C ARIA role=form landmark mapping
← Previous message | Next message →
Hi Bryan, some issues/potential issues:
The form element is problematic (i think) because it is not generally used
by developers as a part of visible UI, so it does not always represent the
form user interface area. (an example of this is use of form to wrap all
elements in a page, a pattern that is not uncommon in legacy pages using
ASP).
I have heard and noted (anecdotal) that the form element can effect the way
AT interact with content, both controls and static content.
I would suggest (as has been suggested) that the form element only be
exposed/announced when it has an accessible name. This will help ensure
that a) legacy and future use of <form> without thought to its semantics in
relation to UI, will not negatively effect users. If it is the case that
the form landmark semantics are desired to be exposed, the descision to
apply an accessible name will result in a wilful desciion to expose as a
landmark in the UI.
--
Regards
SteveF
Current Standards Work @W3C
<http://www.paciellogroup.com/blog/2015/03/current-standards-work-at-w3c/>
On 19 April 2016 at 20:53, Bryan Garaventa < = EMAIL ADDRESS REMOVED =
> wrote:
> Hello,
> Currently the W3C ARIA Working Group is looking into a mapping change for
> the 'form' role as documented at
> http://www.w3.org/TR/2016/WD-wai-aria-1.1-20160317/#form
>
> At present, the form role is not mapped in the same manner as other
> landmark roles within the accessibility APIs, which is why certain landmark
> behaviors such as the ability to jump between landmark regions and other
> such features are not automatically available with forms.
>
> The current proposal is to remap form in the same manner as other
> landmarks to enable this functionality in the future, which will
> automatically apply to HTML form elements through their implicit role
> mappings to the ARIA role=form API mapping.
>
> So the question for the public is, do people want forms to act in the same
> manner as other landmarks?
>
> This feedback will be returned to the W3C ARIA WG for further
> consideration of this proposal.
>
> Thanks,
> Bryan
>
>
>
> Bryan Garaventa
> Accessibility Fellow
> SSB BART Group, Inc.
> = EMAIL ADDRESS REMOVED =
> 415.624.2709 (o)
> www.SSBBartGroup.com
>
>
> > > > >
From: Tim Harshbarger
Date: Fri, Apr 22 2016 8:23AM
Subject: Re: Seeking public feedback for W3C ARIA role=formlandmark mapping
← Previous message | Next message →
I agree.
I think there would be value to the user experience to treat a form like a landmark when it has an accessible name, but no value if the form has no accessible name.
I think there might be a couple of cases where treating any form as a landmark might either make the UI a bit more cumbersome to use or where the landmark adds no value.
First, there is the situation where a section with a landmark role may only contain a form. For example, it is not uncommon to see a site search that uses a form. If the site search is also contained inside a search landmark, it ends up being a bit pointless.
We also have situations where a page might contain multiple forms. Again, knowing that this part of the page contains some kind of form content probably helps the user very little.
However, if the user knows that this is the newsletter subscription form, the landmark becomes a whole lot more useful.
Honestly, I could see a lot of value in identifying a form as a landmark when it has an accessible name. It definitely could help on pages where there are multiple "forms" or on pages where we want to provide a more efficient way to navigate to an important form.
Thanks,
Tim
From: Maxability Accessibility for all
Date: Sat, Apr 23 2016 11:30PM
Subject: Re: Seeking public feedback for W3C ARIA role=form landmark mapping
← Previous message | Next message →
+1 to Birkir. I had the same question in mind as Tim has.
Will it not be a conflict for landmarks such as role="search" and <div
role="form" aria-label="search">
Thanks & Regards
Rakesh
www.maxability.co.in
On Fri, Apr 22, 2016 at 7:53 PM, Tim Harshbarger <
= EMAIL ADDRESS REMOVED = > wrote:
> I agree.
>
> I think there would be value to the user experience to treat a form like a
> landmark when it has an accessible name, but no value if the form has no
> accessible name.
>
> I think there might be a couple of cases where treating any form as a
> landmark might either make the UI a bit more cumbersome to use or where the
> landmark adds no value.
>
> First, there is the situation where a section with a landmark role may
> only contain a form. For example, it is not uncommon to see a site search
> that uses a form. If the site search is also contained inside a search
> landmark, it ends up being a bit pointless.
>
> We also have situations where a page might contain multiple forms. Again,
> knowing that this part of the page contains some kind of form content
> probably helps the user very little.
>
> However, if the user knows that this is the newsletter subscription form,
> the landmark becomes a whole lot more useful.
>
> Honestly, I could see a lot of value in identifying a form as a landmark
> when it has an accessible name. It definitely could help on pages where
> there are multiple "forms" or on pages where we want to provide a more
> efficient way to navigate to an important form.
>
> Thanks,
> Tim
>
>
>
From: Tim Harshbarger
Date: Mon, Apr 25 2016 6:58AM
Subject: Re: Seeking public feedback for W3C ARIA role=formlandmark mapping
← Previous message | Next message →
Hi,
My thought is that, if forms are only recognized as landmarks when they have labels, then you would just have to remove the aria-label from the example to ensure that only the search landmark is recognized.
<div role="search">
<form>
...
</form>
</div>
For the type of environment I work in, I think this approach would be more beneficial. On pages where we might have multiple forms within a main or other landmark, the role="form" and aria-label would give us an additional way to help differentiate forms for users. In existing landmarks where we might only have one form, we wouldn't be introducing an additional landmark that has little or no value for the user experience.
From: Bryan Garaventa
Date: Mon, Apr 25 2016 9:56AM
Subject: Re: Seeking public feedback for W3C ARIArole=formlandmark mapping
← Previous message | Next message →
Thanks everybody for your feedback, I'll pass this along for the UAIG meeting tomorrow.
Kind regards,
Bryan
Bryan Garaventa
Accessibility Fellow
SSB BART Group, Inc.
= EMAIL ADDRESS REMOVED =
415.624.2709 (o)
www.SSBBartGroup.com
From: Bryan Garaventa
Date: Wed, Apr 27 2016 9:37AM
Subject: Re: Seeking public feedback for W3C ARIArole=formlandmark mapping
← Previous message | No next message
Okay, as a quick update regarding the future of role=form, we've agreed to map this to landmark for the explicit role, and to only map role=form implicitly on native HTML form elements when it has an explicit label using either aria-labelledby or aria-label.
The relevant meeting minutes are posted at
http://www.w3.org/2016/04/26-aapi-minutes.html#item05
So to clarify, any container element that includes the actual attribute role="form" will be treated as a landmark no matter if it has a label.
However, all native HTML form elements (<form> ... </form>) that don't have a label will act the same as they do now, unless aria-labelledby or aria-label is used to set a label for that form, in which case it will also automatically map to landmark.
Thanks again for the feedback,
Bryan
Bryan Garaventa
Accessibility Fellow
SSB BART Group, Inc.
= EMAIL ADDRESS REMOVED =
415.624.2709 (o)
www.SSBBartGroup.com