Re: Logic behind aria-required not being valid on role=checkbox?

ARIA can allow things HTML doesn't, which is usually the point.

A very similar example — @aria-readonly is allowed on more role types than
HTML @readonly, which is only for text fields.

Based on the definition of @aria-required that Matt King provided I think
this makes good sense. I wouldn't call it a huge world changing
improvement, but it seems useful and will quite possibly "just work" when
the required state is exposed to current screen readers. Even if it
doesn't, the graceful degradation of not speaking it is very good.

Aaron

On Mon, Sep 18, 2017 at 3:14 AM Michiel Bijl <michiel.list@moiety.me> wrote:

> It doesn’t follow the logic as described in the HTML spec[1] though. The
> value is never an empty string. Unless a checkbox’ value is an empty string
> when unchecked I don’t see how we can add this functionality to ARIA if
> it’s not in the HTML spec.
>
> Perhaps I’m missing something here.
>
> 1: https://w3c.github.io/html/sec-forms.html#the-required-attribute
>
>
> — Michiel
>
> On 18 Sep 2017, at 08:47, Matt King <a11ythinker@gmail.com> wrote:
>
> In the case described, the user interface is requiring a specific value.
> The question is whether ARIA should be able to help assistive technologies
> in a case like that. If the AT can identify required fields to the user,
> then, shouldn't the AT be enabled to identify the checkbox that requires
> user action in order for the user to proceed?
>
> The definition of aria-required is:
>
> Indicates that user input is required on the element before a form may be
> submitted.
>
>
> The checkbox Bryan describes meets that definition.
>
> Matt
>
> -----Original Message-----
> From: Michiel Bijl [mailto:michiel.list@moiety.me <michiel.list@moiety.me>]
>
> Sent: Sunday, September 17, 2017 11:20 PM
> To: Matt King <a11ythinker@gmail.com>
> Cc: Bryan Garaventa <bryan.garaventa@levelaccess.com>; ARIA Working Group
> <public-aria@w3.org>
> Subject: Re: Logic behind aria-required not being valid on role=checkbox?
>
> Thing is that the checkbox can’t be required as such because it always has
> a value, it’s always 0 or 1. It’s never invalid. HTML required doesn’t
> dictate the final value. Neither should aria-required.
>
> — Michiel
>
> On 18 Sep 2017, at 03:59, Matt King <a11ythinker@gmail.com> wrote:
>
>
> That seems valid to me. Worth raising an issue.
>
>
> Matt
>
>
> -----Original Message-----
>
> From: Bryan Garaventa [mailto:bryan.garaventa@levelaccess.com
> <bryan.garaventa@levelaccess.com>]
>
> Sent: Friday, September 15, 2017 7:26 PM
>
> To: ARIA Working Group <public-aria@w3.org>
>
> Subject: Logic behind aria-required not being valid on role=checkbox?
>
>
> Hi,
>
> Somebody brought a use case to my attention today regarding the use of
>
> aria-required and role=checkbox, and it seems like a valid one,
>
> however aria-required doesn't appear to be supported on role=checkbox.
>
>
> The use case being that of a License Agreement checkbox where, if the
>
> checkbox is not checked then the form will not submit, so technically
>
> interacting with the checkbox is required to proceed.
>
>
> Would there be any interest in adding aria-required as a supported
>
> attribute of role=checkbox in this case?
>
>
> Bryan Garaventa
>
> Accessibility Fellow
>
> Level Access, Inc.
>
> Bryan.Garaventa@LevelAccess.com
>
> 415.624.2709 <(415)%20624-2709> (o)
>
> www.LevelAccess.com
>
>
>
>
>
>
>
>