WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: html 5 required

for

From: Steve Faulkner
Date: Nov 6, 2017 5:47PM

>
> Now we find in testing, using NVDA 2017.3 with Firefox, NVDA announces
> empty fields as "invalid entry" the first time the user focuses on them,
> even before the error checking that takes place when a user hits the Submit
> button.


This is not due to any screen reader, it is due to the required attribute
being exposed as required and invalid in browser accessibility APIs. The
reasoning behind this is that until a required control has received some
user input, it does not meet its validation constraints and is therefore in
an invalid state.

discussion of this is in a related Firefox bug
https://bugzilla.mozilla.org/show_bug.cgi?idf6544



--

Regards

SteveF
Current Standards Work @W3C
<http://www.paciellogroup.com/blog/2015/03/current-standards-work-at-w3c/>;

On 6 November 2017 at 23:47, Beranek, Nicholas via WebAIM-Forum <
<EMAIL REMOVED> > wrote:

> Hi Diane, I recommend keeping the required attribute. For compatibility, I
> would add the value "required", as well. You're encountering expected
> behavior with NVDA and Firefox because, by definition, since a value is
> required, lack of a value, or null value, would be invalid. I am always
> hesitant to include a hack to appease a browser/AT, and the use of the
> required attribute is a sufficient technique.
>
> Nick Beranek
> Capital One
>
> On Nov 6, 2017, at 5:17 PM, Tomlins Diane <Diane.Tomlins@HCAhealthcare.
> com<mailto: <EMAIL REMOVED> >> wrote:
>
>
> This just came up for us, and we just switched 70+ code files to use the
> html 5 "required' attribute. Now we find in testing, using NVDA 2017.3 with
> Firefox, NVDA announces empty fields as "invalid entry" the first time the
> user focuses on them, even before the error checking that takes place when
> a user hits the Submit button. So I would think this would make things
> confusing for AT users.
>
> We tried adding aria-required="true" but that didn't help since we still
> have the "required" attribute in there too.
>
> If I understand the replies to this thread, the "required' attribute is
> still a patchwork of support between browsers and AT??
>
> Our developers won't be thrilled to back-pedal to take out the attribute
> they just put in :-/
>
> Any possible workaround besides going back?
>
> Thanks!
>
> Diane R Tomlins
> HCA IT&S | Digital Media
> Accessibility SME
>
>
>