WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Placeholder

for

From: Isabel Holdsworth
Date: Mar 12, 2019 5:13AM

I use a combination of DOM and CSS checking and JAWS OCR, which is
better than nothing but alas not foolproof.

On 12/03/2019, Patrick H. Lauke < <EMAIL REMOVED> > wrote:
> On 12/03/2019 09:01, Isabel Holdsworth wrote:
>> From a blind tester's perspective (hope I've got this right Jim):
>>
>> Say a form element has a label and a placeholder, and the placeholder
>> is used to describe the expected data format. If the screenreader
>> speaks the placeholder even after it's been superseded by entered
>> data, then it's more challenging for a blind tester to know that the
>> screenreader is speaking invisible text and that there's a 3.3.2 fail.
>
> 3.3.2 is tricky in any case, as it does not necessitate that whatever
> text acts as a label/instruction is actually programmatically associated
> (so it could just be text that happens to be "near" an input - for
> whatever definition of "near" - but has no proper <label for="...">
> connection or anything, but under 3.3.2 that's a pass; or the form
> control could have explicit aria-label="..." but no actual visible text,
> which would fail 3.3.2 even though it passes 4.1.2 and would be tricky
> to spot unless a tester using just the screenreader both explores the
> overall text of the page - and makes sure they don't land in visually
> hidden text - and then compares notes with what they hear when focusing
> a form field)...
>
> But yes, this wouldn't help in this already very complex to test (mostly
> visual) SC.
>
> P
> --
> Patrick H. Lauke
>
> www.splintered.co.uk | https://github.com/patrickhlauke
> http://flickr.com/photos/redux/ | http://redux.deviantart.com
> twitter: @patrick_h_lauke | skype: patrick_h_lauke
> > > > >