WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: WCAG 2.1 - 1.3.5 - How to capture a violation?

for

From: Jonathan Avila
Date: Jul 25, 2018 7:29AM

Jared, in my message I did not say that the SC required autocomplete -- but that autocomplete was the only technique that I was aware of that was supported to communicate the purpose. Other techniques such as microdata could be used if accessibility supported. The accessible name is generally not seen to fall into that category because of localization and other issues make prevent it from clearly aligning without ambiguity. You'll notice that the list of purposes from Section 7 of WCAG 2.1 comes from HTML's autocomplete list. So while autocomplete is not required by the SC I am not aware of another method to use that will work given the limited browser support for microdata formats. ARIA in the future could provide this information but I am not aware today that it contains what is needed.

Jonathan

-----Original Message-----
From: WebAIM-Forum < <EMAIL REMOVED> > On Behalf Of Jared Smith
Sent: Wednesday, July 25, 2018 9:16 AM
To: WebAIM Discussion List < <EMAIL REMOVED> >
Subject: Re: [WebAIM] WCAG 2.1 - 1.3.5 - How to capture a violation?

Jonathan -

The normative success criterion text does not require the autocomplete attribute. It simply requires that the purpose of the field be programmatically determined if it collects user information. While there is a sufficient technique (which is non-normative) that allows you to meet this SC by using the autocomplete attribute, this is not required - and therefore it would not be a failure if autocomplete is not present in these cases, so long as the purpose is programmatically identified with the field.

Of course 1.3.1 requires that the field information be programmatically determined anyway, so 1.3.5 really just adds a SUGGESTION that you use autocomplete, which does improve the user experience for those using form fillers.

> The accessible name cannot be used to communicate the purpose

Why couldn't it? I see no suggestion from the normative text that the accessible name could not be used. In fact, if you didn't use the accessible name, then you'd likely fail 2.5.3.

Jared