Re: Let's add an approved date field to Failures and Techniques


exactly correct

ALSO -  (to make it harder)   -  the techniques we document are ALSO  NOT THE ONLY ways to pass

So failure is hard to write not just because of the working group created multiple techniques. It’s because there are always other ways of doing things even beyond the techniques that we document.

So, as you point out, failure is not ever because a person didn’t use one technique or another. A failure is because they did something which absolutely and always will cause them to fail to be able to meet the success criterion.

As for ARIA,  since that only works with HTML, there can never be a failure to ever use any aspect of ARIA.    
 There MIGHT be a failure for using some part of ARIA if you are using HTML, but only if there is absolutely no other way of meeting the success criterion except using ARIA,  which is extremely unlikely since ARIA did not exist when WCAG 2.0  was written and that would mean that the working group created a success criterion that could not be met using HTML.

Small note: we never create failures. We only document common failures that we discover. But whether we document them or not they are failures. Documenting them doesn’t make them failures. And not documenting them does not make them not failures.


> On May 3, 2016, at 3:27 AM, Alastair Campbell <> wrote:
> Joshue O Connor wrote:
>> Yes.  As failures are hard to mint, and David is calling out a need for more,  my 'warning' suggestion is maybe a way of meeting the need without doing normative or quasi normative work.
> Surely the reason that failures are so hard to mint is the “multiple ways to pass” approach that WCAG took? 
> (And I’m obviously saying this with plenty of hindsight! I didn’t think of this at the time.)
> If there are 3 techniques to pass an SC, the absence of one of those techniques cannot be a failure.
> If a failure must always be a failure, there cannot be another way to pass. 
> The more technologies (e.g. ARIA) there are available, the more ways there are to pass, the harder it is to create new failures.
> I like the idea of warnings, or at least some way to say ‘this is a common way to fail’ without it being absolutist. 
> It could also provide more context about the technology, e.g. ‘if ARIA is part of your Accessibility Supported list, then if is a failure not to use landmarks for 1.3.1’.
> Cheers,
> -Alastair

Received on Tuesday, 3 May 2016 16:42:42 UTC