WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Does WCAG require that when you click a label it checks the checkbox or radio button?

for

From: Alastair Campbell
Date: Nov 20, 2015 5:55AM

Paul Adam wrote:

> "So WCAG is fine if let's say I make all my buttons operable with the
> Escape key only. No enter or space bar must press escape to activate the
> buttons. That would not fail keyboard operable? If not, WCAG indeed has
> many loopholes."
>

And Bevi wrote:
"As it is now, WCAG is the antithesis of plain language."

I think this thread has highlighted some contradictory requirements for
WCAG guidelines:
- Closing loopholes in the detailed implementation, presumably by being
more prescriptive about solutions. (Would you agree with that Paul?)
- Easy to understand.
- Technology agnostic (a goal of WCAG version 2).

I'm sure there are improvements, but one of three requirements cannot be
met in the normative text. If it is more prescriptive, it will be less
tech-agnostic and probably harder to understand. If it is easier to
understand, there will be more loopholes (I think).

I think the balance comes from what helps in practice. Patrick emphasises
"we however strongly advise that...", and we do something similar where we
present the best practice as the default way something should be done.
Usually we find more basic errors and present the best practice, so the
'technical pass' never comes up. Usually if someone knows enough to add a
label and title they'd know enough to use for/id, the technical-pass
scenario is quite rare.

Perhaps you come across more scenarios where developers are using automated
testing or have a very technical view of accessibility and then implement
to 'conformance' instead of 'accessible'?

If so, I would suggest using persuasion tactics that are less about the
guidelines, and more about how people use things.

From Matt May's framework: "This is the problem and this is how it affects
people and you can solve it and if you don't, these are the consequences."
[1]

The consequences might be non-conformance, but in practice I find the human
scenario more persuasive, e.g. Someone using a slipper mouse (thanks Julie)
will struggle to use this, and all you have to do is add a couple of
attributes.

I'd also like to +1 Mike Moore's suggestion to move it upstream, and get
the best practices into their specs/pattern libraries, so the argument
doesn't come up again.

I guess this doesn't answer the question as such, but my response comes
from an appreciation that closing all the possible loopholes in WCAG,
making it easier to understand, and make it work across technologies could
not happen unless all technological advancement is paused for a few years
whilst we wrangle the guidelines!

-Alastair

1]
https://modelviewculture.com/pieces/articulating-and-advocating-for-accessibility