WebAIM - Web Accessibility In Mind

E-mail List Archives

Re: Minimum contrast edge case?

for

From: John Foliot
Date: Apr 25, 2018 8:23AM

Glen wrote:

> If I go back to the W3C statement of: "The basis for determining
> conformance to WCAG 2.0 is the success criteria, not the techniques", then
> it seems like you'd be ok.

Indeed.

Techniques (whether for passing or failing) are, in W3C parlance,
non-normative (i.e. they are NOT part of the standard). Techniques are
generally common ways of meeting the requirement (or failing it), but the
Techniques provided in WCAG are incomplete and NOT the *only way* of
meeting any of WCAG's Success Criteria. Each Success Criteria is written as
to define the *functional outcome* in a way that can be consistently tested
and verified, but *HOW* you get to that point is left open to the actual
page creator, and is not specified in WCAG itself (by design).

As an example, SC 1.1.1 calls for all non-text items on the page having an
alternative textual equivalent, and NOT "all images need an alt text"
because a) the SC applies to more than just images, and b) there are many
ways of achieving this, including using @alt, @aria-label, @aria-labeledby
(etc.) Techniques are useful to illustrate "good" versus "bad", or to
examine (and thus hopefully better understand) the *intent* of the SC, but
I would caution everyone to not use Techniques as a basis for determining
whether or not a 'page' is in conformance or not.

JF

On Wed, Apr 25, 2018 at 8:23 AM, Isabel Holdsworth <
<EMAIL REMOVED> > wrote:

> Thanks Glen. Yes, that's how it works. The user talks to an admin, who
> sets up their chosen colour scheme for them. The new colour scheme is
> defined in the application's CSS, and it's used on every page. So I
> think we scrape a pass on this one, although the whole scenario makes
> me uncomfortable.
>
> On 20/04/2018, glen walker < <EMAIL REMOVED> > wrote:
> > I'm not seeing how you meet G18. G18 seems to be just explaining in
> > further detail what 1.4.3 says. It just talks about the contrast ratio
> and
> > that your page has to meet it, which is what 1.4.3 says.
> >
> > If your page does not have sufficient contrast, then it doesn't meet
> > 1.4.3. The *user* needs to have a way (tool) to change the contrast
> > themself, and the tool itself needs to be discoverable and have
> sufficient
> > contrast.
> >
> > If you remove the tool because it's distracting, and the page does not
> have
> > sufficient contrast, I don't see how that passes.
> >
> > But maybe I'm misunderstanding your scenario. If the page does *not
> *have
> > sufficient contrast and there is *not *a tool to change the contrast, but
> > the user talks to an admin who can apply a different color scheme so now
> > the page *does *have sufficient contrast, then you *might* be ok. The
> > difference seems to be whether the user can change the contrast or if a
> 3rd
> > party changes the contrast for them.
> >
> > If I go back to the W3C statement of: "The basis for determining
> > conformance to WCAG 2.0 is the success criteria, not the techniques",
> then
> > it seems like you'd be ok. The page does have sufficient contrast, after
> > the admin changed the color scheme. The SC doesn't specify who should
> set
> > the color scheme. Just make sure the new color scheme is applied to
> every
> > page.
> >
> > Glen
> >
> >
> >
> >
> > On Fri, Apr 20, 2018 at 6:03 AM, Isabel Holdsworth <
> > <EMAIL REMOVED> > wrote:
> >
> >> Thank you Glen for such a comprehensive message. You've given me food
> >> for thought.
> >>
> >> So we meet G174 by providing the colour preferences tool for most
> >> scenarios.
> >>
> >> But one of our clients has asked us to remove the tool from their
> >> instance of our application because they think it will cause
> >> unnecessary distraction. So in that scenario there's no tool.
> >>
> >> However, it still meets G18 in that if a user specifies that they need
> >> a conformant colour scheme, the client can set the application up to
> >> display in that scheme.
> >>
> >> So I don't think we can mark this as a 1.4.3 fail unfortunately.
> >>
> >> On 19/04/2018, glen walker < <EMAIL REMOVED> > wrote:
> >> > G18 is a "sufficient technique", and while the definition of a
> >> "sufficient
> >> > technique" says "If you use the sufficient techniques for a given
> >> criterion
> >> > correctly and it is accessibility-supported for your users, you can be
> >> > confident that you met the success criterion
> >> > <https://www.w3.org/WAI/GL/2016/WD-UNDERSTANDING-WCAG20-
> >> 20160105/understanding-techniques.html#ut-understanding-techniques-
> >> sufficient-head>",
> >> > it also says "The basis for determining conformance to WCAG 2.0 is the
> >> > success criteria, not the techniques
> >> > <https://www.w3.org/WAI/WCAG20/quickref/20160105/#about-techs>".
> >> >
> >> > In your particular case, I'm not sure G18 applies to your situation.
> It
> >> > sounds more like G174: Providing a control with a sufficient contrast
> >> ratio
> >> > that allows users to switch to a presentation that uses sufficient
> >> contrast
> >> > <http://www.w3.org/WAI/GL/2016/WD-WCAG20-TECHS-20160105/G174>;.
> >> >
> >> > If you allow the color scheme to change *and* the tool that lets you
> >> > make
> >> > that change is accessible (including meeting the color contrast
> ratio),
> >> > then you're ok. If the tool itself is hidden, then the tool would
> seem
> >> to
> >> > fail WCAG, which means your page does *not* meet 1.4.3. But we'd need
> >> more
> >> > information on what you mean by "hidden". Is it visually hidden such
> as
> >> > display:none? Is it hidden because it's covered up by another object?
> >> Is
> >> > it hidden because it's in an expand/collapse section?
> >> >
> >> > Separate from the tool being hidden, you mention the color preferences
> >> are
> >> > predefined. That in itself is more of a usability issue and not an
> >> > accessibility issue. If the predetermined colors meet 1.4.3, then
> >> > you're
> >> > ok. If you don't like the colors, that's a design decision.
> >> >
> >> > Glen
> >> >
> >> > On Thu, Apr 19, 2018 at 3:03 AM, Isabel Holdsworth <
> >> > <EMAIL REMOVED> > wrote:
> >> >
> >> >> Apologies but here's another "does this fail WCAG2" question? I
> really
> >> >> appreciate you guys' time and wisdom in answering these, and
> hopefully
> >> >> it'll clarify things for others as well.
> >> >>
> >> >> I'm working on a web application that fails 1.4.3 (minimum contrast).
> >> >> To get around this, they provide a tool that allows users to choose
> >> >> from a whole host of colour schemes, some of which are conformant.
> >> >>
> >> >> But in one particular scenario, the tool is hidden and each user's
> >> >> colour preference is predefined and set at supervisor level and can't
> >> >> be changed by the user.
> >> >>
> >> >> It's frustrating that they've taken the ability to choose a colour
> >> >> scheme away from the user, but I guess their alternative still passes
> >> >> under sufficient technique G18 (
> >> >> https://www.w3.org/TR/WCAG20-TECHS/G18.html ).
> >> >>
> >> >> I think I've just answered my own question, but I'd really appreciate
> >> >> your thoughts.
> >> >> > >> >> > >> >> > >> >> > >> >>
> >> > > >> > > >> > > >> > > >> >
> >> > >> > >> > >> > >>
> > > > > > > > > >
> > > > >



--
John Foliot
Principal Accessibility Strategist
Deque Systems Inc.
<EMAIL REMOVED>

Advancing the mission of digital accessibility and inclusion