W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2014

RE: measuring contrast ratio and Windows Clear Type

From: Mitchell Evan <mtchllvn@gmail.com>
Date: Mon, 18 Aug 2014 11:20:38 -0700
Message-ID: <CAK=xW6v9eKh6exrQ4LGaXKBha2KxfEc_e94cgC6W5kQjnF_ObQ@mail.gmail.com>
To: Jan Eric Hellbusch <hellbusch@2bweb.de>
Cc: w3c-wai-ig@w3.org, Joe Chidzik <joe.chidzik@abilitynet.org.uk>
The glossary is normative, and the "Understanding" section is not. So for
conformance, we can measure font color with user agent antialiasing turned

At first this seems counterintuitive. The intent of SC 1.4.3 is to benefit
people who do not adjust their colors. Yet modern desktop and mobile OSes
make it difficult or impossible for end users to adjust this setting --
surely a greater burden on end users than color adjustment.

Nevertheless, I believe WCAG got this right. Note 2 gives us as testers a
specific, repeatable technique to evaluate text color, while end users can
get that same solid text color through a variety of user agent settings.

For example, many varieties of zoom cause "core" text colors to solidify.
And zoom is a setting that we can more reasonably expect end users to know
about, especially on mobile.

As a web tester, if you don't want to mess with the system's antialiasing
setting, you can generally use a DOM inspector to check CSS computed color.
I wouldn't recommend inferring the color from CSS rules, because color is
often inherited from a parent element. Keep in mind that even the CSS
computed color can differ from the actual color, for example if an ancestor
element has opacity less than 100%.

Mitchell R. Evan
Hi Joe and all,

maybe I should put my question a bit differently:

> I don't know if it's possible to detect (via CSS\JavaScript) if a user has
clear type
> enabled or not, but in my view, measuring the contrast via the values
provided in
> the CSS should be all that it is required to check adherence to the
> criteria.

I agree, because a web developer cannot make considerations for all screen
settings a user may have made.

In fact, the glossary in WCAG 2.0 basically says the same:

"Note 2: Because authors do not have control over user settings as to how
text is rendered (for example font smoothing or anti-aliasing), the contrast
ratio for text can be evaluated with anti-aliasing turned off."

But when reading the "understanding" documents for success criteria 1.4.3
and 1.4.6 I discovered a different approach:

"Note 1: When fonts have anti-aliasing applied to make them look smoother,
they can lose darkness or lightness. Thus, the actual contrast can be
reduced. Thicker stem widths will reduce this effect (thin fonts could have
the full stem lightened rather than just the ends). Using larger fonts and
testing for legibility in user agents with font smoothing turned on is

Is that considered a reliable test? Antialiasing will deliver different
results on different user agents.

Should antialiasing be turned on or off for measuring contrast ratios? ...
and for WCAG 2.0 conformance?

BTW: Both suggestions appear on the same page, too (link above).


Jan Eric Hellbusch
Tel.: +49 (231) 33005825 oder +49 (163) 3369925
Accessibility-Beratung: http://2bweb.de
Blog: www.chemnitzer-14.de
B├╝cher, Artikel: www.barrierefreies-webdesign.de

Received on Monday, 18 August 2014 18:21:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:52 UTC