W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > July to September 2012

Re: SC. 1.4.4 Resize Text is a really good tool to detect bad design.

From: Wayne Dick <wed@csulb.edu>
Date: Sun, 29 Jul 2012 14:49:10 -0700
Message-ID: <CAJeQ8SDmxdjVubN9hLthLmT9MTYQu0HkpJfZtNZ_xt3V69Uxgg@mail.gmail.com>
To: Ramón Corominas <listas@ramoncorominas.com>
Cc: "w3c-wai-ig@w3.org" <w3c-wai-ig@w3.org>
Hi Ramon,

Yes CSS is not a perfect test.  But when it works you can get a nice
uniform view.  Sometimes simple text-enlargment causes such distortion
with large headings it get hard to assess what is going on.

Mostly these comments apply mostly to consultants and developers who
have control over the sight.  It is for pre-release testing or
retrofit testing.

My own example of use is not unusual.  I have a simple bludgeon style
sheet for quick reads

		Dark on Tan for Reading

/* Default preferences */
html   {background-color: #AB9B79 !important;
    color:  #000000 !important;
    font-size: 28px !important;
		line-height: 1.3!important;
		font-family: tahoma !important;
html  * {background-color: inherit !important;
        color: inherit !important;
        font-size: 1em !important;
		line-height: inherit !important;
		font-family: inherit !important;

In firefox is use 28px in general, 24px min in my options.
When these fail, I shut off options and do text-only enlarge.

This is not a bad use pattern.  So, it is a pretty good quick-test
pattern.  Only the t,ext-only zoom is required.  Note that my
line-height forces some line-squeezing away, but you can cut out the
line height lines to see some lines squeeze together in bad code.

I do not suggest this as an only test, but it does reveal some
unacceptable rigidity.  You probably do not need the color change, I


On 7/29/12, Ramón Corominas <listas@ramoncorominas.com> wrote:
> Hi, Wayne and all,
> Please note that forcing the text-size in CSS is not a valid test for SC
> 1.4.4. Many pages may be using relative units for containers that will
> not be affected by this settings, so they will "fail" according to this
> test, although they are correct.
> In addition, horizontal scroll does not mean a failure in this success
> criterion, provided that the text is readable when scrolling and no
> overlaps exist.
> Regards,
> Ramón.
> Wayne wrote:
>> I recommend it as a one of those quick tests you might do.  It usually
>> requires about 5 ctl-+ actions to get 200%, or I just hit it with the
>> style * {font-size: 24pt !important}. That's point not pixel (that's
>> very roughly 32px).
>  >
>  > If columns or rows bleed, positioned text cover
>> elements below them or horizontal lists or menus fly off the right.
>> The page is too rigid for 1.4.4 and Level AA.
Received on Sunday, 29 July 2012 21:49:37 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 13 October 2015 16:21:45 UTC