W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2002

RE: Testing of 14.1 (WCAG 1.0)

From: Jukka Korpela <jukka.korpela@tieke.fi>
Date: Tue, 12 Mar 2002 13:58:26 +0200
Message-ID: <621574AE86FAD3118D1D0000E22138A9418557@TIEKE1>
To: w3c-wai-ig@w3.org
goliver@accease.com wrote:

> Anyone ever seriously tested for compliance of an
> entire web site to 14.1 [1]?

The requirement "Use the clearest and simplest language appropriate for a
site's content." can is fundamentally qualitative, so the question arises
how one could _test_ against it.

In fact, the Core Techniques document gives a very useful checklist for it
at
http://www.w3.org/TR/WCAG10-CORE-TECHS/#comprehension
which even refers to the Gunning Fog index. It would also be possible to
develop indexes that estimate the simplicity of word usage, i.e. whether the
words are commonly understood. But any attempts to test for compliance using
such measures would meet the question what the site's content demands. You
can hardly require that a research report on magnetohydrodynamics must have
Gunning Fog index smaller than 7, whereas for a site that tells about
municipial services such a requirement would be quite adequate - though the
limit should preferably be smaller.

Undoubtedly, if you take a Web page and show that some readability indexes
for it suggest that it's really demanding, the author will respond that the
site's content imposes certain requirements.

For different types of pages and sites, it might be possible and useful to
impose specific, objectively computable criteria for complying with 14.1,
e.g. via local regulation or recommendations. It's debatable how this should
be done, especially whether we should start with mild requirements or with
recommendations that set a challenging goal.

Unfortunately, readability measures have mostly been developed for English
only. For this and other reasons, quite a lot of "tests on humans" would be
needed: just ask different people to read the texts and try to find out how
they understood it. Preferably include people with _different_ problems as
regards to reading texts (e.g., children, old people, mentally retarded
people, immigrants, etc.). This is expensive, time-consuming, and not
objective, but it should be done for the most important pages at least. 

-- 
Jukka Korpela
TIEKE Tietoyhteiskunnan kehittämiskeskus ry
Finnish Information Society Development Centre 
Salomonkatu 17 A, 10th floor, FIN - 00100 HELSINKI, FINLAND
Phone: +358 9 4763 0397 Fax: +358 9 4763 0399 
http://www.tieke.fi  jukka.korpela@tieke.fi
Received on Tuesday, 12 March 2002 06:57:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:01 GMT