W3C home > Mailing lists > Public > www-validator@w3.org > April 2004

Re: Web Content Accessibility Guidelines

From: Jukka K. Korpela <jkorpela@cs.tut.fi>
Date: Tue, 27 Apr 2004 00:06:44 +0300 (EEST)
To: www-validator@w3.org
Message-ID: <Pine.GSO.4.58.0404262332540.496@korppi.cs.tut.fi>

On Mon, 26 Apr 2004, Andrew Robinson wrote:

> While we are discussing the design/layout of new validator beta output,
> how about aiming for one of the  W3C Web Content Accessibility
> Guidelines 1.0 Conformance Logos for the validator page and it's output?

Wouldn't it be better to aim at accessibility? No, it's not the same thing
at all. And I don't mean the distinction between actually passing WCAG 1.0
requirements and presenting a logo that claims (almost always incorrectly)
that a page complies with the requirements. That distinction is important
enough, but not the most important thing here. The first thing about
accessibility is that it should never be identified with any specific set
of rules, especially not with a fairly limited set of rules.

> At a first glance (and a quick bobby scan:
> http://bobby.watchfire.com/bobby/html/en/index.jsp), it probably passes

Irrelevant. Bobby is mostly just misleading.

Accessibility conformance logos are worse than useless, even more so than
validity logos, see

Should someone want to improve the real accessibility of validator
reports, the format and order of the reports themselves should be the
first concern. The report should say in the very first statements whether
the document passed validation and if not, how many errors were detected,
followed by simple presentation of the errors. There would be no place for
any advertisements or tips of the day, except perhaps _after_ dealing with
the actual business that the user asked for.

What matters in accessibility (and usability) in validation reports is
basically the same as in query results: selection of the material to be
included, ordering and structuring it properly so that first things come
first, and distinguishing summarizing information from the bulk of data.
Cf. to http://www.cs.tut.fi/~jkorpela/forms/qdfa.html
This is heavily related to the internals of the validator, and far more
important than the general issues discussed in WCAG 1.0. For example,
starting the result page with a new form is misleading. Although the
report tells the number of errors fairly early, that information should
appear _first_ and be accompanized with a summary that at least classifies
the errors by type (e.g., 7 required attributes missing, 5 undefined
attributes, etc.). And showing the source listing by default isn't
helpful. (Have you tried to listen to a source listing?)
Moreover, using the same link text "(explain...)" for quite
different links does not comply with WCAG 1.0 rules.

It is difficult to say how important it is to make the validator's
interface and reports accessible. Surely the most serious barrier to most
people who use them is that they don't understand what validation is about
and don't understand what the messages mean. Next comes the structural
issue I mentioned. Addressing such problems would mean quite some work,
and I'm afraid it would require some essential rewrite of the program
code, so I don't think it's realistic.

But putting some icons there claiming WAI conformance would be worse than

Jukka "Yucca" Korpela, http://www.cs.tut.fi/~jkorpela/
Received on Monday, 26 April 2004 17:07:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:13 GMT