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

RE: Re[2]: Compliance and html validation - how to interpret?

From: Denise Wood <Denise.Wood@unisa.edu.au>
Date: Mon, 29 Apr 2002 00:04:42 +0930
Message-ID: <E1962E8F1DF0D411878300A0C9ACB0F903590BE6@exstaff4.magill.unisa.edu.au>
To: "'w3c-wai-ig@w3.org'" <w3c-wai-ig@w3.org>
Cc: "'Charles McCathieNevile '" <charles@w3.org>
Thanks Charles

I have looked at the site and yes it does work in lower (broken browsers) but
again (as you have noted) there is still a problem relating to the aesthetics.

I take your point about the importance of upgrading browsers within the
University environment and we most certainly do ensure all staff and student
pools computers are regularly upgraded. However we cannot control what happens
outside the university environment and many staff and students do not use
latest browsers from home (even though we do provide a free 'get connected 'CD
containing latest installs amongst other software to all Uni students).

More importantly I want this training site to demonstrate browser independence
and triple-a compliance and am determined to achieve that though it is more
difficult catering for older and broken browsers. The site is not complex so it
should be possible with some effort. Further, it will serve as the basis for
development of University style sheets and templates so it really is worth
persevering I think.

Re the compliance level if tables are used for layout - yes that is also my
interpretation of the guidelines. What does confuse me though is that the W3C
site which, as you have acknowledged, does use layout tables yet displays the
double-a compliance logo. 

Any way thanks for the lead Charles.

Denise

-----Original Message-----
From: Charles McCathieNevile [mailto:charles@w3.org]
Sent: Sunday, April 28, 2002 6:20 AM
To: Denise Wood
Cc: 'w3c-wai-ig@w3.org'
Subject: Re: Re[2]: Compliance and html validation - how to interpret?


Hi Denise,

http://www.alistapart.com/index.html is not triple-A compliant (not even
double-A compliant) but it does use stylesheets instead of tables for layout
and still works on most browsers. (It ain't the prettiest thing in older
browsers, but it works. their article http://www.alistapart.com/switcher.html
describes and links to descriptions of what they did for their site)

It is also valuable to explain to a University that there are good reasons
for supporting upgrades in default systems provided to students, especially
when they are available for free or very low cost, as with browsers. There
are now a wide range of browsers available for all kinds of platforms -
investigating the linux accessibility work might save a fair bit of money on
assistive technology, or provide some valuable research work for students,
for example.

The question of whether something worksin a particular browser could be
approached in the way that libraries do. In most universities I have been to
there are some books which have multiple copies. Different editions,
different binding, etc for the same content. I have never met a library that
is prepared to categorise the same book differently just because one copy is
red and one copy is blue. (I only ever knew one person who classified his
books by colour, and he had an exception case for multiple copies).

For what it is worth my personal opinion is that using tables for basic
layout violates checkpoint 3.1 and thus means the site cannot be double-A
compliant. Even if you decide not to use style sheets instead of tables for
layout (and the place I work for decided that in some instances) it is still
worth trying to apply all the rest of the checkpoints - each of them is
useful on its own.

cheers

Chaals

On Sun, 28 Apr 2002, Denise Wood wrote:

  Just to clarify things a little, the training site is only meant to serve as
a
  basic introduction to Web accessibility - certainly there is no assumption
that
  authors will necessarily be able to create web sites that meet all the
  requirements for triple-a compliance and html 4 validation using stringent
  validation tools such as the W3C html validator. This site in an interim
  measure while we begin the task of setting University Wide standards on Web
  accessibility (which must still be endorsed by the University's Academic
  Board). I am trying to raise the consciousness of staff of the major concerns
  at this stage.

  While we prepare the standards we are also intending to develop corporate
style
  sheets and templates that University staff will be required to use. The aim
is
  to ensure at least a degree of compliance (we can't control everything
authors
  do once they start using the templates) and also trying to establish an
overall
  corporate look for all Uni Web sites. So the intention is to minimize some of
  the work for Uni Staff in the longer term.

  My dilemma is how to ensure that the site that I publish for this
introduction
  to Web Accessibility is a model of good practice - even though it is likely
  staff will not be able to create a similar site without our yet to be
developed
  style sheets and templates. Your advice about a page explaining how we have
  arrived at our claim for either double a or triple a compliance makes good
  sense.

  I'd still like some advice though about what actually constitutes the
  difference in compliance levels (ie does the use of a table for layout even
if
  styles are used for all other markup make the site automatically only level 2
  or 3 compliant, does the use of any deprecated language to accommodate lower
  level browsers make it a level 2 or 3 site even if all other accessibility
  checkpoints are met?). Can anyone point me to a site that is triple-a
  compliant, validates as html 4 transitional AND looks fine in broken browsers
  such as NS 4? If there is a way of achieving triple-a compliance and
validation
  for the site while still retaining an aesthetic look in NS 4 that is the
ideal.
  This would also help me in the development of the style sheets and templates
  that will eventually be prescribed for all Uni staff.
Received on Sunday, 28 April 2002 10:34:49 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:04 GMT