W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2007

Re: Accessibility tests of Australian University homepages

From: David Woolley <forums@david-woolley.me.uk>
Date: Wed, 09 May 2007 11:53:42 +0100
Message-ID: <4641A836.1000402@david-woolley.me.uk>
To: w3c-wai-ig@w3.org

Tim wrote:
> 
> Did you pick a colour blindnes stylesheet or check the accessibility 
> statement?

The standard style sheet results in something very difficult to read for 
someone with normal colour vision!

There is gratuitous use of strong.

There is distracting animation - this surely has to disqualify it from 
AAA status.

class is abused as a style sheet macro, rather than for semantic 
sub-classing, although the result isn't the actual green implied by the 
class name.

No style is better, but there is a weird line (which don't appear in the 
default styling, or at least I can't find them) which is completely 
cryptic. (Using the Publishing index page, from now on.)  This is what 
it looks like in Lynx:

   :   :  > :   :   :   :   :

Lynx also shows the first page almost completely blank except for:

    #Second  level  index  on  Hereticpress.com

(which is the result of link elements) on the top line, and

    Skip Nav

on the bottom line.

The first half on the second page on Lynx is a "please upgrade" paragraph:

    Some  cascading  stylesheet  layout  features  on this page requires
    a browser  that  supports  JavaScript(TM).  Your browser either does

The centre justification results in some really big gaps in the Lynx
rendering of:

    * University    600    Kb : Web    survey    Australian   university
      sites : Updated 7th May 2007

There is an empty division, which looks like it is really an <hr>:

   <div class="HR"></div>

There is no H1 and H2 and H3 are used for what seems to be the same 
actual heading level.

It was written by someone who doesn't understand the HTML script interface:

    onclick="JavaScript:window.location='.....

(Javascript: schemes are proprietary bad practice, but this is actually 
an unreferenced label in an anonymous function, not a scheme at all!)

XHTML is used, but it is served as text/html.  No foreign namespaces are 
used.  On the home page, table is used without tbody; this is an 
appendix C violation, because it results in a text/html parser getting a 
different DOM from a true XHTML parser.

The XHTML does not use UTF (overridden by the HTTP Content-Type, but the 
  true character set is not specified in the XML directive - there isn't 
one.  I might be wrong about that being mandatory, but it is certainly 
good practice.

Even with the print stylesheet, the navigation bar is printed.

Printing (print preview) from the black and gold sheet results in some 
very poor colour contrasts, e.g. bright green on white.

Incidentally, the best reference I've seen on colour blindness (although 
I think the case here was cognitive disabilities, for which gratuitous 
styling and animation is an issue, particularly the use of non-standard 
link colours, although, at least, you retained the underline) is the BT 
Research one.  Although I can no longer find the original paper, this 
web site seems to have the same content:

<http://www.btplc.com/age_disability/technology/RandD/colours/index.htm>

(Note.  People with cognitive disabilities are even less likely than 
most to find the alternative styles menu.)
Received on Wednesday, 9 May 2007 10:53:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:26 GMT