Re: navigation banner in css.

> <a> is not a valid direct subordinate of <ul>, only <li> is.  As given,
> this example is non-compliant because there is only white space between
> links, so the links cannot be easily distinguished.

Let's not be quite so doctrinaire.

<http://www.w3.org/TR/WCAG10/wai-pageauth.html#tech-divide-links>

 Until user agents (including assistive technologies) render
   adjacent links distinctly, include non-link, printable characters
   (surrounded by spaces) between adjacent links.

As I explain in my book, this is an arse-backward requirement in the
first place. The sequence <a></a><a></a> is self-explanatory and
fully demarcated, as is the similar <a></a> <a></a>. So: Jaws
(inevitably-- WCAG's blindness-related requirements are all based on
working around the peccadilloes of Jaws present when WCAG was
written) couldn't differentiate adjacent links. That's not our
problem. It's valid HTML. Similarly, adding those hideous D-links
because user agents didn't support longdesc was also a bad idea
because it too was not our problem.

Now, can anyone demonstrate that current browsers and screen readers
*are still unable* to "render adjacent links distinctly"?

Note that a response of "Well, Product X version 1.0 from 1994,
which one person in a Swiss canton still uses, is unable to
differentiate links" ain't gonna cut it in 2003. We're not talking
about something genuinely novel and complex, like accessible PDF or
Flash. We're talking about interpreting the original HTML spec!

-- 

  Joe Clark  |  joeclark@joeclark.org
  Author, _Building Accessible Websites_
  <http://joeclark.org/access/> | <http://joeclark.org/book/>

Received on Friday, 24 January 2003 19:52:53 UTC