W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2005

Re: remove validity completely?

From: Ineke van der Maat <inekemaa@xs4all.nl>
Date: Thu, 23 Jun 2005 00:11:52 +0200
Message-ID: <031c01c57777$65236840$0201a8c0@inekem>
To: "David MacDonald" <befree@magma.ca>, <w3c-wai-ig@w3.org>

Hoi David,

>I wanted valid
> code, because that is what the spec calls for. But it didn't make my site
> anymore accessible. It was accessible before I validated

look http://www.creatiefwebmaatwerk.nl/aatest/ and see how Opera and Firefox 
place an invalid < (between </tr> and </tbody> differently in the table text 
John sent yesterday. (code is served as text/html) This might be a serious 
accessibilityproblem for visitors with learning disabilities and still more 
when it not a >, but a text or some letters. Perhaps so the meaning of an 
element can change completely. There is not any control where UAs place this 
invalid piece of content.  How can visitors then get control how they want 
to see the page??

The error existed because later I cut the </table>tag and pasted it 
somewhere else, but < did not want to participate in this process and stayed 
where it already were. I corrected the tag but forgot to remove the lonely 
<.  I noticed it when I tested the page and saw the remarkable difference in 
the 2 browsers.

Valid code sometimes may be a requirement for other level 1 issues like 
guideline 3.2:
Any extreme change of context is implemented in a manner that can be 
programmatically identified.

Invalid code may cause extreme changes of context in some UAs and not in 
others when invalid code is handled differently. Probably invalid code in 
deep nested divs make the problem more worse.

Also the forgotten </a> Tina referred to yesterday, is a serious problem 
because links in the text can not be found anymore. I met  this kind of 
terrible error some times and even navigation was not usable, being part of 
that enormous long link through nearly a whole page of  ca. 50 lines.

Ineke van der Maat
Received on Wednesday, 22 June 2005 22:11:49 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:36:25 UTC