Wellformedness vs. validity vs. specification (was RE: Summary of arguments...)

Paul Walsh asked:
<blockquote>
Can you please provide a real example of an assistive technology doesn't 
work as a result of invalid code, where all WAI guidelines pass? As I said, 
if an assistive technology doesn't work properly, it must fail at least one 
guideline, resolve the issue that causes that failure and you have no need 
for validity.
</blockquote>

At 10:38 5/11/2005, Roberto Scano (IWA/HWG) replied:
<blockquote>
Try for e.g. a page with follow elements unclosed:
- table
- td
- tr
- ul
- blockquote
- a
Try also a page with:
- scripting errors
- numbered "id" attribute
- non-sgml charset presented with encoding different than utf-8

And finally test all these with text/html browser and with 
application/xhtml+xml.
And please don't said that there are no accessible Browser that support 
application/xhtml+xml: we are defining new guidelines, and not wcag 1.1 for 
fix wcag 1.0 mistakes (like, for eg.colour contrast for text at level 3 or 
web app that should work without js at level 1).
</blockquote>

The first list (unclosed elements) contains only wellformedness violations 
instead of examples of invalid code. I don't believe that anyone on this 
list who does not require validity at level 1, would want to tolerate 
wellformedness violations. In the last editor's draft before the 30 June 
2005 Working Draft, there was actually a proposal about wellformedness.

The success criterion we had at level 3 was not just about validity against 
a DTD or other formal schema, but about using the technologies according to 
specification, which goes slightly further than mere validity (because a 
DTD or formal schema does not always define all correct uses of the 
document type).

Let's not discuss formal validity of code, but using technologies according 
to specification. I find the latter much more interesting and helpful.

Regards,

Christophe Strobbe


-- 
Christophe Strobbe
K.U.Leuven - Departement of Electrical Engineering - Research Group on 
Document Architectures
Kasteelpark Arenberg 10 - 3001 Leuven-Heverlee - BELGIUM
tel: +32 16 32 85 51
http://www.docarch.be/ 

Received on Sunday, 6 November 2005 18:39:49 UTC