W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2005

Re: Validity as a technique

From: Maurizio Boscarol <maurizio@usabile.it>
Date: Mon, 07 Nov 2005 20:59:48 +0100
Message-ID: <436FB234.9090700@usabile.it>
To: Yvette Hoitink <y.p.hoitink@heritas.nl>
CC: w3c-wai-gl@w3.org

Yvette Hoitink wrote:

>
>>Validity isn't a tecnique: it is a document property. I think 
>>we should say that *validating a page*, in the sense of 
>>"using a code validator to determine validity error", is 
>>necessary but not sufficient (or another approriate formula) 
>>to test the success criteria that require something can be 
>>programmatically determined.
>>    
>>
>
>I slightly disagree: I think writing valid code is a technique. How to test
>for that is a different subject and not a technique as such. 
>  
>

Yes: writing valid code is a techique. But you said "validity": and 
validity is a page property! Should we say "writing valid code is a 
techique--"? I don't think it's appropriate. Let me try to explain.

If a specific attribute or an element is required as a success criteria, 
the validation process can determine if it is present in page or not. Good.
But the required attribute or element *could be present and the page be 
invalid for totally different reasons*. So we can't say that "validity 
is required to determine if a success criteria is programmatically 
determined". It isn't required! The page may be invalid and the success 
criteria still be ok, of course!

Validation checks may say whether the caption is or isn't present, but 
page may be invalid for an unclosed br or for a border attribute!

It looks to me like we're taking the whole for the parts, here. The 
singles success criteria are important, not the validity itself. Some 
validity issues aren't related to success criteria and to accessibiliy 
at all!

So we should say:
"validation (using a code validator to determine validity error) is 
required/necessary/sufficient (as you prefer) to test the success 
criteria that require something can be programmatically determined".

This for me is exact and appropriate. What do you think?

Maurizio
Received on Monday, 7 November 2005 19:49:46 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:40 GMT