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

RE: Validity

From: Andrew Kirkpatrick <akirkpatrick@macromedia.com>
Date: Fri, 4 Nov 2005 06:26:56 -0800
Message-ID: <DC9D05204B1E16419D62C12561C93221063B689E@p01exm01.macromedia.com>
To: "Gez Lemon" <gez.lemon@gmail.com>, "WCAG WG mailing list" <w3c-wai-gl@w3.org>

> arguments put forwards against validity (from the 
> face-to-face meeting in Seattle) can be summarised as:
> 
> 1: Validity isn't essential for accessibility

> Point 1 is concerning, particularly considering that level 1 
> issues tend to be about ensuring that content is 
> understandable by assistive technology. If content is 
> invalid, I don't understand how it could be testable that 
> assistive technology is able to make sense of the content. If 
> something is intended to be machine readable then validity is 
> obviously important, as it needs to be read by software 
> unambiguously. In simple terms, validity is important for 
> assistive technology.

The assumption here is the the assistive technologies actually use the
code.  The DOM and MSAA are more often the source of information that is
actually used.  Does it matter to the browser if the HTML/CSS/js code
rendered into the DOM or MSAA is valid?  Of course the browser can't put
an image equivalent into the DOM if there is no alt attribute, but what
if a web page claiming XHTML 1.0 Transitional uses marginleft="1em" in
the body element - does this make the page less accessible? Less
testable for accessibility?

Not all validation errors are of the same level of importance to
accessibility.  We have specific guidelines and techniques for important
errors, but most validation errors are insignificant.  I've seen pages
with hundreds of accessibility issues that actual users find quite
accessible and useable. 

I agree with the general desire to have valid code in web pages.  Having
simple, clean, semantic markup makes my life as a developer easier, and
I'm sold on valid code on that basis.  Having a requirement for valid
code in WCAG 2.0, when valid code doesn't always benefit users,
undermines the authority of the document.  Web developers will think
that this is an attempt to force a particular view of How The Web Must
Be rather than to ensure that disabled users are served.

AWK
Received on Friday, 4 November 2005 14:27:27 GMT

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