W3C home > Mailing lists > Public > public-comments-wcag20@w3.org > July 2005

On Validity. was : Re: Call for Review: Working Draft of Web Content Accessibility Guidelines 2.0

From: Nir Dagan <nir@nirdagan.com>
Date: Sun, 24 Jul 2005 20:18:51 +0200
Message-ID: <42E3DB8B.6000605@nirdagan.com>
To: public-comments-wcag20@w3.org

Hello all,

In my view validity should be a level 1 success criterion.

1. The rendering of invalid documents is unpredictable. In particular 
this means that there is no way to guarantee that an invalid document 
will be accessible using future user agents.

2. Validity is a corner stone of quality assurance. If one wants to 
assure the quality of invalid documents, one has to define what 
validation errors one allows. Once these allowable errors are defined, 
one actually defined a modified criterion of validity that may be 
incorporated to the formal grammar.

Moreover, it seems that there is a confusion between two different 
1. "Create documents that validate to published formal grammars".
2. "Create documents that validate to formal grammars published by W3C 
(or ISO or some other organization that we like its recommendations)".

I propose that the first requirement will be a level 1 success 
criterion. Note that the example quoted below may satisfy the first 
validity requirement:

"Example 2: invalid content that is accessible
An HTML-based application with JavaScript widgets that can be negotiated 
with the keyboard arrow keys like a client application. It uses tabindex 
on <div> and <span> tags to set focus programmatically to those 
elements. In HTML 4.01 the tabindex attribute is not valid on <div> and 
--- From "Validity and Accessibility" 

The code above may be valid if one creates a custom DTD that modifies 
HTML 4.01 to include the missing attributes.
If in addition the custom DTD is available at some publicly accessible 
URI by a SYSTEM identifier in the !DOCTYPE declaration, the requirement 
is satisfied.

Custom DTDs are useful not only for adding new accessibility features 
not yet standardized, but also to restrict W3C recommendations with 
accordance to need of a particular project. For example if one uses only 
one <h1> per page, as a QA measure, one may validate its documents with 
respect to custom DTD that makes this requirement, and still publish its 
web pages using HTML4 doctype if they validate with respect to HTML4 as 


Nir Dagan
Dept of Economics and Business, Academic College of Judea and Samaria, 
Ariel, Israel  and
Academic Priority Ltd.

No virus found in this outgoing message.
Checked by AVG Anti-Virus.
Version: 7.0.338 / Virus Database: 267.9.4/57 - Release Date: 22/07/05
Received on Monday, 25 July 2005 01:21:39 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:11:05 UTC