Re: Breaking Dependencies - @summary (FW: Call for Review: German WCAG 2.0 Candidate Authorized Translation)

On Aug 4, 2009, at 20:16, John Foliot wrote:

> L. David Baron wrote:
>> I'd like to further understand the definition of contradict here:
>>
>>  (1) Does HTML5 contradict WCAG if it removes a feature whose use
>>  is recommended or required by WCAG?  (I'm pretty sure the answer
>>  to this one is yes.)
>
> Yes, and this is the partial case with @summary today.
>
> Ian's Draft Spec says: "Authors should not specify the summary  
> attribute
> on table elements."
> [
> http://dev.w3.org/html5/spec/Overview.html#warnings-for-obsolete-but-confo
> rming-features ]
>
> WAI's WCAG 2 says: "Use (Using) the summary attribute of the table  
> element
> to give an overview of data tables"
> [ http://www.w3.org/TR/2008/NOTE-WCAG20-TECHS-20081211/H73 ]

What you are quoting is not WCAG 2--the Recommendation. What you are  
quoting is an accompanying Working Group Note. WCAG 2 is at http://www.w3.org/TR/WCAG20/

WCAG 2--the document that has been elevated to Recommendation by the  
Director per Process--does not even contain the word "summary". In  
fact, it doesn't even the word "table" outside notes, examples and the  
title "Table of Contents".

WCAG 2 has been deliberately written to be format-agnostic. As such,  
the idea that the presence or absence of any particular syntax in a  
language could contradict WCAG 2 is utterly ludicrous.

The document you quote (an informative document published as a Working  
Group Note without the accountability of the Recommendation Process)  
says at the top of the page you quote: "Applicability HTML 4.01, XHTML  
1.x". The quoted technique is informative about implementing the  
language-agnostic normative success criterion 1.3.1 in the context of  
HTML 4.01 and XHTML 1.x. Other languages, such as HTML5, PDF or SWF  
may provide different mechanisms for implementing success criterion  
1.3.1.

Success criterion 1.3.1 reads:
| 1.3.1 Info and Relationships: Information, structure, and  
relationships
| conveyed through presentation can be programmatically determined or  
are
| available in text. (Level A)

First, it's questionable to consider a data table to constitute  
"Information, structure, and relationships conveyed through  
presentation", because HTML 5 includes an algorithm for  
programmatically determining the relationships of cells from the table  
model. Second, even if one, for the sake of argument, accepts that a  
data table constitutes "Information, structure, and relationships  
conveyed through presentation", merely adding a paragraph of  
explanation before the table would satisfy "or are available in text".  
Note that it doesn't say that "available in text" needs to be  
"programmatically determined", but to improve upon mere availability,  
the text could be "programmatically determined" by pointing to it  
using aria-describedby or by placing it in <caption>.

I think it would be productive to avoid the line of argument that a  
new Recommendation-track document on HTML5 couldn't "contradict" an  
older informative Working Group Note that itself limits its  
applicability to HTML 4.01 and XHTML 1.x.

-- 
Henri Sivonen
hsivonen@iki.fi
http://hsivonen.iki.fi/

Received on Wednesday, 5 August 2009 06:36:39 UTC