WCAG 2.0: Request for Review and Update on Comments

Dear Frank Reins, Gottfried Zimmermann, and Christian Bühler,

Thank you for commenting on the March 2004 Working Draft of WCAG 
2.0[1].  The WCAG WG published updated working drafts of WCAG 2.0 and 
four supporting documents on 19 November 2004 [2].  We invite you to 
comment on our latest drafts.

Here is an update on the WCAG WG's discussion of your comments about 
WCAG 2.0.  The WCAG WG has 31 issues related to comments from AbI [3].  
Of these, the WCAG WG believes 9 are closed.  Please verify if you agree 
with the WCAG WG resolution on the following issues.  If not, please 
provide more information for the WCAG WG to discuss.

[1] 
<http://lists.w3.org/Archives/Public/public-comments-wcag20/2004May/0003.html>
[2]  <http://lists.w3.org/Archives/Public/w3c-wai-ig/2004OctDec/0169.html>
[3] <http://trace.wisc.edu/bugzilla_wcag/buglist.cgi?keywords=AbI>

Comment #1, issue 832
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=832>

"(16) 3.2.3.1 should be promoted to level 2. Is there any technical  
reason why this would not be possible?"

The text, "The destination of each link is identified through words or 
phrases that either occur in the link or can be programmatically 
determined." was moved to level 2 in the 30 July 2004 WD.

=======================================================
Comment #2, issue 838
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=838>

"(7) Tables are not adequately accommodated in WCAG 2.0. It is our 
understanding that tables are not techniques-specific.
* 1.3.1.1: Should have the explicit requirement for marking of row and 
column headers in tables, either in guideline 1.3 or guideline 2.4. Cf. 
WCAG 1.0 checkpoint 5.1.
* 2.4.3.1: No explicit mentioning of (layout) tables required to be 
linearized. Should be added to WCAG 2.0. Cf. WCAG 1.0 checkpoint 5.3.
* Summaries for data tables should be explicitly required. Cf. WCAG 1.0 
checkpoint 5.5."

The WCAG WG believes that Guideline 1.3  covers tables and that the WCAG 
1.0 checkpoints related to tables are HTML-specific techniques.  The 
success criteria text is:   "Structures and relationships within the 
content can be programmatically determined 
<http://www.w3.org/TR/WCAG20/#programmaticallydetermineddef>."
<http://www.w3.org/TR/WCAG20/#content-structure-separation>

These topics are covered by the following HTML Techniques - many of 
which are under development:

    * 7. Data Tables
      <http://www.w3.org/TR/2004/WD-WCAG20-HTML-TECHS-20041119/#datatables>
          o 7.1 The caption element (optional)
          o 7.2 Summarizing data tables (optional)
          o 7.3 Terse substitutes for header labels (optional)
          o 7.4 Identifying groups of rows (optional)
          o 7.5 Identifying groups of columns (optional)
          o 7.6 Specifying the set of data cells for which each header
            cell provides header information
          o 7.7 Associating header cells with data cells
          o 7.8 Categorizing data cells
          o 7.9 Misuse of the pre element
          o 7.10 Row and column headings
    * 8. Tables for layout
      <http://www.w3.org/TR/2004/WD-WCAG20-HTML-TECHS-20041119/#layouttables>

          o 8.1 Layout tables (deprecated)
          o 8.2 Table structure in layout tables
          o 8.3 Summaries of layout tables
          o 8.4 Linear reading order of tables

As long as we provide clear guidance about transitioning from WCAG 1.0 
to WCAG 2.0 and ensure our HTML Techniques are comprehensive, do you 
think we will have adequately addressed your issues?

=======================================================
Comment #3, issue 839
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=839>

>(8) WCAG 2.0 has no success criteria regarding the contrast in images 
>(or "non-text content"), only for text (guideline 1.4).  This could be a 
>problem for images containing text - the author could claim that 
>contrast is no issue because it is an image (it is "non-text content" 
>per definition in the glossary of WCAG 2.0).
>
We believe that the following wording addresses your concern,
"Any text that is presented over a background image, color, or text can 
be programmatically determined."
<http://www.w3.org/TR/WCAG20/#visual-audio-contrast>

=======================================================
Comment #4, issue 841
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=841>

>(10) 4.2.1.1 & 4.2.1.2: Typo: Should say "(a) through (i)", not "(a) 
>through (j)"
>
We fixed this typo.

=======================================================
Comment #5, issue 849, 850 (duplicate of 849)
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=849>

>(21) 1.4.2.1 & 1.4.3.1: What is meant by ". or the resource provides a 
>mechanism.". Does it mean that the responsibility shifts from the author 
>to the user agent?
>
"provides a mechanism" has been removed in the latest draft.

=======================================================
Comment #6, issue 863
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=863>

>(34) 2.5.2.2: What exactly is "and can be provided without jeopardizing 
>security"?
>
Andi writes:
I can only explain this by providing an example but I don't think this
should be worked into the guideline itself. So the following two examples
that have been added to the proposal above to explain jeopardizing security
and purpose.

Example for jeopardizing security: A Web page requires the user to enter
both a userid and password. If either is incorrect, the user is informed
that there was an error but, for security reasons, is not informed if it is
the userid or the password and suggestions for correcting are not offered.

Example for jeopardizing purpose: An online test identifies incorrect
answers to the user but does not offer suggestions for correcting them. The
purpose of the online test is to test the user's knowledge, therefore,
providing hints on correct answers would go against the purpose of the Web
site.
http://lists.w3.org/Archives/Public/w3c-wai-gl/2004OctDec/0268.html

Proposed wording adopted in 19 November 2004 Working Draft:
http://www.w3.org/TR/2004/WD-WCAG20-20041119/#minimize-error

=======================================================
Comment #7, issue 864
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=864>

>(35) The following WCAG 1.0 checkpoint should be included in WCAG 2.0:
>5.6 Provide abbreviations for header labels.
>
This is also table- and html-specific.  If we cover this in the HTML 
Techniques will this address your concern?

=======================================================
Comment #8, issue 867
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=867>

>(36) We need a specific criterion for skipping over ASCII art in WCAG 
>2.0.  Cf. WCAG 1.0 checkpoint 13.10 Provide a means to skip over 
>multi-line ASCII art.
>
This is handled in the 8 October draft by:
Guideline 1.1 - Non-text content that does not provide information,
functionality, sensory experience and is neither multimedia nor 
time-dependent
interactive content, is marked such that it can be ignored by assistive 
technology.
http://www.w3.org/WAI/GL/WCAG20/#text-equiv

Perhaps refine text to read, "ignored or skipped [bypassed?]"?  Does 
this address your concern?

We look forward to your response and will provide further updates as we 
address other issues.

Thank you,
--wendy

-- 
wendy a chisholm
world wide web consortium
web accessibility initiative
http://www.w3.org/WAI/
/--