W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2004

RE: Guideline 1.4 and 1.5 Issues Summary and proposed resolutions

From: John M Slatin <john_slatin@austin.utexas.edu>
Date: Fri, 17 Sep 2004 11:05:53 -0500
Message-ID: <C46A1118E0262B47BD5C202DA2490D1A0331824F@MAIL02.austin.utexas.edu>
To: <caldwell@trace.wisc.edu>, "WAI GL \(E-mail\)" <w3c-wai-gl@w3.org>

Ben wrote:
<blockquote>
Issue #489. how to objectively measure questionable contrast?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=489> 
Proposed resolution:
OPEN: Added a dependency with Issue #344, which, if an algorithm can be
found, would make measuring contrast objectively measurable. 

</blockquote>

I'm concerned that the proposed resolution implicitly equates
"objectively measurable" with "machine-testable," an equation that we've
said we did not want to insist on in all cases.

If a set of users, including users with low vision and/or color
deficient vision, could all read some piece of text T over some
background B, would that not count as an adequate test?
John



"Good design is accessible design." 
John Slatin, Ph.D.
Director, Accessibility Institute
University of Texas at Austin
FAC 248C
1 University Station G9600
Austin, TX 78712
ph 512-495-4288, f 512-495-4524
email jslatin@mail.utexas.edu
web http://www.utexas.edu/research/accessibility/


 



-----Original Message-----
From: w3c-wai-gl-request@w3.org [mailto:w3c-wai-gl-request@w3.org] On
Behalf Of Ben Caldwell
Sent: Thursday, September 16, 2004 9:05 pm
To: 'WAI GL (E-mail)'
Subject: Guideline 1.4 and 1.5 Issues Summary and proposed resolutions



Hello,

The following list is a summary of the open issues associated with
guidelines 1.4 and 1.5. It includes a series of proposed resolutions for
each issue based on the proposal from Gregg to combine 1.4 and 1.5 that
Gregg sent to the list earlier today. [1]

[1] http://lists.w3.org/Archives/Public/w3c-wai-gl/2004JulSep/0677.html 

-Ben

Guideline 1.4 (visual-contrast) Issues:

Issue #344. Algorithm for testing foreground/background contrast.
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=344> 
Proposed resolution:
OPEN. The working group is still seeking an algorithm that measures
contrast in a way that is accurate and testable enough that we could
include it in the guidelines. 

Issue #377. threshold for grayscale vs. black and white
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=377>
Proposed resolution:
RESOLVED/UNCLEAR: Not clear whether this is still an issue. Marked
status as UNCLEAR, pending clarification from Harvey.

Issue #409. Are "foreground content" and "background content" generally
applicable? <http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=409> 
Proposed resolution:
CLOSED: Proposed changes include definitions of background image and
"foreground" is no longer used. Need to check back with Kynn to see if
this helps address the issue.

Issue #433. Why is visual/audio contrast extended?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=433> 
Proposed resolution:
CLOSED. We are no longer using core and extended and this proposal
incorporates criterion at each level that address the needs of low
vision users.

Issue #454. sugg. test for contrast
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=454> 
Proposed resolution:
OPEN: Associated with Issue #344 since algorithm is not yet complete. A
tool is in development that will be designed to allow testing contrast
in a repeatable fashion based on whatever algorithm is included in the
end.

Issue #467. contrast should be required
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=467> 
Proposed resolution:
CLOSED. We are no longer using core and extended and this proposal
incorporates criterion at each level that address the needs of low
vision users.

Issue #489. how to objectively measure questionable contrast?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=489> 
Proposed resolution:
OPEN: Added a dependency with Issue #344, which, if an algorithm can be
found, would make measuring contrast objectively measurable. 

Issue #509. easily differentiated in all forms of media?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=509> 
Proposed resolution:
CLOSED. This proposal no longer makes a distinction between visual and
audio contrast.

Issue #559. Contrast checkpoint seems to address usability issues rather
than accessibility
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=559> 
Proposed resolution:
CLOSED: The WG feels that contrast is an important accessibility issue
for individuals with low vision and color deficiency as well as for
individuals with hearing impairments.

Issue #590. vagueness of visual and audio contrast
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=590> 
Proposed resolution:
CLOSED. Proposed wording addresses the ambiguities noted in this issue.

Issue #591. comments regarding contrast and defn. of easily readable
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=591> 
Proposed resolution:
CLOSED. The phrase "easily readable" has been removed. Create a new
issue related to the questions about sources and references for 20 db.

Issue #592. what are "standard foreground/background contrast reqs."?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=592> 
Proposed resolution:
OPEN. The informative sections of this guideline have not been addressed
in this proposal and the examples still need some work.

Issue #605. Proposed wording for visual contrast SC 3
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=605> 
Proposed resolution:
CLOSED: revised proposal incorporates suggested plain language revisions
from John.

Issue #677. 1.6 and user configuration vs. default presentation
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=677> 
Proposed resolution:
The ability to turn off background images where there is text over an
image is not included for the following reasons:

1) If it is included, it would have to either be an or clause with the
two existing recommendations at level 2 that propose a minimum contrast
or it would have to be an additional criteria at level 2

2) If it was an OR clause with the two existing items it would then mean
that instead of having a minimum contrast, a company could have text
over a background image where there was terrible contrast as long as you
could turn off the background. Unfortunately, no one knows how to turn
off the backgrounds, so it would be mean that all of the material would
be less accessible than the guideline it its current form. 

3) If it was added as an additional item to level 2 it would mean that
it would never be possible to have text in an image (e.g., a .jpg or
.gif) ever even if the text was very high contrast and alternate text
was provided since it would not be possible to turn off the background
of the text in the image since it was part of both the foreground text
and the background image are part of single .jpg or .gif file. 

4) Hence, the ability to turn off the background is not included in the
guidelines since it would make them weaker or impossible. 


Issue #678. distinction between foreground and background content
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=678> 

1) We are not making a distinction between foreground and decorative.
Often the background information is not decorative. It may be useful for
users to have decorative information disappear, however, we do not have
any guidelines that address this. It is not the intention of this
guideline to address it.

2) The issue of overlapping 2 pieces foreground content would be
addressed by the fact that when the first one is being read, the second
one would be considered background to it. When the second one is being
read, the first one would be considered background to it. Thus, the
contrast requirements would allow foreground content to overlap as long
as the mutual contrast requirements are met. We will add a note to the
testing and techniques documents to make this clear for those instances
where it occurs. 

Issue #797. Level 1 SC Note rewording proposal
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=797> 
Proposed resolution:
CLOSED: Incorporated suggested rewording for note from Andi (IBM
comments) 

Issue #828. Clearer definition of "foreground" and "background"
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=828> 
Proposed resolution:
CLOSED. New definitions have been included in this proposal.

Issue #839. contrast issues for non-text content?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=839> 
Proposed resolution:
CLOSED: Addressed in rewording.

Issue #849. meaning of "resource provides a mechanism"?
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=849> 
Proposed resolution:
CLOSED: "provides a mechanism" has been removed in the proposed text.

Issue #872. "make it easy" is too subjective
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=872> 
Proposed resolution:
CLOSED: The guideline itself uses the term "easy" but this does not
appear in the success criteria which are very objective. Thus, the
guideline is meant to give the general idea for what should be done and
success criteria tell you exactly what is meant and how much must be
done. 

Issue #984. Contrast should be required at minimum level
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=984> 
Proposed resolution:
CLOSED. We are no longer using core and extended and this proposal
incorporates criterion at each level that address the needs of low
vision users. This proposal includes a level 1 that allows the user to
extract the text and to present at any contrast level they desire. At
level 2, it has specific requirements that the default presentation be
at different contrast levels. 


Guideline 1.5 (audio-contrast) Issues:

798. SC proposal for background audio that plays when a page is loaded
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=798> 
Proposed resolution:
CLOSED: Proposed revisions include a new criterion specific to the issue
of background audio.

873. Good advice but should be in style guide, not these guidelines
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=873> 
Proposed resolution:
CLOSED: This is a very important criterion for individuals with hearing
impairment. If the background sounds are too high, it is impossible for
them to differentiate foreground from background and makes the content
inaccessible. At the level 3 (the highest level of accessibility) we
therefore suggest that the background sounds be kept 20db down. It is
not included at a higher level because individuals who cannot hear who
have difficulty telling the foreground from the background can rely on
captions. 

985. Give audio contrast same treatment as visual contrast
<http://trace.wisc.edu/bugzilla_wcag/show_bug.cgi?id=985> 
Proposed resolution:
CLOSED. There is no Level 1 or Level 2 criterion requiring that
background sounds in an audio track be lower than the speech in an audio
track since there is already a requirement that the audio be captioned.
Thus, someone who is having trouble hearing the speech over the
background sounds can turn on the captions. There is no similar
requirement which is over the top of graphics be read aloud. Hence, the
need for Level 1 and Level 2 criteria in dealing with visual text over
graphical information.


--
Ben Caldwell | <caldwell@trace.wisc.edu>
Trace Research and Development Center <http://trace.wisc.edu>   
Received on Friday, 17 September 2004 16:05:54 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:17:58 UTC