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

Re: 1.4 recommeded additions to wording

From: Doyle-Work <dburnett@sesa.org>
Date: Wed, 03 Nov 2004 14:40:31 -0900
To: Joe Clark <joeclark@joeclark.org>, W3C Web Content <w3c-wai-gl@w3.org>
Message-ID: <BDAE9E5F.3E87%dburnett@sesa.org>

Doyle's Post Starts Below -

>> Level 3 Success Criteria for Guideline 1.4
>> 
>> - Audio content does not contain background sounds or the background sounds
>> are at least 20 decibels lower than the foreground audio content, with the
>> exception of occasional sound effects. [V]
> 
> You have no proof whatsoever that this guideline actually provides
> accessibility benefit; there is no evidence at all that it can be met in
> the real world; the figure of 20 dB is purely arbitrary.
> 
> It seems the entire guideline is a sop to anticipated criticisms that
> Blind People Get Good Contrast But Deaf People Don't! Well, that's because
> the disabilities are intrinsically different.

In terms of classroom signal-to-noise ratios (SNR), there is research and
standards that would support WCAG's 20 dB difference.  According to the
American National Standards Institute's (ANSI) published "ANSI S12.60-2002,
Acoustical Performance Criteria, Design Requirements and Guidelines for
Schools" they state: based upon classroom size, recommends that background
noise level not exceed 35dBA, reverberation time (RT) not to exceed 0.6 -
0.7 seconds and a SNR of + 15 dB.  Based on these findings and standards
which were also replicated via an earlier study by the American
Speech-Language Hearing Association, it seems (at least to me) that some
ratio between background and foreground sound ought to be expressed as at
least a starting point.

Realizing this research and standards approach is based upon classrooms, I
feel it is still worthy of presenting a message that signal to noise ratio
in terms of multimedia is vitally important.  There needs to be some
starting place and "any" standard will leave out some but hopefully we'll
accommodate the majority.

I do think we should retain terms that the industry uses when looking at the
above guideline; signal to noise, background, foreground, reverberation,
noise, etc.  

If others have ideas as to how to address this guideline at this level, it
might be interesting to hear alternative approaches.

> As I've explained for a year and a half-- even at the Toronto f2f, where
> Ben later went right ahead and pretended I had never proved my point--
> colour deficiency requires accommodation only in confusable colours. In
> short, foreground and background matter only in the red/green and
> blue/green ranges. After several years of looking, I have found only rare
> examples of confusable colours used in confusable ways on the real Web
> that the Working Group either ignores or hates.
> 
> Accessibility for colour-deficient persons is a question of confusable
> wavelengths and has nothing per se to do with foregrounds and backgrounds.

In my efforts to address color deficiency, I personally have found NUMEROUS
web pages that have used confusable combinations of colors.  A single color
by itself is not, generally speaking, something that presents confusion to a
color deficient individual - we've been down this road before.

Doyle Burnett


Doyle Burnett
Education and Training Specialist
Multiple Disabilities Program
Special Education Service Agency
dburnett@sesa.org
Www.sesa.org
-- 





On 11/3/04 8:05 AM, "Joe Clark" <joeclark@joeclark.org> wrote:

> 
> This has always been on the wrong track and is only getting worse. I
> expect you'll all continue to pretend I never told you that; I've been
> consistent about it for months.
> 
>> Level 3 Success Criteria for Guideline 1.4
>> 
>> - Audio content does not contain background sounds or the background sounds
>> are at least 20 decibels lower than the foreground audio content, with the
>> exception of occasional sound effects. [V]
> 
> You have no proof whatsoever that this guideline actually provides
> accessibility benefit; there is no evidence at all that it can be met in
> the real world; the figure of 20 dB is purely arbitrary.
> 
> It seems the entire guideline is a sop to anticipated criticisms that
> Blind People Get Good Contrast But Deaf People Don't! Well, that's because
> the disabilities are intrinsically different.
> 
>> I had an action item to provide recommendations on 1.4 based on the open
>> issues and my research. Until we come up with an alga rhythm much of this
>> guideline is up in the air. One person recommended an extensive research
>> study. (Bug#996) Ideally that would be a good idea. I wouldn't mind joining
>> some people but it would require a team to conduct the size of study he is
>> suggesting in order to get a statistically meaningful sample size. He also
>> suggests it would require a statistician.
> 
> And I strongly doubt it would prove a benefit. Until such study is done,
> the Working Group must delete its guideline, since it implies here that
> the whole shebang is hypothetical.
> 
> 
>> Who Benefits from Guideline 1.4 (Informative)
>> 
>> Individuals with low vision can easily read characters in the content even
>> if they don't have the wide field of view
> 
> Visual field is unrelated to colour perception.
> 
>> or full range of color perception used by fully sighted persons to
>> separate text from background images.
> 
> As I've explained for a year and a half-- even at the Toronto f2f, where
> Ben later went right ahead and pretended I had never proved my point--
> colour deficiency requires accommodation only in confusable colours. In
> short, foreground and background matter only in the red/green and
> blue/green ranges. After several years of looking, I have found only rare
> examples of confusable colours used in confusable ways on the real Web
> that the Working Group either ignores or hates.
> 
> Accessibility for colour-deficient persons is a question of confusable
> wavelengths and has nothing per se to do with foregrounds and backgrounds.
> 
>> <new> This will also aid
>> comprehension for individuals with cognitive disabilities who benefit
>> from easy discernment of text.
> 
> OK, great, but every sighted person, according to the worldview of this
> guideline, requires "easy discernment of text."
> 
>> Visual contrast also helps individuals with hearing impairments who are
>> aided by clear visual representation of information </new>
> 
> Prove it. This sounds like grandstanding. What does being deaf have to do
> with psychologyof reading in this context?
Received on Wednesday, 3 November 2004 23:40:57 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:32 GMT