Fwd: Need some help

this is a discussion I have been having with Smith Kettlewell Eye Research
Institute regartding the contrast discussions on Issues 9 &10. Gus had some
nice things to say about our work and provided some useful points that we
can use. Sharing with permission so it is in our archives.

Jim
---------- Forwarded message ----------
From: August Colenbrander <gus@ski.org>
Date: Tue, May 2, 2017 at 11:15 PM
Subject: Re: Need some help
To: Jim Allan <jimallan@tsbvi.edu>

See below.

August Colenbrander, MD

---------------------------------------------------------

The Smith-Kettlewell Eye Research Institute and
California Pacific Medical Center, San Francisco
Home:   664 Atherton Ave.
             Novato, CA 94945
Phone:  415-209-9529 <(415)%20209-9529>,  mobile: 415-246-4668
<(415)%20246-4668>
Email:   gus@ski.org
Web:     www.ski.org/Colenbrander
----------------------------------------------------------


------------------------------
*From:* Jim Allan <jimallan@tsbvi.edu>
*Sent:* Tuesday, May 2, 2017 10:09 AM
*To:* August Colenbrander
*Subject:* Re: Need some help

Hi Gus,
comments inline...

On Mon, May 1, 2017 at 11:06 PM, August Colenbrander <gus@ski.org> wrote:

> Dear Jim,
>
> I read your different documents.  I am impressed with the level of detail
> and the discussion.  There is not much I would like to add.  You are to be
> commended for taking on this work.
>
Thank you for reviewing our work. Would you mind if I shared (posted) your
email to the group list?
No problem.

> As a general test, I would suggest viewing the screen from a considerable
> distance (how far depends on the degree of Low Vision that you want to
> accommodate) to see whether all important details disappear at the same
> time.  Details that fade too soon, need to be enlarged.
>
that is an interesting test.
I like simple tests.

> For instance, for the pie charts in 4. https://github.com/w3c/wcag
> 21/issues/9 ,
> I would like to see the print larger.  In general, the print size should
> be commensurate with the chart size.  When you stand back from these
> charts, the print becomes unreadable, long before the pie chart fades away.
>
Hmm, now to define commensurate. For the web, default font size in browsers
is 16px. Authors can and do change that regularly. Are you saying, the font
in the graphic should be at least the same size as the rest of the text on
the page? Or, that there should be some ratio, above a certain minimum,
between the image size and the text within the image?
I feel that if the graphic is considered relevant, it should have the same
readability as the text.
That is a judgement unrelated to the print size the authors choose to use.

Unfortunately, I cannot point you to any studies that relate print size to
> legibility for people with different kinds of low vision.  You may inquire
> whether Gordon Legge (legge@umn.edu) has any useful information based on
> his numerous MN-read studies.
>
We, the Low Vision Task Force, are in contact with Gordon. We think we have
print covered adequately. What we are trying to tease out, is the necessary
contrast of border around buttons, or the box or line that indicate that
the user can type text in a form field. Additionally, the contrast between
adjacent or overlapping items in a chart or graph.
The LVTF thinks (but no research to back up our thinking) the same contrast
as text is adequate. Others, not on the LVTF think (again with not
research) that the contrast for text is too high and should be set lower
for form control lines, and adjacent/overlapping items in a chart.
Many normally-sighted persons could read smaller print or print with less
contrast.  Many designers fall in that category.  You are designing for the
needs for a minority.  Their argument is like saying that a powerplant or a
public transport system can be considered adequate when it can cover the
average demand.  Both have to be able to also cover PEAK demand.  Did you
see the blog that pointed out that the mix-up at the Oscars  was  in large
part due to insufficient contrast on  the envelopes?  What is adequate
under normal circumstances,may not be adequate under stress.  Low vision is
a form of permanent stress.  http://www.visionaware.org/
blog/visionaware-blog/what-the-oscars-can-learn-from-
visionaware-about-print-legibility-and-effective-lighting-for-reading/12
What the Oscars Can Learn from VisionAware About Print ...
<http://www.visionaware.org/blog/visionaware-blog/what-the-oscars-can-learn-from-visionaware-about-print-legibility-and-effective-lighting-for-reading/12>
www.visionaware.org
What the Oscars Can Learn from VisionAware About Print Legibility and
Effective Lighting for Reading

> One point that is important is that there are many types of Low Vision
> with different needs.  I am convinced that contrast sensitivity is very
> important.  We usually miss it, because we do not measure it routinely.
> When it is measured, it is usually measured with threshold tests
> (Pellli-Robson or Mars), while ADL performance always involves
> supra-threshold contrast, supra-threshold sizes and sustainable (rather
> than threshold) performance.  20/20 by definition indicates the ability to
> recognize 1 M print at 1 m, but sustainable reading is best with 1 M print
> at 40 cm.  That is 2.5x above threshold.  The same is true for contrast.
>
Ah. this is useful. - back after an hour - did some searching on Contrast
threshold and related topics. I have several articles to review. And, lots
of math. When we have digested all the the information, we may need some
help converting the 2.5x you mention above and the recently found
information into the formula that is currently used by W3C. Would you be
available?
Certainly.  I am more in favor of actual readability tests (see below),
than of formulas.  I would consider the text legibility rules a given that
is not up for discussion.  The 2.5x value is not sacrosanct.  In any text I
would quote it as "2x to 3x".  It may vary for different tasks.  An early
quote about this performance reserve is from: Whittaker SG, Lovie-Kitchin J,*
Visual requirements for reading,*.  Optometry and Vision Science 70: 54-65
(1993).

For graphics, I would recommend:
>
>    - *If the text is equally intelligible without the graphics*, they
>    could be omitted, or presented in any desired way.  This also applies to
>    decorative graphics.
>
>    - *If the text is better understood with the graphics*, they should be
>    equally visible as the text.  I.e. when viewed from a longer distance, they
>    should be intelligible, as long as the text is readable.
>
> I like it better when the text appears next to the relevant detail, rather
> than as a side bar.  This holds for pie charts as well as for line graphs.
> Matching color hues can be difficult, especially on monitors of different
> quality.
>

Good point. we will consider adding that to our document. Another option
would be lines linking the key to specific areas. Lines with "good"
contrast as they cross multiple colors.
A line across multiple colors will still be recognizable if it has low
contrast on some color.  The brain fills in the missing part.


> Some Low Vision patients do better with *reverse contrast*.  This is
> common among glaucoma patients, who may suffer contrast losses in areas
> where they still have their visual field.  Others benefit from *bolder
> print*.  But clinicians rarely measure contrast, because it does not have
> a place in their clinical decision making.
>
If they patient benefits from bolder print, it seems to follow that they
would also benefit from bolder lines for form input controls. Right?
Correct.  Even if the form outline for input is clearly visible, the thin
cursor often is not.  The recommendation to use a longer viewing distance
could be supplemented with the use of blurring glasses; but this requires
an extra piece of equipment and is hard to standardize.

> AMD patients also may suffer contrast losses that are unrelated to their
> visual acuity loss.  There are those with good acuity and poor contrast,
> while others have poor acuity with relatively normal contrast.  Another
> difference may exist between those who still use a sick fovea (reduced
> acuity, central fixation) and those who use eccentric fixation.  The first
> condition has become more common, since the advent of anti-VEGF therapy.
> For eccentric viewing, increased spacing may be more important than it is
> for foveal viewing.  But those details are beyond what you can cover in
> general guidelines.
>
Many, many conditions (in addition to age) have contrast as a concomitant
issue.
Contrast is often considered a separate parameter from visual acuity.  This
is because we commonly use different charts to test it.  When you look at
the contrast sensitivity curve, you see that there is an interaction:
better contrast yields better acuity and larger letters require less
contrast.  The same is true for visual field testing.  Larger or brighter
test targets generally result in a larger field.  That is why we specify
high contrast for letter charts, and 10 dB for visual field targets.
All  these interactions cannot be covered by formulas.  That is why I favor
a simple experiment to make sure that the graphics are as visible as the
text.

>
>
> In summary, I cannot give you specific research, only some common sense
> suggestions.  If you have more questions, please feel free to contact me
> again.
>

This has been very helpful. Sparked new ideas and avenues for research.
Thank you for taking the time to review and responding.
I look forward to further discussions.

Jim

> Best regards,
> Gus
>
>
> August Colenbrander, MD
> ---------------------------------------------------------
> The Smith-Kettlewell Eye Research Institute and
> California Pacific Medical Center, San Francisco
> Home:   664 Atherton Ave.
>              Novato, CA 94945
> Phone:  415-209-9529 <(415)%20209-9529>,  mobile: 415-246-4668
> <(415)%20246-4668>
> Email:   gus@ski.org
> Web:     www.ski.org/Colenbrander
> ----------------------------------------------------------
> ------------------------------
> *From:* August Colenbrander
> *Sent:* Thursday, April 27, 2017 3:43:59 PM
> *To:* Jim Allan; Donald Fletcher
> *Cc:* Manfred Mackeben
>
> *Subject:* Re: Need some help
>
>
> Thank you, Jim.
>
> I will study the documents.
>
>
> Gus
>
>
> August Colenbrander, MD
> ---------------------------------------------------------
> The Smith-Kettlewell Eye Research Institute and
> California Pacific Medical Center, San Francisco
> Home:   664 Atherton Ave.
>              Novato, CA 94945
> Phone:  415-209-9529 <(415)%20209-9529>,  mobile: 415-246-4668
> <(415)%20246-4668>
> Email:   gus@ski.org
> Web:     www.ski.org/Colenbrander
> ----------------------------------------------------------
> ------------------------------
> *From:* Jim Allan <jimallan@tsbvi.edu>
> *Sent:* Thursday, April 27, 2017 2:40:06 PM
> *To:* Donald Fletcher
> *Cc:* Manfred Mackeben; August Colenbrander
> *Subject:* Re: Need some help
>
> Hi All,
> Appreciate your willingness to help. Items in [] are referents for links
> at the end of the message.
>
> A bit of background. The World Wide Web Consortium (W3C) Web Content
> Accessibility Guidelines (WCAG) working group released WCAG 2.0 in 2008
> [1]. Over the year folks with low vision among others approached the
> working group say the current guidelines do not go far enough to meet the
> web accessibility needs of people with low vision. A Low Vision Task Force
> (LVTF), a sub-group of WCAG, was formed in 2015 to provide input.
> The LVTF drafted a user needs document [2]. We have proposed several
> Success Criteria (testable statements that authors of web content should
> meet in order to be considered accessible). They are
>
>    -  User Interface Component Contrast (Minimum) [3] - for form
>    controls, buttons, and other visual affordances that are actionable (click
>    or enter key)
>    - Graphics Contrast [4] - for charts, graphs, icons, etc.
>
> The information used for the basis of the contrast ratios[5] was developed
> in the 1990's when WCAG 1.0 was developed. These ratios were developed for
> text. The LVTF is getting push back for applying the text contrast
> criteria for non-text objects. The WCAG working group wants to know what
> research the LVTF has to back up the use of text contrast criteria on non
> text objects. Many are assuming the contrast ratios should be lower because
> the objects in question are not text.
>
> Hope the above helps you understand our need.
> Any help or pointers appreciated.
>
> Jim
>
> 1. https://www.w3.org/TR/WCAG20/
> 2. http://w3c.github.io/low-vision-a11y-tf/requirements.html
> 3. https://github.com/w3c/wcag21/issues/10
> 4. https://github.com/w3c/wcag21/issues/9
> 5. https://www.w3.org/TR/UNDERSTANDING-WCAG20/visual-audio-c
> ontrast-contrast.html (see heading - Rationale for the Ratios Chosen)
>
> On Thu, Apr 27, 2017 at 1:16 PM, Donald Fletcher <floridafletch@msn.com>
> wrote:
>
>> Jim,
>>
>> Your project sounds interesting and important - great questions that need
>> to be addressed.
>>
>> Manfred may have some specific suggestions but I would also suggest you
>> run this by Gus Colenbrander.  He is particularly interested in and well
>> versed on these issues.  I have cced him on this note.
>>
>> Thanks,
>>
>> Don
>>
>>
>> Donald C. Fletcher, MD
>>
>> ------------------------------
>> *From:* Joshua A. Miele <jam@ski.org>
>> *Sent:* Wednesday, April 26, 2017 11:29:29 PM
>> *To:* Jim Allan; 'Manfred MacKeben'; Donald Fletcher
>> *Subject:* Re: Need some help
>>
>> Jim:
>>
>> I am copying two extremely smart Smith-Kettlewell researchers studying
>> low vision -- Manfred MacKeben and Don Fletcher. I feel certain they can
>> give you direction in finding the kind of research you need in order to
>> support the extremely important low-vision web accessibility work you
>> are doing.
>>
>> Manfred and Don, please see what you can do to assist Jim. See below for
>> a brief introduction to what he needs.
>>
>> Jim, best of luck, and please let me know if I can be of any further
>> assistance.
>>
>> Thanks to all!
>>
>>
>> --JAM
>>
>>
>> On 4/26/2017 3:35 PM, Jim Allan wrote:
>>
>> > Hi Josh,
>> > I am the facilitator for the Low Vision Task Force for WCAG 2.1 we are
>> > trying to create new Success Criteria for web content for people with
>> low
>> > vision. I need an introduction to some folks at Smith Kettlewell who
>> might
>> > could provide some information.
>> >
>> > We are trying to address what I call the 'dimming' or 'graying' of the
>> web.
>> > We already have contrast criteria for text. We are trying to apply the
>> same
>> > criteria for non-text object - form control borders, table borders,
>> > graphics contrast (like pie charts, not pictures), icons that help
>> > navigation or serve as links, etc.
>> >
>> > We are getting push back on apply the text contrast criteria for
>> non-text
>> > objects. Others want to know what research we have to back up the use of
>> > text contrast criteria on non text objects.
>> >
>> > Any help or pointers appreciated.
>> >
>> > Jim
>> >
>>
>> --
>> ----------------------------------------
>> Joshua A. Miele, Ph.D., Associate Director of Technology Research and
>> Development
>> The Smith-Kettlewell Rehabilitation Engineering Research Center
>>       on Low Vision and Blindness
>> 2318 Fillmore St.
>> San Francisco, CA 94115
>>
>> 415/345-2113 <(415)%20345-2113> Voice
>> jam@ski.org
>> http://www.ski.org/users/joshua-miele
>>
>>
>
>
> --
> Jim Allan, Accessibility Coordinator
> Texas School for the Blind and Visually Impaired
> 1100 W. 45th St., Austin, Texas 78756
> voice 512.206.9315 <(512)%20206-9315>    fax: 512.206.9264
> <(512)%20206-9264>  http://www.tsbvi.edu/
> "We shape our tools and thereafter our tools shape us." McLuhan, 1964
>



-- 
Jim Allan, Accessibility Coordinator
Texas School for the Blind and Visually Impaired
1100 W. 45th St., Austin, Texas 78756
voice 512.206.9315 <(512)%20206-9315>    fax: 512.206.9264
<(512)%20206-9264>  http://www.tsbvi.edu/
"We shape our tools and thereafter our tools shape us." McLuhan, 1964



-- 
Jim Allan, Accessibility Coordinator
Texas School for the Blind and Visually Impaired
1100 W. 45th St., Austin, Texas 78756
voice 512.206.9315    fax: 512.206.9264  http://www.tsbvi.edu/
"We shape our tools and thereafter our tools shape us." McLuhan, 1964