Clarification of duration of flashing or flickering (Success Criteria 2.3.1)

[Phill Jenkins]

Is the duration of the flashing or flickering (i.e., more than one second of time) a factor in determining accessibility compliance to WCAG 2.0
        See Success Criteria 2.3.1 http://www.w3.org/TR/WCAG20/#seizure
        "... do not contain anything that flashes more than three times in any one second period..."

In my special case example, there is a slightly less than 1 second duration of 3 flashes in an Adobe Flash image at the top of a draft Shanghai EXPO 2010 page.  

WARNING - currently it flashes 3 times for slightly less than one second, then repeats several seconds later.
        see http://fr.expo2010.cn/indextest.htm

Assuming the developer adds a capability to stop the repeating Adobe Flash movie, is it still a problem because it flashes 3 times, regardless of the duration (e.g., 1 second, 10 seconds, etc.)?  Or does it need to flash more that 3 times a second of time for more that 1 second of duration?

The "Understanding Guidelines 2.3" further mentions: "likely to cause seizure when viewed even for a second or two."
        see http://www.w3.org/TR/UNDERSTANDING-WCAG20/seizure.html

metrics to consider: Frequency, Duration, and Size (the text only really address frequency and size, not duration)
        frequency:         (or interval) more than 3 flashes per second of time? (3 to 50 Hz)
        duration:        more than one second duration of time? if it flashes more than 3, but for less than 1 second, is it a problem?
        size:                341 x 256 pixel block or larger - see http://www.w3.org/TR/WCAG20/#general-thresholddef

Since my example only flashes 3 times and not more than 3, and it soon may change to not repeat - Guidelines 2.3 says "more than three", then I think I do not have a problem, but I'm not sure - hence the question to the list.  Your informed opinions welcomed.

I also believe the text and explanation can be simplified.  For example:
1. by addressing the three different metrics; frequency, duration, and size.  
2. by addressing the lower and upper limits of each.  For example, if the frequency is 40 times in a second, but only lasts 1 second or less, regardless of size it may be hardly noticeable and not be a problem.  
3. add some more common passing and failing examples.  The two existing ones are good.  

Regards,
Phill Jenkins,

[Gregg Vanderheiden]

On Oct 27, 2009, at 11:42 PM, Phill Jenkins wrote:

Is the duration of the flashing or flickering (i.e., more than one second of time) a factor in determining accessibility compliance to WCAG 2.0
        See Success Criteria 2.3.1 http://www.w3.org/TR/WCAG20/#seizure
        "... do not contain anything that flashes more than three times in any one second period..."

In my special case example, there is a slightly less than 1 second duration of 3 flashes in an Adobe Flash image at the top of a draft Shanghai EXPO 2010 page.  

WARNING - currently it flashes 3 times for slightly less than one second, then repeats several seconds later.
        see http://fr.expo2010.cn/indextest.htm

Assuming the developer adds a capability to stop the repeating Adobe Flash movie, is it still a problem because it flashes 3 times, regardless of the duration (e.g., 1 second, 10 seconds, etc.)?  Or does it need to flash more that 3 times a second of time for more that 1 second of duration?

if it flashes MORE THAN 3 times in less than a second -- then it flashed 3 times in a second.   so it fails. 

but since yours only flashed three time in any one second period - it would PASS.    even if it repeats later it passes unless the repeat is soon enough that there are more than 3 flashed in any one second period

four flashes alone are enough if they occur within a 1 second window.  

(in the television code it would also fail if it ALMOST flashed four times a second  but continued to almost fail for a long time.  We did not carry that into WCAG). 

but more than three flashes in less than a second fails all by itself.  it is enough to trigger seizures. 

The "Understanding Guidelines 2.3" further mentions: "likely to cause seizure when viewed even for a second or two."
        see http://www.w3.org/TR/UNDERSTANDING-WCAG20/seizure.html

In the phrase  "likely to cause seizure when viewed even for a second or two."     the operative word here is "for a second".     the "or two"  should probably be removed since it implies that more than one second may be needed.   For some people that is probably true.  but not for others.    so the sentence should probably just say "likely to cause seizure when viewed even for a second"  since that is accurate.   We will look at this in the group. 

metrics to consider: Frequency, Duration, and Size (the text only really address frequency and size, not duration)

The metric does include duration.  it says   more than 3 flashes in any one second period.    the duration is one second.   the frequency is not stated.  it is a count and a duration.   the size and the intensity are also both specified. 

        frequency:         (or interval) more than 3 flashes per second of time? (3 to 50 Hz)
        duration:        more than one second duration of time? if it flashes more than 3, but for less than 1 second, is it a problem?

duration is any 1 second period.

        size:                341 x 256 pixel block or larger - see http://www.w3.org/TR/WCAG20/#general-thresholddef

Since my example only flashes 3 times and not more than 3, and it soon may change to not repeat - Guidelines 2.3 says "more than three", then I think I do not have a problem, but I'm not sure - hence the question to the list.  Your informed opinions welcomed.

You are correct.  you just pass.

I also believe the text and explanation can be simplified.  For example:
1. by addressing the three different metrics; frequency, duration, and size.  

these are confusing.  frequency assumes the flashes are equally spaced.   so you can't use that term. 

duration is specified.   any one second period.    size is specified -  as an area of the retina.   

2. by addressing the lower and upper limits of each.  For example, if the frequency is 40 times in a second, but only lasts 1 second or less, regardless of size it may be hardly noticeable and not be a problem.  

40 times a second may not be noticeable (in fact your flicker frequency is lower than that ) but it is still very capable of creating a seizure.     In fact 40 times per second is as dangerous as 10.     (20 is the worst)

3. add some more common passing and failing examples.  The two existing ones are good.

yes - I can see this could be helpful. 

 

Regards,
Phill Jenkins,

[Phill Jenkins]

Thanks Gregg for clarification - it was the key word "any" that I missed - as in: "any one second".

But I think there is still a need for an upper limit during that any one second period.  For example, if it flashes 80 times in that one second period I do not believe it is a problem - is it?  

Maybe a clearer discussion between flicker frequency that the developer has no control over verses the flash rate during that one second period would help?

Regards,
Phill Jenkins,

[Gregg Vanderheiden]

Hi Phill

   For hardware guidelines - an upper limit is made.    For software guidelines it was omitted because there is no reason in software to blink something above 30 hz because no one can see it.     The only thing would be subliminal messages and that would require a frame rate of 160 frames per second and it would be clearly visible -- so it wouldn’t be subliminal.    So there would be no reason to flash something intentionally in software at that frequency.     Frame flicker would not meet the intensity provision.   The frames would have to differ significantly back and forth over a large enough area with enough luminosity difference.  

For hardware you can get flicker.   And we are forced to set the upper limit at 50 hz because of line frequencies in Europe -- but that is actually a bad frequency (anything above 3 and below about 75hz is not good).    So no upper limit is set for software because there is not a need for one, and  it would be set different than hardware which could be very confusing. 

Your suggestion for  "a clearer discussion between flicker frequency that the developer has no control over verses the flash rate during that one second period ".      Sounds like a good idea.  If you can send more about what you are thinking of here it would be helpful.  You can do on-list or off.  

Thanks

NOTE:  These are quick feedback notes from me personally to advance the discussion on the IG.   Responses from the working group itself will come more slowly as the topics sent to public comment are logged and addressed in turn by the group. 

Thanks Phill

Gregg

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

Gregg Vanderheiden Ph.D.

Director Trace R&D Center

Professor Industrial & Systems Engineering

and Biomedical Engineering

University of Wisconsin-Madison

 

On Oct 28, 2009, at 6:06 AM, Phill Jenkins wrote:

Thanks Gregg for clarification - it was the key word "any" that I missed - as in: "any one second".

But I think there is still a need for an upper limit during that any one second period.  For example, if it flashes 80 times in that one second period I do not believe it is a problem - is it?  

Maybe a clearer discussion between flicker frequency that the developer has no control over verses the flash rate during that one second period would help?

Regards,
Phill Jenkins,

remember that most of these "standards" are usually a 90th percentile  so of
100 affected people 10 might still not be helped....
  you propably would not be violating the WCAG but you might still
cause a problem in a very small percentage of the people who might be
observing this.

Bob



On Tue, 27 Oct 2009, Phill Jenkins wrote:

> Is the duration of the flashing or flickering (i.e., more than one second
> of time) a factor in determining accessibility compliance to WCAG 2.0
>         See Success Criteria 2.3.1 http://www.w3.org/TR/WCAG20/#seizure
>         "... do not contain anything that flashes more than three times in
> any one second period..."
>
> In my special case example, there is a slightly less than 1 second
> duration of 3 flashes in an Adobe Flash image at the top of a draft
> Shanghai EXPO 2010 page.
>
> WARNING - currently it flashes 3 times for slightly less than one second,
> then repeats several seconds later.
>         see http://fr.expo2010.cn/indextest.htm
>
> Assuming the developer adds a capability to stop the repeating Adobe Flash
> movie, is it still a problem because it flashes 3 times, regardless of the
> duration (e.g., 1 second, 10 seconds, etc.)?  Or does it need to flash
> more that 3 times a second of time for more that 1 second of duration?
>
> The "Understanding Guidelines 2.3" further mentions: "likely to cause
> seizure when viewed even for a second or two."
>         see http://www.w3.org/TR/UNDERSTANDING-WCAG20/seizure.html
>
> metrics to consider: Frequency, Duration, and Size (the text only really
> address frequency and size, not duration)
>         frequency:      (or interval) more than 3 flashes per second of
> time? (3 to 50 Hz)
>         duration:       more than one second duration of time? if it
> flashes more than 3, but for less than 1 second, is it a problem?
>         size:           341 x 256 pixel block or larger - see
> http://www.w3.org/TR/WCAG20/#general-thresholddef
>
> Since my example only flashes 3 times and not more than 3, and it soon may
> change to not repeat - Guidelines 2.3 says "more than three", then I think
> I do not have a problem, but I'm not sure - hence the question to the
> list.  Your informed opinions welcomed.
>
> I also believe the text and explanation can be simplified.  For example:
> 1. by addressing the three different metrics; frequency, duration, and
> size.
> 2. by addressing the lower and upper limits of each.  For example, if the
> frequency is 40 times in a second, but only lasts 1 second or less,
> regardless of size it may be hardly noticeable and not be a problem.
> 3. add some more common passing and failing examples.  The two existing
> ones are good.
>
> Regards,
> Phill Jenkins,
>

-
end
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
occasionally a true patriot must defend his country from its' government
+++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
"They that can give up essential liberty to obtain a little temporary
safety deserve Neither liberty nor safety",    Benjamin Franklin
-   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -   -
   ASCII Ribbon Campaign. . . . . . . . . . . . accessBob
   .NO HTML/PDF/RTF/MIME in e-mail. . . . . . . accessys@...
   .NO MSWord docs in e-mail . . . .. . . . . . Access Systems, engineers
   .NO attachments in e-mail, .*LINUX powered*. access is a civil right
*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#*#
THIS message and any attachments are CONFIDENTIAL and may be
privileged.  They are intended ONLY for the individual or entity named