Re: Updating 2.4, re-revised

On Wed, May 30, 2001 at 11:12:20PM -0500, Gregg Vanderheiden wrote:
> I'm not sure there should be an absolute timeout ban.    But I agree there
> should be something on timing.
> 
> For example,  the case cited had a 15 minute timeout.  I don't know of
> anyone who cannot respond in 2 minutes.      A person may need more than 15
> min to understand a page,  but that is different than not being able to
> respond.  The site could for example,  allow 15 min to pass and then simply
> ask "are you there?  Do you need more time?".   It could also reset the 15
> min each time the person did something like scroll.    In both cases it
> would allow a person to be on line for hours on a single page.  But they
> would have to indicate that they were there every so often.   If they left
> or shut off their computer etc, the system would time them out after 15 min
> of getting no activity or any acknowledgement from them.

I'm afraid this completely misses the point on accessibility for
people with attention deficits. I overstayed my welcome at the Dell
site because, after I put the computer I wanted on my "shopping cart",
my attention was distracted (by a telephone call) and, by the time I
noticed/remembered that I had been on a site with a timeout, it was
too late. An "extension notice" would have, ahem, escaped my notice
too.

It is not possible for people with attention deficits to turn off all
sources of possible distraction. Every timeout is a barrier.
Accessibility requires that timeouts be bypassable.  The "or extend"
clause would be an invitation to erect more barriers, not a way to
meet accessibility needs. And it is not difficult to design bypass
technologies that will still meet legitimate security and economic
constraints. One easy example: to get the bypass cookie, require an
e-mail address, and send a reminder to that address if a pre-set
window expires. I would get the reminder the next time I check my
e-mail, which is several times a day.

I am open to removing the reference to photoepilepsy - accessibility
for people with photoepilepsy  will be assured in any case, provided
that mechanisms for preventing motion are respected. Proposal now:

2.4 Do not limit the time that a user may need to understand or
interact with your content.
        * Provide disabled users with a way to bypass any demand to
          respond within a preset period.
        * Use automatic refresh and delayed redirection only when
          necessary to bring superceded content up to date.
        * Content must cooperate with user agent mechanisms for
          preventing motion (including flicker, blinking, flashing,
          self-scrolling etc) and for control of the rate at which
          motion occurs.

--
				Adam Reed
				areed2@calstatela.edu

Context matters. Seldom does *anything* have only one cause.

Received on Thursday, 31 May 2001 02:04:31 UTC