W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2017

RE: example of accessible captcha?

From: Howard Leicester <howard_leicester@btconnect.com>
Date: Wed, 1 Feb 2017 21:44:42 +0000
To: 'Karen Lewellen' <klewellen@shellworld.net>, 'Andrew Kirkpatrick' <akirkpat@adobe.com>
CC: 'Gian Wild' <gian@accessibilityoz.com>, "'Sean Murphy (seanmmur)'" <seanmmur@cisco.com>, <w3c-wai-ig@w3.org>
Message-ID: <8BD78177BC244FBEA2058C7BA295721C@H30JC4J>
 
Kare's original question (CAPTCHA resolutions) is, of course, generic.
W3C/WAI is aiming to get some coordination together.

Is this (CAPTCHA) not an example where we might be a bit more coordinated?
So, can we at least present the examples sent so far as possibilities as
CAPTCHA solutions, and perhaps run experiements to see which (plural) work
best?
Enormous best,
Howard
(Howard Leicester,
NHS England).



----------------------------
Dr Howard Leicester
+44 (0)7789 965202
www.accessible-info.co.uk


-----Original Message-----
From: Karen Lewellen [mailto:klewellen@shellworld.net] 
Sent: 01 February 2017 20:47
To: Andrew Kirkpatrick
Cc: Gian Wild; Sean Murphy (seanmmur); w3c-wai-ig@w3.org
Subject: Re: example of accessible captcha?

Hi Andrew,
I appreciate and respect your providing this information.
May I ask a question?
Based on say your personal professional experience how likely is it that a
small business owner creating an international e-commerce website that may
possibly reach millions, will automatically understand this need on their
own?
The incident  leading to my starting this thread involved, not a small
business owner, but a very very large Canadian company.
Said company used recaptcha to create their verification,  and it did not
even manifest in their brain cells that an alternative was possible, let
alone required.
Now, if there was a point  in installing recaptcha  that the user reached a
message like this.
"warning! failure to provide an alternative captcha other than the visual
one  chosen may result in  human rights violations in your jurisdiction, or
may be just unfriendly!  do you wish to proceed?"
Then I would feel the wording of 2.0 was enough...but it is possible to use
the program without learning about alternative captcha formats.
Therein may be the true issue.
In the best of all possible universes,  automatically respecting the diverse
ways to use a computer regardless of body uniqueness would flow
effortlessly.  We are not there yet.
So, why is it not the default to create alternatives to visual captchas in
programs that produce them?
Dancing outside of the box...as usual,
Kare


On Wed, 1 Feb 2017, Andrew Kirkpatrick wrote:

> It bears pointing out that WCAG 2.0 allows CAPTCHA, but requires
accessible CAPTCHA.
>
> It seems that some people think that CAPTCHA is always an image-based
test, but that is not the case, it is any method that is used to distinguish
between an computer and a human.
>
> The only reason that CAPTCHA was called out in WCAG 2.0 was that it would
cause a failure if there was an image on a web page that didn't have an
equivalent alternative, and that would defeat the purpose of the CAPTCHA
image. If you do use an image-based CAPTCHA WCAG 2.0 just requires that you
provide alternative CAPTCHA methods for accessibility purposes, which is
just like if you need to provide an alternative to other types of
inaccessible content.
>
> Thanks,
> AWK
>
> Andrew Kirkpatrick
> Group Product Manager, Standards and Accessibility Adobe
>
> akirkpat@adobe.com
> http://twitter.com/awkawk
>
>
>
>
>
>
>
> On 1/31/17, 20:14, "Gian Wild" <gian@accessibilityoz.com> wrote:
>
>> You might be interested in these articles:
>> CAPTCHA: Inaccessible to Everyone: 
>> http://www.sitepoint.com/captcha-inaccessible-to-everyone/
>>
>> CAPTCHA: How to do it right (ie. don't use CAPTCHA!): 
>> http://www.linkedin.com/pulse/20140831232143-25659818-captcha-how-to-
>> do-it-right-ie-don-t-use-captcha
>>
>> -----Original Message-----
>> From: Karen Lewellen [mailto:klewellen@shellworld.net]
>> Sent: Wednesday, 1 February 2017 11:59 AM
>> To: Sean Murphy (seanmmur) <seanmmur@cisco.com>
>> Cc: w3c-wai-ig@w3.org
>> Subject: RE: example of accessible captcha?
>>
>> Well, at the risk of singing the same song again, a shared label does not
a shared experience make.
>> That you do not find an task difficult is absolutely fantastic...for you.
>> It does not mean everyone within the same well label has your tools your
situations, or choices.
>> I was frankly stunned that assurance could get away with such a clear
problem. I Mean what if the person has another reading  challenge?
>> Kicking them out because they cannot read when providing phones that have
no access at all is a tad much.
>> Kare
>>
>>
>> On Tue, 31 Jan 2017, Sean Murphy (seanmmur) wrote:
>>
>>> Interesting, point in relation to text messages. AS I have vision loss
myself, I didn't find it difficult.
>>>
>>> As there are more and more organisations using this as part of their
security mechanism. One example that comes to mind is two step
authentication. I suspect the challenge the example you provided is related
to the assistive technology they had available on the mobile device. If the
technology doesn't permit them to read text messages easily, then the
solution will break for that group of users. Thus isn't full proof.
>>>
>>> The other ideas promoted I need to check out. As this is really a pain
point for accessibility and disable users.
>>>
>>> Sean Murphy
>>> Accessibility Software engineer
>>>
>>> -----Original Message-----
>>> From: Karen Lewellen [mailto:klewellen@shellworld.net]
>>> Sent: Wednesday, 1 February 2017 10:28 AM
>>> To: Sean Murphy (seanmmur) <seanmmur@cisco.com>
>>> Cc: w3c-wai-ig@w3.org
>>> Subject: RE: example of accessible captcha?
>>>
>>> Hi sean,
>>> I personally dislike the text message idea for two reasons.
>>> first, you must provide your cell number, which if the site in question
becomes compromised  creates issues.
>>> second and most important though that method assumes that both a phone
is available, and a phone with easy to access text messaging.
>>> I knew someone experiencing sight loss who got kicked out of the low
cost American cell phone  program because their provider called assurance
wireless  used  text messages to contact members.  The phone provided in the
program had no accessible features so...they lost their service because they
could not  read the screen.
>>> I believe Google is behind recaptcha.  If they no longer encourage the
visual captcha, then recaptcha should not create as much.
>>> I agree totally with you about the audio editions of the challenges for
many reasons.  those are a poor solution  in my experiences as well.
>>> Kare
>>>
>>>
>>> On Tue, 31 Jan 2017, Sean Murphy (seanmmur) wrote:
>>>
>>>> Karen,
>>>>
>>>> I like the concept of sites which use text messages that appear on your
phone. Then you enter in the number they provide as the challenge.
>>>>
>>>> Audio caption solutions in my book are a negative experience because if
you cannot understand the audio output. Then you cannot complete the form.
As most audio challenges are distorted audio in the first place. I and quite
a lot of others find this method of authentication very poor. This is of
course to address the vision impairment community who cannot se the
challenge graphic.
>>>>
>>>> There was an article I read ages ago where I think google had 
>>>> developed a method of not requiring the graphical challenge at all 
>>>> and used a completely different method. I went looking for the 
>>>> article and cannot find it. :-)
>>>>
>>>> Sean Murphy
>>>> Accessibility Software engineer
>>>>
>>>> -----Original Message-----
>>>> From: Karen Lewellen [mailto:klewellen@shellworld.net]
>>>> Sent: Wednesday, 1 February 2017 6:12 AM
>>>> To: w3c-wai-ig@w3.org
>>>> Subject: example of accessible captcha?
>>>>
>>>> Greetings all,
>>>> I seek a site that uses a captcha which does not involve an image.  by
which I mean one using a math problem, or some other  interaction that
differs from the letter number things often used.
>>>> Ideas?
>>>> Thanks,
>>>> Karen
>>>>
>>>>
>>>>
>>>>
>>>
>>>
>>
>>
>
Received on Wednesday, 1 February 2017 21:45:26 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 1 February 2017 21:45:30 UTC