[nfbcs] Google Accessibility was RE: evaluation display of a web page

Wed Nov 6 03:41:41 UTC 2013

Actually, no one should be getting mad at this question as it is a very good
one. My opinion is that the answer is that it is the responsibility of all
parties involved--the manufacturer of the screen reader, the manufacturer of
the browser, and the producer of the websites.
As there is a growing desire to make websites that are more complex than
that which native HTML can support, there seems to be a movement away from
native HTML, which is what most screen readers support the best. The way to
make custom web controls now is to take a HTML element that does not already
have functionality, such as a span or div, and attach one or more event
listeners as well as decorating it with CSS. This clearly does not work for
screen readers, though, as a screen reader has no way of telling that the
element has functionality. The solution to this problem is Accessible Rich
Internet Applications (WAI-ARIA). From http://www.w3.org/TR/wai-aria/:
Accessibility of web content requires semantic information about widgets,
structures, and behaviors, in order to allow assistive technologies to
convey appropriate information to persons with disabilities. This
specification provides an ontology of roles, states, and properties that
define accessible user interface elements and can be used to improve the
accessibility and interoperability of web content and applications. These
semantics are designed to allow an author to properly convey user interface
behaviors and structural information to assistive technologies in
document-level markup. This document is part of the WAI-ARIA suite described
in the WAI-ARIA Overview.

Okay, so why doesn't this solve all of the problems?

1. Even though it is clearly defined how the roles and attributes are to be
interpreted and presented to the user, not all screen readers follow these
guidelines. For example, JAWS allows the user to specify how the accessible
should be determined instead of following the ARIA spec.

2. Even if ARIA is used, a screen reader only gets the information that the
browser passes to it. This explains why there are often differences in how
well the same screen reader will work on a particular website in different
browsers. Chrome Vox actually does not really have this problem because it
has direct access to the HTML without having to work with only what the
browser gives it.

3. Finally, even when the screen readers and browsers work correctly
together, it still does not solve the problem unless the producers of the
websites use ARIA. Not everyone knows about ARIA. Also, those who do
sometimes inadvertently do not quite use it in the correct way or do not
understand it. It is hard to visually test if ARIA is giving the desired
result as ARIA, for the most part, does not modify the visual appearance.

-----Original Message-----
From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Michael Babcock
Sent: Monday, November 04, 2013 3:29 PM
To: 'NFB in Computer Science Mailing List'
Subject: Re: [nfbcs] Google Accessibility was RE: evaluation display of a
web page

I'm going to go out on a limb and ask something that I've been curious
about... Likely this message will make someone mad.
Why is it google's responsibility to make there products accessible?
Why isn't it freedom scientifics or GW Micro's responsibility to make there
screen readers work better? I mean seriously we all fork out $800+ for our
screen readers on top of the price of the computer, (or state departments
do), and when something doesn't work the manufacture of the screen reader
blames inaccessibility from the vendor of the software were trying to use?
Sounds like someone is just pushing the blaim onto someone else to me...
IMO, and this is why I just use a 40 minute demo of jaws, and narrator with
nvda... I don't buy screen readers, and won't pay for one. Now, I understand
that people who write software can do better to make there software more
accessible, however, google is going to worry about making there stuff
accessible with there screen readers (chromevox, talkback, etc), and
honestly jaws and voiceover will take backburner. 
michael
I work from home, and you can to.
http://myownpay.com/

-----Original Message-----
From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Steve Jacobson
Sent: Monday, November 4, 2013 4:00 PM
To: NFB in Computer Science Mailing List
Subject: Re: [nfbcs] Google Accessibility was RE: evaluation display of a
web page

Jim,

One struggle I have is that I think in some cases Google is trying to do
some leading edge stuff that perhaps could be accessible but isn't supported
well by screen readers.  Our only choice to some degree is to ask companies
like Google to slow down, but I really think we need to get a better handle
on what the limits are to current accessibility and when we need to pressure
screen readers and when we need to pressure companies to conform some to
existing standards and good practices.  From what I know of screen reader
development, the problem isn't simply that screen readers don't bother
supporting what might be supported better but that they are having to try to
support so much that is new that they can't keep up.  What I don't think
people recognize is that the more resources one puts into a project, the
more management overhead is also added.  I don't think it is even
proportional, the ratio goes up faster.  By management overhead, I don't
mean people as much as all that has to be done to track changes and test new
features as well as making sure old features are not broken in the process.
I have found, for example, that some of Google's pages work better when one
turns off JFW's virtual cursor or Window-Eyes' Browse Mode.  Unfortunately,
there are still gaps, but it causes me to unsure when I should be
complaining to Google and when it is the screen readers.  I also don't know
how to resolve this adequately.  I really think we as consumers need to
somehow understand this better as we move forward.

Best regards,

Steve Jacobson

On Mon, 4 Nov 2013 13:47:25 -0800, Jim Barbour wrote:

>It is true that Google, and every other web application developer, 
>releases code far more frequently than older PC based software did.
>However, it's still a good idea to let google know when you find thuff 
>that doesn't work.

>The same is true for Apple.

>I don't know why, but we blind folks seem especially unwilling to speak 
>up and let companies know when stuff isn't working for us.  We seem to 
>have the rather toxic idea that "they should know if accessibility is 
>broken and if they don't want to fix it then I'm going to help them."

>Jim

>On Mon, Nov 04, 2013 at 01:40:37PM -0800, Mike Freeman wrote:
>> The problem is that "fixes" may not stick. Google is tinkering with 
>> its stuff constantly. The phrase "if it ain't broke, don't fix it" is 
>> not in their vocabulary.
>> 
>> Mike Freeman
>> 
>> 
>> -----Original Message-----
>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Nicole 
>> Torcolini
>> Sent: Saturday, November 02, 2013 3:16 PM
>> To: 'NFB in Computer Science Mailing List'
>> Subject: [nfbcs] Google Accessibility was RE: evaluation display of a 
>> web page
>> 
>> April and all, if you are having problems with Google products, 
>> please let them know. They may not be able to fix it right away, but 
>> they still want to know and might be able to tell you some kind of 
>> work
around.
>> 
>> -----Original Message-----
>> From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of April 
>> Brown
>> Sent: Saturday, November 02, 2013 2:05 PM
>> To: nfbcs at nfbnet.org
>> Subject: Re: [nfbcs] evaluation display of a web page
>> 
>> Ten years or so ago, I learned HTMl and attempted to code accessible 
>> from W3schools.  They do have Code check.  I don't think it's that 
>> good.  In the last year I have lost most of my vision, and much of my 
>> hearing, so it's even more important than ever!  And I always wanted 
>> to
code accessible.
>> Though, knowing some varying issues, especially with vision, I'm not 
>> 100% sure it is possible to code for every variation.  I may be wrong.
>> 
>> Hi *Susan Stanzel,  It would be wonderful if programs on both ends 
>> could fix the issues to make websites more accessible.  And I agree.
>> I have tried to learn NVDA, and well, learning keyboard workarounds 
>> is ten times harder than HTML ever was!
>> 
>> Hi ***Mike Jolls - Since you evaluate websites for accessibility, can 
>> I ask you a question?  For the last few years, my author website has 
>> been on a Google site.  Are Google websites accessible?  I can change 
>> some of the coding, though much of what I think would need to be 
>> adjusted is not accessible to the page holders that I can find.
>> 
>> Thanks.  Still new to the world of mostly deaf and blind, and the 
>> screen readers that confuse me when they don't just work when I open 
>> the page.*
>> 
>> *
>> 
>> 
>> _______________________________________________
>> nfbcs mailing list
>> nfbcs at nfbnet.org
>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>> To unsubscribe, change your list options or get your account info for
nfbcs:
>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/ntorcolini%40wavec
>> able.co
>> m
>> 
>> 
>> _______________________________________________
>> nfbcs mailing list
>> nfbcs at nfbnet.org
>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>> To unsubscribe, change your list options or get your account info for
nfbcs:
>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/k7uij%40panix.com
>> 
>> 
>> _______________________________________________
>> nfbcs mailing list
>> nfbcs at nfbnet.org
>> http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>> To unsubscribe, change your list options or get your account info for
nfbcs:
>> http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/jbar%40barcore.com
>> 

>_______________________________________________
>nfbcs mailing list
>nfbcs at nfbnet.org
>http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
>To unsubscribe, change your list options or get your account info for
nfbcs:
>http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/steve.jacobson%40vis
>i.com

_______________________________________________
nfbcs mailing list
nfbcs at nfbnet.org
http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
To unsubscribe, change your list options or get your account info for nfbcs:
http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/empoweringtheblind%40iclo
ud.com

_______________________________________________
nfbcs mailing list
nfbcs at nfbnet.org
http://nfbnet.org/mailman/listinfo/nfbcs_nfbnet.org
To unsubscribe, change your list options or get your account info for nfbcs:
http://nfbnet.org/mailman/options/nfbcs_nfbnet.org/ntorcolini%40wavecable.co
m