[nfbcs] Your thoughts on erosion of Windows accessibility

Steve Jacobson steve.jacobson at visi.com
Tue Jun 7 19:10:01 UTC 2016


Deborah,

As someone who has been around a while as well, I share your frustration.
There are some valid reasons for what you see, but in my opinion, there is
also some poor implementation as well.  There are two primary issues.  

First, it has long been felt that the capability to get information directly
from areas of memory that contains information about the screen has
contributed to instability in Windows.  This is how screen readers used to
build what was commonly called an off-screen model, a sort of database of
what and where items were on the screen.  The ability to use such a source
for screen data is still there, but there is more of a process in place to
only give access to that information to certified applications like screen
readers.  It is my impression that although this data still exists at least
to JAWS and Window-Eyes that screen reader developers are taking some of the
emphasis off this aspect of data, and that is partly due to other factors
mentioned in the next paragraph.

The second thing that is happening is that Microsoft has pushed hard to make
information available to screen readers and other software through other
methods, UIA and MSAA being the main sources.  There are now controls that
expose information to screen readers through other methods but do not have
data that is captured in off-screen models as would have been the case in
the past.  The information that is available is dependent upon what
information is provided, so there are times that the same location
information is not provided.  You will see this often in newer software as
you described in your note where the mouse keys simply read very little or
nothing at all, or may read words but be unable to read characters.  Screen
readers sometimes are having to combine what they can get using other
methods with the old method to provide a complete view of the screen, but
there sometimes are holes.  

As a part of this second approach, screen readers get a lot of information
by using add-ins, especially in Microsoft Office.  However Microsoft Office
can disable an add-in if for some reason it delays a response beyond a
timeout period.  This is an example of exactly what you describe.  Somehow,
in an attempt to get more reliable information for screen readers while
making the operating system more stable, we now have an environment that
seems more complex and contains more mysteries than ever.  We need to
understand more or lean on screen reader tech support more to keep things
working as they should.  While I cannot deny that I probably am accessing
successfully more complex software than I did ten years ago, I am dealing
with more crashes or interruptions than I used to.  It is hard to know which
part is Microsoft and which part is caused  by screen readers, but it seems
that both Outlook and Excel somehow know just when I am about to do a save
so they can crash before I do it.  <smile>  I feel that in many ways it is
easier to work around inaccessible software than it is to work around
accessible software that hangs or crashes a lot.  

I hope you find some answers here as well perhaps seeing that you are not
alone in your feelings.

Best regards,

Steve Jacobson

-----Original Message-----
From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Deborah Armstrong
via nfbcs
Sent: Tuesday, June 07, 2016 12:24 PM
To: nfbcs at nfbnet.org
Cc: Deborah Armstrong <armstrongdeborah at fhda.edu>
Subject: [nfbcs] Your thoughts on erosion of Windows accessibility

I'm new to the list but not new to I.T.: I started working as a blind
programmer in the 1980s.

I am wondering if others share my frustration with this trend or if I'm just
off-base. It seems like more and more applications are being made accessible
in Windows at least with what seems to be excessive scripting. A script used
to help make an application quicker to navigate and interact with. Now, it
seems like without a script, some programs are completely inaccessible.

A few weeks ago, JAWS simply stopped being able to read outlook mail. I
could arrow through the list of messages, but when I opened one, it behaved
like documents do in Word when they arrive in protected mode: I couldn't
read anything. I could still read mail with NVDA but it's more tedious to
use in Outlook 2010.

After a lot of research, I discovered the problem was caused by an add-on
for Outlook that had gotten corrupted. I had to remove and re-install it and
then things were fine. The situation was complicated by the fact that the
JAWS scripts depended on the availability of this add-on for their code to
properly read messages. I figured I'd be able to still use Outlook without a
script or an add-on, it just would be a bit less convenient.  I didn't even
know that without a script that depended on an add-on I wouldn't even be
able to read messages at all!

Another example: before Windows 7, any screen reader could review the entire
screen with its mouse cursor. So with an unfamiliar application it was easy
to locate and click on a control, label a graphic or just explore the layout
of screens. I wrote many simple scripts for JAWS when I needed quick access
to some control that wasn't keyboard-accessible.

But, now with Windows 7 the JAWS cursor, WindowEyes mouse cursor, or NVDA
screen review can't explore everywhere. On a friends PC, I found the same
problem with the Dolphin screen reader, so it seems like something happened
in the operating system to prevent screen access software from reviewing the
screen.

What's going on here and why does it seem more difficult to access programs,
especially unfamiliar ones in modern versions of Windows? Am I missing
something or do others feel there is a trend towards more limited
accessibility from Microsoft?

--Debee
(Deborah Armstrong)

_______________________________________________
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%40visi.com






More information about the NFBCS mailing list