[nfbcs] Glink?

[Mike Freeman]

Leslie:

Steve is right (see below). In addition to what he says, permit me to
observe that the software on the mainframe may change without warning and
without you having control over the change, thus breaking any scripts you or
those who wrote scripts for you may have crafted to deal with that software.
Much as you hate the thought, you may find it most efficient in the long run
to take an active part in gaining the information you need, that is, by
actively looking around the screen as opposed to having information given to
you automatically.

Mike Freeman


-----Original Message-----
From: nfbcs [mailto:nfbcs-bounces at nfbnet.org] On Behalf Of Steve Jacobson
Sent: Sunday, April 07, 2013 7:33 AM
To: NFB in Computer Science Mailing List
Subject: Re: [nfbcs] Glink?

Leslie,

As you are finding out, there are at least two distinct parts of this
problem.  The first is to be certain that the screen reader is always
tracking the cursor as we have discussed.  The second, which is more
complex, is whether the software on the mainframe is conveying information
through colors and/or highlighting.  That, of course, can be a problem with
any software, except that in Windows applications, certain things commonly
conveyed such as menus is done using Windows controls.  When connected to a
mainframe, something that might look like a normal Windows pulldown menu is
only seen by the screen reader as color changes since the whole mainframe
screen looks to the screen reader like a single "control".  If you are using
software like this, scripting may be your only solution as you have said,
but figuring out which emulator is the most stable and provides the best
interface with the screen reader is really important, but it seems you have
already figured that out.  

Where some of my caution comes from is that years ago, I did run some
scripts for the old IBM Screen Reader for DOS on an IBM mainframe, and some
were extremely complicated.  They worked well and were really a pleasure to
work with, but I found I often got off on a screen that either wasn't or
couldn't be anticipated.  I eventually dumped all of the complicated scripts
and took the approach of being able to query the screen for certain
information such as messages in certain color combinations, etc.  If you are
looking at getting a single piece of software working on a mainframe, then
scripting for that software is definitely a good thing to pursue.  If, on
the other hand, you are looking for a script that will produce automatic
speech in a number of applications, that is much more of a challenge.

Good luck.  Please let us know if we can help in some way beyond the help
we've tried to give.  In so many cases, the customer is really out there on
their own, but we'll try to help.

Best regards,

Steve Jacobson

On Sat, 6 Apr 2013 13:41:03 -0400 (EDT), Leslie Fairall wrote:

>Hi Steve:

>I believe that scripting will need to be done. We are telnetting into 
>the mainframe. The Braille display is following some screens, but not
others.
>Right now, I'm working on the 508 side of things, so I'm trying to find 
>a workable solution.

>--
>Leslie Fairall
>mailto:fairall at shellworld.net

>_______________________________________________
>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/k7uij%40panix.com