speakup desktop again

[Christopher Moore]

Hello,
I saw this topic on the speakup list recently and followed the 
discussion with interest.  

I'd classify myself as a happy linux user.  When I was working I more or 
less had to use windows to conform to the culture of my employer.  Even 
if linux alternatives were available, convincing the IT staff that 
accessing their network with "non-standard" software would somehow 
compromise their security would not have been 
an easy task.  Now that I am retired, I find that the linux text console 
satisfies most of my computer needs except for web browsing.  So I still 
need to boot up the ms windows to access some web pages when I need to 
order something online or listen to audio streams.  

While reading this discussion, it occurred to me that we already have a 
text-based desktop environment in the form of emacs.  What emacs lacks 
is an up-to-date web browser.  While I've read comments on the w3 
browser, most of them are not favorable.  What emacs does have, however, 
is buffers and robust editing features.  If, there were a way to say, 
run firefox in a X session and control it from an emacs environment, 
would this not fill the gap?  

Let me take this a step further.  We're all aware of the effort to make 
the gnome desktop accessible.  This is a major undertaking and involves 
retooling many applications to use the gtk+ archa tecture to expose 
their contents.  It further involves developing screen readers to 
retrieve this information and speak it in some usable format.  To-date I 
am aware of three such screen readers: gnopernicus, orca and lsr.  Work 
on the first appears to have stopped while work on the latter two is 
still underway.  

This brings me back to the emacs environment.  Much of the work on the 
above screen readers centers on developing speech and braille output 
facilities.  Orca for example, has an on-screen dialog for 
modifying speech, braille an magnification parameters.  If the screen 
readers were designed to communicate with emacs, the spoken content 
would appear in an emacs buffer and the applications could be controlled 
from a text console.  If this approach had been taken, all the effort 
put into developing the braille and speech facilities would have been 
unnecessary.

I'm looking for volunteers to investigate the prospect of 
communicating 
with X applications from a console environment.  While it would be nice 
to use the entire gnome desktop, my initial focus would probably on web 
browsing since there is a definite gap in this area.  

Feel free to comment on the speakup list or to me directly.  If such a 
group were to be be formed it would probably make sense to start a 
separate mailing list or other vehicle for sharing ideas.

Chris