speechd-discuss
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Using graphical interfaces as a blind user, and a little wish (Re: pulse


From: Steve Holmes
Subject: Using graphical interfaces as a blind user, and a little wish (Re: pulseaudio and speech: performance issues)
Date: Thu, 7 Jan 2010 12:51:55 -0700

-----BEGIN PGP SIGNED MESSAGE-----
Hash: RIPEMD160

Due to the length of the previous message on this thread, I felt it
easier to not include it here.  I would like to comment some on this
debate of GUI vs text environments.  First off, I find a blend of
using both environments to give me the best results; you know, best of
both worlds.  I love using Sepeakup in the text console where I use
snownews for reading RSS feeds and the text editors vim and emacs work
great.  I shouldn't forget emacspeak either; it is a little side trip
but makes emacs and its subsystems really come to life.  But I really
need to comment on web browsers.  In the past, I've used elinks quite
a bit but more and more lately, I find many websites merely unusable
with text browsers.  Firefox plus Orca navigate these same sites quite
well.  What makes these work so well and efficiently with Orca is
"Structural navigation."  We can take full advantage of a site that
has good HTML markup in it like headings and tables and other form
controls.  If I want to get to the second table in a website, I merely
hit the letter 't' twice and boom, I'm on the second table.  There is
no way to do this in elinks.  With text browsers, there is no
structural navigation other than numbered links.  If I come up on a
site with twenty links in its navigation banner, I have to get past
all those links before I can start reading any text at all.  Also,
much of the javascript in elinks is very poor and just doesn't cut it
with many modern sites.  So I'm to the point I use Firefox for all my
web needs today.

I also like moving around in the file browser (Nautilus) for mass
renaming and movement of files.  The text equivelants like Midnight
Commander work OK but I never found them to be quite as efficient plus
the text screen readers like Speakup are difficult to work with when
trying to read isolated parts of the screen.  This is where Xwindows
and MS Windows screen readers do come through - they can read the
controls of interest and not anything else unless requested to do so.

Most important in closing is with a good flexible linux system, one
can achieve a reasonable balance using both console and graphical
applications.  And of course, there is no "one size fits all" solution
here.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.10 (GNU/Linux)

iEYEAREDAAYFAktGO1sACgkQWSjv55S0LfHalQCdFdIhN+wg42ZUGLN2NvYwPXUT
6lUAn0ppbCTl2MFlhI8u2/lk6tsiX9ox
=BSsH
-----END PGP SIGNATURE-----



reply via email to

[Prev in Thread] Current Thread [Next in Thread]