This topic is about learning GUI's and I want to honor that as much as possible but Janina touched on a very important piece of the puzzle to getting a good experience with computers. I actually wrote my own screen reader for the IBM P.C. and continued to use it in DOS until 2009 when Vinux came along. If you are a tinkerer and like to write programs which usually also involves lots of debugging, the best screen readers echo every character you type plus read all new input to the screen from the system. If the input is bursty such as from a compiler or assembler, you should hear a sequential playback of what is happening as soon as it can be read out. My home-made screen reader did this pretty well in to an Echo speech synthesizer and speakup does it beautifully, also. The interface feels or sounds natural and you don't have to pump it for screen after screen unless that is what you want. The terminal in orca also works pretty well but not quite as well as the good old command-line but I am still not complaining. The voice in Apple's voiceover is good but there is one glaring problem which I would love to see fixed and yes, it is broken if you can't change the behavior. If you are listening to bursty output in voiceover, there tends to be a short time period after which new text that arrives blows away whatever is being said right then so you just hear shreds of useless babble. I hate to be picky but I'd like to see a switch, so to speak, where the user can select either a speakup-like mode or the current voiceover mode. Both have their moments where they are the best but I can not use voiceover for Unix trouble-shooting, logging and debugging. Speakup is great at all these things as well as very good at ssh logins and the rare telnet session. I have even had speakup do weird things on slow serial connections and there is a parameter one can tweak in which speakup decides that all new input has stopped and it goes ahead to speak what characters have accumulated so far. My old home-built screen reader actually also had a short timer that kept resetting on new characters unless there had been around a tenth or twentieth of a second with no data. I was glad to see speakup come along as my project only worked in DOS and it was getting to be a monster toward the end. It was 8086 assembler with many modules. One could have done much, much better in C but by then, it was time to move on to Linux. Martin Janina Sajka <janina@xxxxxxxxxxx> writes: > I think as a general rule, Cheryl is quite correct. Support for the > terminal in Voice Over, and in every gui screen reader I know is frankly > subpar. We got to the point in DOS where the screen readers, especially > asap and vocal-eyes, were powerful and able to change configurations on > the fly to deal with different screens properly. Even Speakup can't do > that like they did, though I use Speakup far more than any gui, even > Gnome and Orca. > > Janina _______________________________________________ Blinux-list mailing list Blinux-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/blinux-list