Slashdot Mirror
Developing PC's for the Legally Blind?
Jon asks: "My father is legally blind, to the point of not being able to see more then just a vague
shadow, if even that. His one hope has always been to use a computer. I've decided to
build him a PC that would allow him to send and receive email as well as read the weather
to him, yet not require any voice recognition. Input would be done by showing him where
certain function keys are and possible putting special marks he can learn on the rest of
the keys. Has any of the slashdot readers tackeled a project like this before? I've seen several projects, yet none that seem like a good solution, or are not complete. Any insight you have would be greatly appreciated."
I have worked on a computer that had screen readers installed for a user that was legally blind. It would read everything on the screen. "Mozilla. Title Bar. Slashdot. File Menu. File. Edit.." etc. Seemed like it would take a lot of concentration.
XP has some sort of narrator program built in, I believe.
My client is literally blind in that all he can see are vague shapes. He is also a hard-core smoker and his voice is fairly inconsistant.
He uses JAWS which is a program that reads information on the screen. I find it terribly annoying but he likes it's functionality.
He wanted voice input... and I called a number of vendors including IBM's voice project etc... but all of them said NO GOOD for nearly blind users as the training sessions are very long and involved and the user has to READ the screen (impossible).
He received a few weeks ago a flyer about the QPOINTER software product. This product is fairly new and not much is out there. I contacted the company and they said training was fairly short (15 minutes or so). I inquired if I could print out the training sessions and they (very nicely I might add) had their programmers send me the training sessions as text files. I printed them out in 2-inch high Arial Black letters on paper (29 pages worth) and brought them over and said "can you read these" and he said "nope". Great... but we installed the software anyway. The training system allowed me to pause so I would pause it, read a couple words to him and then un-pause and he would repeat them. We trained for about 3 hours and indeed it would open up the browser for him, etc. One problem was that he couln't read the visual cues that QPOINTER pops up - but by placing things in a KNOWN order I was able to get him around that.
However, when it came time to entering URL's etc... we were only able to get about 10% accuracy from the program - mainly due to the fact that his voice, after years of smoking, is terribly inconsistant. We retrained for a couple of hours on another day but to no improvement. Thus - we scrapped the voice input idea since he would have no idea if it recognized words and letters correctly or not.
At this point, frustrated, I realized that since I can type totally with MY eyes closed, he should be able to as well. Unfortunatly, he doesn't know how to type - but his wife used to be a typing instructor so... I went out and purchased him the ergonomic MS keyboard (the one with the huge split down the middle). I did this because that keyboard makes finding the proper keys easier due to the split (you don't need to hunt for the raised bump). His wife is currently going to try to teach him to type, and once he gets that down I'm going to go over the special key functions and try to make it work for him.
This is an area ripe for development. As us geeks get older we will encounter the same problems and we WILL demand a solution. Voice recognition is great, but it doesn't work except as a productivity enhancement (e.g., you still need to be able to read). A bigger screen would not help because no matter how big the letters are, he can't read them. Same for a head-up display. And voice input on a system that is ALSO doing voice output provides it's own set of problems (solved currently by using a headset mic).
Good luck - and let us know if you will, how this went for you!!
Aloha