Talking Linux PDA discussion

[Deborah Norling]

Bob, my husband, wants to build an open-source Linux-based PDA for blind
people. This wouldn't be a commercial but rather a hobby project.

He's designed other open-source hardware before, including a PDP-8 replica
and an MP3 player.

I'm lukewarm about the project so he's hoping to find someone who is more
enthusiastic. In all fairness, my lack of interest is really my own lack of
confidence: I don't have even half my own problems onm my own linux machines
solved. He would expect me to write software for it and I don't think I'm
good enough.

Bob, on the other hand is a dynamite programmer; he just hates writing
software because he's done it to make a living so long. And he doesn't know
linux. When I tried to port the MP3 player firmware to an open-source
compiler, I could solve all the simple problems easily, but every time I got
stuck he had to bail me out. The project did get finished and the code is
available on his site. So I must say I'd be more excited about the project
if I was a smarter hacker.

Below is a post about the project he asked me to share with you'all. Feel
free to email him directly. Any public discussions about it on this list I
will share back with him.

Also if any of you have suggestions about a better list to discuss this on,
let me know.

**********
  First, let me apologize for a rather long and rambling post.  I could have
made it more to the point, but since I'm neither blind nor a regular member
of this list, it might be helpful to know a little bit about me and I
promise that before long it will have something to do with blind Linux
users.  I'm
an engineer that specializes in embedded hardware and software development,
and on the side I also develop kits and projects for electronics and
computer
hobbyists - you can see a few of those at 
www.SpareTimeGizmos.com
Also, many years ago, I was an engineer and later software department
manager at Telesensory, which is where I met my wife, Debee. 
  Debee is a regular Braille 'n Speak user and has three of them, the oldest
of which broke some time ago.  Looking at the dead B&S made me think about
how hopelessly dated its hardware and software is, but also about how nice
the mechanical and industrial design is.  It's just the right size and
weight
and lasts forever on batteries, and by most accounts the Braille keyboard is
easy to type on.  And of course it has no screen, which is a huge win for a
blind
user compared to all the mass market PDAs. 
  Then I started thinking that, with modern hardware, I could fit a Linux
machine in there. I'd keep the same plastic box and the keyboard, and design
new electronics and a new PC board to fit in the same space.  Add a hardware
text to speech synthesizer based on something like RC Systems RC86xx or the
the Winbond WTS701 and you'd have a talking Linux PDA with a Braille
keyboard that could run pretty much any text based Linux application.  With
Linux
and contemporary hardware we could also fit an MMC/SD card slot, Ethernet,
and USB in the same box. 
  But it's not quite as good as it sounds, though.  There are already
several embedded Linux systems and reference designs available that do 90%
of what
we need, but on this size device you're not going to find x86 based
systems.  It's going to be an ARM or an XScale or maybe a Coldfire processor
instead. 
And it's not going to run Debian or Fedora or SuSE or Slackware -- none of
those distributions; it's going to run an embedded Linux, like uClinux.  
  Most of the embedded Linux hardware platforms already come with a suitable
kernel, device drivers, TCP/IP networking, and standard utility programs,
but
any other software, like PIM applications or a screen reader, would have to
be ported.  And that brings us to the whole point of this message - I'm
interested
in working on the hardware part, but I have neither the time nor the
inclination to do the software work required. 
  As I see it, there are two major chunks of work required, not counting the
hardware - 1) a screen reader would have to be integrated into the kernel
and
modified to work with the Braille keyboard and the TTS chip, and 2) suitable
mainstream Linux applications would have to be selected and ported to this
platform.  I'd love to see an open source Braille and Speak Linux PDA, but
I'm going to need the help of other developers to make it happen. 
  I should also point out that a developer is going to need a development
system, and those aren't going to be cheap.  You can assume that the first
couple
of prototypes will cost as much as a PC.  After that, it will depend on how
many are sold. 

I would like to hear from anyone with ideas about making this project a
reality.

Bob Armstrong   
Bob "at" jfcl ( dot ) com