Re: [NTLK] Linux on Newton 2100?

From: Will Leingang (will.leingang_at_navpress.com)
Date: Thu Sep 19 2002 - 21:47:30 PDT


I think you're both on the same page, tear the guts out of Newt OS and
re-build it (ground up) on an embedded Linux kernel. That way, the things we
like about the newton, stay with the newton, and the things we like about
linux, stay with the newton...

I honestly believe that any future newton software/hardware development
should progress in the open source arena. As Compaq/HP have shown, there are
many advantages to developing products along this line. Although a hobby
project at the moment, putting linux on the newton may very well be the
springboard to advancements in current PDA development, including better
handwriting recognition and a very friendly user experience. I'm not talking
about selling linux equipped newton's, but what I am talking about is
building a powerful, scalable, and usable embedded linux OS with newton
technology. This may all sound like a pipe dream, but I plan on demoing a
"proof of concept" OS at a gathering of CRL (Cambridge Research Labs)
engineers and iPaq linux/itsy enthusiasts at an October gathering in
Cambridge Mass. While I'm well aware of this aggressive, possibly insane
project, I don't think a single person would stand in opposition of using
Newton Technology in future embedded Linux UI development.

As far as how my itsy project went, all of the PCB construction was done on
single layer PC construction bread boards purchasable at any Radio Shack.
Parts I couldn't hijack off of an iPaq I bought online, through very
delicate negotiating(usable DRAM was nearly impossible to find). I read the
schematics with Adobe Pagemaker and a free program they provided, when it
was all said and done, I had a dismantled iPaq 3650 and an ITSY that wasn't
very ITSY... For the record, I never actually "finished" ITSY, I got it to a
usable point, but never built into a self contained unit, I'm 19 and lack
the follow-through and skill.

I'm hoping with the experience of some embedded systems development we
should be able to get linux up and running without too much effort, the
problem exists in leaving newtOS stable. On the other hand, most of my
experience has been with the assistance of thorough documentation, this is
my first attempt at blind development.

As of yet, Apple's been no help in releasing HW documentation, it seems all
Newton engineering specs vanished along with the project itself. I guess
you're right, they're not going to help.

Continuing investigation,
Will

-----Original Message-----
From: Karel Jansens [mailto:kareljansens_at_tiscalinet.be]

> My comment to this is that while yes, I agree with this statement,
> there's only going to be so much continued support down the road. I
> honestly would like to see my Newton last me much longer than another
> 15 years and there's not enough programming skill directed towards
> making that happen. The last thing I'd want to see is a bloated
> X-windows screen running FVWM-95, but a Linux core could mean a more
> Newt-centric outer layer. This would give us more say about where
> things could head down the road, including hardware enhancements.
> There's no reason why the entire Newton OS couldn't be re-written as
> open source and still operate the same. Someone crafty enough could
> possibly even keep Inkwell running through the original binary
> support. At the very least, it couldn't hurt to try.

Reading this, one thing struck me: Wouldn't it be smarter to work the
other way around? Meaning: Try to develop a Newton-like User Interface
that runs atop a linux kernel?

-- 
Read the List FAQ/Etiquette: http://www.newtontalk.net/faq.html
Read the Newton FAQ: http://www.chuma.org/newton/faq/
This is the NewtonTalk mailing list - http://www.newtontalk.net/


This archive was generated by hypermail 2.1.5 : Mon May 26 2003 - 12:59:22 PDT