Re: [NTLK] Newton Future: what could be done IMO

From: Paul Guyot (pguyot_at_kallisys.net)
Date: Wed Dec 04 2002 - 16:00:05 EST


À (At) 14:38 -0500 4/12/02, Stephen Swift écrivait (wrote) :
>I don't think we'd be able to do a hardware rebuild or even a PocketPC port
>without some people from the original Newton team. I don't think we'd be
>able to do a hardware rebuild over the internet. You need a team of people
>working together, making the hardware together -- almost start their own
>company -- to be able to pull something off like that. However, a port to
>PocketPC could be done through open source, over the internet, etc. It
>wouldn't be the greatest solution because we'd loose the newts hardware, and
>we'd still have a lack of software as others mentioned (Acrobat, Java).
>However, I think its a great start. We would at least have an option to
>continue to use a "newton" because something other than the old hardware now
>can run the Newt OS. Imagine also if we got tons of people to run the Newt
>OS on the PocketPC. It would catch Apple's attention, and perhaps then,
>they would revive the Newton, and we'd get what we originally wanted --
>modern-day Newtons in production.
>
>Okay, so my last idea is far fetched but, I'd support a port to a PocketPC.
>It would help to have funds for a project like this -- I don't know how the
>project would raise them. It would also help to have people from the
>original newt team. But it's definitely doable with time. And we'd get
>more attention as the project develops. So, how do we start?

Well, I don't think that original Newton developers will help.
Not that they won't be willing to (even when paid) but:
a/ they do other things now.
b/ they don't remember enough details to be really helpful (I noticed
that when I asked several ex-employees, they were only able to give
me a general description about how this or that worked).
c/ they can be scared of being sued by Apple as former employees.

On the contrary to what you think, the amount of knowledge of the
system that we lack is pretty small. Just because we don't need to
know most details. It's not about porting NewtonOS to PocketPC, it's
about emulating it.

We know nearly everything but the way the heart of the Newton works
(in details). But, this heart of the Newton, the voyager chipset, is
accessed through a P-Class. We have symbols. Reverse engineering of
this part will not be very long and is probably easier than the
reverse engineering of the storage mechanism I had to do for ATA
Support.

We don't have to reverse engineer full details. We just need to
figure out how these methods are supposed to behave. Then, we design
a custom ROM Image with Apple's ROM, Apple's REX (either the eMate or
the MP2K REX) and a REX of our own that will implement the
PlatformDriver interface. This new driver will be hooks to the
emulator. We do emulate parts of hardware that are purely memory
based (PCMCIA areas -- this part is obvious, screen -- Brian Parker
already decoded that part). We put all that in a big endian emulator
with MMU. We put the ROM at address 0. The RAM at whatever address it
is (we can redefine that in the REX). And that's all.

I won't say it's not a big amount of work. It is a big amount of
work. But it is doable. Next step is to figure out how to patch the
ROM (or extend the REX) and bootload that into a PocketPC or another
ARM Based PDA with sufficient storage. And then to extend the
NewtonScript layer for new APIs that might be required (like color
icons).

We won't get newer technologies, that is true.
But:
- having NewtonOS in an emulator will definitely fasten development
times, especially development of native code.
- we might be able, as a second step, put NewtonOS in a larger OS
(just like MacOnLinux) without too high performance loss.
As soon as NewtonOS works on PocketPC, there will be more users than
today. Including fresh motivated users who might want to work on the
project or other applications. Just because as we know, NewtonOS is
the best way to handle personal data including to enter it with a
pen. Plus the issue of synchronization & data exchange is being
solved. This would have been the latest reason for people to not
switch.

I wasn't really reasonable when I undertook the ATA project. I think
this project, if run collectively, can be reasonable. Or rather, I
think this is the most reasonable future of the platform. Rewriting
NewtonOS, rebuilding hardware are much more complex directions.

Paul

Will the last person to leave the platform please turn off the backlight?

-- 
NPDS: http://newton.kallisys.net:8080/
Apache: http://www.kallisys.com/

-- This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries List FAQ/Etiquette/Terms: http://www.newtontalk.net/faq.html Official Newton FAQ: http://www.chuma.org/newton/faq/



This archive was generated by hypermail 2.1.2 : Wed Jan 01 2003 - 10:01:46 EST