Re: [NTLK] [wwna-board] Re: [ANN] Einstein Platform 2006.6

Paul,

I agree that we need to focus on the projects that will bring the
Newton forward and especially those that will allow us to even get
that far (such as Einstein, new development tools, desktop
synchronization, etc.). I also know that I have no idea exactly how
large a project developing a new NIE would be (that's part of the
reason I mentioned recreating it using Relativity first, since it'll
be needed for proper backwards compatibility in the Einstein
Platform; granted, you've probably got a better way of implementing
such NIE compatibility and it sounds like you are working on this).
You know your stuff (obviously), so no need to debate NIE development
with you.

However, I'm a little confused as to the true goals of the Einstein
Platform. I see that it does two things:

* Provides desktop Newton emulation for the sake of testing and the
development of new Newton software (this is excellent and has moved
us towards better Newton development tools and more efficient
software development than any other project)
* Provides a way to run the Newton OS on more modern PDA hardware
(great for when original MessagePad hardware starts to fail; okay,
it's just plain cool as well)

However, the latter brings up the most important question: how far
can we go if we still rely on legal MessagePad 2x00 ROMs?

My concern is that we need all the hardware we can get in order to
keep the platform popular enough. Internet access is still reasonable
on MessagePad 2x00 hardware (look at Mail V, Courier, and Riassa, to
name a few), so I don't see the usefulness of that hardware going
away very soon.

The Einstein Platform running on newer PDAs means more people can
experience the ease of use of the Newton OS and hopefully jump on
board, but it's going to be costly as they need to buy two pieces of
hardware (the modern PDA and a MessagePad 2x00) and it's not [yet]
even as fast as the original hardware (although I trust that you'll
get there).

We then also have to support multiple OSes (currently you've done a
good job limiting it to Mac OS X and Linux), network stacks, network
configuration methods (i.e. configuring ethernet, WiFi, BlueTooth,
etc.), and whatnot, to make the new hardware as simple to use as any
old Newton.

Secondarily, I think we're still waiting for the ideal PDA to run it
on, in terms of design, form factor, etc. The MessagePads were close:
the form factor of the MessagePad 100 series was great, the fact that
there was only one physical button on the device helped; but they
need to be lighter and easier to hold. The Zaurus looks nice, but
wastes a lot of the surface area on physical buttons (a trend the
Palms started). The OQO is good in that it's full frontal surface
area is almost entirely screen real estate, but it's not an ARM
processor, so the speed will never be there. The Nokia 770 seems the
closest match so far, but maybe a bit too big. The video iPod (with a
touchscreen addition) is an interesting idea.

But, how far can we go like this? Won't we, at some point, need a
newer Newton OS (or Newton OS compatible OS/environment) that doesn't
rely on Apple's copyrighted ROMs? I think it's going to be hard to
pull in too many new developers and applications with the current
(and projected) restrictions.

I don't intend to come across as negative, I merely would like to
discuss this topic and see what your ideas are, maybe you have plans
in the works already, maybe others have thoughts on this as well. I
believe that the Einstein Platform is going to pull me into Newton
development, but I just currently don't have the time (and it's
definitely not something that would put food on the table; that and
I'm a web developer, so I need a CSS web browser in order to breathe).

Sincerely,

Morgan Aldridge

--
morgant@makkintosshu.com
http://www.makkintosshu.com/
On Jun 2, 2006, at 12:27 AM, Paul Guyot wrote:
[snip]
> For the first question, there are many projects that need to be
> undertaken as I mentioned at the WWNC 2006. Sooner or later, Einstein
> will run at a decent speed on PDAs and we will need state-of-the-art
> software for it. So basically, we need:
> * a new development suite (referred to as the New NTK project).
> * packages using Relativity providing modern, attractive (to
> developers) technologies to Einstein, such as Python, Java, OpenGL,
> etc. I already wrote a simple Python evaluator that works on MacOS X
> and that would work on any PDA with libpython installed. I wrote to
> Sean Luke about the Java equivalent.
>
> There is a list of things we don't need to waste our time on. For
> example, NIE. Please accept that NIE is outdated and that the old MPs
> will never do better than what they currently do, that is internet
> through WiFi, Bluetooth and so on. Rewriting NIE is way too much work
> for little result. NIE backward compatibility in Einstein is already
> underway, and some bits are already available although it's not
> really usable yet.
[snip]
-- 
This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries
Official Newton FAQ: http://www.chuma.org/newton/faq/
WikiWikiNewt for all kinds of articles: http://tools.unna.org/wikiwikinewt/