Re: [NTLK] "Upgrading NIE"

03/giu/06 Paul Guyot wrote:

> Adriano,
>
> Your plan with NIE is obscure and vague and the most concrete part is
> that you're collecting money (http://u.nie.notwen.com).

I am not a software developer, so developing on NIE is obscure to me.

> You do not define what you mean by "Upgrade the Newton Internet
> Enabler (NIE) to meet actual web standards."

I have put the "Upgrade NIE" webpage online as a service to the
Newton community, and i am really in need of help by people like you,
who would gently share their acknowledgement of NewtonOS and NIE.

I hope that you and other developers would agree to post some tech
notes to the comments section of the U.NIE webpage. Here is a link to
U.NIE's rss
feed://haloscan.com/members/rss.php?user=nieupgrade

> Newton Internet Enabler is only part of the problem. It's rusty, old
> and it never delivered its promises. Obviously, NIE was built like
> the Newton communication stack, for Newton to Newton/Mac
> communication, and it doesn't cope with modern networking. For
> example, it deals extremely badly with binary connections. All you
> can do is doing it with arrays of integers, which is memory consuming
> and slow, or with text interpreted as MacRoman, which is often
> inadequate. The constraints are such that Simon Bell dropped POP
> support in its Mail software because the POP protocol is difficult to
> handle at a decent speed with NIE limits.

I completely agree that NIE never delivered what it promised, infact
when i call for an upgrade, i mean that NIE lacks something that
people at Apple couldn't finish.

> NIE is old also means that it cannot deal with IPv6 which will be a
> problem tomorrow.
>
> Besides, the Newton lacks SSL. This makes it useless on most WiFi hot
> spots. SSL wouldn't be part of NIE, but it's very difficult to
> implement considering both the limits of native code on NewtonOS
> (it's hard to port most open source software such as OpenSSL or other
> SSL libraries) and the limits of NIE (SSL is based on binary
> transfers).

On 2004-06-08 Daniel Padilla posted here about a possible solution
for making SSL on Newton. He posted these links too:
http://www.delegate.org/delegate/
http://www.delegate.org/delegate/ssl/

Then there's the NewtSSH project started by Vlad and published here:
http://www.newted.org/users/vlad/projects.html

> Finally, the Newton lacks modern Wifi because it has PCMCIA slots
> that are not CardBus compatible. This is a hardware limitation.

I don't think that the lack of cardbus slot is a limitation for two
reasons:
  - Today is easy to build a WiFi adapter which would run TCP/IP
transparently over the serial port.
  - A cardbus slot can make your MessagePad run out of power in less
than an hour.

> To fix all these problems (except the hardware one) would require a
> HUGE amount of work. While it's theoretically doable, I seriously
> doubt that it will be done. If all these problems are fixed, more
> limitations will appear. You mention web standards (instead of
> Internet standards), so maybe your "Upgrade NIE" plan should include
> a modern web browser capable of logging to webmail applications. And
> flash support.

What do you intend for flash support? Why we would need it?

When recalling the "web standards" i mean the capability of using
java in example.
We have a subset of java on Newton: called Waba, it will run on
NewtScape.
I think that the browsers we already have are really nice for what
NIE can offer actually,
but a big problem (as i already mentioned on the thread about "Html
rendering ...")
is that the Newton doesn't take use of all its power when rendering
images from the net.

Please take a look at the result of the tests published here:
http://tinyurl.com/njwmo

I don't know the exact reasons why the MessagePad is so slow on
renderings. Do you know why?

> You say you love the MessagePad, then please accept it the way it is.
> The MessagePad never connected properly to the Internet and it's
> getting worse because the technology evolves faster than the
> remaining Newton developers can cope with. It's really great we have
> Bluetooth and first-generation Wifi. But despite Simon Bell's
> excellent work, the Newton simply cannot handle a modern mail flow.
> Web browsers cannot render most websites. You mentioned Raissa, but
> it cannot parse modern syndication formats. Etc.

About Raissa, do you mean that it will never be able to parse modern
formats? Or maybe such a lack depends on the time Eckhart had the
chance to dedicate to his wonderful application?

> I believe the only future of the Newton (if any) lies in the Einstein
> project. This is why I work on this and not on NIE or all other
> problems that remain to fix.

I comprehend that Upgrading NIE will be a huge work, but what we need
is not a solution for all the problems of actual version of NIE, but
just a solution for the slow renderings when surfing the internet and
maybe some new features. I wonder if part of the code in Einstein
could be already be used on such update.

> The Relativity project once targetted both Einstein and existing
> Newtons. It could have been a way to run OpenSSL and other open
> source libraries directly on existing MP2x00 and eMates. But I
> abandonned this part of the project in favor of focusing on my
> priority: Einstein itself.

NewtonOS is a powerful system, and it's fast. So as i trust that
you'll be able to make Einstein run at least equally fast as NewtonOS
is, we would need a very powerful device to run it on

MessagePad is never been a toy, and today we still can see its
superiority.
Instead we don't have a dedicated hardware to run Einstein on it (and
probably we will never have).

> Please accept the Newton is nearly dead. Looking at NewtonTalk
> archives, the last software announced for it, except for e-books, was
> "Rocket Science beta 1", published in December 2005.

Upgrading NIE we will certainly give to developers more reasons to
develop software for the Newton.

> A new NIE is not going to happen.

On the U.NIE webpage i said that all the money collected will be
granted to the developer or 'developer's group' who will be able to
publish the fastest and most complete internet suite for Newton OS
2.1 http://u.nie.notwen.com/

With the contribute of the developers who amazed us so many times, i
think that we have good chances to see NewtonOS become much better.

I agree that upgrading NIE will need a huge work, this is why i am
taking donations. More donations will be collected, more the chances
we will have to get the attention of the developers which would
happily work on the project but would also need to bring the bread at
home.

Regards, Adriano

http://notwen.com/

-- 
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/