Re: [NTLK] [ANN] Einstein Platform 2006.6

Morgan,

i am amazed by your acquaintance of NewtonOS, and i really appreciate
your support.

Please would you gently forward your thoughts and tech notes about a
NIE Upgrade
on the comment section of the U.NIE webpage?

Here is the link to its rss feed:
feed://haloscan.com/members/rss.php?user=nieupgrade

Thanks in advance, Adriano

http://notwen.com/

01/giu/06 Morgan Aldridge wrote:

> I believe that what this means, and please correct me if I'm wrong
> (Paul or others), is that one can now write a new NIE (or Blunt)
> implementation that interacts directly with the host operating
> system's native networking APIs through the use of Relativity.
>
> To do this, one would:
>
> - Make a new NIE stub that mimics the NIE 2 API
> - Fill out the stub using Relativity calls to the host OS's
> networking APIs (i.e. no need to emulate Newton-compatible networking
> hardware)
> - Extend the NIE API to provide new features and functionality while
> still providing full backwards compatibility
> - Eventually write a branch of this new NIE that implements a newer
> TCP/IP stack and hardware compatibility for existing MessagePad
> hardware
>
> Of course, this is only one way of doing this, and it's possible that
> there's a better way of doing it. Paul may already be doing something
> of this nature, considering the other work he's putting into the
> Einstein Platform. The major benefits of this method are:
>
> - Compatibility with existing Newton Internet and networking software
> that is written to take advantage of NIE 2
> - Provides a roadmap to developing a new NIE (NIE 3?) for existing
> MessagePad hardware (Adriano's "Upgrading Newton Internet Enabler"
> initiative -- http://u.nie.notwen.com/ -- nudge-nudge, wink-wink, say
> no more)
>
> Obvious downsides:
>
> - May not be able to take advantage of features and/or functionality
> that don't easily fit into the existing NIE 2 APIs.
> - NIE's API may not be the most efficient and/or practical for a
> modern networking stack
>
> Of course, Eckhart Köppen has an advantage here in that he designed
> Blunt (http://40hz.org/Blunt/) and fairly recently, so he doesn't
> necessarily have to conform to an old API (as is evident in his
> development of a new Blunt 2) and the only application I'm currently
> aware of that relies on Blunt is its BlueTooth Setup utility.
>
> Here's hoping we start seeing some development using Relativity
> relatively soon (sorry, I couldn't resist).

-- 
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/