Path: utzoo!attcan!uunet!decwrl!sdd.hp.com!zaphod.mps.ohio-state.edu!sol.ctr.columbia.edu!cica!iuvax!ux1.cso.uiuc.edu!mcs.anl.gov!midway!oddjob!rfl
From: rfl@oddjob.uchicago.edu (Bob Loewenstein)
Newsgroups: comp.lang.forth
Subject: NEON
Message-ID: <1990Oct8.164154.20400@midway.uchicago.edu>
Date: 8 Oct 90 16:41:54 GMT
Sender: news@midway.uchicago.edu (News Administrator)
Reply-To: rfl@oddjob.uchicago.edu (Bob Loewenstein)
Organization: U of Chicago - Astronomy & Astrophysics
Lines: 47


I have been negotiating with Kriya (which is located in the DC area) about
getting NEON into the public domain. As of last week, they agreed to send me a
letter allowing the NEON sources to be shared. However, the stipulation was that
all references to the word 'NEON' be left out, since there is some legal problem
about the use of that name. I have not received the letter yet, but when I do, I
will arrange to release the sources as has been maintained to the present..

The current version of NEON that we use is up to 3.2, maintained by a few of us
at University of Chicago. Because it runs on a macintosh, and we have not ported
it to another machine (although that wouldn't be too difficult), major
differences from Kriya's last released version 2.0 are: multifinder compatible,
support for color quickdraw, works on all macs from mac+ to iici (not tested on
an fx). While it supports all toolbox calls, not all the interfaces have been
written. The sources I'll release include memory, event, window, control,
strings, serial i/o, utilities, device, floating point, and many other manager
interfaces. We also have TeEdit, Color Quickdraw, MacTCP, and other interfaces
as well, but these would be available separately, since they were developed for
applications, and not as generic interfaces. While bugs may exist, I am aware of
none in the 3.2 system.

In a related note, MOPS, Michael Hore's version of NEON will also be available
in the public domain. He took the nucleus and rewrote it as threaded subroutine
calls. He also changed all memory management to use relocatables for less heap
fragmentation. It runs faster that NEON, but doesn't yet support floating point.
Many neon sources will compile into MOPS, but not all. He also has a simple
implementation of multiple inheritence. Since we have a large number of NEON
applications to support, we will probably continue to use NEON rather than MOPS.
On most macII type machines, NEON runs fast, so that speed is not an issue.
Number crunching is best done in other languages, better designed for algebraic
formulation.

I'll post another note when I receive the letter from Kriya, but this negotiation
has been going on for almost a year. I talked to Chuck Duff a few months
ago, and he has helped by writing a letter to Kriya to encourage them to
release NEON into the public domain.

Anyone interested can contact me at

Bob Loewenstein
Dept. of Astronomy and Astrophysics
The University of Chicago
Yerkes Observatory
Williams Bay, Wi  53191

rfl@oddjob.uchicago.edu
414-245-5555