Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site nvuxd.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!gamma!pyuxww!pyuxv!nvuxh!nvuxr!nvuxd!michael
From: michael@nvuxd.UUCP (M.CAIN)
Newsgroups: net.micro.6809
Subject: Threaded interpreted language for OS-9
Message-ID: <101@nvuxd.UUCP>
Date: Mon, 19-May-86 09:03:34 EDT
Article-I.D.: nvuxd.101
Posted: Mon May 19 09:03:34 1986
Date-Received: Wed, 21-May-86 01:05:28 EDT
Organization: Bell Communications Research
Lines: 32

I am giving serious thought to writing a threaded interpreted
language for OS-9 (Coco in my case) for all the usual benefits
touted by TIL enthusiasts: acceptable speed, extremely compact,
very fast "compiler", interactive debugging, etc.  Before I
start, though, I thought I'd ask if anyone else is working on
such a beast.

The following list are the attributes I consider important in
an OS-9 implementation.  Would anyone care to suggest any others?

  - Code is reentrent, hence can take advantage of OS-9
    multi-tasking without multiple copies of the language
    kernal.

  - Code is position independent, so OS-9 can load it wherever
    it pleases.  Similarly, data space for a dictionary, variables,
    etc should be obtained dynamically.

  - Uses the standard OS-9 file system.

  - Provides access to many, if not all, of the OS-9 system
    calls.

A straightforward implementation of FORTH, for example, fails
most of these.

This is a non-trivial piece of coding, so again, is anyone else
working on such a language for sale or public domain distribution?
I'm willing to pay some reasonable price rather than re-invent
this particular wheel.

Michael Cain