Newsgroups: comp.robotics
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!howland.reston.ans.net!spool.mu.edu!umn.edu!vulcan.iss.bnr.com!vulcan!tgb
From: tgb@bnu003 (Tom G. Brusehaver/Consultant Euler Solutions)
Subject: Re: Stamp Suggestion
In-Reply-To: tmornini@sun.parallaxinc.com's message of 13 Mar 1995 19:03:22 GMT
Message-ID: <TGB.95Mar16084110@bnu003>
Sender: news@vulcan.iss.bnr.com
Nntp-Posting-Host: bnu003.cncc.bnr.com
Reply-To: tgb@bnr.com
Organization: cncc.bnr.com
References: <3jovn6$qrh@nntp.crl.com> <24.11694.2500@pcohio.com> <3k24tq$mhm@nntp.crl.com>
Date: Thu, 16 Mar 1995 14:41:10 GMT
Lines: 31


 T> I believe that this is a request for Pascal & C compilers for the
 T> Stamps, but I don't understand the reference to extra systems for
 T> 1 program...

 T> Also, the Stamp engineer, Chip Gracey, thinks that the level of
 T> complicity between the BASIC Language, the tokens and the BASIC
 T> Interpreter preclude the possibility of efficient tokenization
 T> from anything other than the BASIC language itself.

Hmmmm.... all that talk of tokenizations, and Pascal, I remember the
"Good old Days (tm)" when we used to use a Perq.  This beast was a
bitsliced computer with a 4K WCS (working control store).  The WCS had
a loader, and the P-code interpreter, and a little room for users to
extend their application.

When Perq was planning a switch to the Accent OS, there was talk of a
C code interpreter.  I guess people saw prototypes of the C code
interpreter, but I never saw it.

How about someone coming up with a P code interpreter for a
microcontroller, and then I guess the need for a Pascal compiler would
be in order.  Maybe an old Pascal compiler would have a "standard" P
code output (Apple II pascal, or some of the early IBM Pascal's), and
these could be used.

I guess this is sorta done already with ICC.

tommy
--
tgb@bnr.com  home bnu002.uucp!tommyb!tom@bnu002.cncc.bnr.com
