Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site wdl1.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!unc!mcnc!decvax!tektronix!hplabs!hpda!fortune!wdl1!jbn
From: jbn@wdl1.UUCP
Newsgroups: net.ai
Subject: Re: Thus spake the DoD...
Message-ID: <313@wdl1.UUCP>
Date: Mon, 4-Mar-85 23:18:37 EST
Article-I.D.: wdl1.313
Posted: Mon Mar  4 23:18:37 1985
Date-Received: Sat, 9-Mar-85 07:06:27 EST
Sender: notes@wdl1.UUCP
Organization: Ford Aerospace, Western Development Laboratories
Lines: 16
Nf-ID: #R:ssc-vax:-41700:wdl1:1100012:000:1087
Nf-From: wdl1!jbn    Mar  4 19:09:00 1985


    ... ``Don't forget that LISP is a programming environment, not just a
programming language.''  Indeed.  One problem with LISP is that it is usually
hard to separate out the completed program from the development environment
and run it in a production environment; there is a strong built-in assumption
in most LISP systems that the user wants all the programming tools available
at all times.  Some years ago, when Franz Lisp, which has a separate compiler
that generates somewhat nonstandard .o files was being written, I asked one
of the developers if it would be possible to link the .o files with a suitable
library, rather than loading them into the Lisp image containing the reader,
interpreter, the entire library, etc, so that only the library routines needed
by the compiled code would be present.  It was a sort of culture shock to him
that anyone would ask such a question; the idea that one would want to run LISP
programs without everything in the standard system present was totally alien
to his way of thinking.  But this is exactly what you want in embedded systems.