Path: utzoo!utgpu!news-server.csri.toronto.edu!mailrus!wuarchive!uunet!zephyr.ens.tek.com!tektronix!reed!trost
From: trost@reed.bitnet (Bill Trost,Box 607,ext 640,)
Newsgroups: comp.lang.smalltalk
Subject: Re: GNU Smalltalk/Emacs Interactor Problem (and STIX)
Message-ID: <15368@reed.UUCP>
Date: 26 Aug 90 00:37:39 GMT
References: <26d2ff3e.3a44@petunia.CalPoly.EDU> <Fc&3pyt1@cs.psu.edu>
Sender: news@reed.UUCP
Distribution: usa
Lines: 23

In article <Fc&3pyt1@cs.psu.edu>, melling@cs.psu.edu (Michael D
  Mellinger) writes:
"I also had the problem with the shell-mode-map. To get around it, I
simply created a shell (M-x shell) before I ran Smalltalk."

That's the "right answer", more or less.  shell-mode-map is defined in
shell.el, which probably should be "required" (in the Lisp sense) in
smalltalk.el.  I will look at the code and send a fix to Mr. Gnu
Smalltalk.

(Actually, that code deserves some work --- it's not very well in line
with the other computer language modes)

And, in <26d2ff3e.3a44@petunia.CalPoly.EDU>,
  mjohnson@riglow.acs.calpoly.edu (Mark S. Johnson) writes:
"Also, anyone have STIX working on an HP 9000?"

Though I don't know the details of your situation, and know nothing
about HP machines (at least, so far as I can help it :-) ), I will
warn you that both I and a friend at Cornell found STIX to be rather
fragile in the build phase.  If you think you're doing everything
right, and it still won't build, try throwing it all away and starting
from scratch --- I had to do that at one point.