Path: utzoo!attcan!uunet!vtserf!cohill
From: cohill@vtserf.cc.vt.edu (Andrew M. Cohill)
Newsgroups: comp.lang.smalltalk
Subject: Re: PPS overpriced = low marketshare
Message-ID: <343@vtserf.cc.vt.edu>
Date: 3 Oct 90 13:59:46 GMT
References: <1990Sep28.171937@Atherton.COM> <51869@brunix.UUCP> <2734@dataio.Data-IO.COM>
Organization: Virginia Tech, Blacksburg, VA
Lines: 26

Smalltalk can be difficult in a multi-programmer environment.  We have
been using Digitalk ST/V on pcs and Macs here, and have found that we
have to be *very* careful about not stomping on each other's work.

The strategy we developed works fairly well:  as project leader, my
machine is designated as the master source.  We try to keep work
partitioned, so that only one person works on a class or subset of class
methods at a time.  Then once every two or three days, as needed,
everyone brings their stuff to my machine, files it in, and we verify
that in fact everything works together.  Then we file out the entire set
of application classes and methods (gotta get one of those application
browsers--this part is a nightmare), and everyone goes back and copies
the whole new set of stuff over their old stuff.

We have also found it to be pretty useful to program in pairs part of
the time.  It not only cuts down on overlapping code problems, but ST is
so compact and powerful that we seem to get more done at times
(together) than if the two of us work separately.  In my experience,
this never works well with conventional languages because you spend so
much time just typing, editing, debugging syntax errors, and similar
stuff, so one person ends up just sitting around.
-- 
|          ...we have to look for routes of power our teachers never       
|              imagined, or were encouraged to avoid.   T. Pynchon          
|Andy Cohill                    
|703/231-7855        cohill@vtserf.cc.vt.edu            VPI&SU