Re: build_campus, raison d'etre?

[Seth Teller <seth@graphics.lcs.mit.edu>]

the raison is that will wanted to learn how to write a scripting
language & interpreter.  i attempted to dissuade him, in favor 
of csh, but he had his mind made up.

that particular design decision was not well-justified, in my view.
A, the project was/is not about scripting languages.  2, scripting
shouldn't have been on the critical path (i.e., there were
more important things that should have been done first, such
as debugging basic functionality).  and D, it was clear that there
would not be time to implement the language well.  thus its 
"crippled" status.

it's probably best to remove & replace the scripting component with
something more mainstream and well-specified.  (if you use perl [not
my preferred choice], please comment it heavily.)

	To: bmg@graphics.lcs.mit.edu
	Subject: build_campus, raison d'etre?
	Date: Tue, 28 Jan 2003 02:25:49 -0500
	From: Luka <luka@MIT.EDU>
	
	any idea, anyone, why it was considered appropriate to write the
	command interpreter build_campus (walkthru/mit/src/build_campus)
	to process a "run script" which is effectively just a crippled
	shell language (plus an uncertainty layer that comes from running
	it through an underspecified interpreter)?
	
	does it perhaps provide some internal parallelism functions, or
	was it meant to improve cross-platform behavior, or what?  i'm
	not seeing any good reason for it, but i could be overlooking
	something.  so help me out.
	
	this code appears to come from will leiserson (wml), 2002/06.
	
	-pl