Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site eddie.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!allegra!eddie!gds
From: gds@eddie.UUCP (Greg Skinner)
Newsgroups: net.cse,net.singles
Subject: Re: Value of Computer Science degree
Message-ID: <214@eddie.UUCP>
Date: Sun, 16-Feb-86 17:03:11 EST
Article-I.D.: eddie.214
Posted: Sun Feb 16 17:03:11 1986
Date-Received: Wed, 19-Feb-86 01:05:56 EST
References: <4514@kestrel.ARPA> <3407@nsc.UUCP> <4588@kestrel.ARPA> <256@hropus.UUCP> <394@ccivax.UUCP>
Followup-To: net.cse
Organization: MIT Lusers and Hosers Inc., Cambridge, Ma.
Lines: 50
Keywords: Computer Science, Bullshit about degrees
Xref: watmath net.cse:600 net.singles:10322

> From: rb@ccivax.UUCP (rex ballard)
> An MIT (BSEE) graduate once told me that some day, a program will be written
> that solves all problems.  He was actually serious!

Apparently, this was one of those MIT Course 6 people that "can't
program themselves out of a wet bag", because if they had learned
anything in their theoretical computer courses, they would have known
that such a program is mathematically impossible.  However, I could
excuse an MIT BSEE of such a statement, since they don't take all the CS
classes, but I would be surprised to hear a CS major say that.

> EE programs are really 'practical math' degrees.  A lot of time is
> spent doing proofs of things like "DeMavre's Theorem".

DeMoivre's Theorem, acutally.  I think it is debatable whether or not
someone should theoretically understand a concept and be able to apply
it in practical situations.  Most of the MIT undergraduate curriculum
(not just Course 6, but the entire engineering school) is based upon
teaching the theory of the subject and applying the theory in some
practical situations.  Of course all situations cannot be covered in the
classroom (like porting vax code to 68000's, since typically you do all
your assignments on a single machine).  However, there are programs
designed to give undergraduates exposure to solving practical problems
in both research and production environments.  The Undergraduate
Research Opportunities Program gives undergraduates the chance to do
research work in the lab groups of certain professors.  Many companies
have agreements with MIT to allow MIT students to do cooperative work
and research at the company, as part of a combined 5-year BS/MS program.

> A good engineer can mathmatically prove that 1+1=3.

A good mathematician can tell you that such a proof requires division by
zero, or some such incorrect assumption.

I'm not going to go into a long-winded flame about whether it is better
to have a degree or not to have one, or if people without degrees are
better, or what is important in getting a product out.  As the original
poster said, different companies need different individuals to do
different things.  I agree that programs that are written should be
fast, portable, etc. but I won't deny the importance of understanding
the principles behind sorting, code optimization, etc.

[Sorry for cross-posting to net.singles, but I wanted the people there
to see it too.]
-- 
It's like a jungle sometimes, it makes me wonder how I keep from goin' under.

Greg Skinner (gregbo)
{decvax!genrad, allegra, gatech, ihnp4}!mit-eddie!gds
gds@eddie.mit.edu