Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!swrinde!elroy.jpl.nasa.gov!jarthur!ucivax!orion.oac.uci.edu!ucsd!ucrmath!koufax!rhyde
From: rhyde@koufax.ucr.edu (randy hyde)
Newsgroups: comp.sys.apple2
Subject: Re: math background and HLL
Message-ID: <14522@ucrmath.ucr.edu>
Date: 17 May 91 16:58:08 GMT
References: <3581@kluge.fiu.edu>
Sender: news@ucrmath.ucr.edu
Reply-To: rhyde@koufax.ucr.edu (randy hyde)
Lines: 45

>> I would strongly recommend every CS major have a BS in math....

Why not chemistry, physics, or electrical engineering?

While mathematics is very important background material for computer
science, keep in mind that every math class a student takes is one less
computer science class they take.

Obviously the more education someone has, the better (all-around)
they're going to be.  But most students need something like 180
quarter-units to graduate.  If you make them take another 4 units of
math, that's 4 less units of something else they're going to take.  If
you've got free units laying around, making a student take more math is
a good idea (e.g., I'd like to see CS students here at UCR take
probability and combinatorics; they already take a minimum of three
years of math).  On the other hand, I have *personally* never used
differential equations in the past 15 years.  Oh, I know where it could
be applied in computer science, but I've never had the opportunity to
use it.  Combinatorics and probability, OTOH, I've used all over the place.

What it comes down to is this-- from your perspective you think students
need more math.  This is probably based on the type of work you're
currently doing.  A programmer who works under a chemist probably feels
CS students should have more chemistry.  Ditto (with physics) for a
student working under a physicist.  I personally work in fields dealing
with real time systems and controlling electronic devices, I feel CS
students should have more electronics.

The purpose of a typical university (at least here in Ca.) is not to act
as a technical school.  Rather, we attempt to provide students with a
wide basis of knowledge and convince them to start thinking things
through.  Someone who can think can *learn* other stuff as appropriate; 
someone who has been trained to do one particular thing is a technician,
not a scientist.

Of course you could argue that mathematics is the ultimate discipline
which teaches people to think, and you would be right; but experience
counts for a lot as well.  We can't simply teach students math and leave
it up to them to figure out the CS on their own.

BTW, most students leaving UCR can, at least, read and understand a
proof.  Between the discrete mathematics, data structures, analysis of
algorithms, and automata theory courses, they see a lot of proofs.  As
for creating their own proofs, we usually require that of grad students,
not the undergrads so much.