Path: utzoo!attcan!uunet!ogicse!mintaka!snorkelwacker!usc!orion.oac.uci.edu!uci-ics!ucla-cs!gds@oahu.cs.ucla.edu
From: gds@oahu.cs.ucla.edu (Greg Skinner)
Newsgroups: comp.edu
Subject: Re: DiffEq - heave-ho (Was: What math would you require?)
Message-ID: <33153@shemp.CS.UCLA.EDU>
Date: 17 Mar 90 16:09:26 GMT
References: <8368@pt.cs.cmu.edu> <32769@shemp.CS.UCLA.EDU> <1652@cs.rit.edu> <70155@kean.ucs.mun.ca>
Sender: news@CS.UCLA.EDU
Reply-To: gds@cs.ucla.edu (Greg Skinner)
Organization: UCLA Computer Science Department
Lines: 19

In article <70155@kean.ucs.mun.ca> mquigley@kean.ucs.mun.ca (Martyn Quigley, E3044, x8656) writes:
>I second the motion, but why not scrap all the calculus?  Presumably
>we've all got Macsyma or Maple or some such.  I find it most 
>interesting that so many remarks on this thread favour the retention 
>of mathematics which is perfectly capable of being mechanised - by 
>computer!

Even if "we all had macsyma or maple" which we do not (at least *I*
don't, and I could sure use it), it is important to learn the
fundamentals of calculus because not all calculus problems are
amenable to symbolic computation (or numerical, if you have those
tools on hand).  In addition, the concepts covered in calculus are
necessary for the understanding of other types of mathematics
important to computer scientists.  (For example, expressing a
recurrence relation in terms of a function of the derivative of
another, to relate their behavior.  In case you're interested, this
came up in my analysis of algorithms class last fall.)

--gregbo