Xref: utzoo comp.edu:2733 comp.software-eng:2578
Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cornell!uw-beaver!apollo!perry
From: perry@apollo.HP.COM (Jim Perry)
Newsgroups: comp.edu,comp.software-eng
Subject: Re: Re^2: CS education
Message-ID: <472bd511.20b6d@apollo.HP.COM>
Date: 1 Dec 89 20:26:00 GMT
References: <1989Nov21.172751.3078@world.std.com> <7183@hubcap.clemson.edu> <450@cherry5.UUCP>
Sender: root@apollo.HP.COM
Reply-To: perry@apollo.HP.COM (Jim Perry)
Organization: Hewlett-Packard Apollo Division - Chelmsford, MA
Lines: 56

In article <450@cherry5.UUCP> murphyn@cell.mot.COM (Neal P. Murphy) writes:
>billwolf%hazel.cs.clemson.edu@hubcap.clemson.edu (William Thomas Wolfe, 2847 ) writes:
>
>>...
>>   such application domains, and it would behoove us to present the
>>   student with software development problems which are relevant to
>>   the problem domain(s) in which the student intends to specialize.
>>...
>
>Why must students specialize? (Not why *would* they, but why *must* they?)

Bravo!  This discussion thread has tended to an extremely limited view
of the world, pitting computer science against mathematics on one side
and software engineering on the other.  (Assuming here that the goal
is to produce good software engineers; other parts of the thread have
argued different goals).  There's a lot more to day to day engineering
than is reflected in a discussion of whether or not the compiler/OS
course should be required or what to teach in its place.

In solid practical terms a greater emphasis on English would benefit
more engineers than any particular new SE course.  If more engineers
could produce decent English prose, productivity would be
significantly increased.  As it is, there's a considerable tendency 
among engineers and CS/SE students to downplay writing (and general
communication) skills.  This is extremely counterproductive.  

Much of the day to day work of software engineering involves
communication.  A design specification must, by its nature, be clear
and accurate.  The most painstakingly-designed, correct, efficient
program, with no comments, is worth only a small fraction of the same
program well documented.  The cost of a program is not in its
implementation but in its maintenance.  The world's best engineer
could go to make a change to the undocumented program and lose
hours or days analysing it, time that could have been saved with a
much smaller investment of time by the writer.  

I'm firmly of the opinion that there's much too much specialization in
American undergraduate education.  College is, for most of us, the
last chance to be exposed to a broad palette of educational
opportunity.  To waste that time in narrow pursuit of a particular
discipline, whether as a precursor to graduate study, or as
preparation for the job market, or even out of personal preference, is
a waste of an opportunity.  I consider this true at the level of
choosing CS over history; I would *certainly* counsel avoiding a
choice between CS and SE (while acknowledging that they are different,
or at least represent different, possibly overlapping, regions of a
spectrum).

By the way, if I were asked to pick the one course that had the most
direct influence on my day to day software engineering over the past
ten years, there would be no question, it was the compiler course.
The second was probably The History of the English Language.  No :-)
-
Jim Perry   perry@apollo.com    HP/Apollo, Chelmsford MA
This particularly rapid unintelligible patter 
isn't generally heard and if it is it doesn't matter.


Brought to you by Super Global Mega Corp .com