[Edu-sig] Computer science without all that "heavy math" stuff...?

Rob Andrews rob@jam.rr.com
Fri, 20 Jul 2001 08:46:38 -0500


# -----Original Message-----
# From: edu-sig-admin@python.org [mailto:edu-sig-admin@python.org]On
# Behalf Of Alan Gauld
# Sent: Friday, July 20, 2001 5:53 AM
# To: Danny Yoo; edu-sig@python.org
# Subject: Re: [Edu-sig] Computer science without all that "heavy math"
# stuff...?
#
#

<snip />

# The CS courses have
# split into those that have become so esoteric in their
# algenbraic/algorithmic/computational(in
# the math sense)
# focus that they are of little pracxtical relevance and
# those that have simply tried to produce cannon fodder
# for the programming/software develoment industry (the
# subject of my own rant).

In all honesty, all the Computer Science and Software Engineering people I
know call me to work on their hardware, home LANs, and a great many
technical problems. (Most of the time they spend a few days really
destroying things before they admit that they're in over their heads. They
scoff at my love of Python while expressing frank amazement at my ability to
handle the practical tasks that leave them baffled.

My best friend had to get through Calculus III, write a compiler, generate
working binary and hex code, and many other arcane facts while working on
his CS degree. And compared with the other career programmers I've met, he's
the most capable with actual computers and networks. But he's still
basically clueless in comparison with a host of barely-literate PC jocks I
have known. However, my IT skills (learned almost exclusively hands-on,
since my Psychology and Philosophy degree are only of peripheral value) seem
Liliputian in comparison with the advanced material they know (and actually
do use).

#
# The problem is that as industry builds ever computing systems of
# ever greater
# complexity there are
# no graduates available who can understand how or what
# its about. Industry in effect has to bear the cost of
# training which in other engineering disciplines is done at
# univversity level.
#
#
# The telecoms indiustry was in a similar position about
# 20-30 ytears ago with telcos providing huge amounts of
# basic theory simply because EE coursew didn't teach the
# networking/switching
# stuff needed. As the Telecomms
# sector exploded colleges reacted to that and there are
# now specialist telecomms degrees available in most
# places. The Computing industry is cryiong out for
# the same kind of change to happen.
#
# > It seems to ask "What good is education?"

Maybe colleges and universities can produce quality IT workers if they try.
I have no idea. When hiring new technicians, I can honestly say that I have
never met a trade-school IT graduate of any clear value as an employee. I
have had to sadly turn people away at the door because I do not have the
time to train someone with A+ certification on how to replace a video card
or locate updated drivers on the web.

In at least one major corporation with its head office in the Jackson,
Mississippi area, written policy dictates that new hires (for Software
Engineer positions) with CS degrees be hired at a lower pay scale than those
with SE degrees, and related salary caps apply. The official statement is
that this applies regardless of actual technical skill. Apparently, the
desire is not to have to train the CS grad in engineering concepts (like a
variant of my denial of the trade/vocational schools).

#
# I don't think so, I think it asks why is the education
# so impractical? When the subject name would suggest
# it should be one of the most practical of subjects
# to study! Nobody expects a course on prehistorc music
# to be practical but CS students might be expected to
# emerge with a degree which teaches them about
# computing in the modern context rather than a
# purely theoretical understanding of mathematical
# computation and irs application to machines.
#
# > If CP4E ever becomes widely implemented,
# > will educators also have to defend what they teach
# > on the basis of direct application?
#
# My understanding of CP4E was that its justification was precisely
# the programming
# was a practical skill and didn't neeed lotsof esoteric background
# math etc.
# Indeed my concern with CP4E is that it may not in fact be
# possible to really
# teach programming without at least basic math.
#

I am actually taking the opposite approach. Having not been exposed to much
math, I use Python and other programming languages to teach me math. It
provides some meaningful context, enabling me to see why mathematics is
important and can be useful. Every cheezy little script I can write (and the
ones that fail, as well) teaches me more about computing and how it works. I
actually find that some of the advanced theory I learned even in Psychology
and Philosophy becomes tangibly useful when I sit down to code. The obscure
theoretical material is important, even if not visibly so. And I mostly
program for simple pleasure.

# The real problem as I see it is that CS has not apparently
# grasped that computing
# as a subject is about much more
# than programming.
#
# >I don't think application training is what computer
# > science should primarily be about.
#

<snip />

#
# Education serves two purposes - to inform and to train.
# The IT industry needs the training aspect and isn't
# getting it. The academic community, if it wishes to
# be paid to inform, needs to provide the training too.
# In this it is singularly failing at present. If the
# current trend continues I can fortesee many large
# businesses withdrawing the current high level of
# funding/grants/sponsorship
# given to CS departments
# because they are simply not seeing the required
# return on investment!

Is there any real truth to what I've heard along the lines that "really
qualified people working in Software Engineering often hesitate to take a
massive pay cut to share critical skills in the University"?

-Rob

"I do remain confident in Linda. She'll make a fine labor secretary. From
what I've read in the press accounts, she's perfectly qualified." -- George
W. Bush
Useless Python: http://www.lowerstandard.com/python/