Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!agate!eris.berkeley.edu!mwm
From: mwm@eris.berkeley.edu (Mike (I'll think of something yet) Meyer)
Newsgroups: comp.lang.misc
Subject: Re: Which language to teach first?
Message-ID: <1989Aug25.222558.23054@agate.uucp>
Date: 25 Aug 89 22:25:58 GMT
References: <1501@shuksan.UUCP> <2633@cbnewsc.ATT.COM> <1514@shuksan.UUCP>
Sender: usenet@agate.uucp (USENET Administrator;;;;ZU44)
Reply-To: mwm@eris.berkeley.edu (Mike (I'll think of something yet) Meyer)
Organization: Missionaria Phonibalonica
Lines: 59

In article <1514@shuksan.UUCP> scott@shuksan.UUCP (Scott Moody) writes:
<I agree that a lot of different languages is best but for short 
<quarter courses, learning a new language can be difficult.

In that case, the education of those taking the course has been sadly
neglected (assuming they don't have to pick up a completely new way of
looking at things, of course. This assumption will be unstated by
applicable for the rest of the article). Knowing what languages are
available for use on a system, what their strengths and weaknesses
are, how to use them effectively, and how to get them to cooperate in
a single program are important parts of being a programmer. Not as
important as algorithm design, but not far from it - because the tools
you have will affect the algorithms you choose.

<Remember
<that it usually takes at least 1 quarter to learn Pascal, and even then
<half the people are not experts yet (barely proficient).

But that includes trying to teach the system the language is being
used on, as well as some basic programming skills. After you've been
through that, you should be able to pick up almost any Pascal-like
langauge well enough to write code in with a week of classwork and the
reference manual. You won't be an expert, but you'll be able to get
the classwork done.  At least, that's how I had it done to me. The
intro course taught any of Pascal, Algol, Fortran or Cobol. A later
course did a quick overview of lots of languages, concentrating on how
they were different from the others. After that, instructers taught
courses in whatever language they felt like - and if it wasn't covered
in those two courses, they spent a week on it, and had you buy the
refernce manual. Some let you write in any language you wanted to that
met certain guidelines (e.g. - has to have structured programming
constructs). Some did both, teaching in a preferred language, but
letting you write what you wanted.

Those advocating a language because that's what industry needs are
making a fatal mistake - they're pushing for technological
obsolescense for the students. If you train somebody for the currently
"hot" language in industry, there's a good chance they'll have trouble
finding a job using it in ten years. Think about what was going on in
languages ten years ago: Pascal was the "hot" language, and employers
were looking for people to write the stuff. Was there an Ada compiler?
How about Modula II? Those seem to be where the Pascal people have
gone. C was being used a little, but there wasn't much demand for it
in industry. Now, people are moving to C++.

If your goal is to have the highest possible placement rate after
graduation, go ahead and teach a single language. While you're at it,
drop the liberal arts and mathematics requirements, so you can cut the
time to a degree by a couple of years. If your goal is to provide
people with training that will be usefull to them throughout a carreer
as a programmer, teach them multiple languages, and teach them how to
learn new languages.

	<mike
--
The sun is shining slowly.				Mike Meyer
The birds are flying so low.				mwm@berkeley.edu
Honey, you're my one and only.				ucbvax!mwm
So pay me what you owe me.				mwm@ucbjade.BITNET