Path: utzoo!mnetor!uunet!husc6!think!bloom-beacon!gatech!mcnc!ecsvax!hes
From: hes@ecsvax.UUCP (Henry Schaffer)
Newsgroups: comp.edu
Subject: Re: What is CS? (Was re First languages)
Message-ID: <4871@ecsvax.UUCP>
Date: 5 Apr 88 03:02:21 GMT
References: <3684@medusa.cs.purdue.edu> <568@abcom.ATT.COM> <612@imagine.PAWL.RPI.EDU>
Organization: NC State Univ.
Lines: 52

In article <612@imagine.PAWL.RPI.EDU>, jefu@pawl18.pawl.rpi.edu (Jeffrey Putnam) writes:
> In article <4859@ecsvax.UUCP> hes@ecsvax.UUCP (Henry Schaffer) writes:
> >In article <607@imagine.PAWL.RPI.EDU>, jefu@pawl23.pawl.rpi.edu (Jeffrey Putnam) writes:
> >> ...
 
> >  Does teaching students to use a spreadsheet or a math manipulator
> >(e.g. TK-solver or MathCad, ...) meet your requirement for what 
> >teaching programming is supposed to do?
 
> I dont think so - though i have certainly considered it.  I would
> not favor spreadsheets or math programs for a couple reasons.  First,
> these programs are in fairly general use and it is likely that people
> will use them at some point naturally, and i believe it is a good idea
> to expose them to something that they might not normally see.  But 
> more fundamentally, i would prefer a "real" programming language as 
> it is an opportunity not just to teach programming, but to teach something
> about the way computers work, the way programmers work, the way
> algorithms are described and implemented, and how they can use this
> stuff to actually build programs that will solve problems they might
> encounter.  
 
  We might get into some controversy over what is a "real" programming
language -  Is an object oriented programming language a "real" one -
how about hypertalk - or some of the 4GL database languages?  While
I have a gut feeling that only the procedural languates are "real"
(they certainly do relate to the way computer work - they kind of
reflect it) but wonder if we shouldn't raise above that. 1/2 :-)

  I think that the criterion "how they can use this stuff to actually
build programs that will solve problems they might encounter" is a
very strong justification for using the spreadsheet as the specimen
language!

> I was thinking of either a lisp family language (to minimize syntax
> woes) or something like ICON but with graphics extensions.  I would
> focus on graphics (given an appropriate system) or string processing -
> primarily non-mathematical stuff to ease math anxiety.  Graphics, 
> even given math anxiety, would be nice because you can _see_ what
> is going on and there are some interesting things to do that are
> quite accessible.  String processing is of fairly general interest
> as it doesnt involve math and there are some fun problems that
> are also accessible.
 
  I agree that non-mathematical applications are a necessity.  Why
burden the student with an irrelevant subject area if you are supposed
to be teaching about programming.  (That reminds me about how I had to
learn about electrical circuits in order to pass calculus - but that
is another story.

> jeff putnam  

--henry schaffer  n c state univ