Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!brutus.cs.uiuc.edu!ux1.cso.uiuc.edu!lees.cogsci.uiuc.edu!resnick
From: resnick@lees.cogsci.uiuc.edu (Pete Resnick)
Newsgroups: comp.sys.mac.programmer
Subject: Re: The programming CULT (WOW!)
Message-ID: <1990Aug8.233604.26207@ux1.cso.uiuc.edu>
Date: 8 Aug 90 23:36:04 GMT
References: <1990Aug6.172135.27287@midway.uchicago.edu> <PETE.90Aug6142343@titan.rice.edu> <4153@rodan.acs.syr.edu>
Sender: usenet@ux1.cso.uiuc.edu (News)
Organization: University of Illinois at Urbana
Lines: 58

isr@rodan.acs.syr.edu ( ISR group account) writes:
>Re: carpenters, film ritics, Civil engineers, and filmamakers:
>One for the (possibly) CS-trained analyst/designer
             ^^^^^^^^^^^^^^^^^^^^^
[others deleted]
>The system designer/analyst type is like an architect.
>   They can perform all the basic tasks, allthough not each one
>   quite as well as the narrow specialist. They also must know
>   what is on the cutting edge of new techniques and materials,
>   while they have to understand the properties, usage, and
>   reasons behind these, they aren't interested in the details
>   (besides maybe a 1-time curiosity) of how it was proved that
>   concrete X actually does turn a nice blue when heated to the
>   right temparature
>-- 
>Mike Schechter, Computer Engineer,Institute Sensory Research, Syracuse Univ.
>InterNet: Mike_Schechter@isr.syr.edu isr@rodan.syr.edu Bitnet: SENSORY@SUNRISE

And so the thread goes on.....

I think Mike has brought us back to the original problem: the best
"architects" (read programmer analyst/designers) that I have encountered
are rarely if ever CS-trained, and those that have been trained that
way are rarely good programmers due to that training. Granted the
brick layers need not be CS-trained and the academics should be. The
contention of the side of the argument that I have been on throughout
this thread is that the architects needn't (maybe shouldn't) be CS-
trained either.

CS-training, at least in the big places that I have seen, is limited
to idealized engineering, where all we care about is how fast the
code runs and how nifty the mechanism is that it is using. People in
this part of the field still do not understand the importance of
portable, readable, pretty, or modifiable code. Inevitably, it is the
best brick layers who learn how to do this and then become architects,
or it is those of us who came in from the outside. CS-training is
great for theoretical problems in CS, but I would never want to get
the code for the implementation of their solutions from the CS people
themselves. (No flames.....there are lots of CS people who are great
architects and coders; I just don't think that they learned a lick
of it in CS classes.)

[Strange thing......CS people hate VAX/VMS, Macintosh, and generally
Pascal programming. ("Strong typing is for weak minds" and "Why should
I deal with System service calls when I can do it all myself in
assembler?" come to mind). Philosophers and people who have had to
fix other people's code know why!]

pr - BA in Biology, BA in philosphy, MA in The Program in Philosophy
     and Computer and Systems Sciences (PACSS), PhD student in Philosophy.
     (since everyone else is giving their credits......)

--
Pete Resnick             (...so what is a mojo, and why would one be rising?)
Graduate assistant - Philosophy Department, Gregory Hall, UIUC
System manager - Cognitive Science Group, Beckman Institute, UIUC
Internet/ARPAnet/EDUnet  : resnick@kant.cogsci.uiuc.edu
BITNET (if no other way) : FREE0285@UIUCVMD