Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site burl.UUCP
Path: utzoo!watmath!clyde!burl!rcj
From: rcj@burl.UUCP (Curtis Jackson)
Newsgroups: net.cse
Subject: various recently-discussed topics
Message-ID: <1093@burl.UUCP>
Date: Wed, 5-Mar-86 04:31:44 EST
Article-I.D.: burl.1093
Posted: Wed Mar  5 04:31:44 1986
Date-Received: Fri, 7-Mar-86 03:31:19 EST
References: <1092@burl.UUCP>
Reply-To: rcj@burl.UUCP (Curtis Jackson)
Distribution: net
Organization: AT&T Technologies, Burlington NC
Lines: 67
Summary: 

a) 4.0 GPAs -- I essentially agree with the contention that a 4.0 doesn't
   necessarily qualify someone to do a good job in the real world.  It can
   often mean someone who is too focused, too tunnel-vision about require-
   ments, and downright unimaginative.  However, I must also admit that we
   'stole' one of my best friends from AT&T in Denver to work on this project,
   and he (with his 4.0 BSCS and 3.9? MSCS) has been a boon to us -- besides,
   he's a great drinking buddy!  He is none of the things mentioned above.

b) BA vs. BS -- At the University of MS, a BACS was simply a BSCS that required
   less math and only 30 hours of CS courses instead of the 42 required for
   a BSCS.  The rest of the extra hours were largely electives.  Anyone who
   wants to try to tell me that a BACS *from Ole Miss* better prepared anyone
   for anything is welcome to try -- translation:  remember not to make
   sweeping generalizations about degree programs.

c) A lot of universities used to teach FORTRAN as their introductory course.
   Now, they usually teach PASCAL or some other structured language.  Progress,
   right?  Well, not necessarily.  Seems that, as the baby boomers left
   college, new sources of enrollment were needed.  The next preying ground
   of college recruiters?  Junior colleges.  And all these junior college
   students would be real bored having to take FORTRAN that they had already
   had in junior college -- and the courses at a lot of colleges are scheduled
   so that low demand courses are spaced to correspond with high-volume
   semesters.  That meant that if the junior college students came in and
   started in on PASCAL they would be one semester ahead of, and out of sync
   with, the bulk of the other CS students.  This was the lecture I received
   from my department head when asked why FORTRAN was suddenly no longer
   required.

d) I started with FORTRAN my first semester.  Then came summer, and a co-op
   job where I learned HPL, HP-Basic, and Tektronix Basic.  Then PASCAL,
   COBOL, and MACRO-11, among others.  I was ready for anything.  Then
   came my intro to C -- our Minicomputers I prof told us to write a simple
   sleep queue scheduler/manager in C, compile it into Unix assembler on
   our PDP-11/34, run it through an ed script he would supply to translate
   it into MACRO-11, download this to the PDP-11/03 running RSX-11, and have
   it run without any editing of anything after the original C program.
   Finally, someone got up enough guts to raise their hand and tell him that
   none of us knew C (it wasn't taught at the university).  He said, "Oh, yes,
   get this book [holding up a copy of K&R] from the bookstore, $12.  You
   have 2 weeks."  Now THAT is the way to teach a computer language!
   In case anyone is interested, the real trick was to initialize an array
   of ints with octal PDP-11/03 hardcode for an interrupt and trap handler
   and, after setting up your sleep queue, to 'goto' the name of the array
   from inside your idle loop.

e) On hackers vs. non-hackers -- I don't want any hackers working for/with
   me -- not true hackers.  I don't want the person who spent his/her time
   in school finding out how everything in the world works, I want the kind
   of person who was relatively in the mainstream but who strived (and
   hopefully succeeded) to keep two or three steps ahead of it.  The kind
   of person who, when the AI prof suggested that we now try to apply a
   heuristic to our bi-directional search, had already done so two weeks ago
   and was now three chapters ahead in the book working on some other aspect.
   The kind of person who said, "I don't like this parser generator we have
   to use for our compiler course; I'll write my own and see if I can get
   future classes to use it instead."  I realize that this last example in
   particular impinges on the realm of the hacker, but I want people around
   whose primary motivation is to improve the working environment; not to
   satisfy their own curiosity.

'Nuf said,
-- 

The MAD Programmer -- 919-228-3313 (Cornet 291)
alias: Curtis Jackson	...![ ihnp4 ulysses cbosgd mgnetp ]!burl!rcj
			...![ ihnp4 cbosgd akgua masscomp ]!clyde!rcj