home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group93b.txt
/
000009_icon-group-sender _Fri Apr 23 11:36:35 1993.msg
< prev
next >
Wrap
Internet Message Format
|
1993-06-16
|
4KB
Received: from owl.CS.Arizona.EDU by cheltenham.CS.Arizona.EDU; Fri, 23 Apr 1993 08:38:54 MST
Received: by owl.cs.arizona.edu; Fri, 23 Apr 1993 08:38:51 MST
Date: Fri, 23 Apr 93 11:36:35 +0200
From: karczma@univ-caen.fr (Jerzy Karczmarczuk)
Message-Id: <9304230936.AA23022@univ-caen.fr>
To: icon-group@cs.arizona.edu
Subject: The popularity of Icon Never Ending Stooooory...
Status: R
Errors-To: icon-group-errors@cs.arizona.edu
Dear Icon Masters,
To resume, Icon has acquired less popularity that it deserves because:
A. It is free, or even worse (as you should buy the book), it is cheap.
So, the people do not appreciate.
B. It is not being taught early enough. Not used in introductory programming
courses.
C. There are plenty of other languages that are already popular, and people
are resistant to changes.
Perhaps you wouldn't mind taking into account some other factors.
1. There are dozens of other languages *which are NOT popular* as well.
Perhaps we shouldn't be so catholic? We are the elite and not the
popular prophets, or are we? How many new religious sects, sorry, I
wanted to say *programming languages* are born every day? YES, I know
that we are the best, but...
2. In the computing science community the tendencies fluctuate, from time
to time we want something really powerful and universal, as Algol68
(yes, no need to check my address, I am an European), or (pardonnez le
gros mot) PL/I, but more often people like the simplicity. Look at the
successes of LISP despite its syntactic austerity and quite primitive
semantics.
Why almost all physicists still use FORTRAN? You might be tempted by
the obvious answer: the are forced to reuse the terabytes of existing
library code, so the disease perpetuates itself. But this answer is
incomplete, I assure you, I was a physicist myself before moving
(evolving? degenerating?) into comp. sci.: they are reluctant to learn
something too complex!
While using FORTRAN you are not tempted by some new, elegant ways of
coding your algorithms. You are so disgusted by the programming that
you don't risk paying too much attention to it, and you feel well
that you are still a physicist inside your heart.
So, perhaps Icon is too nice and powerful? It is a drug to be avoided?
3. Something serious this time. I think that we need more APPLICATIONS.
The Newsletter article about stems in Finnish language was very good.
I transmitted immediately the information about it to our linguists,
and I suggested that I could give a seminar about the possible uses of
Icon in the domain of automatic language processing. But, somebody asked
me about some other practical examples, and when I dumped on paper
some elegant modules from the Icon library, the reaction was:
"I wanted something PRACTICAL!"
Funny, but there are people who do not consider a parser generator to
be something practical. It seems that the number of existing tool building
facilities is so large that there is a tendency to judge the quality of
the language by the number of closed, finished and polished products.
Not kross.icn, but theUltimateScrabblePlayer.icn.
Jurek Karczmarczuk Lab. d'Informatique
Universite de Caen
Caen, Normandy, France.
karczma@univ-caen.fr