home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group93b.txt
/
000046_icon-group-sender _Tue Apr 27 08:45:32 1993.msg
< prev
next >
Wrap
Internet Message Format
|
1993-06-16
|
2KB
Received: by cheltenham.cs.arizona.edu; Sun, 2 May 1993 06:17:44 MST
Date: 27 Apr 93 08:45:32 GMT
From: mercury.hsi.com!mlfarm!cs.arizona.edu!icon-group@uunet.uu.net
Subject: Re: "Find that Niche!" contest?
Message-Id: <9304270845.AA09147@ruls41.LeidenUniv.nl>
Sender: icon-group-request@cs.arizona.edu
To: icon-group@cs.arizona.edu
Status: R
Errors-To: icon-group-errors@cs.arizona.edu
I don't know whether finding a 'hot' or 'cool' application
written *in* Icon would be so convincing. Of course,
any good application written in Icon *supports* the
language by means of illustration and a sort of
"see what you can do in this language"-proof.
However, I think the main power of Icon lies in the
possibility of reducing development time. I have
implemented several psycholinguistic models using
Icon in 2-3 hours, including a handy command-line
interface. What struck me as one of the advantages
of Icon in that context is that I could call procedures
by their string values, so after defining a new procedure
in the model, the user interface did not have to be
adapted; the new procedure could be called from the command
line right away!
This sort of 'trics' usually surpises programmers in a positive
way. I had a colleague (C programmer) turn whitish when he
saw me scramble (randomize) a file with:
procedure main(ar)
infile := open(ar[1]) | &input
lines := list()
every(put(lines,!infile))
every(?lines :=: !lines)
every(write(!lines))
close(infile)
end
Now this rather innocent kind of program is so easy to
write in Icon that you'd have to run it a thousand times
a day to justify writing it in C.
The unique thing of Icon (IMHO, of course) is that it
reduces development time so drastically that it actually
pays to use it, even when the performance is interpreter-like.
Maybe what would help to convince programmers if they are
potential Icon enjoyers is a document that does not only
highlight Icon's features, but also compares these features
with other programming languages like awk() en C(++). If
such a document is written well, it could be quite convincing.
Jan
----------------------------------------------------
Jan de Ruiter
Leiden University
Dept. of Information Science for the Social Sciences
The Netherlands
ruiter@ruls41.leidenuniv.nl