home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.cs.arizona.edu
/
ftp.cs.arizona.edu.tar
/
ftp.cs.arizona.edu
/
icon
/
newsgrp
/
group97b.txt
/
000012_icon-group-sender _Thu Jul 3 15:19:30 1997.msg
< prev
next >
Wrap
Internet Message Format
|
2000-09-20
|
3KB
Received: from kingfisher.CS.Arizona.EDU by cheltenham.cs.arizona.edu; Tue, 8 Jul 1997 08:44:30 MST
Received: by kingfisher.CS.Arizona.EDU; (5.65v3.2/1.1.8.2/08Nov94-0446PM)
id AA24966; Tue, 8 Jul 1997 08:44:30 -0700
Date: Thu, 3 Jul 1997 15:19:30 -0500
Message-Id: <199707032019.PAA04723@segfault.cs.utsa.edu>
From: Clinton Jeffery <jeffery@segfault.cs.utsa.edu>
To: jthall@lucent.com
Cc: icon-group@cs.arizona.edu
Subject: Re: Scripting paper
Reply-To: jeffery@cs.utsa.edu
Errors-To: icon-group-errors@cs.arizona.edu
Status: RO
Joe Hall asks about a paper from www.sunlabs.com that
> talks about Perl, Tcl, Java and C as scripting languages.
> Where, if at all, do we think Icon fits into this picture?
Joe, I wonder how many different answers you will get; different people
use Icon for different purposes. I agree with a lot of Gordon Peterson's
comments. I'll post a description of using Icon for CGI scripting, along
with a pointer to my CGI library, in a separate message. Below I make a
few additional comments on where Icon fits in compared with scripting
languages.
I think the author of that paper is smart, but patently biased and focused
on pushing his own product. He is a commercial vendor of a scripting
language that he invented (Tcl) that is good at gluing and bad at data
structures and algorithms. From this he concludes that all scripting
languages are good at gluing and bad at data structures and algorithms.
So according to him Icon, being especially good at data structures and
algorithms, must not be a scripting language. But, it isn't a systems
programming language, either. What is it? It is a very high level general
purpose programming language, used occasionally for scripting, but more
often for writing applications.
Nevertheless, many or most of the author's arguments for "scripting
languages" are true of Icon, and the reason for this is because he
is citing general advantages of higher level languages, but claiming by
omission that scripting languages have a monopoly on these properties.
The paper's author knows about Icon (and other higher level languages
besides "scripting languages"), he just doesn't understand them, or else
discounts the value of language features that make it easier to write
complex algorithms and data structures, since that is not what his product
does. This is really unfortunate, since his language (Tcl) certainly has
other nice properties, and there is no reason it had to be so lacking in
support for general-purpose applications development.
Clint Jeffery, jeffery@cs.utsa.edu
Division of Computer Science, The University of Texas at San Antonio
Research http://www.cs.utsa.edu/research/plss.html