Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!snorkelwacker!bloom-beacon!A.GP.CS.CMU.EDU!shivers
From: shivers@A.GP.CS.CMU.EDU (Olin Shivers)
Newsgroups: comp.lang.scheme
Subject: ML syntax
Message-ID: <9009200443.aa24008@mc.lcs.mit.edu>
Date: 20 Sep 90 08:40:53 GMT
References: <9009190401.aa01540@mc.lcs.mit.edu>
Sender: root@athena.mit.edu (Wizard A. Root)
Organization: The Internet
Lines: 18


    Date: 18 Sep 90 17:48:51 GMT
    From: Cliff Click <news@rice.edu>
    Subject: Virtues of ML syntax

    I've done some Scheme programming, darn little ML programming and scads
    of C/Fortran/Pascal/etc programming.  I find ML syntax to be incredibly 
    unreadable, while Scheme & C syntax isn't that bad.
One of the authors of *The Definition of Standard ML* is a friend of mine,
and I have heard him complain about getting tripped up by little oddities
in the syntax. If the syntax causes trouble for one of the guys that actually
defined the language, that's an indicator to me that maybe it could
have been better designed.

I like lisp syntax myself, because you can extend it with macros.
Notational engineering, while certainly abusable, is an important
tool in a software engineer's bag of tricks.
	-Olin