From mboxrd@z Thu Jan  1 00:00:00 1970
Received: (from weis@localhost) by pauillac.inria.fr (8.7.6/8.7.3) id VAA15004 for caml-red; Tue, 10 Oct 2000 21:24:03 +0200 (MET DST)
Received: from concorde.inria.fr (concorde.inria.fr [192.93.2.39]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id UAA14492 for <caml-list@pauillac.inria.fr>; Tue, 10 Oct 2000 20:55:50 +0200 (MET DST)
Received: from hebe.or.intel.com (hebe.or.intel.com [134.134.248.4])
	by concorde.inria.fr (8.10.0/8.10.0) with ESMTP id e9AItnj00224
	for <caml-list@inria.fr>; Tue, 10 Oct 2000 20:55:49 +0200 (MET DST)
Received: from ichips-ra.pdx.intel.com (ichips-ra-hme3.intel.com [10.7.7.35])
	by hebe.or.intel.com (8.9.1a+p1/8.9.1/d: relay.m4,v 1.31 2000/08/22 00:15:13 dmccart Exp $) with ESMTP id MAA05802
	for <caml-list@inria.fr>; Tue, 10 Oct 2000 12:08:07 -0700 (PDT)
Received: from dhpc0010.pdx.intel.com (dhpc0010.pdx.intel.com [10.7.21.33])
	by ichips-ra.pdx.intel.com (8.9.1a/8.9.1/d: internal.m4,v 1.2 1998/11/09 19:18:37 iwep Exp iwep $) with ESMTP id LAA28076;
	Tue, 10 Oct 2000 11:55:37 -0700 (PDT)
Received: from ichips.intel.com (johnh@localhost [127.0.0.1])
	by dhpc0010.pdx.intel.com (8.9.1a/8.9.1/d: client-ra.m4,v 1.1 1998/12/24 19:00:55 jamesw Exp jamesw $) with ESMTP id LAA25863;
	Tue, 10 Oct 2000 11:55:36 -0700 (PDT)
Message-Id: <200010101855.LAA25863@dhpc0010.pdx.intel.com>
To: caml-list@inria.fr
cc: John Harrison <johnh@ichips.intel.com>
Subject: Re: Undefined evaluation order
Date: Tue, 10 Oct 2000 11:55:34 -0700
From: John R Harrison <johnh@ichips.intel.com>
Sender: weis@pauillac.inria.fr


In the old ML compiler used in Cambridge LCF and HOL88 (which I guess was
a kind of ur-CAML), the order of evaluation was genuinely changeable, e.g:

   #let f x y = (x,y);;
   f = - : (* -> ** -> (* # **))

   #f (print_int 1;100) (print_int 2;200);;
   21(100, 200) : (int # int)

   #(I f) (print_int 1;100) (print_int 2;200);;
   12(100, 200) : (int # int)

We just ended up saying in HOL's DESCRIPTION manual:

   "Due to optimizations in the ML compiler, the order of evaluation
    may vary."

Gerard Huet might have been responsible for the optimizations that led to
this behaviour, so perhaps he knows more...

Cheers,

John.