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 JAA22749 for caml-red; Sat, 7 Oct 2000 09:38:59 +0200 (MET DST)
Received: from nez-perce.inria.fr (nez-perce.inria.fr [192.93.2.78]) by pauillac.inria.fr (8.7.6/8.7.3) with ESMTP id NAA17107 for <caml-list@pauillac.inria.fr>; Fri, 6 Oct 2000 13:19:28 +0200 (MET DST)
Received: from mailserver.cli.di.unipi.it (crudelia.cli.di.unipi.it [131.114.11.37])
	by nez-perce.inria.fr (8.10.0/8.10.0) with ESMTP id e96BJRj22860
	for <caml-list@inria.fr>; Fri, 6 Oct 2000 13:19:28 +0200 (MET DST)
Organization:  Centro di Calcolo - Dipartimento di Informatica di Pisa - Italy
Received: from fire.cli.di.unipi.it (fire-ext.cli.di.unipi.it [131.114.11.52])
	by mailserver.cli.di.unipi.it (8.9.3/8.9.3) with SMTP id NAA30441
	for <caml-list@inria.fr>; Fri, 6 Oct 2000 13:19:22 +0200
Received: (qmail 18826 invoked by uid 7794); 6 Oct 2000 11:19:10 -0000
Received: from carlotta.cli.di.unipi.it(131.114.11.15)
	via SMTP by crudelia.cli.di.unipi.it, id smtpda18812; Fri Oct  6 13:18:32 2000
Received: from localhost (bernardp@localhost) by carlotta.cli.di.unipi.it (8.8.5/8.6.12) with SMTP id NAA01844; Fri, 6 Oct 2000 13:18:18 +0200 (MET DST)
X-Authentication-Warning: carlotta.cli.di.unipi.it: bernardp owned process doing -bs
Date: Fri, 6 Oct 2000 13:18:18 +0200 (MET DST)
From: Pierpaolo BERNARDI <bernardp@cli.di.unipi.it>
To: Ken Wakita <wakita@is.titech.ac.jp>
cc: caml-list@inria.fr
Subject: Re: Undefined evaluation order
In-Reply-To: <20001006110211J.wakita@is.titech.ac.jp>
Message-ID: <Pine.GSO.4.00.10010061259150.1439-100000@carlotta.cli.di.unipi.it>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: weis@pauillac.inria.fr


On Fri, 6 Oct 2000, Ken Wakita wrote:

> Scheme also has its evaluation orders of function parameters
> undefined.  I remember one of the reasons for this decision is to
> leave possibility for parallel evaluation of the arguments.

This is not correct.  In scheme, parallel evaluation of the arguments is
explicitly disallowed (unless the compiler can prove that evaluating them 
in parallel does not have any observable effect, of course).

The real reason of the unspecified order in Scheme, is that when the first 
reports on scheme were being drafted, there already were implementations
which evaluated arguments in different order.  And this has never been
fixed afterwards.

(People interested in the history of Scheme can check the archives of the
scheme authors mailing list at MIT).

Gxis,
  Pierpaolo Bernardi