Newsgroups: comp.lang.scheme
Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!think.com!barmar
From: barmar@think.com (Barry Margolin)
Subject: Re: eqness of procedures
Message-ID: <1991May8.063427.25012@Think.COM>
Sender: news@Think.COM
Reply-To: barmar@think.com
Organization: Thinking Machines Corporation, Cambridge MA, USA
References: <9105080039.aa09228@mc.lcs.mit.edu>
Date: Wed, 8 May 91 06:34:27 GMT

In article <9105080039.aa09228@mc.lcs.mit.edu> shivers@bronto.soar.cs.cmu.EDU (Olin Shivers) writes:
>In the revised report that I have (3.95)
>	(eqv? (lambda (x) x) (lambda (x) x))
>is explicitly undefined -- implementations can return either true or false.
>
>However, this is never spelled out for
>	(eq? (lambda (x) x) (lambda (x) x))
>Has this been settled?

(eq? x y) implies (eqv? x y), which implies (not (eqv? x y)) implies (not
(eq? x y)).  So, if those two procedures can be non-eqv, then they can be
non-eq.

-- 
Barry Margolin, Thinking Machines Corp.

barmar@think.com
{uunet,harvard}!think!barmar
