Newsgroups: comp.lang.scheme
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!yeshua.marcam.com!usc!elroy.jpl.nasa.gov!decwrl!netcomsv!netcom.com!netcom15!vanmeule
From: vanmeule@netcom15.netcom.com (Andre van Meulebrouck)
Subject: Re: parsing lambda calculus
In-Reply-To: gregs@delphi.ccs.neu.edu's message of 10 Sep 1994 13:37:02 GMT
Message-ID: <VANMEULE.94Sep11134228@netcom15.netcom.com>
Sender: vanmeule@netcom.com (Andre van Meulebrouck)
Organization: NETCOM On-line services
References: <VANMEULE.94Sep10053951@netcom7.netcom.com>
	<GREGS.94Sep10093703@delphi.ccs.neu.edu>
Date: Sun, 11 Sep 1994 20:42:28 GMT
Lines: 25

In article <GREGS.94Sep10093703@delphi.ccs.neu.edu>
gregs@delphi.ccs.neu.edu (Gregory T Sullivan) writes:

[. . .]

Thanks for the response--much appreciated.

>2 notes: (1) the convention is that the body of a lambda extends as far as
>possible, so, in your example, any reduction will take place within the
>outermost Lf.  

Hmm.  Well, I got the same result you got, but I had never heard of
the "convention" you mention. The word convention makes me nervous.
I'd like to think things are more rigorous than to be based on a
convention.  I'd like to think the parsing of lambda calculus is a
well defined art with parsing rules just like Boolean algebra has
(i.e. AND binds tighter than OR).

Maybe it's just a matter of how you wish to think about things, but I
think using the rule: "application binds tighter than abstraction"
[Gunter 92: Semantics of Programming Languages] subsumes the
convention you cited.


-- 
for personal data:  finger vanmeule@netcom.com
