Newsgroups: comp.lang.scheme
Path: cantaloupe.srv.cs.cmu.edu!das-news.harvard.edu!news2.near.net!MathWorks.Com!europa.eng.gtefsd.com!howland.reston.ans.net!usc!elroy.jpl.nasa.gov!decwrl!netcomsv!netcom.com!hbaker
From: hbaker@netcom.com (Henry G. Baker)
Subject: Re: Scheme _in_ Emacs?
Message-ID: <hbakerCvqAD8.D2F@netcom.com>
Organization: nil
References: <BLUME.94Sep5110232@beth.princeton.edu>> <HARRISON.94Sep5193413@sp10.csrd.uiuc.edu> <34gn3k$jh3@nkosi.well.com>
Date: Tue, 6 Sep 1994 21:58:19 GMT
Lines: 18

In article <34gn3k$jh3@nkosi.well.com> djello@well.sf.ca.us (Darius Bacon) writes:
>
>Rather to my surprise, it turns out that a dynamically-scoped language
>can have a properly tail-recursive implementation.  There are two parts
>to the problem: keeping the control stack from growing on a tail call
>(which is easy), and keeping the representation of the environment from
>growing without bound in an infinite chain of tail calls (which is more
>interesting).  Actually, keeping the environment size bounded is also
>easy if you don't care about efficiency; but we want something not more
>than a constant factor bigger and slower than a straightforward shallow-
>binding representation.
> ........

A number of these issues are covered in my paper:

"The Buried Binding and Dead Binding Problems of Lisp 1.5".  ACM Sigplan
Lisp Pointers V,2 (Apr/Jun 1992), 11-19.

