Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!think.com!snorkelwacker.mit.edu!ai-lab!life!tmb
From: tmb@ai.mit.edu (Thomas M. Breuel)
Newsgroups: comp.lang.c++
Subject: Re: Implementing LISP in C++ (type discrimination)
Message-ID: <TMB.91Apr22163043@volterra.ai.mit.edu>
Date: 22 Apr 91 20:30:43 GMT
References: <1991Mar8.024331.14235@searchtech.com> <27D7F621.2F5@tct.uucp>
	<17238@cadillac.CAD.MCC.COM> <1991Mar12.221015.22144@aero.org>
	<14342@hacgate.UUCP> <1991Apr17.233653.25149@dsd.es.com>
	<DAVIS.91Apr22105508@barbes.ilog.fr>
Sender: news@ai.mit.edu
Organization: MIT Artificial Intelligence Lab
Lines: 18
In-reply-to: davis@barbes.ilog.fr's message of 22 Apr 91 08:55:07 GMT


   From: davis@barbes.ilog.fr (Harley Davis)
   In article <1991Apr17.233653.25149@dsd.es.com> bpendlet@bambam.dsd.es.com (Bob Pendleton) writes:
      It's called reference counting. It's been used since the dawn of time.
      Reference counting usually uses more total time than traditional
      garbage collection. The other cost of reference counting is the
      storage used by the count associated with each object. 

   One of the most important problems with reference counting is that it
   doesn't collect all the garbage.  In particular, circular data
   structures never get collected.  [...]

Reference counting is, however, ideal for garbage collecting large,
non-recursive objects like arrays of type real, complex, bit, etc.
Any other kind of garbage collection will have much higher overhead
in this case, and you can never build circular data structures out
of them.