Return-Path: <J.Crowcroft@cs.ucl.ac.uk>
Received: from EDRC.CMU.EDU by B.GP.CS.CMU.EDU id ad21224;
          19 Sep 94 13:09:26 EDT
Received: from venera.isi.edu by EDRC.CMU.EDU id aa01093; 19 Sep 94 13:04:28 EDT
Received: from quark.isi.edu by venera.isi.edu (5.65c/5.61+local-18)
	id <AA15348>; Mon, 19 Sep 1994 09:37:34 -0700
Received: from bells.cs.ucl.ac.uk by quark.isi.edu (5.65c/5.61+local-16)
	id <AA21034>; Mon, 19 Sep 1994 09:37:27 -0700
Received: from rodent.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.07068-0@bells.cs.ucl.ac.uk>; Mon, 19 Sep 1994 17:35:38 +0100
To: David Cheriton <cheriton@DSG.Stanford.EDU>
Cc: end2end-interest@isi.edu
Subject: Re: multicast transport...
In-Reply-To: Your message of "Mon, 19 Sep 94 09:14:50 PDT." <199409191614.JAA05126@Pescadero.Stanford.EDU>
Date: Mon, 19 Sep 94 17:35:36 +0100
Message-Id: <1132.779992536@cs.ucl.ac.uk>
From: Jon Crowcroft <J.Crowcroft@cs.ucl.ac.uk>



 >of packets on that first address.  If they are missing a packet, they subscribe
 >to another address for the retransmission.  Then, the joining is proportional
 >to the number of retranmissions.  
 
David 

you have it right - your scheme is best when most packets get there -
my scheme is ok if most packets are lost:-)

 jon