From: "Robert A. Costner" <pooh@efga.org>
Date: Mon, 29 Sep 1997 07:55:46 +0800
To: cypherpunks@cyberpass.net
Subject: Re: Remailer Attack
In-Reply-To: <89bca20c4c4e7e17eddcd9e11f3b5f69@anon.efga.org>
Message-ID: <3.0.3.32.19970928194755.033d2350@mail.atl.bellsouth.net>
MIME-Version: 1.0
Content-Type: text/plain



At 05:10 PM 9/28/97 -0400, (Monty Cantsin) wrote:
>If the downtime statistic reflects connection failures it isn't that
>useful.  This sort of problem is relatively common on the Net and is
>handled automatically.  I strongly suspect it really does reflect
>disappeared messages.

A missing message may just be a delayed message.  Almost all remailers on
Raph's list are at a 99+% uptime rate.  This figure is probably computed
based on a ping to rlist and a reply through the reordering pool.  I would
suspect that Raph waits for the latency period of the remailer.  If the
ping does not return, then uptime is less than 100%.  My suspicion is these
pings are "missing" because they are near the end of the list of pings and
just stuck in the reordering pool, or stuck in the normal connection
failure process.  I've personally had messages in the reordering pool for
over six hours.

Since almost all of the remailers on Raph's list are at 99+% uptime rates,
then I would daresay no messages get lost.

Even if you wish to assume that remailers lose messages due to bugs in the
software, a payment system is not likely to fix that any more so than the
current group effort.  Public domain, freeware, and shareware software has
often had higher quality than commercial software.

  -- Robert Costner                  Phone: (770) 512-8746
     Electronic Frontiers Georgia    mailto:pooh@efga.org  
     http://www.efga.org/            run PGP 5.0 for my public key