From: cjl <cjl@welchlink.welch.jhu.edu>
Date: Wed, 3 Aug 94 14:53:10 PDT
To: Jidan <yusuf921@raven.csrv.uidaho.edu>
Subject: Re: Remailer traffic analysis foiling
In-Reply-To: <Pine.3.87.9408031432.A17383-0100000@raven.csrv.uidaho.edu>
Message-ID: <Pine.3.89.9408031733.A14476-0100000@welchlink.welch.jhu.edu>
MIME-Version: 1.0
Content-Type: text/plain



On Wed, 3 Aug 1994, Jidan wrote:

> 
> for total anon post/mail
> How workable is setting up remailers with psudo-cooperation
> so that when it recieves an anon mail it waits 20 or so min
> and then randomly sends copies of it to 5 other remailers of which
> the original reciever randomly decides which 1 of the 6 will post 
> and the rest simply discard. 
>   a 5 fold increase in traffic will make it harder to analize if
> 80% is just noise

I think that sending many copies of the same message sounds like a good
way of making sure that it ends up being monitored by some 
alleged surveillance net.  Sending dummy messages is another matter. 

A fivefold increase in traffic may or may not have an impact on analysis, 
depending on your assumptions about the adversary's capabilities.  Anyway,
you still have a message of fixed size going in one end, coming out the 
other, and landing in someone's mailbox.  The superfluous messages may in 
fact be easy to identify if they are addressed to bit.bucket@dev.null.  

C. J. Leonard                     (    /      "DNA is groovy"
                                   \ /                - Watson & Crick
<cjl@welchlink.welch.jhu.edu>      / \     <--  major groove
                                  (    \
Finger for public key               \   )
Strong-arm for secret key             /    <--  minor groove
Thumb-screws for pass-phrase        /   )