1994-06-10 - Re: Remailer Chaining Security?

Header Data

From: ghio@cmu.edu (Matthew Ghio)
To: N/A
Message Hash: 3a9939370b6e180f0eab0522d464a524ec91b45a6b9d6814193ae141160bb59e
Message ID: <9406100847.AA20331@toad.com>
Reply To: N/A
UTC Datetime: 1994-06-10 08:49:48 UTC
Raw Date: Fri, 10 Jun 94 01:49:48 PDT

Raw message

From: ghio@cmu.edu (Matthew Ghio)
Date: Fri, 10 Jun 94 01:49:48 PDT
Subject: Re: Remailer Chaining Security?
Message-ID: <9406100847.AA20331@toad.com>
MIME-Version: 1.0
Content-Type: text/plain


nobody@shell.portal.com wrote >>>

 | What level of security is achieved by the practice of "chaining"  |
 | messages through various "anonymous remailers"?  Do these servers |
 | not keeps logs by which a post could be traced back through the   |
 | chain to the source?  If there is a REAL level of security        |
 | afforded, then the question arises "how much is enough"?          |


Yes.  If you use only one remailer, and that remailer keeps logs, then it
makes it trivial for the remailer operator to find out who sent the message.
If you chain remailers, then it makes it more difficult to trace it back,
but the first remailer can see the source and destination, which means someone
could find out if they looked.  However, if you chain and use PGP, it requires
the cooperation of all remailer operators involved.  If someone really tried to
get a lot of remailer operators to cooperate in tracing a message, someone's
going to make a stink about it.

Now there is the traffic analysis problem, and the time-correlation problem.
This can be solved by using the slower (UUCP) remailers.

"How much is enough" is something you must decide for yourself.


BTW: Which remailers keep logs?  Mine does...  I think Chael Hall and Hal
Finney do.  And catalyst doesn't (or didn't awhile ago).  What about the rest?





Thread