NODE 254cabc5Mailing list software losing posts
Matthew J Ghio <mg5n+@andrew.cmu.edu>Sat, 26 Feb 94 17:22:59 PST
The list software seems to be having problems again. I sent the below
post to the list three times, and it never got sent out. But since
another post from me just showed up, maybe it's working now... (Let me
know if you got more than one copy, because I sure didn't.)
-----
To: cypherpunks@toad.com
Subject: Re: lists of U.S. cypherpunks and Tentacles.
CC: Matthew J Ghio <mg5n+@andrew.cmu.edu>
In-Reply-To: <9402260131.AA08270@toad.com>
References: <9402260131.AA08270@toad.com>
Date: Sat, 26 Feb 1994 11:34:54 -0500 (EST)
From: Matthew J Ghio <mg5n+@andrew.cmu.edu>
mg5n+eaibiubkxb58z84cy2iaf9r61u26ra5x26mc0h@andrew.cmu.edu wrote:
> Please add me to your list. As you can see, I clearly have
> a US-based email address. :)
hehe... This didn't come from my site tho. It looks like a clever port
25 hack. However, the return address is valid.
I suppose this points out another problem with the US export laws: How
can you know where a message came from? As the above example shows,
there is really no way to know where the person you're communicating
with is. In other words, you can't not break the US export laws. All
you have is someone's word that they are in the US or not, but you
really can't know for sure. Something to mention in your letters to
Rep. Cantwell.
> -Ibiu
Cute... using part of your encrypted address as a handle? I never
thought of doing that... hmm...
NODE 55c87c92Re: Mailing list software losing posts
rarachel@prism.poly.edu (Arsen Ray Arachelian)Sat, 26 Feb 94 22:58:15 PST
Why bother with something as obvious and complex as an encrypted
address with a + in the middle. If I were Joe Foreign_Guy I would
simply get an account somewhere in the USA, there are plenty of
public access unix systems that allow you a free month or so,
do the request for the crypto software, and immediatly put a
.forward file in my directory. I do this because I can no longer
support the high price of calling the USA, and thus want my mail
sent to a machine on the net that is sitting on my desk here in
sunny (insert_foreign_country).
This way, the author has not broken the law by sending the software
to anysite.com, and I haven't either because all I did was to tell
the unix box to forward my mail out of the country. Such a setting
isn't illegal, neither is sending crypto software via email to a
USA site.
Legally who is to blame? Neither "I" nor the sender broke the law
although the software has been sent. If I move from the USA to
another country and arrange a deal with my post office to send me
ALL my mail to wherever I am and pay them in advance for the
service plus agree to pay for whatever forwarding costs, who is to
be blamed if Joey_CryptoAuthor sends me a disk with a ton of
crypto software in an unlabled box, and the Post Office does not
check its contents, but exports it?
Neither I nor Joey_CryptoAuthor broke ITAR. Not really. Not
intentionally. But who gets blamed?
NODE eb2cd283on international transmissions
hughes@ah.com (Eric Hughes)Tue, 1 Mar 94 14:08:16 PST
The situation:
-- non-USA person retains a USA-based email service
-- this person forwards mail to non-USA machine
-- this person requests crypto software be sent to the USA email address
-- another person sends software to the USA address
-- the forwarding works and the software is shipped outside the USA
>This way, the author has not broken the law by sending the software
>to anysite.com,
This is correct, since the sender of the crypto was told that the
address was a US place. If, however, the sender of the crypto knows
that it will be automatically forwarded outside of the country, the
they become liable because they have prior knowledge of the
consequences of their actions.
>and I haven't either because all I did was to tell
>the unix box to forward my mail out of the country.
Such a direction is not improper _per se_, but the combination with a
request to have crypto software sent to that address means that the
requester has prior knowledge that the request will cause crypto
software to be exported across US boundaries. And that prior
knowledge creates liability.
>Neither I nor Joey_CryptoAuthor broke ITAR. Not really. Not
>intentionally.
Incorrect. The person who sets up forwarding with the intention of
moving crypto software automatically outside of the country is in
violation.
But since I am not a lawyer I feel compelled to point out that the
cost of extradition of random people to the USA for trial under USA
laws is expensive and difficult, the most recent high-profile example
being Noriega. In addition, detection of such an action will be
difficult at best, and near impossible to prove if encryption is used.
Proof that software was exported in encrypted mail would require at
least the following:
-- a copy of a particular piece of mail claimed to contain encryption
software
-- evidence that this particular piece of encrypted mail did in fact
contain encryption software
-- evidence that a particular piece of encrypted mail was sent outside
of the country at a particular time and between two given machines
So, someone has to supply the authorities with a copy of the mail,
with a decryption of the mail, and with mailer logs evidencing a
transmission across USA borders.
Mailer logs are typically purged after a week or two. So if the
intermediate machine has purged logs and the .forward file is gone,
there will be precious little direct evidence left of an actual
transmission.
If the encryption is addressed to only the receiver, and if the sender
did not keep a record of the session key, only the receiver can
provide the session key. The session key is necessary to show that a
given piece of encrypted mail is an encrypted copy of a particular
piece of software.
And unless the NSA or the intermediate machine or the sender provides
a copy of the particular piece of mail, there's no fact in evidence
that any software was actually sent.
Of course, if the sender is out to sting you and ther intermediary
provides logging information, one might get screwed. But then again,
all intermediaries would have to cooperate, were there more than one.
And finally, I have written so much only to point out that legality
and enforceability are two very different things.
Eric
NODE 6ed6a38eRe: on international transmissions
rarachel@prism.poly.edu (Arsen Ray Arachelian)Wed, 2 Mar 94 23:31:57 PST
<<And finally, I have written so much only to point out that legality
and enforceability are two very different things.>>
Another scenario. Suppose Person A has a unix account they don't
use very often. Say once every few months. Suppose person X
is a hacker and breaks into Person A's account. Person X then
sets up a small program that acts as a remailer, *BUT* in a very
specific way. This remailer looks for a particular message,
specifically the requested cryptographic software sent from
person B who is a crypto author. Person X is outside of the
USA and has grabbed many many accounts and has installed
similar programs on each machine, which strip out all headers
of forwarded mail, and encrypt/stego all traffic through
them.
Person X wants to get his hands on a new, but very strong
cypher or interesting program that he doesn't have. So
through his captured accounts whose profieles match person
A's, he manages to set up the transaction and receive the
software.
The remailer programs have a very specific code in them
that when triggered will remove all trace of their existence
within the captured accounts. (ie: the program rm's itself
and overwrites the original .login file to the one person X
has hacked.)
In other words, he gets his hands on the software, sends
the kill code to one mailer which sends it on to the next
and kills itself, all mailers die.
The resulting trail is a long, but quickly disappearing one.
Nevermind that the odds are that Person X willbe caught
before he manages to set up his "network" and all that.
Cliff Stoll is/was an exceptional sniffer with a lot of
time on his hands talking to deaf ears until he talked
through a megaphone and was taken seriously.
The law in theory has been broken. In practice, Person
C, F, and N, are pretty much out of luck, unless they
get some other breaks leading to person X. :-)
Now suppose that Person A = Person X. Would you say that
person A has an aliby that his account had been hacked in
and he wasn't aware of it until a few months later his
password mysteriously appeared in 2600 or Phrack or
elsewhere and a few people told him of the "breech"
of security? :-)
Okay, that's it, time for me to get some sleep. Too
many theoretical, useless schemes. A true sign of
sleep deprivation.