From: dat@spock.ebt.com (David Taffs)
Date: Wed, 4 May 94 09:43:03 PDT
To: Carl_Ellison@vos.stratus.com
Subject: Re: TLAs, etc.
In-Reply-To: <199405041527.LAA03247@transfer.stratus.com>
Message-ID: <9405041640.AA06509@helpmann.ebt.com>
MIME-Version: 1.0
Content-Type: text/plain



   From: Carl_Ellison@vos.stratus.com

   > 
   > Jim Miller says: 
   > > 
   > > My hypothesis:  The TLAs could shut down the cypherpunks mailing list  
   > > (as it now exists) by dragging all the U.S. list members into court.   
   > > The TLAs would probably lose the case, but they would still do a lot  
   > > of damage to the lives of the U.S. list members.
   > > 
   > 

   ...

   Actually, we might try inviting such prosecution -- e.g., with each of us
   posting source code for some algorithm to the list.  This is so clearly
   publication (ala newsletters on paper) that the case would never hold but
   it wouldn't hurt to have a court rule.

It might be interesting for a group to publish crypto code (or other
potentially illegal bit strings) using something like DC-NET, where
each person contributes to each bit of information.

For example, maybe 100 people publish random bit strings, and when
XOR'ed all together you get some bitstring which might be illegal to
export, such as crypto source code.

They couldn't possibly prosecute any subset of the 100 people, because it might
be the case that the 100'th person is the one who XOR'ed all the other strings
with the source code and published that.

Thus, the only possibility would be to prosecute all 100 people at
once, and each could point the finger at any one of the other 99. It
is hard to believe that a jury would convict under these
circumstances, at least without more evidence of an actual
conspiracy. To help the situation, each of the 100 could publish
another bit string, which when XOR'ed to the first, produced some nice
GIF, which of course might have been their intent in the first
place. People could publish both halves in either order, marked A or
B, and so it could appear to be pure happenstance :-) that all 100 B
halves, when XOR'ed together, produce compilable source code.

At any rate, tracing to a particular person would be impossible, and a
large subset of the group could actually be completely unaware of the
final product. Any one of the 100, if aware ahead of time of what the
other 99 would publish (or aware of what the XOR of the 99 would be),
could slip in the real source code in the middle of the message stream.

But, in the famous words attributed by the late RMN to himself, "but
it would be wrong"...