From: Black Unicorn <unicorn@access.digex.net>
Date: Thu, 19 Jan 95 16:56:32 PST
To: "Paul J. Ste. Marie" <pstemari@erinet.com>
Subject: Re: Does encrypted equal safe?
In-Reply-To: <9501172355.AA12246@eri.erinet.com>
Message-ID: <Pine.SUN.3.91.950119195241.1505A-100000@access4.digex.net>
MIME-Version: 1.0
Content-Type: text/plain


On Tue, 17 Jan 1995, Paul J. Ste. Marie wrote:

> Date: Tue, 17 Jan 95 18:55:30 EST
> From: Paul J. Ste. Marie <pstemari@erinet.com>
> To: Eric Hughes <eric@remailer.net>, cypherpunks@toad.com
> Subject: Re: Does encrypted equal safe?
> 
> At 01:28 PM 1/17/95 -0800, Eric Hughes wrote:
> > ... Meaning is subjective.  If I see encrypted text, am I to be held
> >responsible for having seen through an encryption for which I hold not
> >the key?  Merely because someone knows a transformation into a
> >disapproved form does not mean that I do. ...
> 
> Which is exactly why the encrypt on receipt or decrypt on delivery ideas 
> won't work.  You have to be provably ignorant of the data.
> 

I must disagree.

This hinges on the REASON for encrypting the data.  In my model, data 
that arrives at the haven unencrypted is unwelcome, and is encrypted to 
be used as traffic "noise," not for security.  Any unencrypted data is 
undesireable, it opening the door to kiddieporn by mail tactics.  
(Government sends user A kiddie porn, then arrests user A for kiddie porn 
possession.)

An automatic encryption of all unencrypted data, the key to which is 
randomly generated and destroyed, allows the traffic to foil analysis, 
while prevents the operator from being subjected to plant frames.

Decrypt on arrival is hardly defenseable in this context of course.

> 
>     --Paul J. Ste. Marie
>       pstemari@well.sf.ca.us, pstemari@erinet.com
> 
> 

-uni- (Dark)

--
073BB885A786F666 nemo repente fuit turpissimus - potestas scientiae in usu est
6E6D4506F6EDBC17 quaere verum ad infinitum, loquitur sub rosa    -    wichtig!