From: Matthew Ghio <ghio@myriad.alias.net>
Date: Fri, 8 Nov 1996 13:42:50 -0800 (PST)
To: cypherpunks@toad.com
Subject: Re: Why is cryptoanarchy irreversible?
Message-ID: <199611082141.QAA32413@myriad>
MIME-Version: 1.0
Content-Type: text/plain


Well, this is becoming a rather long thread, but I'll just make what I
think are the key points:

If, as is presumed in this discussion, society will become divided into
two somewhat distinct groups of people, the crypto-anonymous group and
the traceable-identifiable group, then the predictable outcome is that
members of the traceable-identifiable group will become increasingly
victimized by members of the crypto-anonymous group.  People who are
easily identifiable and tracked are easy prey.

For example, if you can see when John Doe is logged in at home or
logged in at his office, then you can burglarize his home while he is
at the office and vice versa.  If, as some government agencies would
like, all motor vehicle records were online, a criminal could go
'shopping' on the internet, pick out the make and model he wants,
look up the owner's address and go steal it.  There are lots of other
examples, such as seeing how much money is in various people's bank
accounts, then target those who have money, etc...
(Identifiable people are also likely to be targets of spam and other
annoyances, and I suspect the proliferation of spam is encourgaing
a great number of people to explore the possibilities for anonymous
posting.)

This situation is going to put an increasing amount of pressure on
those in the traceable-identifiable group to change their ways.  They
will have two choices: either they start using crypto, remailers, etc
to protect their privacy and protect themselves from criminals, or
they can demand more identification and restrict their interactions
with others such that they only associate with suitably identified
persons.

The net result of this is that it's easy (and economically desirable)
for people in the 'traceable-identifiable' group to join the 
'crypto-anonymous' group, but vary hard for anonymous personas to
interact with the identified group.  As such, the number of people
demanding identification of their correspondents will decline.
(Another way to look at it, is that crypto-anonymity is a high-entropy,
and hence stable, situation, whereas tracable-identifiability is a 
low-entropy and metastable state, which spontaneously degrades into
the former in the absense of a constraining force.)

This is why cryptoanarchy is inevitable.