From: Adam Back <aba@dcs.ex.ac.uk>
Date: Wed, 14 Jan 1998 07:29:10 +0800
To: rdl@mit.edu
Subject: Re: (eternity) autonomous agents
In-Reply-To: <199801131244.HAA22857@the-great-machine.mit.edu>
Message-ID: <199801132138.VAA00407@server.eternity.org>
MIME-Version: 1.0
Content-Type: text/plain




Ryan Lackey <rdl@mit.edu> writes:
> You would be expected to comply with efforts to stamp out illegal
> activities.  Laws may be passed (e.g. CDA) which would require
> technical measures to prevent illegal activity even if you are a
> common carrier.

Governments are always trying to coerce people into complying with
technical measures designed by the likes of GCHQ, NSA, and various
government sell-outs.  Spammers have surived so far because spam is so
hard to stop; stopping SPAM is an inherently hard problem, mostly
because of open access SMTP relays which is basically a historic
accident, and also because spammers make use of accounts that it pays
them to treat as disposable.  (I like to think hashcash,
http://www.dcs.ex.ac.uk/~aba/hashcash/ is a viable solution for spam,
though it too must be widely deployed before it would be practical).

We can perhaps learn something from the services and environments
which allow spammers to flourish, we could emulate the lack of
authentication, and identification in SMTP protocols and
implementations, to design attractive new services which are hard to
police by design.

This is the advantage of the internet protocol designer.

Distributed web services as Ryan is prototyping is one attractive new
service.

The challenge is deployment.

Adam