1998-01-13 - Re: steganography and delayed release of keys (Re: EternityServices)

Header Data

From: Adam Back <aba@dcs.ex.ac.uk>
To: tcmay@got.net
Message Hash: 49b18bc6ffe489fd448aa6ac6c5cdff18c728a56e74be562371d4b84c750a63c
Message ID: <199801132241.WAA00454@server.eternity.org>
Reply To: <v03102803b0e08ed02241@[207.167.93.63]>
UTC Datetime: 1998-01-13 23:31:00 UTC
Raw Date: Wed, 14 Jan 1998 07:31:00 +0800

Raw message

From: Adam Back <aba@dcs.ex.ac.uk>
Date: Wed, 14 Jan 1998 07:31:00 +0800
To: tcmay@got.net
Subject: Re: steganography and delayed release of keys (Re: EternityServices)
In-Reply-To: <v03102803b0e08ed02241@[207.167.93.63]>
Message-ID: <199801132241.WAA00454@server.eternity.org>
MIME-Version: 1.0
Content-Type: text/plain




Tim May <tcmay@got.net>
> At 6:46 PM -0800 1/12/98, Adam Back wrote:
> >One tactic which could protect a USENET newsgroup operator from child
> >porn prosecutions is if he had no practical way to recognize such
> >materials until after it was distributed to down stream sites.
> 
> Who are these "USENET newsgroup operators," anyway? (A few newsgroups are
> moderated, by individuals or committees, but the vast majority are not.)

I meant USENET site operators (so that would be administrators plus
the people who decide policy).  

Even moderation need not be a fatal problem with steganography, if we
post the key to decrypt the stego encoded message after the article
has had a chance to be distributed.  (And presuming that our mimic
function is good enough to fool the moderator.)

For unmoderated groups, we have a much easier task: that of avoiding
undue attention or cancellations until the message has propagated.

> Newsgroups get removed from university and corporate newsfeeds, or by
> nations, and Adam's ruse would not stop them from continuing to do so.

In the extreme we can try to use mimic functions to post textual
information seemingly on charter for whatever groups are remaining at
a given time.  Binary data is obviously much easier to hide data in,
but is in any case a natural target for omitting from feeds due to
volume.

Unfortunately good quality textual steganography encodings are I think a
hard problem for reasonable data rates.  One advantage in our favour is the
massively noisy and incoherent garbage which forms the majority of USENET
traffic.  Plausibly mimicing an alt.2600 or warez d00d message, or a
`cascade' seems like an easier target.

Adam






Thread