Truth Machines and Open Networks

John K Clark (
Sun, 15 Mar 1998 08:25:11 -0800 (PST)

-----BEGIN PGP SIGNED MESSAGE----- (Damien R. Sullivan) On Sat, 14 Mar 1998 Wrote:

>John, you're being at least as dogmatic and simplistic as you accuse
>Brin of being.

Throughout historic times and until just a few years ago the secret keeper
and the secret finder were pretty evenly matched, that is no longer true.
Yes, the underlining reason for this is indeed simple and it's one of the few
things I will admit to being dogmatic about, the power of exponential growth.
In general, if I increase the complexity of my encryption algorithm so that
it takes me twice as long to encrypt, I have made it not twice, but billions
of times more difficult for an attacker to break. The government has access
to more powerful computers than I do, but not that much more powerful.
A child's 50$ video game has enough computing power to encode messages that
would stump the governments largest super computers for centuries.
For the first time in human history 2 people on opposite sides of the planet
can communicate instantly in total privacy, and soon they'll be able to make
enforceable contracts without anybody else knowing about it, and without one
even knowing the physical location of the other.

>Earlier you talked about splicing into the network "they can't be
>watching me all the time". But they can watch the network cables.

Watch all cables all the time, all of the billions of miles of them?

>For that matter, the cables themselves might be 'smart'. "Help! I'm
>broken! Come look at me!"

But I already know the cables have that property, and I know exactly how they
determine that they've been tampered with, and I know how they communicate
that signal to Big Brother, and I know how the signal is authenticated as
valid, and I've read all the internal memos that the designers of the cable
wrote when they were working on them and worrying about weakness'.

>the projected openness isn't based on voluntary transparency, like
>altruism; it's based on snooping technology being better than
>secrecy technology.

I know, that's why it won't work.

>How will you use this secret?

I'd probably start by adding 5 or 6 zeros to the balance of my bank account,
then I'll make people I don't like look like they've committed some very
serious crimes, then I'll give myself a new identity with full documentation,
then reveal to the everybody what I've done (except for the name of my new
identity), destroy the world's confidence in the network and watch the entire
rotten system collapse.

>a world of total encryption might well be vulnerable to a new >algorithm or a quantum computer.

If a practical Quantum Computer is ever built the world will change so
radically that the fact that your encrypted mail can now be read will be the
last thing on your mind.

>And then there's the TEMPEST arms race.

I don't worry about Tempest because unlike wiretapping it's expensive, easy
to jam, must be done at close range and is not well suited for mass
surveillance. All I need to do is ask when the Tempest truck will be in my
driveway, or just look out the window.

John K Clark

Version: 2.6.i