GAMES: The "Crypto Anarchy Game"

[tcmay@netcom.com (Timothy C. May)]

Seth Morris has written extensively about his ideas for crypto games
that would teach the essence of crypto and possibly get into more
advanced areas, such as digital money, DC-Nets, information markets,
various kinds of attacks, and so on. All the usual stuff.

Though I've already written a lot today, I feel compelled to comment.

At the very first Cypherpunks meeting, in September 1992, about 20 of
us played the "Crypto Anarchy Game" for most of the afternoon. The
goal was explicity the same as Seth Morris is discussing: to make
concrete the various strange ideas associated with the vision of
digital economies, anonymous transfers, reputations, and so on. The
reactions were pretty good.

We also played the game for a couple of hours at our second meeting,
in October 1992, with an even larger group--and many new faces. This
second playing was somewhat less successful, for reasons I will
speculate on below, and we've never really considered playing it a
third time. 

Why we haven't, and the "return on time invested" are important
issues. Frankly, it takes far too much time to prepare, and the
players are smart enough (they were bright adults, after
all....targeting children or novices might have a different payback,
but then they'll be almost completely lacking in the basics, which
slows things down even further).

I'll share some views on what we did, what we learned, and the value
of games/simulations in general.

Let me also note that this issue has--like so many things on this
list--come up a couple of times. Geoff Dale, for example, proposed a
Cypherpunks game/simulation area in a virtual world at the Illuminati
BBS. I have no idea how it's coming; haven't seen Geoff here in
months. Best wishes to Seth or anyone else who actually implements
such a system.

1. What we did in the "Crypto Anarchy Game." (sung to the tune of "The
Crying Game," and also dealing with spoofing and false appearances).

Eric Hughes and I prepare fake e-money (Monopoly money), envelopes,
etc. Participants played various roles, assigned randomly. Some were
drug dealers, some were CIA Counter Intelligence agents, looking for
moles and information brokers. Valuable information was also
distributed. 

Various publically-visible transaction regions existed (akin to the
anonymous pools we have now). Remailers were simulated by envelopes
within envelopes, with each remailer choosing his postage, latency,
etc.

(These are things we could simulate easily, but still don't have built
into actual remailers!)

The game went on for several hours. Often chaotic, with lots of
messages lost (humans are fallible and don't run complicated protocols
very well).

2. What we Learned.

- chaos and confusion, as noted above.

- computer support needed badly (but this is not an easy task, or a
very rewarding one....laptops? Newtons? answers are unclear)

- it was sort of fun, but the lessons got driven home fairly early
and, after that, not much new learning took place

- getting into even more sophisticated areas would have required even
more effort and computer support, for marginal learning

* My major conclusion: Few people will put the effort into playing
such a game, even with computer support. (and generating the computer
programs to support players would be a nontrivial task--partly because
the protocols are so fluid and ill-specified).

3. The Value of Games and Simulations in General

- makes the protocols more real

- sometimes it uncovers hidden assumptions or provokes new ways of
thinking

4. But is it worth it?

I don't think so. Thought experiments provide nearly the same
benefits, can be done with scattered groups, and require far less
suppport.

Reaching children and less computer-oriented folks will be tough. I
don't see that it will "sell" anyone on the value of crypto. Most
folks already understand locks and keys and similar things. Is crypto
all that different?

The really interesting stuff--digital money, DC-Nets, etc.--is too
abstract for most people, anyway.

This is all I'll say for now. Good luck to Seth or anyone else, but I
can't see many Cypherpunks lining up to build such a game. Those who
wish to should, ideally, live near each other and try their own
"manual" version of crypto games before planning an automated version.

I think you'll find that intelligent folks won't have much patience
and nonintelligent or "differently interested" folks will not want to
play.

And I have seen a cipher-oriented game at a local Macintosh software
store. I don't recall the title, but it involved solving a cipher to
advance to the next level. Not exactly the stuff we concentrate on.

--Tim May


-- 
..........................................................................
Timothy C. May         | Crypto Anarchy: encryption, digital money,  
tcmay@netcom.com       | anonymous networks, digital pseudonyms, zero
408-688-5409           | knowledge, reputations, information markets, 
W.A.S.T.E.: Aptos, CA  | black markets, collapse of governments.
Higher Power: 2^859433 | Public Key: PGP and MailSafe available.
"National borders are just speed bumps on the information superhighway."