As the token cryptoanarchist around here, I've been lurking way too long,
mostly because I'm working on this financial cryptography conference we're
doing in Anguilla next week. However, A lot of good stuff has gone by on all
of these groups, and I think it's time I put my oar in and earn my keep a
bit before I enter the maelstrom of next week's FC98 conference, and lose
my chance to say anything here until it's all over.
The first topic I'd like to talk about is something which is more general
than my ostensible commercial focus in these discussions, and, after I've
said my piece here, I'll go back to the commerce list and pay more attention
to that end of things. I have a little more to say there on what you can do
with the technology of money on public networks, though I'll drop a few
hints here to get people thinking about them.
My observation about networks in general is a rather obvious one when you
think about it: our social structures map to our communication structures.
As intuitive as it is to understand, this observation provides great insight
into where the technology of computer assisted communication will take us in
the years ahead.
Because of Moore's Law and its effect of collapsing the price of
semiconductors by half every 18 months, our telecommunication architectures
have changed from hierarchical networks, where it's cheaper to add lines
than it is to add expensive switching nodes, to geodesic networks, where it
is ever-exponentially cheaper to add microprocessor switches instead of now
relatively more expensive transmission lines.
This isn't new. In fact, it's outlined in Peter Huber's landmark 'The
Geodesic Network', written in 1986 as a report for Judge Harold Greene as
part of the Modified Final Judgement which broke up American Telephone and
Telegraph, and with it the US telephone monopoly. I believe the original
version is still available from the US Government Printing Office, and I
know that you can order a revised edition from Peter Huber's law firm in
Washington. Huber himself is now a famous technology analyst from the
Manhattan Institute and a Forbes columnist, among other things.
In 'The Geodesic Network', Huber observed that because the network was
becoming more and more geodesic, competition in telecommunications was
becoming much easier. That's because switching, a scarce thing which had
heretofore caused economies of scale and resultant 'natural' monopoly, was
becoming cheaper and cheaper to build, and thus causing *dis*economies of
scale in the telephone markets.
One can almost hear Huber doing a little heavy lifting from the Marines in
report's conclusion, which was, essentially, 'Deregulate 'em all, and let
God sort 'em out.' It's nice to see that we're finally getting to see
deregulation of the 'last mile' of the US telephone network 10 years after
As it is, it took *me* almost 10 years to realize something else about
geodesic networks. It's something which required me getting back on the
internet 4 years ago, after not being there since grad school, and
discovering that financial cryptography -- that is, the cryptographic
protocols for internet payment -- was much more important than the project
management software I had wanted to sell on the net at the time.
My realization was, if Moore's Law creates geodesic communications networks,
and our social structures -- our institutions, our businesses, our
governments -- all map to the way we communicate in large groups, then we
are in the process of creating a geodesic society. A society in which
communication between any two residents of that society, people, economic
entities, pieces of software, whatever, is geodesic: literally, the
straightest line across a sphere, rather than hierarchical, through a chain
of command, for instance.
This seems like a very simple truth these days. A 'motherhood', as people in
American business like to say. But, once you start thinking about the world
in the terms of geodesic networks versus hierarchical ones, the world
changes. A Buckminster Fuller version of satori, if you will, though I'm
sure Bucky didn't think of human society in geodesic terms, at least from
what I've read of his work. His 'World Game', for instance, is primarily
about the hierarchical centralization and redistribution of resources in an
industrial fashion. But, as it was, Bucky Fuller had discovered a geometric
archtype which was deeper than even his capacious understanding of its
implications had gotten him before.
So in light of this observation, for fun, let's look at human history in a
few paragraphs. :-).
Humans first lived in small groups on the African savanna. An artifact of
this life is the fact that most people can't have serious emotional
relationships with more than about 12 people, depending on how you define
serious. :-). Think of it as the carrying capacity of the human 'switch',
and things get interesting. These small groups communicated geodesically.
When you wanted to talk to someone, you went up and talked to them. Then we
developed agriculture and its resulting food surpluses, people tended to
congregate at the crossroads of trade routes, and that's where the first
cities began. Civilization means, literally, 'life in cities', remember?
Once we had large groups of people in a single place, we had lots of
information to pass around, but we also had expensive humans 'switching'
that information who were only able to trust about 12 people at any time.
So, we had to develop hierarchical 'networks', social organizations in other
words, to move that information around. Notice we finesse the whole trust
problem by using the entire hierarchy as one entity in everyone's
trusted-person list. That's why people die for king and country, for
instance, instead of just their family hunter-gatherer clan.
So, we can now see the ancient city-state as a hierarchy of power,
economics, whatever. We can also see ancient empires as a hierarchies of
city states, and so on. Notice that the size of any given hierarchy in
geographic terms is determined by the *speed* of communications it posesses.
Athenian triremes were very secure ways to move goods and information in a
relatively lawless Agean. Roman roads and galleys didn't just haul goods
quickly, they moved information as well. Staged Mongol riders could carry
messages across their own short-lived empire from a capital near China to
the gates of Warsaw in as little as 14 days. Napoleon invented his
10-mile-an-hour stagecoach and highway system for exactly the same reason,
and could almost legitimately call himself an emperor for the feat alone.
That brings us to the modern nation state, which, I claim, is entirely the
result of industrial communications technology. That is, you have
increasingly faster communications, from sailing ships to trains to
telegraphy and finally telephony, but you still have humans switching
information. That gives you larger and larger communication, and thus
social, hierarchies. Up until the automation of telephone switching --
paradoxically brought about a demand for universal service in exchange for
that ultimate industrial hierarchy, the US telephone monopoly -- things just
kept getting bigger and bigger. One could even see the increasing size of
government in this century as an 'antihierarchy' funded by the forcible
confiscation or political extortion of economic rents from the large
industrial hierarchies where industrial society's money was being made in
the first place.
For a tasty little digression, Marxism then can be seen as simple
anti-industrialism, and an intriguing validation of Bertrand Russell's
comments about the similarity of Marxism and the feudal aristocracy it hated
so much. Hegel can't come to Marx's rescue here at all, because, for all
it's anarchistic pretensions, Marxism can now be seen as merely
industrialism's hierarchical antithesis, and not something 'beyond
capitalism'. Besides, trading has been around since the savana itself. It's
hard to imagine something antithetical to trade -- and have the result be
human, anyway. :-).
Okay. Now let's look at the future, shall we? Oddly enough, the 'future'
starts with the grant of telephone monopoly to AT&T in the 1920's in
exchange for universal telephone service. When AT&T figured out that a
majority of people would have to be telephone operators for that to happen,
it started to automate switching, from electricomechanical, to electronic
(the transistor was invented at Ball Labs, remember), to, finally,
semiconducting microprocessors. Which, Huber noted, brought us Moore's Law,
and, finally, that mother of all geodesic networks, the internet.
So, seen this way, using the hierarchy-to-geodesy synthesis (speaking of
Hegel :-)), a lot of things jump out right at us. Let's look at financial
operations, for example.
One can see, for instance, that the thing we call disintermediation in the
capital markets is in fact a process leading to something I call
*micro*intermediation, where large human decision hierarchies, like the New
York Stock Exchange, or money center banks, are being outcompeted by large
integrated proprietary computer networks, like the NASDAQ interbrokerage
network, or Fidelity Investments here in Boston. Yet, these financial
versions of big dumb bulletin boards, which still need humans to operate
them on behalf of the customer, will themselves be replaced someday by
smaller, more specialized and automated entities operating in increasingly
smaller market niches, and, we aren't just talking about financial
'shovelware', with database-driven web forms, either.
Someday, for instance, a couple of portfolio managers from Fidelity could
strike out on their own peculiar investment specialty, and set up a web
server to handle their investor relations, but in a way that financial
operations people thought was obsolete decades ago. Using financial
cryptgraphy like David Chaum's blind signature protocol, our portfolio
managers could just issue digital *bearer* certificates, right over the net
to their customers, representing shares in the portfolio they manage, rather
than keep track of all a given client's transactions in a database for
posterity. Even more fun, using the digital bearer *cash* they get from the
sale of those certificates, they could turn right around and instantly buy
debt, equity, or any derivative thereof, in digital bearer form, of course,
without waiting for any transactions to settle through a clearinghouse of
any kind. Why? Because knowing that you've digitally signed a unique blop of
bits and honoring the promises those various outstanding blops represent is
a whole lot easier, faster, and, of course, cheaper than keeping track of
every transaction you make for seven years, or whatever your friendly nation
state says you have to do so they can send somebody to jail if that person
lies to you. And, of course, digital bearer settlement is *much* faster than
waiting for all those book-entries to percolate through various
clearinghouses, banks, brokerages, and other financial intermediaries in
order for a trade to clear and settle.
Financial cryptography is a direct consequence of Moore's Law. You can't do
it without computers, and, more important, lots of cheap computers on a
network. But, you can do a lot of very neat things with it, as we've seen
above. In fact, the protocols of financial cryptography will be the glue
which holds a geodesic economy, if you will, together. And, of course, as
Deke Slayton put it, 'No bucks, no Buck Rogers.' No geodesic economy, no
I joke about VISA being replaced someday by an innumerable swarm of very
small underwriting 'bots' whose job it is to form an ad hoc syndicate which
buys the personal digital bearer bond issue you floated for today's lunch.
In a geodesic market, the one-to-many relationships of hierarchical
book-entry-settled industrial finance, like checks and credit cards, becomes
to the many-to-one relationship of the geodesic digital-bearer-settled cash
and the personal bond syndicate.
But, what, you ask, do I do when someone defrauds me? The neat thing about
using financial cryptography on public networks is that you can use the much
cheaper early-industrial trust models that went away because you couldn't
shove a paper bearer bond down a telegraph wire. In short, reputation
becomes everything. Like J. Pierpont Morgan said 90 years ago,
'...Character. I wouldn't buy anything from a man with no character if he
offered me all the bonds in Christendom.' In a geodesic market, if someone
commits fraud, everyone knows it. Instantly. And, something much worse than
incarceration happens to that person. That person's reputation 'capital'
disappears. They cease to exist financially. Financial cryptographers
jokingly call it reputation capital punishment. :-). The miscreant has to
start all over with a new digital signature, and have to pay through the
nose until that signature's reputation's established. A very long and
expensive process, as anyone who's gone bankrupt will testify to.
So, you don't need biometric identity to stop non-repudiation. Translated,
that means that since you're moving secure digital bearer certificates over
an insecure private network like the internet, and not moving insecure
debits and credits over a secure private network like the SWIFT system, you
don't need audit trails to send someone to jail if they make the wrong book
Instead, you trust the issuer of a given piece of digital bearer cash, say,
and not the person who gave it to you, just like you trust the issuer of a
given currency today. Biometric identity is orthogonal to reputation in, um,
'cypherspace'. And, of course, a financial intermediary like the above
issuer of digital bearer cash is not about to destroy its reputation for the
sake of a very small transaction like the one you're doing, any more than
the Fed would demand 6 one dollar bills in exchange for one five dollar bill
just to make an extra buck. Well, not since they started listening to
Friedman, anyway. :-)
Microintermediation means what it says. Financial intermediaries never go
away. You can't have markets, much less efficient ones, without financial
intermediaries buying things low and selling them high. Renting their
reputations to ensure transaction liquidity, in other words. This is at the
essense of Von Mises' 'Calculation Argument' against planned economies, and
the defunct economy of the ex-Soviet Union is mute testament to that
particular economic truth.
Moore's Law, I like to say, operates like a surfactant of information,
breaking great globs of concentrated information fractally into smaller and
smaller bits, like so much grease in soapy dishwater. Capital, for the most
part, can now be converted into information and instantantly bought or sold,
or, more to the point, instantly settled and cleared in digital bearer form,
in increasingly smaller and smaller bits, by smaller and smaller and
increasingly more automated financial intermediaries. Microintermediated, in
What we get is a world where anything which can be digitized and sent down a
wire will be auctioned off in real-time in cash-settled markets. Stuff like
capital we've seen, but lots of other things, which are not immediately
intuitive. Machine instructions -- teleoperated or not. Software of all
kinds including entertainment and art. Bandwidth; I talk about a router
saving enough micromoney out of switching income to buy a copy of itself.
Maybe even adjudication and physical force, someday. After all, who says we
have to buy violence from the local force monopolies we now call nation
states, especially if we can get it cheaper and better -- and possibly in
smaller amounts -- in a competitive auction market? Curioser and curioser,
as Alice used to say...
I mean, the nation-state's just another hierarchical artifact of industrial
communication technology, right? Besides, If everyone's paying for things in
cash and no book entry taxes can be collected because there aren't any book
entries, then, as someone said on a Harvard Law School list a few years ago,
'What happens when taxes become a tip'? Of course, there are various
cypherpunks out there who say things like 'Write softare, not laws.', which
should make those folks on Mass Ave in Cambridge more than a little nervous
So, welcome to the geodesic future. Not hoping to attract the wrath of the
famous curse, isn't it an, um, interesting place?