The Theory of Everything, PNAS 
A recurring theme on complexity discussion lists is the question
what actually makes a system complex. Laughlin and Pines approach this
question from a fundamental physics perspective: For traditional physicists
the reductionist goal is to take a system apart, look for the most fundamental
building bricks and the laws of how they interact and write those laws down as
mathematical models. The ultimate goal is that those models then eventually
lead to a "theory of everything (TOE)". Going back up the scales to build
objects of everyday experience like clouds and brains then is a mere matter of
building complicated objects. In that view an organ is not different from a
very complicated machine. The position of the authors is that for all
practical purposes the existing quantum theory and classical physics
constitute already a TOE. But they claim that it is not really relevant, that
even if the TOE would change it still could lead to the same phenomena on a
macroscopic level.
An implication of this claim is that something fundamentally new has to
happen as one moves up from elementary particles to brains. They give an
argument that already for very small microscopic systems a TOE would in
principle not be able to calculate the behavior of the system and that
fundamentally new properties emerge. Thus Laughlin and Pines understand
"emergence" in a philosophically strong sense, that the emergent properties
are indeed unknowable just from the behavior of the microscopic components.
Some of the emergent properties are universal in the sense that they are
independent of the details of the microscopic theory. They form the basis for
a rigorous theory of self-organized complex systems that is not reductionistic
but whose fundamental entities are the universal properties. The examples
given by the authors are from the area of superconductivity and other
macroscopic quantum effects. But they also mention turbulent patterns from
fluid dynamics ordering phenomena in liquid crystals (like LCD displays). They
call those self-organized states "protectorate, a stable state of matter whose
generic low-energy properties are determined by a higher organizing principle
and nothing else".
They see a new empirical non-reductionistic science on the horizon of the
new millennium that is rigorously based in observations and experiments and
does not depend on speculative deductions from microscopic first principles.
At the same time it is not purely phenomenological but has a foundation of
universal properties at mesoscopic and macroscopic/observable scales. Laughlin
et al. discuss self-organization specifically at mesoscopic scales in the same
issue.
The Theory of
Everything, R. B. Laughlin, David Pines,
PNAS, Vol. 97, Issue 1, 28-31, January 4, 2000
The Middle Way, PNAS
Traditional, reductionistic science and especially physics tried
to explain the world by looking at the smallest, (sub-) microscopic and the
largest, cosmological scales. The scales of our direct experience seemed to be
understood in principle and it was to be only a matter of time and computer
effort before all phenomena at those intermediate scales can be fully
understood. Laughlin & Pines discuss in a separate paper the reason why
this view is fundamentally flawed because of emergent properties of complex
systems. In Laughlin et all the researchers focus on a scale smaller than
accessible to direct observation but large enough that the number of particles
involved in the organized pieces of matter is large enough to be beyond
explicit calculation from the constituent parts. These are scales at which
biological "miracles" happen. Phenomena that can be categorized and described
but not explained in terms of detailed mechanism. Examples are bio-molecules
such as enzymes and intra-cellular structures that are responsible for
dramatically complex biological processes that are reproducible and can be
modified but not derived from the properties of their constituents. The
authors discuss the possibility of new and undiscovered organizing principles
that have universal properties in the sense that they are independent of
specific details of the underlying microscopic properties. They call those
properties "protected" properties of matter.
Some of those principles are related to the concept of energy landscapes:
Similar to a ball that will roll down slopes in a natural landscapes until it
comes to rest at a local minimum it is visualized that the constituents of
mesoscopic systems will arrange themselves until they reach a state of minimal
energy, or a local minimum in the energy landscape. One complication is given
by the fact that these landscapes are not two or three-dimensional but they
can be of a very high dimension (dimensions here has to be understood not as
the dimension of our physical space but the number of different, independent
modifications the system is allowed to undergo.) Just from the combinatorics
of the system it is clear that there are a large number of possibilities how a
system can reach a minimal energy state such as a stable configuration of a
folded protein. Biologically interesting complex structures typically are not
those that have an absolute minimum in energy (when a biological system
reaches that state it usually is dead) but those that are meta-stable. One
universal mechanism that can lead to an interesting structured state is known
among the experts as "frustration": Subsystems are in a state where they would
move in one direction that would decrease the energy of one interaction but
increase the energy because of different interactions. Phenomena like those
can generically lead to non-exponential dynamics and aging. Physical systems
with such properties are glasses and materials that will be used for quantum
computers. That is an indication that a new generation of (nano)-computers
will not only be very complicated but also truly complex.
The Middle Way,
R. B. Laughlin, David Pines, Joerg Schmalian, Branko P. Stojkovi, and
Peter Wolynes, PNAS, Vol. 97, Issue 1, 32-37, January 4,
2000
-
The study by Zaror and Guastello is concerned with the particular
case of leadership studies, and was conducted to determine whether a nonlinear
dynamical process for leadership emergence would generalize from North to
South America. Many of the leadership styles that are known in the
Americentric research literature are also evident in the other cultures. In
spite of the similarity of styles and constructs, however, cultures appear to
differ with regard to how the styles are expressed and preferred by
subordinates at large.
The researchers identify interaction between leaders and subordinates as
characteristic for the degree of leadership displayed: A high-quality
interaction would be characterized by four principles -- loyalty, respect,
contribution, and positive affect -- which comprise a single indicator of
leader-member exchange (LMX). High ratings on LMX have been associated with
work outcomes such as individuals' work performance, job satisfaction,
satisfaction with supervision in particular, role clarity and conflict, and
leader-member agreement, perceived organizational support, and commitment to
the organization,
The evolving principle of self-organization is that the local dyadic
interactions give rise to global phenomena. Not unimportant is the finding
that interactions among all organization (or group) members are relevant to
the global outcomes. This latter point is pivotal in the emergence paradigm
adopted here, which is based on the rugged landscape model of
self-organization and leadership emergence.
The authors claim that they were able to identify a "swallow-tail
catastrophe models" that describes the transitions involved in the emergence
of leaders. These models have generic properties for non-linear dynamical
models and are used to describe sudden transitions (catastrophes) in a
geometrical framework.
The "swallow-tail catastrophe model" depends on three control parameters
and although the authors were not able to clearly identify them they think
that they are related to asking questions, making jokes, initiating a line of
discussion, clarifying points made in a discussion, diffusing conflict,
following others, and "gate keeping".
Self-Organization and Leadership Emergence:
A Cross-Cultural Replication, Gonzalo Zaror, Stephen J. Guastello,
to app. in Nonlinear Dynamics, Psychology, and Life
Sciences, Vol. 4, No. 1 (January, 2000).
A Rugged Landscape Model for Self-Organization and
Emergent Leadership in Creative Problem Solving Groups, Stephen J.
Guastello, Joanna Craven, & Benjamin R. Bock, unpublished
abstract
Hacker group: The future of war is information, CNN
It seems that one emergent structure in all social systems is the
collective violence in organized war. At times war was "the mother of
inventions" and arms races pervaded all aspects of human creativity. Therefore
it comes as no surprise that as society moves into an information age warfare
keeps up with this trend to create "information warfare". This was one of the
topics discussed at the recent 16th annual Chaos Communication Congress in
Berlin. Frank Rieger, a member of the notorious Chaos Computer Clubs that was
responsible for some of the early Internet break-ins discussed some recent
examples of information warfare as part of military operations like the air
raids of NATO planes against targets in Serbia. He mentions the use of
intercepted cell phone calls to make demoralizing calls to families of
soldiers. He also brought up the old speculation that the bombing of the
Chinese embassy in Belgrade was an information-attack launched by the Serbs.
Apparently he didn't mention defensive systems against satellite-guided bombs
and cruise missiles that would simply send out a fake GPS signals as featured
in a recent James Bond movie that would guide them to false targets.
Rieger expects that in the future it will be difficult to identify an
information attack that is disguised as a communication glitch. Spreading of
false information is also an aspect of information warfare although with the
high redundancy of information on the Internet counter measures will also be
improved for instance by a fast verification by trusted and independent
sources. On the other hand if one considers the recent drop in Taiwan stock
values after the politically provocative statement of President Lee of
state-to-state relationships with China one can also expect that media
announcement can be timed and spread rapidly by the Internet that may lead to
economically destabilizing swings in financial markets. Since all of these
examples deal with interacting intelligent agents it is to be expected that
methods from complex systems and non-linear control of chaos will play a
significant role in information warfare.
Hacker group: The future of war is
information, Douglas F. Gray , CNN,
December 30, 1999, 18:13 GMT
-
Both the Internet and the Word Wide Web are examples of global,
self-organized structures of information and communication networks. This does
not mean that national and private organizations did a lot of planning and
active organizing to set these structures up but there was note one single,
global organization that made the decisions about how this global network
would be designed and implemented instead it evolved through a series of
"Request for Comments (RFC)".
History took a different course in the area of computer operating systems
that are now used to access the Internet and WWW. Some historians will explain
one day why users of personal computers got "locked-in" to an operating system
that started as "Quick and Dirty Operating System (QDOS)" (later the "Q" was
dropped when it turned out that it was not that quick). Today, one single
company dominates the global market of personal computer operating systems in
a way that gives advanced users and software developers little space to adapt
and customize the operating system.
The conflict of a user between an increasingly integrated and complete but
restrictive software environment and an open, adaptive operating system has
been shifting lately towards the second category. VA Linux Systems Inc., a
developer of systems and services for the alternative Linux operating system
will create a hosting service for open source software projects. "The
Sunnyvale, California company, which went public last month with the biggest
first-day gain in IPO history, will unveil a service called SourceForge for
the open source community. With this new service, VA becomes an application
service provider, or ASP, for open source projects and corporations. "
Linux and unix are popular in academic circles but are not as user friendly
and application rich as the current versions of the windows system. The 200
employees and 3000 software developers that registered on the SourceForge Web
site (growing at a rate of 25% per week) will try to change that situation. If
that rate continues we can expect some interesting phase transitions in the
not too distant future.
Breaking news (Reuters/Taiwan News, 1/7/2000): China will ban the
use of Windows 2000 in its vast government institutions. Instead, the use of
"Red Flag Linux", a verion of Linux developed by Chinese researchers will be
the operating system of choice for official government business.
VA Linux Unveils New
Service, Wired News, Reuters
,7:12 a.m. 4.Jan.2000 PST
Expansion of the Genetic Alphabet, Nature
Excerpts: "The genetic book of life is a monotonous tome: it is
written in just four letters. Now US chemists have shown that they can expand
the language of the genes by adding a new letter." (...)
"McMinn and colleagues made up-by hand-a single strand of DNA in which the
synthetic base was inserted four characters from the end. They paired this up
with a strand of complementary bases that stopped just before the synthetic
base, leaving a dangling stretch of five unpaired bases.
The researchers then showed that a natural DNA-making enzyme from the
bacterium Escherichia coli was able to generate and incorporate a second copy
of the synthetic base into the unfinished strand, pairing it up with the
first. Unfortunately, however, this artificial pair seems to disrupt
completion of the second strand, probably because the shape of the pair
differs from that of a natural base pair.
Nonetheless, this work, reported in the Journal of the American Chemical
Society1, shows that the genetic code can be given a foreign character, which
is not readily confused with the four natural letters. Now the question is:
can it be used to say anything new?"
Expanding the book of
life, Philip Ball, Nature, 12/24/99,
McMinn, D.L., Ogawa, A.K., Wu, Y., Liu, J., Schultz,
P.G. & Romesberg, F.E. Efforts toward Expansion of the Genetic Alphabet: DNA
Polymerase Recognition of a Highly Stable, Self-Pairing Hydrophobic
Base J. Am. Chem. Soc. 121, 11585 (1999).
Sex Determination in Malaria Parasites, Science
Sex ratio is one of the directly accessible parameters that can be
observed in populations of different species and that plays a direct role for
the species' evolutionary fitness. Whereas in most mammals a symmetric ratio
of 50-50 is maintained -often with surprising stability- in other species the
sex-ratio is used adaptively. For instance ants take advantage of the facts
that only a few males are needed for reproduction and that female individuals
are more valuable for the survival of the species. Paul et al. investigated
the adaptive changes of the sex ratio in Malaria parasites. They came to some
interesting conclusions that might have a direct effect for immunization and
treatment of human Malaria:
"For the malaria parasite, a single haploid cell gives rise to a clone
producing both females and males. The ratio of males to females is important
in understanding and controlling disease because transmission of malaria in
nature depends on sexual union of the parasites and because more males are
formed in lethal infections. Paul et al. have found that the frequency of the
sexes is affected by the host hematologic state. Treatments that induce
erythropoiesis result in a shift to male parasites, which leads to decreased
reproductive success. This finding may provide new approaches in malaria
control as well as new considerations in therapy, as the antimalarial drug
chloroquine inhibits erythropoiesis. "
Sex Determination in Malaria
Parasites, Richard E. L. Paul, Timothy N.
Coulson, Anna Raibaud, Paul T.
Brey,
Making the Right Connections in the Brain, Science
Topoisomerases, enzymes that are critical for successful DNA
replication have an unsuspected role in neural development: Yang et al. find
that mice with mutations of that enzyme have very specific defects in the
nervous system. It seems that these enzymes are responsible not for the
development of the neurons themselves but in how they connect to other neurons
with the help of their outgoing axons. More specifically the researchers could
show in their experiments on mice that axons of motor neurons would not reach
skeletal muscles, and axons of sensory neurons did not enter the spinal cord.
As a consequence mutant mice with this deficiency were born with the inability
to breath normally and died shortly after birth.
Scientifically these results are exciting evidence for the modular
organization of biological systems as discussed by Hartwell et al. (see ComDig
1999.9.1): The genes that lead to the formation of the topoisomerase enzymes
are not involved in the development and location of cells (neurons in this
case) but specifically in the communication and networking among these
cells.
It is known that in yeasts and Drosophila one type of topoisomerase is
critically important for segregation of intertwined pairs of newly replicated
chromosomes. In yeasts, the enzyme is also involved in relieving torsional and
flexural strains in DNA. It appears as one of the miracles of molecular
biology that one enzyme can be involved in different organisms in a number of
seemingly unrelated chemical processes that are in their effect on the
development of the organism highly specific.
DNA Topoisomerase II and Neural
Development, Xia Yang, Wei Li, Elizabeth D.
Prescott, Steven J. Burden, James C. Wang , Science, Volume 287, Number
5450 Issue of 7 Jan 2000, pp. 131 -
134
Long-Term Weight Control, Science
One of the primary functions of life is to make sure that the
living organism lives long enough to have a chance to reproduce. That includes
a regulation of the food intake at a level that prevents starving as well as
the effects of too much food intake. In mammals it seems that the controlling
factor that will trigger behavior that makes sure that these goals are reached
are connected to the pleasure center in the brain. Depending on the energy
level of the body the intake of the same kind of food can either be highly
pleasurable (after being starved) or almost painful (for instance in a hot-dog
eating contest).
It is known that this communication system inside the body consists of
several sub systems. A chemical subsystem uses the hormone Leptin that is
produced by fat cells and that signals when the body has sufficient energy
reserves. It has been known that leptin influences the hypothalamus in the
brain but the exact way in which it communicates with the brain's pleasure
center was not well understood.
Fulton et al. studied this problem in a series of smart experiments with
rats. The worked with rats that had electrodes implanted in their pleasure
centers that they could activate by pressing a lever. This technique is known
as Brain Stimulation Reward (BSR). They observed that the effect of
electrically stimulating the pleasure center was significantly enhanced for
hungry rats, which means that their BSR value was increased during food
deprivation. The researchers then injected leptin into the brains of the rats
and they observed that as a result the efficiency of brain self-stimulation
was greatly reduced. It is interesting that this effect lasted for as long as
four days after leptin infusion. This indicates that leptin is involved in the
long-term energy balance of the body and does not depend on short-term
fluctuations of sugar levels in the blood or in the gut. These results suggest
that leptin might play a central role in long-term weight control.
Modulation of Brain Reward Circuitry by
Leptin, Stephanie Fulton, Barbara Woodside,
125, Volume 287, Number 5450, 2000, 125 - 128
Equilibrium Regained: From Chaos to Mechanics, Science
The original meaning of the Greek word "Chaos" was that of
emptiness and absence of any order or structure. The picture of a gas at
thermal equilibrium, i.e. a state that it will take on eventually in an
isolated container if it is left by itself is a good approximation of that
original form of chaos. It is quite different from the concept of
deterministic chaos that has entered main stream science about twenty years
ago and that is full of structures and fractal patterns. An old question for
some theoreticians was if it is possible to "go-back" to an equilibrium type
of chaos if one starts off with a deterministic, non-linear dynamical systems.
For a few cases that transition could be rigorously proven for instance in the
case of Sinai's billiard. Other examples that are studied numerically often
suffer from the problem that tiny islands of stable order survive the
onslaught of chaos. They can lead to trapping regions that keep the system
from reaching equilibrium for a long time.
Some of the earliest numerical attempts to tackle that problems, the
Fermi-Pasta-Ulam model, took place at a location that traditionally hosted the
fastest computers in the world, the Los Alamos National Laboratory. This
tradition is continued in a recent attempt to bring chaos back to
equilibrium:
"Statistical mechanics provides a powerful method for understanding systems
at or near equilibrium. Egolf now shows that a theoretical system far from
equilibrium, a chaotic, coupled-map lattice, can display equilibrium-like
properties. The length scales associated with chaotic behavior in this system
are much smaller than those used in the averaging processes needed to recover
macroscopic properties. Thus, a dissipative chaotic system, observed on a
sufficiently coarse scale, can recover features associated with equilibrium,
such as Gibbs distributions, ergodicity, and detailed balance."
Equilibrium Regained: From Nonequilibrium
Chaos to Statistical Mechanics, David A.
Egolf, Science, Volume 287, Number 5450, 2000, 101 - 104
-
The most of complex dynamical systems are multistable. This means
that for the same values of parameters, the system state may be either chaotic
or regular (periodic). Such a phenomenon appears to be common for a variety of
complex systems: electronic circuits, lasers, geophysical models, mechanical
and biological systems. An important problem of controlling these multistable
systems is to device a mechanism, which can make the system jump from one of
the coexisting states to a regular state where the effect of nonlinearity is
minimal. For example, in the medical science, the undesirable states may
appear because of various dynamical diseases of the brain, such as,
Parkinson's disease and epilepsy. There, the objective would be to make the
system jump from those undesirable states to a normal state. Practically, it
is not often possible to change the system parameters and one must consider an
external control, which would lead to the transition from one state to the
another. Apparently, a proper change in initial conditions might be
appropriate in this situation. Experimentally this can be realized in the form
of a short external impact. However, the probability that the system hits into
a coexisting basin of a given state, depends strongly on the phase and
amplitude of the external force. In medical practice a short-lived external
force is used for the defibrillation of a heart. A high voltage impact
sometimes allows the heart to change its state from chaotic to regular one.
Unfortunately, it does not happen always if the instant of impact (phase) and
the amplitude of the voltage are not properly chosen. The situation would be
even more complicated if sufficient information about the true nature of
multistability were not available, very often due to the dearth of a good
theoretical model.
Pisarchik and Goswami offer another approach, which allows annihilation of
one of the coexisting states, so that the system jumps to the remaining state.
In the earlier mentioned method, a short strong external impact just switches
the system from one stable state to another. Then the system may stay on the
selected attractor until some instability or noise switches the system back to
the previous attractor or to different state (in case of multistability). The
new method implies the application of slow small harmonic perturbations with
properly chosen amplitude and frequency to the available system parameter.
This makes the system a little different from the unperturbed system so that
the new system does not already have the undesirable state, i.e. becomes
monostable.
In spite of the simplicity of the method, the physical mechanism is rather
complicated. Numerical simulations and experiments with a laser suggest that a
periodic modulation to one of the system parameters can make a periodic state
chaotic via a period-doubling route at a much smaller control frequency, where
the original periodicity acts as a carrier. Notice that no qualitative change
of the behavior would have occurred if the value of the system parameter were
increased by the same magnitude as that of the control amplitude but without
periodic modulation. In a monostable system a slow parameter modulation
destabilizes a periodic orbit in a sense that the negative leading Lyapunov
exponent increases approaching zero. In the presence of another stable
attractor, the former attractor looses its stability and the system switches
to the remaining stable state. These results may have several applications
from the viewpoint of a control of bistable systems. For example, in the case
of epilepsy or Parkinson's disease, a slow periodic modulation to one of the
important physiological parameters (e.g., by a variation of magnetic or
electric fields applied to the head) may annihilate the disease state. Some
medical experiments can be interpreted in the frame of the attractor
annihilation (for instance, experiments on fluctuations in tremor and
respiration of patients with Parkinson's disease.
Contributed by: A. N. Pisarchik
Annihilation of one of the coexisting attractors in
a bistable system, A. N. Pisarchik and B. Goswami, Physical Review
Letters (2000) (accepted for
publication).
-
Hal Varian and Carl Shapiro discuss their book Information Rules:
A Strategic Guide to the Network Economy and some general principles of
"information economy" where "the costs of production are so low that the
effects of high fixed costs and low marginal costs are sharpened or
accentuated". One consequence of that is a "strategy for price discrimination
or dissociating the prices charged from the costs of production and simply
charging each customer a price that he is willing to pay. " A similar strategy
is already in place for airline tickets: "The marginal cost of putting someone
in an empty seat on that plane is next to nothing -maybe the cost of a few
bags of peanuts and pretzels."
Another phenomenon that is not novel but takes on new extremes is the
"lock-in" of customers, "the result of hidden or unanticipated costs of
switching from one technology or brand to another." One example is "the ink
jet printer. These printers are selling at incredible prices, $150 to $200 for
very high quality equipment that looks like a great deal. Then the first ink
cartridge runs out and you find that the next one is going to cost you $60
-for a $150 printer." Creating "lock-in" is presumably also at the basis of
the success of Microsoft and VHS video format as well as the e-commerce custom
of giving away information goods and services for free initially until the
customer is locked in.
Software patents is another emergent strategy that gives an advantage to
companies who focus on getting rather unspecific patents instead of going
through the more expensive process of detailed software development and
testing. All the basic concepts of the information economy network effects,
lock-in, positive feedback, and switching costs are all naturally described in
the concept of complex adaptive systems.
Information Rules: A Conversation with Carl
Shapiro and Hal Varian, Ernst & Young
Center for Business Innovation, Perspectives on Business Innovation,
Issue 3
Information Rules : A Strategic Guide to the
Network Economy ,by Carl Shapiro, Hal R.
Varian , Harvard Business School Press,
1998
-
I felt cheated. I picked up the book about the highly publicized
but inscrutable Prediction Company expecting to learn how a clever group used
complexity to crack market codes. And found no evidence, no experiments tried
and succeeded or failed, nothing that could be repeated in any scientific
sense of confirmation. But I did find an entertaining tale of a quest by a
couple of people I knew only slightly engaging in a journey similar to one I
had taken.
Complexity should fit the market model almost exactly. If markets are not,
with their rush of information and adaptive agents (us) running around
satisfying themselves, nothing else is. But saying that complexity is a way of
thinking about markets and as a source of sympathetic metaphors is one thing.
Trying to use complexity to develop hard models that can predictably beat the
markets is another.
The attraction of using quantitative methods for market analysis has been
popular since the late 1960's when academics offered insightful notions about
how markets behave. The aftermath of a market crack, post-1967, always seems
to sharpen the minds for alternatives of "better ways." And academics were
unscathed by dismal performance measures as the nifty-fifty moved into the dog
house.
But these quantitative measures were very crude by most statistical
standards. Even though they worked, it was generally agreed, as the author
attributes to one of the books commentators, they were the Model A's of an
emerging industry. Mostly they were addition, subtraction, multiplication and
division and very little of division.
A form of elitism developed among the quants and academics that the tools,
crude as they were, worked because the rest of the market actors were not very
bright. Maybe it could be that all one had to do was throw all the market data
into a big machine, search for all the correlations that would emerge and
assume that this "research" would remain stable enough in the future to
produce systematic, predictable profits. The principal founders of Prediction
Company, Doyne Farmer and Norman Packard were susceptible to this
temptation.
At first these energetic scientists wanted a different kind of company,
sort of a commune dedicated to demonstrate superiority and produce funds to
support more research. In the end, the firm structure and behavior seem little
different than any other firm organized in the money business in this heady
period.
The book implies that three or four years of research failed to produce any
stable results. However, Predication succeeded in obtaining ever more funding
from well capitalized sources who wanted to believe that contemporary
complexity methods would be the alchemy of the day. There were little other
major research paths going on in investments that offered even the promise of
an edge. The book describes the promotional work done by the founders and a
handful of colleagues on a stream of investors who arrived wanting to be sold.
And they were.
Throughout the inevitable crises of company startups, the founders manage
to take long vacations, often at inaccessible places. The conflict between
wanting their independence and making a financial killing was also in play.
And the ability to back test, to make the data speak (I cannot resist one of
my favorite quotes from "How to Lie With Statistics"...."If you torture the
data long enough, it will confess to anything.") and to solve the inevitable
operational problems of dealing with expensive security settlements, half
holidays in Hong Kong and the like, were all exciting elements in the
story.
You will like the heroes although you will not know what they did except
shake the money tree for investment in their company. You will empathize with
their internal struggles between academe and commerce. And you will learn
about derivatives, forex and market structure.
And you will conclude that the question the Predicators asked themselves
five years ago...could complexity be used to make money in markets, remains a
uncertain, challenging and worthy pursuit. It is my hope that these new
entities who rely on the inherent open system characteristics of complexity
will be more forthcoming in sharing their results in the spirit of the field.
The prospect of turning lead pencils into gold money may be a driving force
for some, for others it is a test of drilling complexity down to specific,
testable applicatioms.
Contributed by Dean Le Baron
The Predictors
by Thomas A. Bass, Henry Holt & Company, Inc,
1999
