Out of Control
The new Biology Of machines , Social
systems, and the Economic World
By Kevin Kelly
HIVE MIND
As a bee hive , the
emergent property will rule. Simulated
penguins and bats in Batman are like a flock: One. Vivisystems are the result of melding wet and dry ware. The have unpredictable emergent
properties. More is different. Temperature cannot exist in one thing. The atom distinct, has given way to the
hive/net age. Nets are non-guided yet
resilient. They are hard to crash and
flexible, yet redundant and impersonal.
Nets are unbounded. Out of
control growth requires nets.
MACHINES WITH AN
ATTITUDE
The made v. the
born. Robot destroyers that swarm. Put a guinea pig behind the controls. Mobots v. Staybots. The ultimate movement ant used independently
“minded” legs. No CPU. This is like an economy. Program each individual with rules and watch
it grow. Soviet command economy
tripped. PG 40. Soda can collector rules. There is no I for the human, and the bee
hive, and the corporation. But society
nets it will also appear to have a direction.
But....?
Millions of dumb pieces
adding to one smart illusion is how biology works.. It must be bottom up.
Command economies fail. Reverse
engineering is harder than emerging solutions.
The machine doesn’t need to use memory recreating the world view in it’s
head. The world represents itself
. The machine just processes
contigencies. A mind in a vat, without a body, controlling robot is
immortality. Self-supporting room
vaccuumers, light changers, door openers create a living house. The house has a mind. Nature insists we are
apes. Artificial evolution insists we
are machines with attitudes. We will
make machines that run with wet and we will use computers more.
ASSEMBLING COMPLEXITY
A prairie replanted
grows weeds. The missing elements was
fire. The ecosystems were sensitive to
order planted and worked towards stability.
Order points out you need assembly instructions.
COEVOLUTION
What color is a
chameleon in a mirror? Would it endlessly cycle like fashion or settle to a
middle like democracy. Its a hard
experiment. The green it went was
fear. The blue eye feedback loops of
the lizard. The mirror and lizard
became a system. Middle ages man had no mirror. Now we do, TV, Paper trails, polls. Feedback systems defy cause and effect. The prime cause is gone.
Butterflies/Catepillars
coevolve with milkweed. They Coevolve
and are Codependent. Much coevolution
is parasitic. As in the arms race competition
often breaks into the necessity to cooperate to succeed. Neither wants to eliminate the other. Life
systems are instable constantly shifting.
The key of life is reproductive instabillity. The life effects the non-living.
The life and atmosphere / environment in a coevolutionary dance. Both are unstable chemical factories. Human/machine, rock/life weed/butterfly chameleon/mirror. These are false differences. Under what conditions will cooperation
emerge from a world of egoists? Hobbes
said only with central authority. Rules are different in zero sum and non-zero
sum gain models. Control and secrecy
are counter productive. As companies
compete and announce new product, adaptation and improvements are announced by
another’s company. They co-evolve.
THE NATURAL FLUX
Forest, nor economics
peak at an equilibrium. Variance of
success increases richness in nature.
The longer the food chain, the greater the interdependency, the more
susceptible to collapse. The less
species, the more connected they are.
A Michigan lot left alone
will turn into a hardwood forest.
An organism is tightly
bound. It staves off invaders and is
susceptible to them.
Ecosystems, on the
other hand, are loose and permeable.
Ecologies are temporary networks.
They mix with each other, try each other on promiscuously.
Evolution is a tight
web: Growth via being inextricably
bound.
Ecology a loose one.
Ecology is community evolution in history.
Evolution can be synthesized.
Ecology cannot.
To create diversity
don’t make a complex creature. Make a
simple creature and a complex environment.
Life is like a flood that flows into cracks. Non-living things don’t live nearly as long as living. Living is
so complex due to adversity that it can’t die.
Non-living is too stupid and simple to kill life. Life will eventually fill the universe. Ed Fredkin says the universe is a
computer. Life are information
processors. Life conquers matter. Mind conquers life. If not us.
Mind via life will sprout again.
Life to machine is an artificial distinction. Both have self-organized change.
Life is becoming becoming, on to complication.
EMERGENCE OF CONTROL
James Watts invented
the fly ball governor to control the output of steam engines. Its one of the
first non-biological circuits. The regulator gave the machine a self and had
“purpose”. Gun controls on navy ships
made a regulatory system that relied on the interface of man and machine. One simple control beats a complex set of
interrelated controls. Price for
example Justice is its own cause and
effect. If a cause causes cause and
causes a cause... Its recursive.
Circular loops will cause spinoffs, they are fertile. Self-merges from itself. Natural selection causes organisms that
cause natural selection. Its not a one
way cause and effect. It causes itself
. Matter is almost free. since the
steam engine we’ve gone bigger and now smaller and smaller to pull out matters
information.
CLOSED SYSTEMS
Natural coral reef-like
closed systems start turbulently and then settle. Closed microbial ecosystems are like our own. The microbes do most of the work. Big flashy mammals are window dressing. Shepeler put himself in a closed algae box
for 24 hrs. The stench drove him
out. He used enough algae to supply
oxygen for someone his size.
Dogs have been in for 7 days. The biosphere ecology was synthetic, but so
is California. We, Chile and Australia
removed the ancient herbivores to make room for bovine.
POP GOES THE BIOSPHERE
Bio needed humans as
pruners. A thing missing from biosphere
was turbulense. 1/2 of the CO2
makes it into the atmosphere. Noone
knows what happens to the rest?
1) increased tree mass
2) Put into soils by
microbes or
3) put into
oceans.
The biosphere lets us
study CO2 cycles with control.
CO2 levels cycle on a daily and seasonal basis. Oxygen levels dropped from 21% to 6% (equal
to llahsa in TIbet).Where did it go? To
combat rising CO2 they made an “intentional” season in a
savannah. To increase CO2,
they put garden scraps in a compost. In
such a system you cannot maximize the yield of an individual species. You can
only maximize the system.
INDUSTRIAL ECOLOGY
Smart offices will
happen before smart homes. They will
have memory : which book you put where.
Where you are. You carry the
light switch with you. The most profound
technologies are those that disappear.
The voice activation pin head. When the phone rings the stereo goes
quiet. Office voice mail knows I’m not
there yet because the parking spot is empty. When you pick up a book your
favorite reading light goes on. The TV
tells you there is a movie version of the book. Visitors will be a bit frozen out. People will have more or less friendly homes. We already can’t use eachother’s
phones/computers/palm pilots.
Idiosyncratic tech configurations to us. Worn spot , on a bed or book are emergent. Many individual actions make them. Where the best place or most confusion is
will be noted. Iron to machine to car
to scrap yard and back. Its an
ecosystem. Tree to paper to earth
again. Bio tech to Bio to... Fuel energy comes from the H not the C. The C is work and pollution. Natural gas is 20% C, Oil 33%, pure H is
best.
Nature will not move,
so it must be accommodated. Nature -
which is larger than us and our contraptions - sets the underlying pace for
individual progress, so the artificial will have to conform to the natural in
the long term
NETWORK ECONOMICS
Cyberspace is “where”
you are when you’re talking on the phone. They shouldn’t be called computers,
they should be called connectors. The
network is more important than lone computers.
ATM credit cards, fax, remote diagnosis and fixing of machines. Jeans bought in California trigger inventory
orders in Idaho and dye orders from india.
Orders will drive fashion, not designers. Adaptable factories follow an item through the factory via a
barcode. Batches of one with shipping
and parts need networks. Future
companies maybe less one company than a specified manufacturing process your
order puts into play. Distributed,
decentralized, collaborative and adaptive.
--Distributive:
Employees and offices all over. Cells
of 10-12.
--Decentralized:
Robber barons controlled all. Now
business moves so fast owning a vast infrastructure or factories as a
liability. Out source. Hertz rents, others repair, others
deliver. Governments outsource too.
---Collaborative: From to competitor to collaborator you form
an “Strategic alliance”.
-----Adaptive: As
companies disappear into cyberspace they start to become like software. This is worrisome as bugs may cause it to
crash as butterfly wings make ecosystems crash. Object oriented programs are program pieces that do a specific
function , therefore, localizing errors and personalizing function are
possible. And thus programs can grow
into different niches. This is
neo-biological. Be like natures neural
nets, redundancy. Engineers, not
mystics, are bringing cosmic consciousness.
E-MONEY
Information can be
hidden in cassettes, anonymity can be assured.
Crypto anarchy means no taxes. Privacy wins over full disclosure and
government control. As printing did to
guilds and the church, encryption will do to corporations and governments. Corporate espionage.
Buying the reputations
of anonymous business partners is important.
Anonymity is essential for those who do non-legal things: abortion, pot,
whistleblowers.
We have gone from
cypherpunks to cryptopunks. Online
signatures and ids are needed too.
There is less anonymity now.
Mail was anonymous. The war for
privacy is on. Anonymous “lending” of
films.
Encryption of taxes removes government (the only hope
of central control). In typical
economics, increase supply decreases value.
But the wider encryption or the net go, the more value they have. There
is freeware that gives you peoples encryption code. Encryption means you can trust where you got it from (i.e.: it
wasn’t tampered with).
Newspapers are flat fee. You pay no matter how much you read. Edison wanted a flat fee on electricity because he thought
accounting would cost too much.
Eventually a good meter was invented (10 years). Encryption until use is meter. Success via meter or computer use is being
copied. Not scarcity like old
economics, but abandoning of copies thrives.
But use, not copies, is
countable. How many web hits. Garner money not from copies of software
made , but times it s used. Radio is
like this. Companies send them free
copies of the music, then they pay if they play it on the air.
Trailers and cheapness
will convince us to pay before we use.
A dirty limerick for $0.10.
Weather $1 a month. Digital cash
leaves no trail. This is instead of ATM
cards or passing pictures of dead presidents.
Digital cash is like phone cards (you buy them , use them, toss them).
To avoid theft such
cards would/could use pins and you could have a duress code. This system is private. This could be used
for phones, parking meters, toll booths etc.
Any company could make
a “currency”. Another function of
government gone. Instantaneous bills, no more interest lost. Underwire (not underground)
sub currency sub currencies acceptors may flourish. Again government wants full disclosure, this is hidden
money. Transnationally accepted money
will pass-by nations. Like travelers
checks.
Encryption wins because
it is the necessary counterforce to the nets runaway tendency to link. Without some force of disconnection, the
world would freeze up in an overloaded tangle of unprivate connections and
unfitered information.
GOD GAMES
God as buddy is
new. Of old god he was feared and
appeased. You didn’t want to not be a god.
My generation cant start games without narratives or fixed goals. Kids do. We raise and love our own grown
worlds. They reflect our care and ideas like children . Sim city is more
controllable than sim earth. These god games
grow as gardens do.
Dolls to theatre is
virtual reality. Coca cola is “the real
thing” umberto eco precedes baudrillard.
“They map is not of the territory, the map is a territory.” Idols is an artificial layer of meaning to a
statue. Icon is all the way. We live in the hyper real. They have dinner without reference to
media. Two people can go into the
virtual reality world. It is real. Virtual reality tv. You said the static backdrop as one and the
foreground ad instructions. The
military runs 2-D simulations of many potential battlefields. You can enter any player and outcomes
change. 300 can enter them individually and play
collectively against the computer. You
play from different countries. Our men
in the Persian Gulf war had fought it 6 times before. 90% had intensive simulation experience of it. The attack on Baghdad at night weren’t as
hard as the video. Remote control tanks
exist. Unmanned aircraft were big in the Persian Gulf War. Supertankers practice in every dock in the world. People spend time in these “ships” with
friends at Disney funded game centers.
Ads to the military read “simulate before you build”
“simulate before you buy” “simulate before you fight”. Enter the plane and check out the dial
placements before one is built. World
wide remote unmanned armies. A team, a
superorganism.
People hang out in multi user places to have
parties. Multi users can add
features. A super organism grows. People note they are different on the net
than in “real life” This makes them
question who they are. Australia banned
these text worlds cause they were taking up too much space.
Virtual businesses will start.
Sim city.
Text mud and military. SIMNET
will meet.
Yahweh had been able to
do the rest of his creation in his head, but this part “man” was a simulacra of
Yahweh. SO Yahweh endowed the model the
same type of creativity he himself possessed.
Free will and creativity meant an open-ended world with no limits. Anything could be imagined anything could be
done. This meant that the man-thing
could be creatively hateful as well as creatively loving (although Yahweh
attempted to encode heuristics in the model to help it decide.
A few bold man-things
have had a recurring dream: to make a model of themselves. Is this sacrament or blasphemy?
To be a god, at least a
creative one, one must relinquish control and embrace uncertainty. Absolute control is boring. The new and novel only come from letting go. To win, you must surrender the seat of power
to the mob below.
IN THE LIBRARY OF FORMS
Based on Borges’
library story, Sims generates an evolution of a picture using all possible
pictures. This is done by stripping the
art down to numbers. This is
surprisingly natural for the computer.
The program looks for patterns.
He selects the best and cross breeds them with others that
“evolve>” IT shows “life” evolves
without a creator. Dawkins was shocked
when non-plant life forms evolved from his program. The computer contains every possible pattern in potentiality. The biomorph was to be symmetrical,
recursive and genes (algorithms were included).
People do art by
mutations they like. (breeding 30%,
choosing 70%). There is a mutation intensity knob. Animals can be bred.
Polygamy. He made antiparents. Mutate away. This is a less analytic co-creation (with the computer all logos,
buildings, plate designs, appliances, movies, books. Guide the growth. How like this is making a sentence? There is the 1st word may
possibilities for the second, generate/select.
I
wrote this book by finding it. Who
knows if it is my book or merely the one almost my book. I made it by quotes from the library of
Borges.
ARTIFICIAL EVOLUTION
Butterflies run from
leaf cutters.
A bird exists that eats
the fleeing insects. Other insects
follow to lay eggs in the bird droppings.
Evolution duplicated from computer RNA viruses. The 80 bit virus mutated. The 80 attracted parasites that couldn’t
live without them (45 bits). They spawn
hyperparasites. But then totally new
creatures evolved They replicated in 22
instructions. How? Engineers couldn’t do a self replicating 22. But the program did.
When he turned the mutation function on it invented
sex. The system had punctuated
equilibrium. Evolving spreadsheets find
the typical patient, then run mutations and see how many fit and come up with
variable profiles. It will climb to a
peak and then survey.
Living organisms solve problems by sex and mutation. Many are set to the goal. The closest approximates mate. Why was there no interest in these genetic
algorithms? He traces it back to Von
Neumann’s creation of the modern OS from ENIAC. It allows you to store information and to come back to it and
flow chart another way. If then, check
prior, go dictated path. But the 1950s
John Holland started to think of these as AI machines. He found a 1929 book that looked at the
evolution of butterflies as a mathematical system. Serial computers are slower with more info and a longer flow
chart. Hillis tried to make a parallel
processors dreamt up by Holland. He
wanted to make AI that “would be proud of him”. Quickly it created a great sorting program. When the algorithms fought it went
faster. This showed that evolution and
learning are related. They created
answers like neurons. S hasn’t made
consciousness yet. But records of
genealogy are kept and generations go quickly in computers. People didn’t
believe Ben Franklin had put monstrous lightening in a jar, that they were the
same. Today folks don’t believe in evolved
intelligences in a jar. 250 years
later?..
Bioengineering
found stuff that happens to trigger reactions, then biomimicing , but its sooo
hard. Now: evolution. Generate billions and try em. Antibodies are the complementary shape to a
toxin. Impotent surrogates that
activate immune responses (vaccines) could be evolved. They would be like the best designed
synthetic molecules except we would understand them. We ask the molecule about itself. You select results you like isolate and let them breed, just like
guided evolution for all else. People can alter the evolved computer code. But the result is anemic. They were good for individuals, but died
before too many generations. Death is a
teacher to the flock/superorganism in computer code, the gene is the phenotype,
therefore, lamarkian evolution is possible. Lamark blows doors off Darwin. It injects learning. Ants communicate by pheromones. Virtual ants (VANTS) tried to find the
shortest route to many cities. Their
pheromones evaporated slower on shorter routes, therefore, they were
reinforcing. Learning goes from one
v-ant to all. Parallel computers are
hard to program. Perhaps evolution,
“the natural way to program” will write the code. Make a navigation problem, then create virus to crash it. One bad line of serial code can crash
it. There for much code is protection.
Evolved code can adapt, but we give up control because we don’t understand it.
THE
FUTURE OF CONTROL
In
traditional animation, all knowledge of physics has to come from the animator’s
head. Jurassic has weight friction,
kinetic energy etc, programmed in.
Virtual leaves blowing down the street are following wind patterns.
Automatic
fabrication technology (AFT) goes from CAD to prototype. You don’t print a picture, you print the
part out of strong plastics. Virtual
flowers are grown via algorithms. Life
has infiltrated computational cinema.
Mickey mouse is one of the ancestors of artificial life. Animators get to know a character as they
see it move. Heads can be animated and
the feathered have to respond to physics.
Animation is tied to censor suits now.
Humans have 200 moving joints.
We/animals have antagonistic instincts.
The phone rings your hands are dirty and there’s a knock on the door all
at once. You stall. Then “please wait” the behavior may have
several sequential components. Via a
data bank of character expressions and movement some have you in a story where
you can interact with bit characters.
But characters must have destinies and freedom. This has been done with underwater
creatures. You interact with them as
virtual reality. They can be dinosaurs
and learn via algorithms and have their own agenda. 2003 Mickey’s only instructions “go in the room and retain your
hat”. 2005, “Friend with Donald, but
more and more irritated with him. Can’t
change this cause the code taught itself.
2009. Mickey is a hot commodity. He is a good actor and does all his own
stunts. As we are projected into the
computer we work with the prima donna cause he is Mickey. We’ve less control. We manage our
co-control. Human did autopilot. But we got bored now the computer monitors
the pilot.
AN OPEN UNIVERSE
A program to balance a broom on a skateboard
evolved. But it was inelegant. It isn’t code a drunk human would
write. It’s alien code. But we make simplicity a virtue. Not it.
We have faith in simple elegant order.
We can work with f=ma. We think
serially (with a parallel brain).Computers give us new ways to thing
(navigate). We shouldn’t tell machines
the way to solve a problem. Give them a
problems and let them figure it out.
Machines usually do a specific task.
Evolution solves lots of problems lots of ways. The only machine that reprograms itself now
is the brain. A recursive program that
generates novelty. Natural selection of
solutions. Evolution can open entirely
new places to get to .
All biomolecular novelty has evolved to create
molecules that solve problems they already know how to resolve. This guy was in a hang glider crash and felt
his body reboot. Artificial life is a
verb, not a noun. Look at what it
does. The artificial part is the
natural, not the life. We can’t get
universal principles our example (all life is carbon based life). Biology tears down living systems. A life must start with basics to watch life
emerge. Hyper life is in fast state of
evolutionary change. An acorn is a few
lines of code. Life as it is and life
as it could be. Life wants to break out
into crystals and wires and silicon. Embryonic cells are robots in disguise
made to make more life. We are a
conduit of llife expanding in the open ended universe. By mid-century we could destroy all life
(a-bomb) now we can create it (a life).
The
first extraterrestrial life was spotted on the moon in 1969.
THE STRUCTURE OF
ORGANIZED CHANGE
Textbooks on evolution
emphasize change at the expense of stability.
Evolution of objectivity, economies, laws: all is evolutionary. He’s pushing evolution being synonymous with
slow change. Revolution = fast human
change. Industrial plastics,
electricity, nano revolutions. Daily
revolutions increasing in speed.
Evolution is blind in direction with no historical record. Info
goes from gene to body, but not back.
If it could it’d be lamarkian. I
need longer legs! To do things you need to know how to manipulate your own
genes Robots can do this. Flexibility allows us to live in the Andes
and many generations the genes would be better adapted. The behavior does move the genes (move to
the Andes, straying form the natural environment).
Learning
and evolution combined are more effective than either alone. Both are the causes of evolution. Learning and evolution lay the ground work
for culture. Those who can learn
quickly (ie have genes for memes)
survive in a culture. Culture has a
symbiotic relationship to biology! The
reason we have brains that can produce culture is that we have a culture that
can produce brains. Evolution’s path
goes like this:
1) self
generating things happen
2) Replication
allowed for mutation /change
3) Genes,
developed phenotype.
4) bodies
chose their own niches (learning)
5) Symbolic
memes become possible
6)
self-directed evolution.
Evolutions job is to
create all possible possibilities to take over niches.
POSTDARWINISM
It took 50 years after Darwin to put evolution on a
firm footing. Darwin said every stage
in evolution must be useful. And he had
trouble explaining necessary interim steps.
Science needs to be theory to experiment. Computers allow another method: simulation.
No new species has been observed in science. Fossils show it, but don’t prove natural selection. Steven Gould says the change isn’t in the
fossil record cause its really quick.
AI stalls too. It hits a peak
and then nothing new happens.
Symbiosis is natural normal in microbiology: Intercell genetic transfer via plasmids. Some
biologists think large chunks of human DNA were inserted by virus. Human diseases being escaped human
virus. This is not based on the
individual as 2 cells become one and its not incremental (both blows to
traditional Darwinism). Branches on the
tree of life intersect. More a thicket
of life. Natural selection is the editor, not the
author. Origin not variation of
species was darwin’s question.
Mutations aren’t random: it’s internal selection. Mutation correction takes a lot of
energy. Correction is selective. The Krebbs cycle is our energy cycle. It has been for millions, therefore, it gets
mutation correction priority. Some
random changes are more equal than other.
Homeobox genes regulate entire areas of vertebrate. This isn’t heresy.
His next idea, that purposeful mutations occur is. Needed mutations happen more than not
necessary, mutations by 100 million
times in e. coli. But until a mechanism
is found few biologists will follow the radicals.
A phenotype must grow before a gene mutation can be evaluated. This can take years. Computers have no lag time or body. Our development mimics evolution. We have gills. Evolution is conservative, retaining what has worked and building
on it. Computer changes are
discrete.
Monsters,
like drosophilia with legs for antennae do occur. Science of mutant monsters is called teratology. The egg is the mother of us. It is affected by stress, age, mutations,
hormones of momma. The egg particularly
effects the eggs (which has had decades in mom’s body (environment). This transgresses, Lamark like, the view
that info only goes gene to body. We
should be suspicious of the idea of one way non-feedback genetic transmission. It happens no where else in nature.
Small
changes early lead to big changes later.
Environmental stresses would lead to more monsters. This would account for the discontinuity of
species in the fossil record.
To
science historians, Darwin’s biggest deal is continuity between us and other
species. Discontinuity is an
illusion. But by and large, species are
distinct and stable and persist or don’t. Species hang between one and 10
million years. Institutions also clump
(churches and banks) required to adopt too much institutions die.
The genome being
conservative/hemmed in means that when intensive breeding for characteristics
happens side effects come along. Even
genetic mutations are buried in the population at large. Artificial evolutions on computers have a
little less history. They conserve
quickly.
But
longer runs will teach us a lot about the dynamics of evolution. After a point is bilateral symmetry
necessary? AI people will take over
evolution studies. This upsets biologists.
But biologists gave us evolution.
Since then?
Artificial
evolution folks, showed it isn’t a biological process. Its
a tech, math, info and biological process. A law of physics.
Evolution comes by multiple causes.
At different times and levels different types of causes are
emphasized.
Among
the dynamics are 1) symbiosis 2) non-random 3) satationism (cluster jumps) 4)
bias towards certain forms (separate limbs, bilateral etc.)
THE BUTTERFLY SLEEPS
Genes are controlled by
regulatory genes. Genes are controlled
by genes. Like the net, math is
evolving. Newtonian math was suited to
the linear physics, but…Net math requires computer power. Many magnetic fields are the same net. In net math, counter intuitively, a small
cause can have a disproportionately large effect. The butterfly effect. Run
with 10,000 variables networked many times he found they settled into “basins”
Order out of chaos for free. The number
of basins is the square root of the # of genes. This is the same in cells.
Its not gene selection, but gene architecture that determines
function.
In
the lap game a circle of people all sit on each other’s laps. This is
tautology. Living systems are like
this. Strong loops are like Escher’s
that go up to the bottom. Like chicken
function spawn functions. Communication
functions create democracy. This has
the previous characteristics from molecules to democracy. Can a machine make a machine of more
complexity? Self organized order in
evolution systems from transistors to computers to virtual reality. Kaufman found the most adaptable system
didn’t have too many, nor too few connections.
10 per was optimum, no matter size.
Too many connections and woo many control your work. In flocks there is locked repetition. (*$# too many connection and gas with too
few. There is a sweet spot between and
at the edge of chaos. Its oscillation
is self-reinforcing.
RISING FLOW
Earlier studiers of heat had no unifying theory. Then Carnot came up with basis for
thermodynamics law. Biology has no
overarching principle. There is co
mplexity, but not rule for it. Now law
of rising order. After the big bang
there is entropy and something else that creates order in galaxies and life to
higher levels of order. Like a crest on
a wave, life. The direction of progress
seems too anthropocentric. A history of
racism and out place in the universe (periferal) have killed optimistic
progress as a belief. At 4 billion all that
survive are equally evolved for their niches.
If we videotaped the course of evolution and played it backwards to
aliens, could they tell if it was going forward or backward? Yes.
It is directional. NewDarwinism
says evolution is unpredictable. Post
Darwinism see it as inevitable when evolution shook off god they thought they’d
gotten rid of purposes and direction.
But machines have purpose and directions (*$# GOD. A –life acquires an agenda of its own. Systems, given initial conditions, may be
deterministic & unpredictable.
Deterministic systems maybe teleological. If you don’t like purpose or direction, you may claim
tendency.
There are seven trends:
1) irreversibility:
Animals don’t regress to past niches, hard won attributes aren’t given up. 2)
increasing complexity (seems to be true) 3) increasing diversity. Species types don’t go up. Their disparity does. 4) Increases in population. And super groups
(hives) add to the count. 5) increasing
specialization. Specialized cells and organisms rarely go back to the more
general functions. 6) Increasing
codependence . Early stuff are directly
dependent on a rock. Plants are tied to
the earth, animals to specific plants, man to animals, but we get less tied to
environs and more to each other 8) Increasing evolvability. Certain A Life innovations are
evolutionarily pregnant with possible evolutionary routes. Evolution selects for things that can
evolve. Like animal segmentation,
variability (which would survive and make more variability). Since life came from non-life, evolution
preceded life. And the dynanism of change
is necessary to continuation.
A
summary of evolution’s evolution may be hypothesized as follows. In the beginning, evolution started as
varying self-replication that produced enough of a population to induce natural
selection. Once population bubbled up,
directed mutation became important.
Next symbiosis became a major mover and shaker feeding off the change
produced by natural selection. As forms grew larger, the constraints of form
set in. As genomes grew in length, internal selection began to rule the
genome. With the cohesion of the gene,
speciation and species level selection kicked in. With organisms of sufficient complexity, behavioral and somatic
evolution emerged. Eventually, when
intelligence came on the scene, Lamarckian cultural evlution took over. As we humans introduce genetic engineering
and self-programming robots, the make up of evolution on Earth will continue to
evolve.
As
life becomes more hierarchical – genes, cells, organisms, species – evolution
shifts its work.
Evolution
breeds more evolution. Life is the substratum on which it runs.
PREDICTION
MACHINERY
Chaotic
systems are high dimensional and predictable in the short run generally
only. Chaos is not randomness. In guessing roulette you don’t need perfect
prediction to win. Humans can predict
the very near future. How about
computers? We have general guidelines
to evaluate the profitability of a strategy, we don’t plan for every
eventuality. They run past situations
the prediction machine guesses 20. They
tell it which is correct, it learns to discern causes. The machine develops intuition. Strategy in the Iraq war was based on high
dimensional computer predictors that keyed in on a power. Stocks and war = world? Most long term predictors are wrong, short term
right. In predictions, some things are
invariants. Then the model growth
curves: Rise, peak and fall. Cycles, like generational cohorts taking
different jobs than their predecessor.
Adaption is change in the service of the non-changing. Without some prediction we die. All facts are now being digitized and
charted. Complex systems are those
first talk to themselves. About what? Usually the future. Since the 70s we’ve had pictures of the
whole globe. Its wounded computers
model future trends for pollution, population, consumption and see
disaster. But such models don’t get the
requisite complexity and never stabilize at another pattern. They just crash.
WHOLES,
HOLES, AND SPACES
In
1959 the MACY conferences brought hundreds of scientists came to a conference
where the topic was “self-organizing systems – how an organism bootstraps
itself to life”
The
speaker laid out 3 principles?
Memory
(computers do this), differentiation (ie the random mutation of biology),
subordination (or executive function).
Why
did cybernetics die?
The
funding went to artificial intelligence and it failed.
The
computers weren’t advanced enough then.
This
hole has been filled by amateurs.
Science
itself is a system of individual out of control nodes without a center.
He
is fascinated by the holes and wholes of knowledge. These often lead to Kuhn’s paradigm shift.
These
come from the anomalies that lead to questions. The end of this book is a list of questions.
Obvious
things make the best starting points for questions. Such as; why is it when we see something the word comes up? Staircases, no matter how twisted bring up
the word staircase.
QUESTIONS:
What “Happened” when
the word “emergent” was used. What
ignorance does the use of the word mask?
If life grows more
complex, why? And is complexity more
efficient than simplicity? What is
complexity? Is a brain more complex than an economy or cucumber?
What is the minimum
variety for processes such as self-organization, evolution and learning and
life?
What cannot be
simulated?
Is the simulation fake
or real on its own?
The prairie meadow
makers asked “How far back can you compress a meadow into a seed? At what level are the parts not the sum of
them?
Are the parameters of
stability different for a more complex system?
Why do things get
stable? Why do things go extinct?
What makes a good
god? Can evolution create its own
teleological purpose?
The Institute of
Science Information has a program that finds like citations in articles. Then you can see who else uses an idea. Does the idea produce offspring? Does it die?
This swarm of hypertext
is a very large document. The
individual authors are contributing, but perhaps unaware of the outcome
coming. But sooooo many links. It is easy to get lost with everything
having the same weight, and no center.
In this late age of print, writers and readers still conceive of all texts, of text itself, as located in the space of a printed book. The conceptual space of a printed book is one in which the writing is stable, monumental and controlled by the author. Electronic print is fluid and interactive.
Blackboards encourage
modification and updating. Quill pens
carefulness and attention to grammar, tidiness and controlled thinking. A printed page: rewritten drafts, proofing,
introspection, editing. Hypertext
creates: telegraphic, modular, nonlinear, cooperative works. They are swarmed. It is, says Brian Eno, The way we organize our thoughts and the
way the world itself must be organized.
The space of knowledge in ancient times was a dynamic oral tradition. Rhetoric was structured as poetry. The texts were debated.
When the printed page
came it got monumental. Gone was the
role of the reader. The unalterable
progression of ideas across the pages cave an impression of authority.
The internet is the
largest functioning anarchy in the world.
Not just chaotic, but perpetually reorganized.
Knowledge in print
shaped the very idea of the canon, which implied fundamental truths, from which
knowledge progressed, but never retreated.
Hypertext is
fundamental to post modernism in which the reader challenges the author for
control of the writing space. Meaning
is a multiple swarm of interpretations.
A differential network traces references to something other than itself
always. Signs that point to signs.
AN aggregation of
fragments is the only kind of whole we have now.
There is no central
keeper of knowledge. Just curators of
particular views. I am becomes
hallow. We are conglomerates of views,
votes and spendings.
Orwell was totally
wrong about this, computers kill authority.
How large is the
collection of all possible thoughts?
How much do humans occupy. AI
will probably have access to more and think differently than us.
THE NINE LAWS OF GOD
1) Distribute being
The spirit of a
beehive, the behavior of an economy, the thinking of a supercomputer and the
life in me are distributed over a multitude of smaller units. The sum is more than the parts. Something from nothing comes from smallers
interacting.
2) Control from the
bottom up
When everything is
connected to everything in a distributed network, everything happens at
once. A mob can steer itself, and in the territory of rapid, massive and
heterogeneous change, only a mob can steer.
Its like the economy.
3) Cultivate
increasing returns
That is because
whatever works is given more resources.
All large sustaining systems do this.
Economics, biology, computer science and human psychology.
4) Grow by chunking
The only way to make a
complex system that works is to begin with a simple system that works. Attempts to instantly install a highly
complex organization inevitably lead to failure.
5) Maximize the fringes
Heterogeneity
creates. Uniform systems need huge
changes to maintain. Diverse
heterogeneous systems do thousands of microrevolutions daily.
6) Honor your errors
Out side of
conventional wisdom there are errors.
That is where all substantial
advances are to be found. Error
and correction are a part of all creation.
Evolution is systematic error management.
7) Pursue no optima:
have multiple goals
Simple machines can be
efficient, but complex adaptive machinery cannot be. Have many masters. An
adaptive system must trade off between exploiting a know path of success or
diverting resources to exploring.
8) Seek persistent
disequilibrium
Neither constancy nor
relentless change will support a creation.
Equilibrium is death. No
equilibrium is too. Keep stable, but
growing.
9) Change changes
itself
Evolution is about how
an entity changes. Deeper evolution is
about how the rules for changing change.
The hallmark for the
mechanical age was mechanical design.
The hallmark of the neo-biological age is the return of the organic.
·
The intensely biological nature of the coming
culture derives from five influences:
·
Nature will continue to be the infrastructure of
human experience.
·
Machines will become more biological in
character.
·
Technological networks will make human culture
even more ecological and evolutionary.
·
Engineered biology and biotechnology will eclipse
the importance of mechanical tech.
·
Biological ways will be revered as ideal
ways.
In the coming age all
will be born more than made.
OUR INDIVIDUAL SOULS -
FROM PLOTINUS
If you think our soul is
part of the greater soul. It is
wrong. It is the greater soul. Because your analogy is with the physical. But because milk and paper are white doesn’t
make them part of white. They are white
. 2 is but isn’t really a part of 10. Biology is a part of science but not a piece
of science. If we were said to be mini
replicas of the soul, we would be independent.
Not so? The eye and ear don’t
have separate soul types. They feed
differently because they’re different organs.
But they inform the soul. The
soul is a unity of everywhere, but with different functions. The soul is universal and rational. The soul comes as an explanatory/ expression
of intelligence and like this rational laws of nature is undying and
undifferentiated. How has the soul made
the world, and why, . While individual
souls govern each part of the world?
THe intelligence is the law of nature, the soul its movement.
We say body comes into
a soul, but this is a convenience. For
never has there been a body without one.The cosmos is ensouled. The soul is
after nature. This imitates it.It
creates bodies. The shape the cosmos too is a result of soul comint into
it. Everything soul gives to body is an
image of its life. Thus When people
make idols soul is attracted into them. There is life in rocks to the extent that they can hold it. All
with soul contemplates the intelligence.
Thecosmos has all it needs and is limited only by form. ----calty it goes
towards and away from the form of form (one)via intelligence insoul. Soul mutates into appropriate forms. They don’t do so deliberately, it is as
water going downhill. Longing and
gravity are the same. intelligence.