[Kabar-indonesia] Science: Why Aliens Love A Good Moon; Universal Law of Networks; Predictability

[Joyo at aol.com]

5 articles: 

- Life is unpredictable - get used to it

- Why Aliens Love A Good Moon

- The net reloaded -- Is there a universal law of
  networks, or is it all about how they are linked up?
  The answer could mean life or death for the internet

- Distorting the web -- Does scale-free network theory
  hold for information on the World Wide Web? 

- US Congress starts to believe in climate change

----

New Scientist magazine [UK] 
issue 2558 / 01 July 2006

Life is unpredictable - get used to it

By Michael Bond

When did you first notice the way humans deal with
unpredictable events?

I was brought up in Lebanon, where we always recreate
memories, revise experiences and read more into them
than necessary. During the war there, when I was 15, I
wanted to be a philosopher. While I was hiding in
basements I read William Shirer's Berlin Diary: The
journal of a foreign correspondent 1934 to 1941. It
made me realise three things about the people around
me: that they were always predicting (wrongly) that
the war was going to be "solved" soon; that they
seemed confident about their estimates for the future
even though crazy events were happening all the time;
and that after the crazy events had happened, people
acted as if they were predictable. I realised that you
can find an infinity of narratives to fit your data.

Why do we all fall into this trap?

Our brains operate on autopilot most of the time. In a
primitive environment, that was fine, since most
random variables we encountered then were what I call
"type one" randomness: things like the throw of a die,
a person's height or weight, or the number of calories
we consume on a particular day - the randomness
disappears under averaging and is measurable and
well-behaved. The variation in the calories we consume
each day disappears when you add up the calories we
consume over time; the throws of a die will average
out.

Today, however, most random variables we encounter are
"type two" randomness: socioeconomic variables that
can impact the total in a disproportionate way. I call
these unpredictable ones "black swans".

Can you give an example?

Individual wealth is a good example. Bill Gates's
contribution to the wealth of the world is so huge you
cannot discount it. The first world war was "type two"
randomness - no one could have predicted its magnitude
the year before it happened. The same goes for the
invention of the computer. There was not much "type
two" randomness in the primitive environments in which
we developed our intuitions. The few extreme events we
did encounter were not repetitive enough for us to
learn from them, and they were often so catastrophic
that they terminated the populations exposed to them.
Sadly science has not been too interested in this type
of randomness. Scientists do not want to tell you what
you don't know, and it devalues the field of
statistics, given that the statistics we use are based
on "type one" randomness.

How bad we are at predicting "black swans"?

Our track record is quite dire. Look at the net,
computers, lasers. The internet was designed as a
military system, not for chat rooms. The person who
first marketed computers didn't think he would sell
more than five. The laser was designed by a physicist
who had no idea how it might be used. You can't even
forecast something that would affect us tomorrow -
revolutions, wars, epidemics, political changes,
economic variables.

But don't people make predictions based on history all
the time?

As well as our ability to concoct empirically flawed
narratives to explain past events, there are biases in
history that we don't seem to be aware of and that
make us overestimate the causal links between events:
for example, when you see only the winners and not the
losers. When you look at the fossil record, you see
only the species that left a fossil. You cannot make a
generalisation of all species just from fossils - you
have to take into account the species that left none.
History has a lot of hidden pockets. You can't take it
any more seriously than a visit to a museum.

Does science suffer from this?

Yes, because of the way research gets published.
Research that yields no result does not make it into
print. The problem is that a finding of absence and an
absence of findings get mixed together. Science as a
discipline is very sceptical, but scientists are not
necessarily so. It is exactly like financial markets.
Markets can be extremely rational while dominated by
entirely irrational individuals. It leads to
overconfidence.

What are the consequences of this overconfidence?

It's massively dangerous. Let me give you an example.
I was invited to give a talk on prediction to civil
servants at the Woodrow Wilson International Center
for Scholars in Washington DC. I told them that their
social security forecasts were dangerous to society.
At the end of the session a gentleman from the
audience wanted to speak to me privately. He showed me
oil price forecasts by his department for 25 years
ahead: in January 2004 they estimated it would be $27
a barrel; six months later they revised it to $74 a
barrel. If you had to double your forecast six months
after making it, wouldn't you realise something is
wrong with your forecasting? So we're depending on
flimsy government forecasts. And there's a track
record to our prediction failure: think of someone
making an inflation forecast in Germany in 1913.

My point is that a forecast is irrelevant unless you
have an error rate on it. But if this happened, these
people would realise there was no point in
forecasting, because their error rate would be so
monstrous. It wouldn't be a problem if we knew our
limits in prediction. People are naturally
over-optimistic, which is good because it helps us get
by in the world. It's fine for individuals and
corporations, but it's very bad for societies and
governments when policies are set by such wildly
unfounded predictions.

So successful forecasting is just luck?

Yes, and you need a lot of luck to forecast things
accurately. There are enough people predicting crazy
events for one of them to get it right. After a crazy
event, such as 9/11, you will always find someone who
predicted it out of luck, but they'll think they
predicted it by skill.

"Enough people predict crazy events for one of them to
get it right"

Should we change the way we view the world?

We cannot help being fooled by randomness. We're too
impressionable. I was in London when the second
terrorist attack happened and I automatically behaved
like anyone else, ducking for safety. Then I realised
that my biggest danger in London came from my jet lag
and being used to traffic driving on the other side of
the road. We should worry about preventable sources of
death. I should worry more about how much sugar I put
in my tea than whether I am going to be hit by
terrorists. The key is not to try to stop being a
fool, but to be aware of when it matters not to be a
fool. If you can't do anything about a problem, it's a
waste of time analysing it.

---

Nassim Nicholas Taleb is an applied statistician and
essayist specialising in the problems of uncertainty.
He has been a derivatives trader in New York and
London and is currently a professor in the sciences of
uncertainty at the University of Massachusetts at
Amherst. His book Fooled by Randomness (Random House,
2001) was the bête noire of Wall Street for its
conclusion that success is usually the result of luck
rather than judgement. His new book, The Black Swan,
will be published by Random House later this year. In
addition to his scientific and literary interests,
Taleb's hobby is to poke fun at those who take
themselves and the quality of their knowledge too
seriously.

-----------------------------------------------------------------

New Scientist magazine [UK] 
issue 2558 / 01 July 2006

Why aliens love a good moon

By Marcus Chown

IN A distant planetary system, a dozen giant moons
buzz around a planet five times the diameter of
Jupiter. Several of the moons are monsters as large as
Earth, their surfaces continually rocked by violent
earthquakes and volcanic eruptions that would put
Krakatoa to shame. Perhaps they seem a vision of hell.
Yet a growing number of researchers believe such moons
could be cosmic oases. "They might be the most likely
places to find life in the galaxy," says Caleb Scharf,
an astrobiologist at Columbia University in New York.

"Exomoons might be the most likely places to find life
in the galaxy"

Although no one has ever seen a moon orbiting an
extrasolar planet, observing the satellites around
Jupiter and Saturn has led some astronomers to think
that "exomoons" capable of supporting life are almost
certainly out there. What's more, Scharf believes that
life could thrive much farther away from stars than
planetary scientists ever thought.

This is great news for astrobiologists hoping to find
hospitable worlds outside the solar system. Life is
abundant on Earth because our planet lies at just the
right distance from the sun for liquid water. This
so-called "classical habitable" zone is
disappointingly narrow. The heat from the sun would
boil any water on the surface of a planet orbiting
slightly closer than Earth, while any water on the
surface of Mars, which lies farther away from the sun,
would freeze (see Diagram).

Now the idea that life will only be found in the
habitable zone is looking a little simplistic. Many
planetary scientists who have been studying Jupiter
and its four largest moons Ganymede, Callisto, Io and
Europa think that despite Europa's frozen surface,
this moon could be one of the most promising locations
for life in our solar system.

That is a dramatic consequence of the four large moons
orbiting very close to Jupiter and each other.
Gravitational interactions between the moons are
strong and this sets up orbital "resonances". For
instance, for every time Ganymede circles Jupiter,
Europa completes two orbits and Io four. This means
the moons regularly end up in the same arrangement
relative to one another and repeatedly yank each
other, causing their orbits to be elliptical. In such
orbits, they swing in close to the huge mass of
Jupiter, which puts them through enormous squeezing
and stretching, warming their interiors through a
process called tidal heating.

"Consequently, we see Io with many active volcanoes,"
says Scharf, "while Europa very probably has a giant
ocean beneath its global ice sheet." That has big
implications for the hunt for extraterrestrial life,
not only on Europa but also in other solar systems.

Beyond the zone

"Since planetary scientists recognised the possibility
of life on Europa, they have realised that satellites
of giant planets could be habitable," says Richard
Greenberg of the University of Arizona's Lunar and
Planetary Laboratory in Tucson. "Such bodies, which
lie far outside the habitable zone, may be even more
common platforms for life than planets in the
habitable zones."

Tidal heating makes Jupiter's moons far warmer than
you would expect from the sunlight falling on them.
"Exactly the same thing will happen in moon systems
around extrasolar gas giants," Scharf says. His
calculations have revealed that exomoons as large as
Earth could experience at least 100 times as much
heating as Io. Scharf thinks this combination of tidal
heating and warmth from the parent star could keep
exomoons cosy enough for water to remain liquid.

"I believe it could double the size of the habitable
zone around a star," he says.

There could well be an abundance of such worlds.
Astronomers have so far spotted more than 170
extrasolar planetary systems, and all of them contain
gas giants up to 10 times the mass of Jupiter. If
these planets are anything like the gas giants in our
own solar system, they could each harbour several
moons almost as large as Earth. That means the tally
of exomoons in these planetary systems alone could run
into thousands.

In April, Scharf published the results of a study of
74 gas giants orbiting more than 90 million kilometres
out from their stars
(www.arxiv.org/abs/astro-ph/0604413). At this
distance, which is almost as far out as Venus is from
the sun, Scharf reasoned that any putative moons
around the planet would be able to resist the
gravitational force of their parent star and remain
stable. Much closer and the star's tidal forces would
quickly rip the moon apart. By examining the mass of
each planet and its proximity to its star, he
calculated that many of the gas giants he studied
harbour regions where tidal heating would make any
moons warm enough for liquid water.

Assuming watery exomoons do exist, could life really
arise on them? One problem would be the violent
upheavals of the habitat. A moon experiencing 100
times the tidal heating of Io should have much more
volcanic activity, and that could reduce a habitable
world to nothing. Scharf sees no reason to worry,
however. He points out that observations made by the
Hubble Space Telescope and by NASA's Voyager and
Galileo probes show that the volcanic activity on Io
is patchy. Some places are in severe turmoil, while
others are relatively quiet. "There may be quieter
locations on giant exomoons where life could exist,"
Scharf says.

If many exomoons around gas giants are as large as
Earth, their sheer size would boost the prospects for
life. Only the gravitational pull of such a large body
would be able to hold onto a thick, sheltering
atmosphere. What's more, a large moon is more likely
to have a magnetic field. Just as Earth's magnetic
field deflects the energetic particles spewed out by
the sun, so a moon's magnetic field will protect life
from damaging radiation. "Big moons will be the safest
places," says Scharf.

One day we might discover whether he is right. We can
already detect planets around nearby stars if they
pass in front of the parent star. Later this year, the
French space agency CNES plans to launch the Corot
telescope. It will simultaneously monitor 12,000
stars, looking for small dips in a star's brightness
when a planet passes in front of it. NASA aims to go
one better in 2008 with the launch of Kepler, a space
telescope that will look for such "transits" in
100,000 stars at once.

The planet hunters might find even more than they
hoped for. According to Jean Schneider of the Paris
Observatory in France, massive moons tugging a gas
giant first one way and then another would alter the
precise timing of a transit. "If the moon is as large
as the Earth, it will also make its own detectable
transit," he says. By observing such effects,
Schneider believes that Corot should be able to detect
moons with diameters 50 per cent bigger than the Earth
as long as the planets orbit close to their parent
stars.

In picturing habitats for life, our horizons may have
been too limited. Scharf maintains that we should
expand our search beyond looking for planets that
match Earth. "I'd hate to miss spotting life because
we abandon systems with no terrestrial planets." 

-------------------------------------------------------------------------

New Scientist magazine [UK] 
issue 2558 / 01 July 2006

The net reloaded

Is there a universal law of networks, or is it all
about how they are linked up? The answer could mean
life or death for the internet

How the internet resists attack

By Kim Krieger

IT WAS the late 1990s, and a group of physicists had
it all figured out. A universal rule seemed to explain
a vast range of behaviours in social, biological and
computer networks. Everything from how ecosystems
evolve to how the internet works seemed to follow the
same statistical pattern, called a "power law". What
came as a surprise was that this simple law implied a
deep underlying principle for all these networks.

Researchers began working with power-law models of the
internet and other systems - and that's when they made
a startling observation. If their models were correct,
eliminating the most highly connected computers would
cripple data flow. Their work appeared to show that
the internet and systems like it had an "Achilles
heel", and that contrary to popular belief a few
carefully targeted attacks could bring down the entire
network. The finding spawned a whole branch of
research known as "scale-free" network theory (New
Scientist, 13 April 2002, p 24). The name, which
derives from the idea that members of a power-law
network have no "typical" number of connections, has
since graced the covers of the most prestigious
scientific journals.

Now a growing number of biologists, mathematicians and
computer scientists are complaining that the idea has
been overhyped, and that the power-law pattern does
not reveal anything fundamental about what makes these
networks tick. John Doyle, an expert in control and
dynamical systems at the California Institute of
Technology in Pasadena, is among those who dismiss the
idea that scale-free theory can make useful
predictions. "The problem isn't hype, the problem is
it's wrong," he says. Researchers like Doyle are
developing more sophisticated tests of power laws and
models of what they mean. They have proposed theories
that take into account the evolution, design and
structure of specific networks, and their ideas have
led to statistical methods for modelling forest fires,
protein-protein interactions and other biological
phenomena.

At heart, power laws are simple. If you plot the
proportion of "nodes", or members of the network,
having a certain number of connections versus that
number of connections, a power law appears as a curve
shaped like a suicide ski slope: declining steeply at
first, and then ever more gently. This reflects the
fact that most nodes have only one or two connections
to others, while just a handful of nodes have
hundreds, even thousands of links.

So where does the controversy come from? In physics,
power laws give powerful insights into simple systems
like phase transitions from liquids to gases. The
systems that researchers began referring to as
scale-free, however, were more complex. "It was a
fascination for many of us," says physicist
Albert-László Barabási of the University of Notre Dame
in Indiana, a leading author of the original
scale-free papers. "So many networks have absolutely
nothing to do with each other, but they all end up
being scale-free."

Perhaps they should not have been so impressed, says
Michael Mitzenmacher, a computer scientist at Harvard
University. "I think that's a sort of lack of
historical knowledge." He says the notion of a power
law is ill-defined, and what, if anything, it
signifies outside of simple systems has been debated
for the past 80 years.

The most popular model to explain why power-law
distributions occur in networks is known as
preferential attachment - the idea that in general
well-connected things tend to garner ever more
connections. The first widely cited appearance of this
model was in a paper in 1925 that described the
power-law distribution of species among genera. It
also proposed an explanation: genera with many species
were more likely to have a random mutation in one of
the species that then spawned a new one, so genera
with many species added more species faster than those
that were species-poor. In 1959, in the journal
Information and Control, Benoit Mandelbrot, famous for
his work on fractals but also prolific in statistical
analysis, confronted Nobel prizewinning economist
Herbert Simon in a heated debate over whether the idea
of preferential attachment has any validity. The
argument is still going strong today, as Mitzenmacher
pointed out in a 2004 paper in the journal Internet
Mathematics.

"Power laws don't mean anything. The devil lurks in
the details of the network"

Preferential attachment has earned itself the most
play on the World Wide Web, where search-engine
companies Google and AltaVista have used ideas from
scale-free theory to justify their system of ranking
the most connected web pages at the top of their
search results. Scale-free thought, however, doesn't
go much further than noting the existence of these
highly connected pages and predicting that they should
become ever more highly connected.

The problem lies in making the leap from scale-free
statistics to the underlying process that determines
how a particular network behaves. "Power laws don't
necessarily mean anything," says Mark Newman, a
physicist at the University of Michigan in Ann Arbor.
"There's this tendency for people to see a power law
somewhere and assume that this process is going on.
This is a logical fallacy, like saying, 'Bears like
honey, my wife likes honey - therefore my wife is a
bear.'"

Consider the router network that directs data through
the internet, and which was the subject of a 2000
Nature paper, co-authored by Barabási, that brought
scale-free thought into the mainstream. The router
network is not unlike a traditional telephone network,
only with lines stretched over the globe connected to
routers that receive information packets from one link
and send them out on another. If you are sitting in
London and send an email to a friend in Washington DC,
your email gets routed from your computer to your
internet service provider (ISP), which may send your
email to a data router in New York, which passes your
email to a router in Washington, which tosses it to
your friend's ISP and on to your friend's computer.

In this whole sequence, the only computers linked to
more than a few others are those operated by the ISPs,
which will connect to hundreds or thousands of
subscribers. The major data routers will typically
have two to six links. If a graph is plotted showing
how many of the internet's computers have a given
number of connections to other computers, the vast
majority will pile up on the far left, linked to just
a few others. Trailing out at the far right of the
graph will be the handful belonging to the ISPs. The
resulting curve, which falls steeply to start with and
flattens out at its tail end, represents the power
law.

Knock out the net

After finding that this power law described the
statistics of internet routers, Barabási and
colleagues used a theoretical network with the same
proportion of highly connected routers to model the
net, and from these models came the idea that
eliminating highly connected routers could shut the
net down. Doyle argues that this approach, while
superficially attractive, ignores a simple fact: the
highly connected routers are ISPs on the edges of the
network, close to end users (Proceedings of the
National Academy of Sciences, vol 102, p 14497). Take
down highly connected routers around the US, and
you'll knock out ISPs that serve users in certain
neighbourhoods. It would, for sure, be an annoyance
for localised clusters of internet users, but the
majority of traffic flowing around the world would
continue unscathed. The bulk of internet data - be it
financial trading, web surfing or massively
multi-player online games - will flow unimpeded,
through routers that have only a few links each.
"There are so many routers that you'd have to destroy
a ridiculous number of them before you'd really
cripple the internet," Doyle says.

The router example reveals the weakness of scale-free
models as a predictor of how a system will behave.
They simplify systems, leaving out the details in
which the devil can lurk. "The approach of scale-free
models was diametrically opposite to the types of
models that are truly useful, which are grounded in
specificity," says Evelyn Fox Keller, a historian of
science at the Massachusetts Institute of Technology.
A useful model would specify what the nodes do, where
they are in the network and how their connections
work.

A few researchers have proposed ways to model
power-law networks to make more useful predictions.
While this has raised some hackles among supporters of
scale-free theory, even Barabási concedes that the
original ideas are not the whole story. "It is
absolutely correct that there are lots of other
properties of the networks that are just as important
as the scale-free: some are fractal, some are not,
some preferentially attach, some don't," he says.
Research on networks has evolved considerably in the
seven years since scale-free ideas made their
entrance, he says.

One leading alternative is known as "highly optimised
tolerance", or HOT. It originated in the late 1990s,
and is the basis for a more realistic internet model.
Its most vocal proponent is Doyle. "Real engineering
and real biology are really complicated. Yet we want
simple models," he says. "With HOT we're trying to
explain, in as simple models as the scale-free models
that are more faithful to the specifics of the domain,
what is general about complex networks."

The key idea of HOT is that networks evolve according
to what they are designed to do and their physical
constraints. Real systems behave differently at
different size scales and locations. Take biological
cells and tissues. At one level they look like a sea
of proteins. Zoom out and you see organelle structure.
Zoom out further and you see bunches of cells stuck
together to form a particular tissue. HOT's
recognition of complex systems' "self-dissimilar"
structure sets it apart from scale-free theory, which
treats systems as the same if they have the same
statistics.

Yet both theories share an appreciation for power
laws. HOT uses them to gauge which aspects of the
system are important. The exponent in the power law
controls the steepness of the curve, and is in turn
determined by the specific goals and constraints of
the network. In the simplest type of HOT model, called
the "profit-loss-resource" model, a complex system is
boiled down to a conflict between the resource and the
loss, and HOT assumes there is one optimal way to set
up the system.

As an example, take the management of forest fires. If
the profit is timber, the loss is the amount of land
that burns and the resource is land used for fire
barriers, the essential trade-off is planting trees
versus protecting them with fire barriers (Proceedings
of the National Academy of Sciences, vol 102, p
17912). Another application of HOT is working out how
to create an optimally navigable website, in which the
conflict is between small file sizes for fast
downloading and minimising the number of clicks before
the user finds the desired information. The router
configuration of the internet is more complicated -
there are multiple goals, such as speed and volume of
information flow - but Doyle says HOT can treat it as
an optimisation problem too.

Where do networks go from here? The work that Doyle
and others are doing is "generally excellent",
Mitzenmacher says. "They're trying to define what it
means to be scale-free." However, he cautions that
although HOT is alluringly useful, optimisation
arguments cannot be the whole answer any more than
preferential attachment was. If the universal law of
networks is wrong, it seems, researchers must continue
to press on for better models, better testing and
better understanding - all seasoned with a healthy
dose of scepticism.

------------------------------------------------------------------------

New Scientist magazine [UK] 
issue 2558 / 01 July 2006

Distorting the web

Does scale-free network theory hold for information on
the World Wide Web? In contrast to physical internet
routers, well-connected web pages do appear to collect
links faster than less well-connected pages. Now
researchers from Carnegie Mellon University in
Pittsburgh, Pennsylvania, are studying how search
engines like Google have altered the statistics of the
web. When users enter a search query, most never
explore beyond the first two pages of hits. If they
link from their own site to one of the pages they
find, it will be one of the top-linked pages. By
listing the same top 10 or 20 pages whenever a given
query is received, search engines elevate those pages
to celebrity status, and they gain links at an
inflated rate. That distorts the distribution of links
and makes it difficult for new pages to crawl up the
stack.

This distortion makes for an interesting problem in
network theory. A power law can no longer fully
describe the connectivity of the web. Some researchers
have suggested that putting randomness into search
results could increase the usefulness of search
engines by highlighting pages that would otherwise be
at the bottom of the heap. Whether this would work is
uncertain, but predictive models might provide an
answer.

All in your genes

What do power-law statistics say about complex
biological systems? "If you realise that something has
a power law property, you don't know yet whether or
not that tells you how the network evolved," says
Chris Wiggins, an applied mathematician who works in
computational biology at Columbia University in New
York City.

Take networks of regulatory genes. Right now,
researchers have little grasp of which DNA segments in
a cell interact to control the rates at which genes
are transcribed and expressed. The best they can do is
to make a table, Wiggins says - "just a list of which
genes are talking to which other genes". It appears
there might be hubs: certain genes that talk to many
other genes. How significant these hubs are to the
overall network is unknown, however. Wiggins's team
approaches gene networks as a machine-learning
problem: they plug in every bit of data they know
about a set of genes and let a computer tease out
factors that are relevant to the network's behaviour,
whether it's highly connected genes, an abundance of
triangle-shaped links, or other such patterns. They
have recently applied this technique to characterise
protein-protein interactions in fruit flies
(Proceedings of the National Academy of Sciences, vol
102, p 3192).

---------------------------------------------------------------------------

New Scientist magazine 
issue 2558 / 01 July 2006

US Congress starts to believe in climate change

IN WASHINGTON DC they are starting to believe. A report commissioned by 
Congress from the National Academy of Sciences has concluded that the controversial 
"hockey stick" graph of global warming is real, and that the spike in 
temperatures has probably been caused by human activity.
“The report concludes that the 'hockey stick' graph of global warming is real”

The NAS report says that the past few decades have been the warmest in the 
past 400 years and that it is "probable" that the last 25 years have been the 
warmest since AD 900. Sherwood Boehlert, chairman of the House science 
committee, requested the report in November 2005 in response to the political debate 
around the work of palaeoclimatologist Michael Mann of Penn State University at 
University Park. Mann's work examined average temperatures over the past 1000 
years. When he plotted the results they showed that for the first 900 years 
there was little variation - like the shaft of a hockey stick - but that there 
has since been a spike of massive warming - the blade of the stick.

The NAS report finds that conclusions drawn from Mann's reconstructions of 
warming bear weight, but it says that there is too little evidence to be 
statistically certain that a given year or single decade stands out as unusually warm.

Some people try to promote the idea that the recent warming is unconnected 
with human activity, says Martin Rees, president of the Royal Society in London. 
"This report renders that notion much less plausible."

------------------------------------------
Joyo Indonesia News Service
------------------------------------------