JOURNAL: TURBULENCE

A short PBS interview with Benoit Mandelbrot and my compatriot, Nassim Taleb, on the current financial/economic crisis.

It underlines my ongoing fear that this crisis will recast the world at a fundamental level. Why? Our simplistic, slow, and fractured 20th Century control system isn't capable of stabilizing a financial/economic system of this complexity/speed/size once it becomes very turbulent. As a result, the global system will follow its own course, dictated by its hypercomplex internal dynamics and feedback loops, destroying everything that gets in its way.

Like Nassim, I hope I'm wrong.

For the engineers and pilots out there, our current situation is akin to trying to fly an F-22 at the edge of its performance envelope with only cables and pulleys for control inputs. NOTE: cables and pulleys are the classic control system for 20th Century aircraft. The pilot inputs a control movement on a pedal or the stick and the cable/pulley system translates the input into a movement of the control surface (elevator, etc.). Direct pilot control is possible because the plane wants to fly -- i.e. is stable. In contrast, in order to get high performance, modern designs are made to be unstable. As a result, modern aircraft require control system inputs every 1/32 of a second or more, all of it done automatically. If these control inputs aren't made, the plane will rapidly exceed its design capacity and lose structural integrity.

One way to look at it: We are all Iceland now.

Comments

[George]

Paul Virilio on the current financial crisis

"[...]A statement by Hannah Arendt was the banner of your demonstration: "progress and catastrophe are the two faces of the same coin". Is this where we have come to with the 'crash of the stock exchange'?

Paul Virilio:

Well, of course. In 1979, at the time of the mishap at the Three Mile Island nuclear plant in the U.S., I did mention the occurence of an "original accident" - the kind of accident we bring forth ourselves. I said that our technical prowess was pregnant with catastrophic promises. In the past, accidents were local affairs. With Tchernobyl, we have entered the era of global accidents, whose consequences are in the realm of the long term. The current crash represents the perfect 'integral accident'. [...]

http://cryptome.info/virilio-crisis.htm

[The Dark Avenger]

There's a book on this subject,
The Logic of Failure:
http://www.amazon.com/Logic-Failure-Recognizing-Avoiding-Situations/dp/0201479486
The Chernobyl atomic-plant explosion, observes Dorner, was entirely due to human error involving the breaking of safety rules by a team of experts who reinforced one another's puffed-up sense of competence. This German psychology professor believes people court failure through sloppy or ingrained mental habits, whether the mistakes involve cleaning dead fish out of a garden pool, adding rooms to a schoolhouse, launching economic development programs in Africa or forecasting oil prices or the scope of the AIDS epidemic. Things go wrong, according to Dorner, because we focus on just one element in a system complicated by interrelationships; we apply corrective measures too aggressively or too timidly; we ignore basic premises, overgeneralize, follow blind alleys, overlook potential side effects and narrowly extrapolate from the moment, basing our predictions of the future on those aspects of the present that bother or delight us the most. This ingenious manual will assist problem-solvers in all fields.

[David Hodgson]

The system is far too complex for us to control it :) Once a system enters into a chaotic mode then any semblance of control is impossible, primarily because of non linearity, the impact of any given input to the system is absolutely unknowable, and it is the height of hubris to suggest anything else :)

[oldefortran]

John's likening to the F22 is quite apt... as aviation was where the term 'pilot error' originated, and also where much of modern control theory sprung from, strongly due to the forced necessity of war. Once systems become too complex, or too chaotic- we either anticipate, cope and adapt, or there is system failure. More often than not it takes significant failure to realize we should do the first two better- not a very resilient philosophy.

In a nutshell, that's the basic thinking behind the Cognitive Systems Engineering and Resilience Engineering movements that grew out of a post Three Mile Island demand for change. Technology change gives rise to the need to think about human factors and system safety. Dörner and his work, as mentioned above is a great example of this thinking, as are the works promoting the study and patterns of work learned over the last 30 years by the likes of Jens Rasmussen, David Woods, Eric Hollnagel, and Kim Vicente.

There are anticipatory movements for positive change- we see this push toward reigning complexity in healthcare with the patient safety movement. We see this with promoting the 'new look' on human error, i.e. Jim Reason's 'swiss cheese model' where we assume systems have multiple layers of defense, but all of them have holes- Sydney Dekker's insights into conducting fieldwork into such failures so we don't get stuck into traditional mindsets that past successes breed future success. Learning from failure IS military history, and there has been an interesting growing movement toward such concepts being incorporated into mil engineering psychology and leadership programs (such as research areas promoted by Col. Larry Shattuck).

Sociotechnical failures are breakdowns in adapting and coping with complexity. Being successful means that groups and individuals must design resilient systems that are grounded in being able to recognize and adapt to dynamic and changing worlds with the high potential for surprise. It all comes down to how groups and organizations are able to create foresight- and unfortunately, it seems that these (critical) lessons learned still aren't reaching all the ears they should.

[oldefortran]

John's likening to the F22 is quite apt... as aviation was where the term 'pilot error' originated, and also where much of modern control theory sprung from, strongly due to the forced necessity of war. Once systems become too complex, or too chaotic- we either anticipate, cope and adapt, or there is system failure. More often than not it takes significant failure to realize we should do the first two better- not a very resilient philosophy.

In a nutshell, that's the basic thinking behind the Cognitive Systems Engineering and Resilience Engineering movements that grew out of a post Three Mile Island demand for change. Technology change gives rise to the need to think about human factors and system safety. Dörner and his work, as mentioned above is a great example of this thinking, as are the works promoting the study and patterns of work learned over the last 30 years by the likes of Jens Rasmussen, David Woods, Eric Hollnagel, and Kim Vicente.

There are anticipatory movements for positive change- we see this push toward reigning complexity in healthcare with the patient safety movement. We see this with promoting the 'new look' on human error, i.e. Jim Reason's 'swiss cheese model' where we assume systems have multiple layers of defense, but all of them have holes- Sydney Dekker's insights into conducting fieldwork into such failures so we don't get stuck into traditional mindsets that past successes breed future success. Learning from failure IS military history, and there has been an interesting growing movement toward such concepts being incorporated into mil engineering psychology and leadership programs (such as research areas promoted by Col. Larry Shattuck).

Sociotechnical failures are breakdowns in adapting and coping with complexity. Being successful means that groups and individuals must design resilient systems that are grounded in being able to recognize and adapt to dynamic and changing worlds with the high potential for surprise. It all comes down to how groups and organizations are able to create foresight- and unfortunately, it seems that these (critical) lessons learned still aren't reaching all the ears they should.

[johnrobb]

Interesting Roubini comment (he's the best analyst in the financial sector right now):

``Systemic risk has become bigger and bigger,'' Roubini said at the Hedge 2008 conference. ``We're seeing the beginning of a run on a big chunk of the hedge funds,'' and ``don't be surprised if policy makers need to close down markets for a week or two in coming days,'' he said.

[mindlessrabble]

Open Source philosophy is an answer to these problems. The concepts of group effort enabled by complete transparency will work. To spite all the warnings to the contrary Wikipedia has worked through all the issues and thrived. JBoss, Apache and Postgres have built tools which everyone can use.

Unfortunately, the current system does not see any path to migrate from central control to open source while maintaining the power and rewards given to those at the top of a closed source/non-transparent system.

It the financial markets had been more transparent the crisis never would have happened. And it is the most non-transparent portions of the system (derivatives and oil markets) that have driven the system to collapse.

We have to bring about transparency first. This will undoubtedly lead to a massive creative destruction of the closed source system. However, it is inevitable (as was the industrial age). The longer it is delayed. The steeper the cliff we go over in the end will be.

[Alex J.]

When was the economy in "control" like an aircraft? Bretton Woods? Paulson and the rest are in the process of bringing more control to the center, making things less resilient. The more things are controlled from the center, the more opportunities the center has to screw things up. The more things are controlled by politics, the more people do what pleases the powerful instead of using their own particular knowledge.

There's already an Open Source philosophy for the economy. It's called Laissez-Faire, or just "the market".

[Duncan Kinder]

"We are all Iceland now"

We ought to remember that Iceland was settled by fellows, the Vikings, who encountered a system a great deal more turbulent than a F-22, the North Sea.

Sailors built up extensive lore to deal with their turbulent systems, e.g. "Red sky at morning, sailors take warning; red sky at night; sailors delight."

Part of building up resiliency will entail building up similar sorts of lore.

[The Dark Avenger]

When your only tool is a hammer, you tend to see a lot of nails that need to be taken care of:

http://rawstory.com/news/afp/Terrorist_tweets_US_Army_warns_of_T_10252008.html

The report by the 304th Military Intelligence Battalion, posted on the website of the Federation of American Scientists (FAS), examines a number of mobile and web technologies and their potential uses by militants.

The posting of the report on the FAS site was reported Friday by Wired magazine contributing editor Noah Shachtman on his national security blog "Danger Room" at wired.com.

The report is not based on clandestine reporting but drawn from open source intelligence known as OSINT.

A chapter on "Potential for Terrorist Use of Twitter" notes that Twitter members sent out messages, known as "Tweets," reporting the July Los Angeles earthquake faster than news outlets and activists at the Republican National Convention in Minneapolis used it to provide information on police movements.

"Twitter has also become a social activism tool for socialists, human rights groups, communists, vegetarians, anarchists, religious communities, atheists, political enthusiasts, hacktivists and others to communicate with each other and to send messages to broader audiences," the report said.

Hacktivists refers to politically motivated computer hackers.

"Twitter is already used by some members to post and/or support extremist ideologies and perspectives," the report said.

"Extremist and terrorist use of Twitter could evolve over time to reflect tactics that are already evolving in use by hacktivists and activists for surveillance," it said. "This could theoretically be combined with targeting."

The report outlined scenarios in which militants could make use of Twitter, combined with such programs as Google Maps or cell phone pictures or video, to carry out an ambush or detonate explosives.

"Terrorists could theoretically use Twitter social networking in the US as an operation tool," it said. "However, it is unclear whether that same theoretical tool would be available to terrorists in other countries and to what extent."

Besides Twitter, the report examined the potential use by militants of Global Positioning Systems and other technologies.

"GPS cell phone service could be used by our adversaries for travel plans, surveillance and targeting," it said, noting that just such uses have been discussed in pro-Al-Qaeda forums along with the use of voice-changing software.

"Terrorists may or may not be using voice-changing software but it should be of open source interest that online terrorist and/or terrorist enthusiasts are discussing it," the report said.

END

It would seem to me that with the exception of the voice-changing software, all of the above could be useful tools for promoting resiliency, especially in cases of an emergency or natural disaster, as with the July LA earthquake.

This has been in the works for a while, in 2004 I was in the Philippines and read a book about the culture of text messaging there, one example was a bank robbery that was planned and executed almost entirely by text messages.

[Loan Systems]

Hi ppl! I have made a blog about loans. Please welcome http://toploansystem.com