For me, THE most influential book of the first half of 2008 is definitely "The Black Swan: The Impact of the Highly Improbable" by Nassim Nicholas Taleb. Craig Janet reminded my with his post yesterday of this book, and the fact that it was due time to give it also some blog time on Project Shrink.

Photography by JL2003.

When you have seen only white swans in your life, you think "all swans are white". But it might be that you haven't seen every swan on the planet yet, and that a black swan exists, but only that it is very rare. Taleb defines a "Black Swan" as an highly unlikely event, but with an enormous impact when it occurs.

Two main themes of the book stand out. First the fact that people don't think black swans exist, simply by having seen only white ones. Some months ago, I referred to Taleb while explaining that

"… it is the human need to categorize everything. We just have to put the world around us in neat boxes. Taleb coins the term "Platonicity" for this phenomenon — "the focus on those pure, well-defined, and easily discernible objects like triangles, or more social notions like friendship or love, at the cost of ignoring those objects of seemingly messier and less tractable structures." We put a label on an event and use that knowledge to reason about the future. We use this mechanism on everything, including people."

We need to put a value onto everything unknown. Otherwise we are becoming restless. We use stereotyping and labeling for filling an unknown mental void. As Edward Tufte explains:

"If you know nothing, take the average or use persistence forecasting. To describe something, observe averages and variances, along with deviations from persistence forecasting. Understanding, however, requires causal explanations supported by evidence." "Average" is meant both in the statistical sense and in the wisdom-of-crowds sense. "Persistence forecasting" is, for example, saying the tomorrow will be like today (and is often a hard forecast to beat)."

And the use of this kind of "average", a stereotype, a deduced label, is exactly the core problem of forecasting events. This brings me to the second main theme. Taleb spends a large part of the book explaining the difference between assuming a Gaussian distribution or a Fat Tail distribution for events or topics. It is really worth to think about this, so please read the next couple of lines, even if you think it will get boring…

Pareto And Gauss

If two people earn together $ 1million per year, how many does each of them make? Most people would think half a million each. This would reflect a Gaussian way of thinking. When considering the Pareto principle ("20% of the people have 80% of the wealth") you would rather expect a division along the lines of 100.000 and 900.0000. Pareto's law is a type of Fat Tail distribution. This example is metaphorical. The central theme is that the Gaussian is centered around an average, and the chance of deviation from this average reduces quickly. With a fat tail there is not a real average, only events that are likely to occur, and event that are unlikely to occur, but who's effects are enormous in comparison with the likely ones.

Another example. You have a set of bugs to resolve and to plan. In a Gaussian way of thinking you would calculate the average time spent on a bug the previous time and use that as a metric to estimate how long the bug fixing will take. Assuming a fat tail you only know that the most bugs can be resolved very quickly, but only a handful of bugs take all the time available. If you pick a bug randomly, you have a a big chance you have one you can solve in a minute. But you also have a small chance it will take you a week.

With traditional project planning we use the Gaussian mindset.

But if I look at projects, information distribution, the importance of employees, skills needed, bugfix time, motivation, authority and many other aspects are not evenly distributed. They seem to follow more a fat tail occurrence.

Fat tail distributions are associated with randomness. The more events follow a fat tail distribution, the more random they are. The more random events a system has, the more complex it appears.

Yes, I am suggesting that the appearance of this type of distribution is linked with complex systems. And resilience in a complex system is an important way to cope with a world following Fat Tail Distributions.

That is why Tom Peters, THE management GURU, started chanting about "resilience" directly after reading the Black Swan.

I hope you get a glimpse of how the "Black Swan" ties the limits of the mind together with complexity and resilience.