Mark Buchanan (1) (1961–)

I finally completed reading this book and found it to be very insightful. It is interesting how a wide variety of different occurrences and phenomenon can be described in an abstract manner. I remember that I started the book fairly quickly, but got caught in the middle and plodded along for awhile, before quickly reading the remaining chapters.

This book is dated in parts, as it was written and discusses issues in the early 2000's. This doesn't detract much from the overall message of the book though.… (més)

Che libro interessante! Finalmente un giusto equilibrio tra semplificazione e rigore. Un'ottima appendice con approfondimenti anche piuttosto tecnici, e argomenti di estrema attualità. L'idea principale dell'autore consiste nel fatto che la complessità sociale sia un fenomeno collettivo, emergente, e non costituito semplicemente dalla somma delle complessità dei singoli individui. Un "Must have" per chiunque si sia mai chiesto perché dopo aver fatto un'ora di coda in autostrada scopriamo poi che non c'era nessun incidente, oppure come sia possibile regolamentare i flussi di dieci, venti o anche centomila persone che si sono riunite tutte nello stesso luogo (ad esempio, per un concerto). E tante altre cose, dalla sociologia alla geologia all'antropologia. Che ci sia una speranza che, finalmente, le "scienze sociali" entrino in contatto con il metodo scientifico?… (més)

I have been fascinated by critical point phase transition phenomena since I read Stanley's book on Phase Transitions back in college, probably 1974. In graduate school I worked on renormalization groups and critical phenomena. I even spoke at a small conference - I was the warm-up act for Stuart Kaufman! - on universality in critical point phenomena.

One the one hand, all this baggage prejudices me in favor of the topic of this book. On the other hand, it gives me a good background to scout for errors in the presentation. I must say, I found this book delightfully clear and accurate! Of course it doesn't go into every detail, but there is quite a bit of detail here.

I do tend to fall in love with theories like this, self-organized criticality. Sometimes that puts me on a popular bandwagon. Sometimes the area stays esoteric: will tree decomposition (see e.g. Hans Bodlander) ever take off?

This book really is a sort of popular cheerleading. One of the real challenges with this stuff is that there is a lot of noise in the tails of statistical distributions: extreme events are rare, so any measurement is going to have a lot of uncertainty. Looking at these logarithmic plots and fitting a line, which will work well if the tail is fat - it is a bit like reading tea leaves.

This books dates back to 1991 or so. The self-organized criticality bandwagon likely didn't have much momentum at the time. But some methodological caution is definitely needed. Buchanan does admit that fat tails are not as ubiquitous as his title suggests. But the challenge of determining where this model is useful and where it is not, he doesn't really address that.

Anyway it is a great book. The phenomena it discusses is definitely real and important. It describes this simply, clearly and accurately. I don't think the book asks for anything more than high school mathematics.… (més)