Does anyone know a popular science book about, how should I put it, statistical patterns and distributions in the universe. Like, what kind of things follow normal distributions and why, why do power laws emerge everywhere, why scale-free networks all over the place, etc. etc.
Sorry for ranting instead of answering your question, but “power laws emerge everywhere” is mostly bullshit. Power laws are less ubiquitous than some experts want you to believe. And when you do see them, the underlying mechanisms are much more diverse than what these experts will suggest. They have an agenda: they want you to believe that they can solve your (biology, sociology, epidemiology, computer networks etc.) problem with their statistical mechanics toolbox. Usually they can’t.
For some counterbalance, see Cosma Shalizi’s work. He has many amusing rants, and a very good paper:
Note that this is not a one-man crusade by Shalizi. Many experts of the fields invaded by power-law-wielding statistical physicists wrote debunking papers such as this:
Thank you, I never knew this fallacy has its own name, and I have been annoyed by it since ages. Actually, since 2003, when I was working on one of the first online social network services (iwiw.hu). The structure of the network was contradicting most of the claims made by the then-famous popular science books on networks. Not scale-free, (not even truncated power-law), not attack-sensitive, most of the edges were strong links. Looking at the claims of the original papers instead of the popular science books, the situation was not much better.
You could try “Ubiquity” by Mark Buchanan for the power law stuff, but it’s been a while since I read it, so I can’t vouch for it completely. (Confusingly, Amazon lists three books with that title and different subtitles, all by that author, all published around 2001-2002.)
Does anyone know a popular science book about, how should I put it, statistical patterns and distributions in the universe. Like, what kind of things follow normal distributions and why, why do power laws emerge everywhere, why scale-free networks all over the place, etc. etc.
Sorry for ranting instead of answering your question, but “power laws emerge everywhere” is mostly bullshit. Power laws are less ubiquitous than some experts want you to believe. And when you do see them, the underlying mechanisms are much more diverse than what these experts will suggest. They have an agenda: they want you to believe that they can solve your (biology, sociology, epidemiology, computer networks etc.) problem with their statistical mechanics toolbox. Usually they can’t.
For some counterbalance, see Cosma Shalizi’s work. He has many amusing rants, and a very good paper:
Gauss Is Not Mocked
So You Think You Have a Power Law — Well Isn’t That Special?
Speaking Truth to Power About Weblogs, or, How Not to Draw a Straight Line
Power-law distributions in empirical data
Note that this is not a one-man crusade by Shalizi. Many experts of the fields invaded by power-law-wielding statistical physicists wrote debunking papers such as this:
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.21.8169
Another very relevant and readable paper:
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.11.6305
That gives a whole new meaning to Mar’s Law.
Thank you, I never knew this fallacy has its own name, and I have been annoyed by it since ages. Actually, since 2003, when I was working on one of the first online social network services (iwiw.hu). The structure of the network was contradicting most of the claims made by the then-famous popular science books on networks. Not scale-free, (not even truncated power-law), not attack-sensitive, most of the edges were strong links. Looking at the claims of the original papers instead of the popular science books, the situation was not much better.
You could try “Ubiquity” by Mark Buchanan for the power law stuff, but it’s been a while since I read it, so I can’t vouch for it completely. (Confusingly, Amazon lists three books with that title and different subtitles, all by that author, all published around 2001-2002.)