Fabulous Nonfiction Roundup: Data Miners need a theory too, Spanish Flu, Autonomous Cars

I've been reading more nonfiction again lately, with some keepers.

Judea Pearl is a famous computer scientist from Stanford; he won the Turing Award in 2011.

For years he's been studying the causal theories behind "big data" and says that data by itself isn't good enough. You need the theory driving your question to come out with sensible answers.

He's also come up with a way of representing the mathematics of causal theories graphically, for example:

This book is well worth reading for anybody involved in "big data" ... four stars!

I've been interested in autonomous cars for several years, and just read a compelling account by an insider:

Burns was a GM exec who wound up working for Google's WayMo and got to sit in on meetings with Anthony Levandowski and Travis Kalanick when the wrestling over intellectual property was going on resulting in a WayMo/Uber lawsuit.

(One especially crazy anecdote: A. Levandowski's bonus from WayMo on his way out: $50 million+ ... some reports say the total was $120 million. Sheesh.)

Overall this was very interesting ... I can't wait to ride in one of these!

Finally this month ...

This is a nonlinear narrative on the flu epidemic from several angles ... one question considered is why the epidemic, which lasted from 1918 to 1920 in several waves, doesn't rank higher in world disasters even though it killed millions?

The technology wasn't good enough to count all the victims until much later, and the ability to even diagnose this viral disease wasn't developed until decades later.

I read most of this in one sitting ... four stars!