Monday, March 21, 2016

Review: The Half-Life of Facts

I picked up The Half-Life of Facts hoping to get some insight. There's an old joke (repeated in this book) that the medical school exam questions are the same year after year, it's just that the answers are different. I expected charts and statistics, and analysis of whether the large number of researchers in various disciplines have led to an increasing churn of results, and whether that makes the rush to publish more likely to lead to false theories being promulgated.

In practice, this book did little more than touch on bits of those threads, and provided little insight. In particular, it feels like the book's full of anecdotes and stories, rather than statistics and deep analysis. Worse, the author seems to have a very shallow view of technology, blithely talking about Moore's law, not realizing the lack of Dennard Scaling past the 45nm process, for instance, has brought CPU improvements down to a crawl.

I finished reading the book out of sheer bull-headedness, but I wouldn't recommend it.

Post a Comment