That’s all fine and we’ll, but I’m not mentioning Taubes for his carbs hypothesis (btw, he goes to great pains in Good Calories, Bad Calories to make sure the reader understands that his hypothesis could be wrong because it is also lacking the necessary type of study to demonstrate it’s effectiveness). I’m linking to him because he has important stuff to say about the limits of epidemiology, in general, and the terrible state of nutrition science, more specifically.