Visualizing Bayesian Inference [link] — LessWrong