In defense of subjective probability. I’ve been pondering the phenomenon, and it goes something like this: we go through university being taught a frequentist perspective, which involves deriving hypotheses from data collected, accepting or rejecting the null hypothesis, and accumulating information. But what if there is another perspective which does not deal with data as a prior, as a given set from which you deduct your hypothesis? This is the logic of plausible reasoning. It's uncomfortable from the frequentist point of view. How do you determine the priors? Who is to say if they are correct? biased? is it too subjective? . . . Here, Polya argues poetically for the potential for plausible reasoning to have rigour and leverage subjectivity. I think this is an incredible book. I highly recommend it if you are interested in Bayesianism or conditional probability.