A brilliant book which made me think that this is how science is really done. One of the questions is whether there is a message at all. How this message is first suspected (due to non-pseudorandom behaviour of a fraudulent table of random numbers) itself is a fresh idea. I found the various hypotheses in this book itself to be a very nice. IIRC, there is also an "opera singer" hypothesis that it is not the content of the message which matters, but its frequency.