An interesting point: If someone predicts that the probability of an event occuring (e.g. the Earth being sucked into an LHC caused black hole) is 10^-9 or some such incredibly small number, then the likelihood of there being an error in the paper is almost sure to be much, much larger than the estimate itself. But if that’s the case, then does estimating the number even make sense to do?
3 years ago
January 29, 2009
