On Fri, 08 Jun 2018 17:45:58 +0000, Peter Pearson wrote: > What bothered me was my feeling that a "reasonable observer" > would expect the random-float population to be much larger than 2**32, > and the probably-collision-free sample size to be accordingly much > larger than 2**16, which is, after all, small enough to appear in many > applications.
People's intuition about probability is crap. Hence the Birthday Paradox I already linked to, and the Monty Hall Problem, which you can google for, and the Prosecutor's Fallacy: http://www.conceptstew.co.uk/pages/prosecutors_fallacy.html which sometimes (often?) leads to real miscarriages of justice: https://en.wikipedia.org/wiki/Sally_Clark -- Steven D'Aprano "Ever since I learned about confirmation bias, I've been seeing it everywhere." -- Jon Ronson -- https://mail.python.org/mailman/listinfo/python-list