If you generate enough data (eg 60 pages worth), you are bound to come up with a fair amount that fits. maybe thats why there are so may believes in in depth astrology. If they look into it deeply enough they are bound to find bits that fit like a glove. Very easy to forget about or explain away all the bits that don't quite fit, or are completely contradictory to your nature. I am sure if you did the long report based on one of the fictional people, I would find something that equally sounded very like me. In science if we do experiments with hundreds of statistical tests, if we used the ordinary tests, many of them would come up positive without there being an actual true phenomenon, just a statistical blip. This is called comprimising your confidence levels (also known as a "fishing expedition" which is very bad science). In these cases you have to choose between changing your level of proof to be very high for each individual test, or limiting the questions you ask beforehand and making just a few testable predictions. In the silly pseudo experiment we did here, I really did the second. I limited the prediction to one - that my short report generated from the site would be more recognisable to me as me than 3 other reports. You did not raise any objections to that test before the experiment, only once it had failed to support what you expected. It is very human to move the goalposts after the fact, but not very good science.

Bagpuss