Alan Kay <alan.n...@yahoo.com> writes: > When I said "even scientists go against their training" I was also > pointing out really deep problems in humanity's attempts at thinking > (we are quite terrible thinkers!).
I think a quite modest improvement would be more powerful calculators. For example, we already augment our arithmetic when we go shopping, why don't we augment our statistical ability when making judgements? We can be bad at estimating probabilities, but we're far worse at combining them correctly; for example the common error of not taking into account population bias when interpreting test results[1]. If we carried around a statistical calculator app with a simple UI (eg. graphs, tiles and sliders; not RPN) then we can enter numbers individually and have the device carry out the tricky combinations for us. When we don't need to estimate the numbers, eg. if they're given in a paper, report, magazine, etc. then a machine-readable representation like a barcode could be given alongside. This would allow us to intuit the values graphically, rather than slipping up on our parsing of the representation. [1] http://yudkowsky.net/rational/bayes/ Cheers, Chris _______________________________________________ fonc mailing list fonc@vpri.org http://vpri.org/mailman/listinfo/fonc
