TDD, when normalized for time spent writing tests, has not been found to be any better than the 'normal' way of writing tests afterwards. It's interesting how m

TDD, when normalized for time spent writing tests, has not been found to be any better than the 'normal' way of writing tests afterwards. It's interesting how much of programming lore falls apart when you actually try to measure the impact.