You can't measure engineering productivity over a span of a couple of weeks. You can't compare the output from week X to week Y unless the test conditions and s

You can't measure engineering productivity over a span of a couple of weeks. You can't compare the output from week X to week Y unless the test conditions and subjects are completely in-sync. We all know that would be impossible. Here, the author claims there is a change in productivity, all the way down to a %. There is nothing scientific behind this; you might as well let your cat pick the numbers. They will be just as valid.