> We would need to do an experiment to verify your claims. Unfortunately, software development productivity (and solution quality) is notoriously difficult to

> We would need to do an experiment to verify your claims. Unfortunately, software development productivity (and solution quality) is notoriously difficult to measure and perform experiments on. It is also expensive, time-consuming, and their conclusions tend to be shaky at best. The experiments' applicability is also heavily influenced by the uniqueness of each individual person. Technology just moves too fast. I'm afraid anecdotal evidence is probably the best we usually can do when talking about how we work.