Sometimes the most important experiments are the ones that fail, because they teach us what doesn't work.