> In general we look for a new law by the following process: first we guess it, then we compute the consequences of the guess... and then we compare those compu

> In general we look for a new law by the following process: first we guess it, then we compute the consequences of the guess... and then we compare those computation results to experiment. If it disagrees with experiment it is wrong. In that simple statement is the key to science. It doesn't make a difference how beautiful your guess is. It doesn't make a difference how smart you are, who made the guess, or what his name is. If it disagrees with experiment its wrong, that's all there is to it. That elides a lot of complexity, mostly around the validity of the experiment design and execution. Confounding factors, confirmation bias, selection bias, p-hacking, etc. can all skew the results in ways both overt and subtle. The following quote from the movie 'Thank you for smoking' may illustrate: "This is where I work, the Academy of Tobacco Studies. It was established by seven gentlemen you may recognize from C-Span. These guys realized quick if they were gonna claim cigarettes were not addictive they better have proof. This is the man they rely on, Erhardt Von Grupten Mundt. They found him in Germany. I won't go into the details. He's been testing the link between nicotine and lung cancer for thirty years, and hasn't found any conclusive results. The man's a genius, he could disprove gravity." So no, 'if it disagrees with experiment it is wrong' is a gross oversimplification. Not to mention that there may be very good reasons that an experiment can't (or shouldn't) be done, or that various 'natural experiments' must be relied upon.