No, that's very different. Yesterday I proclaimed that our experimental colleagues do an excellent job of avoiding false positives, because they use great scien

No, that's very different. Yesterday I proclaimed that our experimental colleagues do an excellent job of avoiding false positives, because they use great scientific practices. Sabine would agree with me on that. This article is not about doing data analysis in experiment, it's more about how we should choose the theories that guide our choices in experiments in the first place. I disagree with Sabine because I see her as painting an unfair picture of theorists, as if we were all completely unreflective and she was the only one in the world to think seriously about theory selection. Most of her vague suggestions are things that people (sometimes, hundreds of people) have been already doing for decades. That breaks the cardinal rule of science popularization: don't trash people or take credit for insights without giving a careful, fair picture of everything that is going on, because the scientists affected don't have the right of reply.