I think of the core ideal of LW as "getting better at thinking, understanding the world, and acting effectively on it". "Rationalism" and "Rationality" are good summaries of this idea, though they do have some PR problems as the post says. "Bayes" is just too restrictive: not only it leaves out a lot of things as Raemon says, it also ties our identity too much to a particular epistemology, however powerful. If it turned out that a different epistemology is better than Bayesianism, this should not destroy LW.