Hot take: the LessWrong community is doomed to fail, because it’s original memetic code is written by a libertarian individualist. Yud focuses on rationalism as a singular activity of debiasing oneself, which is ultimately doomed to yield astonishingly diminishing results.
Modern theories on the origins of reason pose that the purpose of reason is not to independently find truth, but to evaluate and author arguments in debate. Hence, much like pair programming makes better code, pair reasoning ought to be a discipline. Indeed academic debate is a very natural medium for creation of good ideas.
A much better saying that “what do you think you know and how do you think you know it” would be “who did you hear it from and who benefits from you believing it.” A philosophy of truth-as-social-construct both ties into the paradigmatic theory of science and the skepticism of accepted truths that is crucial to good social idea generation.
In fact, science is a poor fit for a prototype of rationalism, since science is a highly involved and highly specific process of discerning the mechanics of reality in a way that makes it very easy to profit of (c.f. the practice of engineering) and therefore hard to question w.r.t. truth-as-social-construct (give or take global warming denialists.)
A much better prototype for rationalism is an unholy union of mathematics and postmodern philosophy, favoring discourse and communal creative problem solving over solitary reasoning. An idea-economy based on social deconstruction, rather than derision of one’s intellectual forbearers and lessers.
Also, queers and unashamed socialism (c.f. who told you socialism was bad and who benefits from you believing it?,) because it scares off the undesirable crypto-fascists and ancaps.
Discuss.