fill in and rework intro to "Consilient Cultural Worldbuilding"

We need to acknowledge the "measurements screen off demographic
categories" point first, before we move on to explain the "... but
Bayesian reasoning still works on demographic categories" thing. As
originally written, I'd expect people to get angry and stop reading
before I addressed the point two grafs later.

I somewhat regret the jab at Yudkowsky and his "rationalists" getting
cut in the shuffle, but it's probably for the best.