Registered to post this.

I was linked to the Sequences and was going through them, mostly impressed, when I hit this post.

Eliezer's assessment that the human species can be clearly divided into exactly two sexes, and that dealing with the one you are not a member of is like dealing with an alien species, struck me in an extremely analogous way to how Robert Aumann's Orthodox Judaism struck Eliezer: a usually intelligent person buying wholeheartedly into a local cultural construct that, to my fairly simple observation and deduction, should be assigned very negative log odds.

I've assigned a considerably lower weighting to my Bayesian updates from the Sequences since. I very nearly stopped reading but realized that wasn't the right plan (since, after all, Aumann's agreement theorem is still true).