Sigh, another misreading of my post:
actually strongly disliked his verbose and provocative writing style at first but was drawn in by the content. I don't agree with on everything but steel manning his missteps as much as I steel man the missteps of authors in Academia or pundits of the New York Times have found his models much better and superior at giving good predictions about political outcomes (for example the Arab Spring).
I was talking about using Moldbug's model to make predictions, not using his predictions! Though the latter will often correlate with the former.
http://unqualified-reservations.blogspot.com/2013/01/noam-chomsky-killed-aaron-swartz.html
Summary: Moldbug on the Aaron Schwartz affair. Power is a very real thing with real consequences for activists, yet many people don't understand the nature of power in modern times. People like Noam Chomsky get great fame doing bad epistomology about who has power, and as a result do great harm to idealistic nerds who don't read between the lines to selectively target their attacks at weak institutions (Exxon, Pentagon) instead of strong ones (State, academica incl. MIT).
Here he returns to a theme that is one of his real contributions to blogospheric political thought: that victory in political competitions provides Bayesian information about who has power and who doesn't. If your worldview has the underdog somehow systematically beating the overdog, your epistemology is simply wrong - in the same way, and to the same extent, as a geocentrist who has to keep adding epicycles to account for anomalous observations.
This means that activists like King, Schwartz, and Assange are only effective in bullying the weak, not standing up to the strong (despite conventional narratives that misassign strengths to institutions). When such activists stop following the script, and naively use the same tactics to attack strong institutions, reality reasserts itself quite forcefully: