I feel like most of the value I got out of the minicamp in terms of techniques came early. This is probably due a combination of effects:

1) I reached a limit on my ability to internalize what I was learning without some time spent putting things to use. 2) I was not well mentally organized -- my rationality concepts were all individual floating bits not well sewn together -- so I reached a point where new concepts didn't fit into my map very easily.

I agree things got more disorganized, in fact, I remember on a couple occasions seeing the 'this isn't the outcome I expected' look on Anna's face and the attempt to update and try a different approach or go with the flow and see where things were leading. I marked this responsiveness as a good thing.

As for your ugly it's important to note that was a casual discussion among attendees. I suppose this highlights risks from a general increase in credibility-giving by close temporal association with other new ideas you're giving credibility to? Example: I talked to a lot of curious people that week about how Valve's internal structure works, but no one should necessarily run off and establish a Valve-like company without understanding Valve's initial conditions, goals, employee make-up, other institutions, and comparing them with their own initial conditions, goals, employees, institutions, etc.