In response to the blog post Nyk writes:

I am of the opinion that “politics is the mindkiller” rule is bad because it allows some unrecognized aspects of mainstream politics (i.e. Universalism) to slip under the radar. Many Universalist ideas are ‘no longer politics’ in the same way that they are ‘no longer Christian’, allowing them to bypass any red flags that might be triggered with politics and religion respectively.

James_G responds: "Echoing Player of Games: my imagined forum design is federalist, like the style of government I favour; LessWrong exhibits democratic degringolade, as does today’s West."

Also, the karma system on LW has all of the bad characteristics of demotism, and the fact that such a system of votes of equal value was chosen in the first place again seems to point to demotist bias. I would very much prefer moderation similar to the dictatorial one of Razib on GNXP.

Konkvistador (me): Razib's harsh style does indeed create a comment section well worth reading.

My opinion is that groupthink is already quite strong on LW at this point in time; I am not sure how it was in the past. Their preference for philosophy and pure reason (rather than experimental science) is immediately obvious to me as an outsider; also, some of them seem obsessed with a few topics while losing sight of other important issues. I presume that is because of their (or should I say our?) highly atypical psychological profile: many Asperger types (some borderline, but many quite beyond that), and heavily risk-averse types. There have even been articles calling for the sabotage of scientific research into computing and AI as long as they consider their current pet obsession (Friendly AI) to be not 100% implemented in a safe manner. I for one believe it becomes impossible to achieve anything at all if you crawl into a hole due to fear of inadvertently creating paperclip maximizers. At some point, you have take some risks and go past analysis paralysis.

James_G responds: "I can’t fault this."

Zack M. Davis criticizes James_G's approach of viewing humans as a collection subagents:

The subagents idea is interesting, but it seems like a metaphor at best. That humans are an incoherent kludge of partially-conflicting values is indisputable, but to say that they meaningfully factorize into subagents seems like a much stronger claim; I don’t understand what is gained by speaking of a dominant hedonistic utilitarian subagent coexisting with ideological upstart subagents, when one can just say “I value (or ‘this brain contains parts that value’, &c.) pleasure, and antivalue pain, and I also value these-and-such political goals, but not quite as much as I antivalue pain.”

Is this a mere semantic quibble?—possibly, but near the end of your “Beyond Moral Anti-realism,” you seem to want to attribute your writings to your hypothesized hedonic utilitarian subagent, and in this post you write that “the strongest sub-agent in an educated mind is usually a hedonic utilitarian[;] [i]deally, all parties to a discussion claim to share the same utility function,” and it seems unnecessary; I don’t need to suppose that my values factorize in any particular way, nor disparage any of them as mere inferior upstarts, in order to be eager to cooperate and discuss ideas with smart, sane people who happen to like things I find distasteful or abhorrent.

I strongly agree with this.