This is a confused model of turning money into results.  

I think there's a missing element from the description of "strong favorable signals must be costly".  In fact, they must be costly TO FAKE.  They can be cheap when honest.  

I think you have to accept the truth of your observations (it's not tautological if it requires evidence to support, and you happen to have that evidence).  A whole lot of humans have difficulty with impulse control, and with subverting their natural reactions and desires in order to appear friendlier and more vulnerable than their inside view of themselves supports.

Worry about looking like an idiot is a VERY fine balance to find.  If you get desensitized to it, that makes it too easy to BE an idiot.  If you are over-concrerned about it, you fail to find correct contrarian takes.

'notkilleveryoneism' IMO is a dumb meme.  Intentionally, I presume.  If you wanted to appear smart, you'd use more words and accept some of the nuance, right?  It feels like a countersignal-attempt, or a really bad model of someone who's not accepting the normal arguments.

not downvoting - it's already at -10, which is about right IMO.  There's nothing exactly wrong with it, but it's a lot of text and detail for fairly trivial content.  If it were a reaction to some specific ongoing debate, or if there were a much clearer summary followed by examples, and if the title were less misleading (it has nothing to do with most philosophical questions, certainly not mine, and is not about why there is no answer, but about why many discussions are pointless whether or not there is an answer), it could be a useful post.

I'd prefer to limit or simply remove strong votes, or scale them to the number of total votes on a given post/comment.  It's overwhelming to get strong votes as the first few votes.  Of course, it's unimportant to get strong votes on already-heavily-voted items, so I think just doing away with them is best.

Does "crisper" mean "more correct", or just "more detailed, but with the same accuracy"?  And is it correct to be more conservative (counterintuitive or no)?  Assuming it's a problem, why is this a problem with minmaxing, as opposed to a problem with world-modeling in general?

I suspect, for most decisions, there is a "good choice for your situation" range, and whether you call it conservative or aggressive depends on framing to an extent that it shouldn't be a primary metric.

I don't post much, but I comment frequently, and somewhat target 80% positive reactions.  If I'm not getting downvoted, I'm probably not saying anything very interesting.  Looking at your post history, I don't see anything with negative totals, though some have low-ish scores.  I also note that you have some voters a little trigger-happy with strong votes (18 karma in 5 votes), which is going to skew the perception.

I recommend you not try to learn very much from votes.  It's a lightweight indicator of popularity, not much more.  If something is overwhelmingly negative, that's a signal you've crossed some line, but mixed and slightly-positive probably means you're outside the echo chamber.  

Instead of karma, focus on comments and discussion/feedback value.  If someone is interested enough to interact, that's worth dozens of upvotes.  AND you get to refine your topical beliefs based on the actual discussion, rather than (well, in addition to) updating your less-valuable beliefs about what LW wants to read.

I don't think this is core to alignment, though it's probably a good idea overall.  Making it easier to kill or threaten to kill people is anti-human-friendly on it's own, even if all the agency involved is from a subset of humanity.

More importantly, I don't know that anyone who disagrees is likely to engage here - "let's agree" doesn't move very far forward unless you can identify those who need to agree and why they don't.  I'd start with how to overcome the argument that some humans (which ones depends on who you ask) need to be stopped in their harmful actions, and killing or threatening to kill is the most expeditious way to do so.  Without that underlying agreement, it's hard to argue that safer (for "us") mechanisms are wrong.

You're incorrect to put zeros in the right column.  Following an ought that is incorrect is a cost.  And then you need to factor in probabilities and quantified payouts to decide what to optimize.

Yeah, I am.  But more bottlenecked on cost and flexibility (ability to shift from use case to use case) than directly on capability.