Cool potential feature doable with just a prompt: guide two users through determining fair sharing via shapley value calculations

These mostly sound like good things to me. Especially if, as somewhat implied iirc, they are outputs of some high level equilibria defenses.

My impression is that so far the kinds of people whose work could be automated aren't the kind to navigate the complexities of building bespoke harnesses to have llms do useful work. So we have the much slower process of people manually automating others.

Tails come apart. The most successful actresses are not the most attractive women.

The connection is in the deciding whether or not to regularly participate. Said didn't affect my decision here that much, but I'm way way above average in ability to dismiss criticism that I feel is non central.

Fwiw, my interaction with lw and more broadly the rationalist scene in the Bay area was most of what formed my current stance that communities that I want to participate in operate on white lists, not black lists. This is such a fundamental shift that it affects everything about how I socialize, and made my life much better. Banning someone requiring a post of this effort level predicts that lots and lots of other good things aren't happening, and that is mostly invisible.

Still finding this post useful. LLM suggested some mnemonics

"Wave" (Words, Assertions, Values, Execution)

"Team" (Terminology, Evidence, Attitudes, Methods)

"Maps" (Meanings, Assertions, Principles, Strategies)

"Salt" (Strategies, Attitudes, Labels, Truth-claims)

"Save" (Strategies, Assumptions, Values, Evidence)

I like the memorability of 'salt' as in 'salty'

I also like mapping these to the four causes to help embed it further

Value-Final

Label-Formal

Facts-Material

Strategy-Efficient

This reminds me of the quip that a lot of white cultures don't seem to have a signifier for 'real talk' and are very hamstrung on going down simulacra levels as a result.

I'd guess first few chapters of The Evolving Self

You are ignoring Kegan's experimental setup which lead to the clustering in the first place. I recommend putting his books into an LLM and asking questions about how the model was generated.