Jesse Mazer

@vkargatis @zaelefty Why are you so confident Zach has the same understanding you have and doesn’t believe sex is one member of a special class of natural kinds? Just because you and he are “on the same side” in this debate in other ways, or have you seen him say anything specific reflecting this?

@vkargatis @zaelefty But it’s not a wholly objective assessment, yet Zach used the word “objective” to refer to the sex binary in each of the quotes I gave, and doing a quick search I couldn’t find any quotes where he talks about utility of sex binary (including calling it ‘useful’).

Personality is a spectrum. Sex is not.

@vkargatis @zaelefty And that's because the cellular automaton universe is wholly determined by an underlying rule set that makes no reference to *any* higher level patterns--our physical universe is similar.

@vkargatis @zaelefty In a cellular automaton it's very useful to call a select set of black/white patterns in a 5x5 region a "glider" (en.wikipedia.org/wiki/Glider_(C…) than it is give names to other random black/white patterns in a region of same size, but no basis for sharp distinc bt "real"/"unreal" pattns

@vkargatis @zaelefty Yes, no spectrum by definition, but could equally well use definitions that still distinguished reproduction involving meiosis from other forms, but not defined in terms of a sex binary. Would Zach say a description using diff definitions involves objective material reality?

@Lasermazer @zaelefty We assess the sex of an individual (if it has one!) based on the developmental pathway of gamete-specific gonads. Secondary sexual characteristics are not used to nail down that assessment, tho ofc they're highly correlated. There is no "spectrum" of gonads.

@ArgonGruber @morallawwithin you can think of some additional conditions you can think of that would prevent the conclusion that it's irrational to two-box, or are you appealing more to unknown unknowns to argue that "somehow" the conclusion is prevented even if you can't imagine how?

@ArgonGruber @morallawwithin It's a thought-experiment, just like the original Newcomb problem--are you saying that even if it worked as specified (you could step into machine at later time T2 and step out at earlier time T1 with memories of T2 intact, and the timeline constrained to be self-consistent),

got to teach about Newcomb’s problem

@ArgonGruber @morallawwithin You don't think retrocausality entitles us to say that Omega putting 1 million in box B in the past is a "consequence" of me deciding to 1-box, and Omega leaving B empty is a "consequence" of me 2-boxing?

@Lasermazer @morallawwithin Without a theory of how time travel works, it’s unclear why we would believe that time travel prevents 2-boxing from dominating 1-boxing. Why would time travel change whether & which facts are consequences of other facts?

@forces_at_work What does “unconscious as a social system” mean if not referring to psychology? Would anything short of fully planned production be unconscious in this sense?

womp womp womp

@Sacrifice_Isaac Peirce’s version of pragmatism has not evaporated and it also dispenses with anything not scientifically testable or mathematically provable in the infinite limit, just with a more holistic notion of what it means to test that doesn’t require any strict verification procedure

This is the naivete of the Logical Positivists, and why they inevitably evaporated from serious philosophical discourse. Some of them will claim, adamantly, with a little bit of cheerfulness and comedy, that “Metaphysics is Dead!” How so? Of old age? By some cruel murder? 2/9

The denial of metaphysics is still, of course, an engagement with metaphysics. Just as Nihilism, when it denies objective morality, is still participating within the field of ethics. 1/9

@ArgonGruber @morallawwithin *If* you're correct that 2-boxing still dominates in case with time travel I'd say that's a case where choosing dominant strategy is irrational, though looking at definition there, why couldn't "consequence" of your decision include fact of how box was filled in the past?

@Lasermazer @morallawwithin True, the questions are distinct. The dominance principle might be false. I think it’s true. Do you think the dominance principle is false? Can you think of any examples where there are 2 moves, A & B, & A dominates B & A should not be chosen?

@ArgonGruber @morallawwithin Technical question of what strategy would weakly dominate is distinct from philosophical question of whether it is rational to always follow the strategy that weakly dominates, even in a case like this. Calling it “causal” DT is a misnomer if true retrocausality changes nothing!

@Lasermazer @morallawwithin Would 2-boxing weakly dominate 1-boxing even if Omega had a literal time machine? I think so. The mathematics of the relationships between 1-boxing & 2-boxing are analytic. It is analytically true, given the terms in which the problem is described, that 2-boxing dominates. Yes?

@deepfates Many assume that you could tack on continual learning to a net that’s already been trained to abilities greater than or equal to current LLMs via supervised learning, will be interesting if that doesn’t work and you need unsupervised learning to start from baby-like state

continual learning is a huge problem for alignment and kind of messes with everybody's business model and battle plan at the same time huh

@boguswaves1 @the_autonomia Note part in first draft at marxists.org/archive/marx/w… where he says it is "Precisely because it is contemporaneous with capitalist production" that Russian commune might skip capitalism, and might not need "a long Western-style incubation of mechanical industry"

@boguswaves1 @the_autonomia I think his argument was that the first society to achieve industrial technology has to go through capitalist stage (so would apply to Russia in alternate history where it was first), later societies can take advantage of more advanced tech on hand for diff transition

Passage from the late Marx that shows the Progressivism that I think plagues his work: "They know that in order to work out their own emancipation, and along with it that higher form to which present society is irresistably tending by its own economical agencies, they will have

@ArgonGruber @morallawwithin Since in this case we assume a fixed timeline, at the moment of your choice there is already a determinate truth about whether box B contains money or not, and in either scenario "box B empty" and "box B full" you will get more money if you 2-box

@ArgonGruber @morallawwithin So you would 2-box even if Omega had a literal time machine and after you stated your choice, Omega went back and time and put a million dollars in box B iff you said you'd 1-box? (assume fixed timeline as David Lewis talks about at home.sandiego.edu/~baber/metaphy…)

@ArgonGruber @morallawwithin You can also get 1-boxing from EDT, I haven't seen any cases where I'm convinced it necessarily gives different recommendations from what FDT advocates would recommend. Would you 2-box even if you were a software entity and Omega might have run an earlier exact duplicate of you?

@morallawwithin 2-boxers unite! One issue I have with FDT is its uneasy relationship to Kolmgorov’s 2nd axiom, Normalization. There seems to be no good accounting of what the elementary elements are for FDT.

@ankkala It'd depend on choice of reference frame, in general relativity no large frames are inertial but at scale of galaxy GR is well-approximated by "linearized gravity" which does have them, so this could be from the perspective of an inertial observer at rest relative to center

"a hundred million years ago, there were dinosaurs on the other side of our galaxy" is a sentence that sounds made up until you realize it's true