roystgnr
1mo ago
IMHO a better solution to the "fruit from the poisonous tree" rule would be "the criminal defendant can be in prison when the criminal cop is too". Two crimes get two sentences, not zero. Making one sentence contingent on the other would be sufficient to fix the bad incentives.
In this case, though ... do we even need to imprison the "defendant"? "A confidential informant said he was MS-13" got him held without bond after he was arrested for loitering, but never got a conviction. "The cops think this gang-member-turned-snitch is very trustworthy now" is a good place to start an investigation but surely it's not a good enough place to end one; police informants are sometimes themselves motivated more by base incentives than by a newly-acquired love of honesty and justice.
roystgnr
1mo ago
I'm a consequentialist with a complex value system that isn't trivially compressed.
Wait, how is this incompatible with utilitarianism? A large chunk of the Sequences was an attempt to convince people that, despite Von Neumann-Morganstern being a theorem about rational values being expressible as a utility function, human values still aren't easily compressed into a trivial utility function. It was a key lemma in service of the proposition "if you think you have a simple function representing human utility and you're going to activate ASI with it then You're Gonna Have A Bad Time".
As an aside, this is where I most differ from Yudkowsky on the current race to AGI: he seems to think we're now extra-doomed because we don't even fully understand the AIs we're creating; I think we're now fractionally-doomed for the same reason. The contrapositive of "a utility function simple enough to understand is unsafe" is "a safe utility function is something we won't fully understand". I don't know if stochastic descent + fine-tuning for consistency will actually derive a tolerably human value system starting from human text/audio/video corpuses, but it's at least possible.
roystgnr
1mo ago
But "talking" to them does not give the impression they are better at reasoning than anyone I know who has scored >50% on USMAO, IMO, or the Putnam.
They are still improving very quickly, and I don't see the rate of improvement leveling off. Gemini 2.5 recently answered with ease a test question of mine that Gemini 2.0 (and, honestly, everything prior to Claude 3.5) had been utterly confused by. But I admit that they're definitely lacking in reasoning skills still; they're much better at retrieval and basic synthesis of knowledge than they are at extrapolating it to anything too greatly removed from standard problems that I'd expect were in their training data sets.
Still, can we take a step back and look at the big picture here? The USAMO is an invitation-only math competition where they pick the top few hundred students from a bigger invitation-only competition winnowed from an open math competition, and the median score on it is still sub-50%. The Putnam has thousands of competitors, but they're typically the most dedicated undergrad math majors and yet the median score on it is often zero! How far have we moved the goal posts, to get to this point? It's the "Can a robot write a symphony?" "Can you?" movie scene made manifest.
roystgnr
1mo ago
If you think that Presidential democracy was a mistake
Do you think 5 is less than?
That's not a coherent question, right? You have to have two numbers to be able to talk about whether one is less than the other. 5 is less than 6. 5 is not less than 4.
But the same applies to any question of the form "Was X a mistake?" Was Presidential democracy a mistake compared to remaining part of the British empire? Probably not - the colonists did have some legitimate grievances. Was Presidential democracy a mistake compared to a Parliamentary democracy with a Prime Minister? Maybe, but not obviously so; we can see the cracks in parliamentary democracies too, today.
Was Presidential democracy with first-past-the-post voting a mistake compared to an approval-voting system? Here I'd opine the answer is clearly "yes", but when the Constitution was ratified Condorcet had just barely started publishing on voting theory, and Arrow and Duverger were a century away from being born, so I can hardly fault the Framers for lacking the benefit of hindsight here.
They did try to leave us with a mechanism for changing the Constitution to fix their later-identified mistakes, which has been very fruitful in the case of some other mistakes, and which you'd think would be sufficient in general... but the trouble with changing a mistake in the mechanisms by which people and parties gain power is that, almost by definition, the people and parties in power have strong incentives to want that change to not be made. If you're a partisan demagogue whose route to election has been "take advantage of your polarized base, plus a few moderates who can be convinced that the opposing partisan demagogue is more awful", why would you want to make it easier for challengers within your ideology to run against you and simultaneously make it likely that you'll face less-awful opponents from other ideologies?
roystgnr
1mo ago
FDR's supporters did object to Court-packing, which is why it didn't happen.
Well, there may have been other reasons too. It's hard to read minds, but even if the court's streak of "growing wheat on your own land for your own animals' consumption is interstate commerce"-level nonsense wasn't done out of fear of court packing, the resulting "feds can do whatever they want" situation still made court packing effectively unnecessary.
roystgnr
1mo ago
That's a very good point. We try to make 66% or whatever the de jure floor for "overturn this rule", but if the median voter is tempted enough or unprincipled enough then 51% remains de facto sufficient for "wilfully misinterpret this rule and get away with it".
roystgnr
1mo ago
That's a fascinating rabbit hole to go down: How the ‘Star Trek’ Punch Became the Worst Fight Move on TV.
In hindsight I also love the fact that, out of all the unrealistic sci-fi show things that nerds loved to geek out about, I never saw Star Trek fight choreography come up. Do the fleet sizes in Star Wars make tactical and economic sense? Oh, we can rant and debate about that for hours. Is the apparent motion of stars while the Enterprise is at warp consistent with the canon speeds the ship is at? Of course not, but here's a fan theory explaining why. Now, is clasping your fists together to punch with both at once a great fight move? What, why would you even ask a question about fisticuffs instead of something we know more about like teleporters??
roystgnr
1mo ago
The best fiction accurately reflects the world in most ways, in order to explore the implications of deliberately changing the fictional world in one or a few ways. If you don't want to make any changes, just write a history or a biography, and it'll be more useful. If you want to make an unlimited set of changes, that might be aesthetically evocative but it's not going to be interesting - at some point your attempts to create a world primarily from your own mind only tell me about your mind, not about any worlds.
Looking at "Aliens, Terminator 2, Buffy, Xena, ... the various Star Treks, The Matrix, ..." (I never watched Dark Angel, nor enough Farscape to comment), most of them actually came off pretty well in this sense? In rough order from best to worst (by this criterion; I still love DS9!):
Aliens had Vasquez get pretty buff, but it doesn't save her any more than it does the buff male marines who outnumber her. Newt is a survivor because she was the best at running and hiding; Ripley is buff enough (in part due to the Alien backstory) that it's important to the story, but the importance is "she can carry that huge gun/flamethrower", not "she could overpower a man". She's capable of as much courage as the men, but she doesn't let that drive her into extreme risks until maternal "must save Newt" instinct forces it on her.
Terminator 2 relies even more heavily on backstory here. Out of context, "the woman is the super tough super buff one" seems like pure proto-Woke, but the backstory is that this is the same woman who was a fragile damsel-in-distress who barely survived the previous movie, was turned into an utterly driven person for a decade and a half by her experiences there, and hasn't had anything to do during that time except plot and scheme and exercise in a mental hospital. That "experience creating strength" character arc isn't denied to the males; Miles Dyson and John Connor don't have time to get buff, but they both gain emotional fortitude very quickly.
In The Matrix, "the girl is one of the super tough super buff ones" is a natural consequence of the deliberate changes, no weirder than the same situation for Keanu, who at this point is closer to gangly "Bill and Ted" Keanu than to "training with Taran Butler for John Wick" Keanu. Their avatars have super powers when they're in the video game simulation.
In Buffy, "the girl is the super tough super buff one" is the deliberate change, specifically justified in-story by ancient magic. There's a wide variation in physical skill among the non-magically-powered girls and women, but they generally don't fare as well.
In Xena, "the woman is the super tough super buff one" is pretty much only justified by "she's exceptional", unless there was more to the story I've forgotten? But IIRC "exceptional women are physically competitive with exceptional men" was just played straight here.
In the 90s Star Treks, there wasn't a ton of physical combat, but when there was they generally played the gender differences straight too. Perhaps the most common example was Major Kira fighting, and while I can definitely believe that a trained and experienced guerrilla can disarm a random idiot with a hand weapon in close quarters, by the time they get to "mostly holding her own while surrounded by three Klingon warriors" it's clear that plot armor knows no gender.
roystgnr
1mo ago
I've made no secret in this forum of my attachment to truth as a terminal value, but experience has dispelled in me the conviction that this is a preoccupation of most people
Yesterday's news was, "NASA reveals astronauts’ return 'would not have happened' without Trump’s intervention"
By "NASA" here, we mean "Press Secretary Bethany Stevens, appointed a couple weeks ago", so hopefully I'm not indicting our top space-administration minds when I point out that this is a obvious bold-faced lie. The decision to bring back the Starliner astronauts in the Crew-9 capsule, as finally happened a week or so ago, was made last August, months before Trump was elected, much less took office. The Crew-9 launch was performed with two empty seats, reserved for their return as part of this plan, last September.
How can someone appointed to the job of "understand and explain what NASA is doing" be such an utter, unbelievable failure at understanding and explaining what NASA is doing? Well, that's probably why she was appointed.
or indeed something they ought to preoccupy themselves with in the first place.
Exactly! Imagine if, in Stevens' previous job as Ted Cruz's press secretary, she had been very assiduous about explaining that the Crew-9 return had been all planned out during Biden's term, and that the only change in plans during Trump's second term was that SpaceX took a little longer than planned to get the Crew-10 capsule ready and so the rotation was delayed a bit. Does she get praised for her commitment to truth and accuracy, and get her promotion more promptly? Or does NASA instead end up with a different press secretary who isn't such a killjoy?
The interesting thing about Brodski's story, that makes it not another case of "believing lies can be strategically useful", is that he is one of those few people who specifically and deliberately tries to avoid that, and yet one of the "useful" lies still bit him. When people like Stevens or the Boston Globe tell obvious falsehoods, it's good to wonder which of them fell for a dumb idea vs which of them are just being strategically deceptive, but Brodski would have to be playing the long game indeed to post a deep dive into how dumb he was. In his case, I'd like to hope that @pigeonburger had the right idea, that "if it had been a big issue for him I would assume he would have ended up on the correct opinion faster". Indeed the easiest way to fail to answer a question is to fail to truly ask the question, so you'd think Occam's Razor says we're done here. But maybe now I'm the one not paying attention to evidence? E.g. questions of politics and religion have no shortage of dedicated investigators, and yet many major questions don't see those investigators converge toward a single answer, or into a set of different-but-compatible-answers, or even to a state of humble explicit uncertainty.
Perhaps the key phrase there is "politics and religion"? Our ancestors may have all been through too many generations wherein anyone who announced "My epistemic credence is 90% on your side but still 10% on the other" had a good chance of ending up with their bodies 90% on one side of a blade and 10% on the other. The strategically deceptive thing to do in such cases is to keep your solid Bayesian reasoning private and just express false certainty publicly, but humans aren't as good at tricking each other unless we first trick ourselves, and either way why bother hanging on to good reasoning habits you can hardly ever use? Just be part of the tribe. You might get a promotion out of it, and if you were smart enough to ditch those good reasoning habits beforehand then you don't even have to feel ashamed afterwards.
roystgnr
1mo ago
I admit that my "60%? 66%?" bit was partly a crack at the subtly different levels needed to ratify this or end a filibuster over that, but was partly just an admission of ignorance as to what the "right" level to require would be. My ideas here aren't quite graduating to the status of "brainstorming" vs just "spitballing".
New spitball: we could probably push the arbitrariness back a step, as well as avoiding being too stuck on an outcome from a singularly bad era, by by just requiring anything that overrides an old law to have half the margin that the old law passed by. They got 66%? You need 58% to undo it. Only 60%? You just need 55%. The smaller margins should be much less rare than the larger ones, but still rare enough to make planning for the future less capricious.
I would only want something like that in concert with a "House of Repeals", though, an institution that could only initiate bills to repeal old laws, not to add new ones, because otherwise it would also end up reinforcing the existing "hysteresis of laziness" problem where laws linger long past obsolescence only because lawmakers have little incentive to prune them.
I'm not sure it's healthy for political life to move from "I believe that free speech of this form is good because..." to legalistic arguments about "The first amendment says..."
Yeah. XKCD famously mocked the "it's not illegal for me to say this, so hooray for me" line of reasoning once or twice, but the converse of "freedom of speech is all about what the government can do and so my attempts to squelch it privately are just fine" is pretty awful too.
But this feels like a mirror of the US founding fathers' arguments. In that case, the Federalist fear of "a Bill of Rights will be used as an excuse to deny the limitations on power in the body of the Constitution" and the anti-Federalist fear of "those implicit limitations will be easy to overreach so even redundant explicit limitations on the most important types of overreach" turned out to both be true, and so in hindsight the anti-Federalist position was least bad. Even if without a Bill of Rights we might have taken longer to get to the point where legal precedent says "the Constitution says the feds can do anything they want, Madison just misspelled that as 'commerce' for some reason", we'd still have gotten to that point, and then that would have been the end of their limitations entirely.
Similarly, it sucks that some people think the First Amendment is the be-all and end-all of free speech, but I have trouble imagining that the same people would have otherwise become principled defenders of unrestricted public debate. I think they hammer on "The first amendment says" primarily because it's literally the most authoritarian position they can take without freaking people out by outright saying they want to weaken the Bill of Rights. Take away that Schelling point and most of them would just take even more authoritarian positions with less opposition.
roystgnr
1mo ago
Macbooks excel in construction quality, battery life, and especially battery-life for high performance tasks, though that last isn't very important for word processing. I don't like the OSX UI, but that's not very important for word processing either. The only thing I'd caution is that something about my Macbook keyboard just doesn't "feel right" to me; I can't touch type quite as fast and I make more errors. I'm not sure if this is something real about the tactile feedback, or if I'd just get used to it if I used my laptop more frequently and other keyboards less frequently.
roystgnr
1mo ago
I've been considering giving Ground News a try, but I'm still on the fence. I'd love to hear from any current users of it.
It's just an aggregator, though, which means it's no more of an improvement over its sources than statistics allows. It tries to help readers avoid left-vs-right bias, but I'm not sure what can be done to avoid journalism's more general biases toward clickbait, outrage porn, novelty, oversimplification, etc.
roystgnr
1mo ago
Whenever I notice a public complaint that a rule isn't being enforced, the conversation goes something like
Complainant: "Why didn't you mod this?"
Mod: "Why didn't you report this?"
Complainant: something in between "Oops!" and "It's not my job to educate you!", depending on whether they're about to improve or just flame out.
Obviously from a normative sense I'm on the side of the mods here, because asking volunteer users to report violations as they see them is less burdensome and more scalable than asking volunteer mods to proactively police every comment posted, but from a positive sense maybe being more isolated now is making the system work less well? Judging by vote scores, even here many people have trouble understanding that the standard is not supposed to be "click the down arrow if you disagree", and I suspect "don't report rule violations if you don't disagree" is a equally popular misunderstanding. Reddit always provided an influx of people of all political persuasions who stumbled onto the subreddit, which was probably a good source of not-all-false-positive reports that's dried up here.
roystgnr
1mo ago
Technically, because a little hysteresis is beneficial in any case where a continuous (and worse: continuous plus noisy) input is used to determine a discrete output where switching between outputs is costly. We wouldn't even consider using a keyboard without a "debounce" filter; it might be reasonable to consider using a government with one.
Politically, for the same reason as we have a supermajority requirement for Constitutional amendments - to permit governments to, slightly, bind the hands of their successors. In the case of binding commitments it would also be sufficient to simply have a population who were mostly able to avoid hitting "defect" first in an iterated Prisoner's Dilemma, but here we are.
At least we're getting better. We had a government that legalized slavery for small children for nearly a century, yet now the thought of something similar happening again is worthy of use as a reductio ad absurdum. Maybe that's the paradox you're pointing out, though? The distance from "we should amend the Constitution to prevent voters from legalizing slavery" to "we can't imagine voters legalizing slavery" wasn't too great. Perhaps it's pointless to speculate about amending the Constitution to force voters to not make any game-theoretic screwups, if we couldn't actually get such an amendment through until we practically have an electorate who wouldn't screw up regardless.
roystgnr
1mo ago
how do you make democracy work if you permit governments to bind the hands of their successors?
With legislative supermajorities. If you can get 51% agreement on something, good for you, but there's no reason to expect that to bind others once a slight shift of political winds leaves you at 49%. But if you can get 60% (or 67%?)? That might be something worth hanging on to for longer, if it's not so soundly refuted that support drops to 40% (or 33%).
roystgnr
1mo ago
The IRS announcements were a few weeks ago: a bit under 10% of their current staff (the "probationary" employees) immediately (or at least when the lawsuits shake out) laid off, with a target more like 50% in the long term.
roystgnr
1mo ago
I've been noticing misandry against men in Western culture for quite sometime, but now it looks like boys are targets too
Arguably the primary targets are boys just becoming men. From an example published a few days ago:
not a single white American man born after 1984 has published a work of literary fiction in The New Yorker (at least 24, and probably closer to 30, younger millennials have been published in total).
(after similar anecdotes about 9 other prestige outlets)
The chief editor of the New Yorker is still a white American man, mind you. He replaced a woman in 1998 (back when that was still more unremarkable than Problematic) and he's probably still safe there today. If you try to take away an old man's job then you're certain to engender conflict with a powerful man. If you take young men's jobs before their careers really get started, the young men tend to just go away and find a different career. It might take a decade before people even start to notice.
roystgnr
1mo ago
On the one hand, the Comanche territory didn't come within 500 miles of Austin until after they had Spanish-descended horses to conquer it with, and it's pretty ridiculous and racist for the bleeding-hearts to bemoan the fact that Texas doesn't belong to those conquerors rather than to their conquerors' conquerors. I'd paraphrase it as "Those Natives are all alike, right? Who cares when one of their tribes is dispossessed by another of their tribes; it's like flipping a two-headed coin!", except that would suggest a verbal rationalization and I doubt anyone made it past a non-verbal gut feeling.
On the other hand, despite the ridiculous intention of the land acknowledgement, I feel like the Comancheria people would have been the sort to appreciate what the land acknowledgement means de facto. "You took this land from us, and you're not giving it back, and you're so confident about your conquest that you're willing to rub it in, at public festivals, while our descendants live among you? Impressive. Kudos!"
roystgnr
1mo ago
roystgnr
1mo ago
IMHO the mechanical mirrors are pointless; large lenses are really the only things phones lack. My DSLR is just old enough that mirrorless options were still kind of new. We also got a Nikon 1 around the same time, for portability, but unlike the DSLR that one's been completely obsoleted by our phones.
I'm not a fan of the current state of computational "photography", though. Detecting motion between multiple frames and trying to stack and deconvolve to get a sharp still image, that's fantastic, but when we reached the point where there's a "upsample moon photos using a neural net trained on moon photos" step, we'd lost the plot. If I wanted data from existing photos rather than my own photos then I'd be using the web browser, not the camera.
WMBF tend to look like they have more euro ancestry
I like your theory for this much better than mine, but I do have my own:
"Momma's baby, daddy's maybe" is an eons-old problem, and combining the lower certainty of paternity with the (typically) higher violence of males suggests there's been a long history of some nasty evolutionary selection pressure for kids to resemble their dads more than their moms. Infanticide in non-human primates and many non-primate mammals isn't uncommon, even many human "civilizations" took our damn time before deciding to punish it, and before agriculture it may have been a common form of birth control in tough times.
I'd guess even a couple million years of this would mostly affect which genes control appearance (or mediate other genes controlling appearance? there's not that much Y chromosome...), though, not habits. My interracial marriage is N=1, but while my kids look much more (75/25?) like me than like my wife, their personalities are maybe 55/45 on average and no more than 40/60 in one case.
roystgnr
1mo ago
My family splurged for a DSLR a decade or more ago, but now it basically only gets pulled out when we need the 50-300mm lens for distant shots, or maybe once or twice a year when a few shots are so important that they're worth the extra hassle. We used to pull it out for low-light photography too, but at some point phone image sensors got so sensitive that it makes up for not having half a pound of glass in front of them.
Oh - I do still use the DSLR body with a telescope adapter. I tried an eyepiece-to-phone adapter for that, but the quality wasn't nearly as high. Maybe I just need to find a better one.
roystgnr
1mo ago
The one opposing "everyone in the Big Yud singularity doomerist community"? The opposition itself isn't a deal-breaker (though it's clearly at least a non-central example), but the word choices to maximize emotional reaction at the expense of clarity are.
I was hoping someone would at least point out an interesting source being paraphrased. You see ML papers that talk about the infinite-width limit of neural networks, and sometimes that's just for a proof by contradiction (as OP appears to be attempting, to be fair), and sometimes it leads to math that applies asymptotically in finite-width networks ... but you can see how after a couple rounds of playing Telephone it might be read as "stupid ML cult thinks they're gonna have infinitely powerful computers!"
So then we're back to the status quo of zero sentences in N% of cases, but we get justice in 100-N%? Since N will be less than 100 that still sounds like an improvement.
More options
Context Copy link