@RobertLiguori's banner p

RobertLiguori


				

				

				
0 followers   follows 0 users  
joined 2022 September 04 21:34:07 UTC
Verified Email

				

User ID: 165

RobertLiguori


				
				
				

				
0 followers   follows 0 users   joined 2022 September 04 21:34:07 UTC

					

No bio...


					

User ID: 165

Verified Email

Can you talk a bit more about what you mean by a modern food environment? As far as I know, fatness isn't evenly distributed across populations, and it's not that hard to find subgroups and cultures with much less obesity than we observe as the baseline in America.

How much is a very large part? Can you list off a few specific metrics in which you believe that black Americans underperform, give the relative numbers for some other races, and then estimate where you'd expect the needles to be absent the legacy entirely? Can you also provide what you'd expect the numbers to look like for other races in a hypothetical no-discrimination-at-all environment?

I mean, you do have some kind of "X specific discrimination which happened in these specific ways at these times caused these gaps, which I expect to persist for this many generations.", right? It would be very silly to claim that you had any idea what the expected but-for-slavery-and-Jim-Crow outcomes of black Americans relative to all other colors of American would be absent that kind of comparison, after all.

Let's get some actual numbers around this, both in terms of how you see things now, and how you'd expect to see things differently with a few specific point interventions.

What would be more secure; a system with no wifi, or a system with maximal wifi, scanners everywhere, and a Dominator pointed at everyone entering the factory at all times?

Heck, why not just corral people entering into the factory into cells loaded with scanning equipment but no net connection, have them be scanned, send a courier to take that data to a network node, drop it off, and come back with the vibe shift kill orders for that crop of entrants that day?

I also feel like I need to balance what (presumably) characters are saying about the cost and difficulty of full scans, versus the fact that full scans were conducted on a hostage in a hostage situation, multiple times and repeatedly, and came back quickly enough for them to be relevant in the moment-to-moment crisis. I know that there's bullshit under-the-hood reasons for why you can't, e.g., announce "Hey, we're limiting job and romance horoscopes to once per month per person and re-allocating those MIPS towards crime-screening, and we're also re-allocating 3% of the industrial economy into putting a Dominator on every street corner - for all of our protection.", but you should need an actual cover for that reason, because if people start asking questions and then stop trusting the Sybil system en masse, your society is kind of fucked.


I also feel like we're at kind of the same problem I've been discussing with the fancy boarding schools. Are they meant to be Amish-esque? Because if not, then "Yeah, a cop went to one of those schools finally and wiped out like a third of the students. What fools they were, for investing in money and power and influence and not carefully purging their children of wrong-think! Why would they think that the ability to temporarily dodge the judgement of the all-seeing Sibyl would protect them when one of those criminal bastards wandered within its gaze? Did they really think they could bribe or threaten a good, honest, upright sub-100 citizen?" should be the attitude we see; how do you coordinate as an elite to dodge the judgement of a world-enveloping digital system that can kill you on a whim?

The Sybil system, as described, should be an entirely new way of organizing society, and one that should utterly demolish previous forms of influence and corruption that have been basically universally-present. And even if it is not at all as described, either everyone believes in it and the system should be able to instantly break the power of any previously-elite groups, or the system should basically be purely the judgement of society and thus to be avoided at all costs no matter what people say externally, and we should have people recognize that, unless the point is that everyone in the society is double-thinking super-hard at all times.

Why?

Because that's not what the law is, as we here and now define the law. Like, you could also have a pre-crime society with a totalitarian dictator where it was the written law and the whole of the law that what he says goes...but that's not pre-crime, that's just the dictator with extra steps. Likewise, society outsourcing its ability to designate people to pick on to an AI system is something, I guess, but that's not the law (again, as we understand the law.) That's just, I dunno, racism with extra steps? Classism, maybe? Or...ah, life in your choice of post-revolutionary totalitarian shithole, where anyone who commits a public impiety (or is just the least conspicuously pious when the bloodlust gets up) gets stoned or gulaged, and everyone who's not in the clergy or Party knows that they're on thin ice and need to police themselves accordingly.

I am not a sociologist who's studied Japan specifically, but one facet I've heard repeatedly is that lots of aspects of society are stultifying, with unwritten rules and expected codes of behavior, and avoidance of direct conflict. And turning those unwritten codes of behavior into something that is being parsed by a theoretically-objective system feels like a cleaner metaphor. The guy playing his speakers on full blast in the crowded train? And who absolutely refuses to take a hint or read the room, in that or any other similar situation? What's his crime index like, if he does actually care about the law-as-written, and the law-as-written expected people to get with the program due to social pressure?

Either you formalize the social codes, and you should have the cops sternly warn the manager in episode two for not sticking to the recommended duration and quantity of Accepted Workplace Bullying as defined in Psycho-punk OSHA (since if they've got algorithms for prediction and they're not just magic bullshit, there should be a huge amount of regulations and research about what makes the mean person more or less likely to crime). But the point is that there was no rule saying it was that guy's turn in the barrel and that he can sue to demand Victim Overtime instead of poking back (or alternately, him cheerfully pointing out that he's got a rock-bottom score himself because taking very-precise actions that just happen to result in his co-workers dying in hilarious industrial accidents isn't technically illegal any more than their harassment is).

Now, you can absolutely have a Sybil system that looks only at "Is this person going along with society?" And, if you're in a place that has a clown-shows justice system like Japan, then specifically using the term 'criminal' for someone who's gotten the cyberpunk equivalent of a K9's flag is probably making a point. But that's not fulfilling the promise of an actual pre-crime system that can actually predict crimes and thus eliminate them before the crime actually happens and the harm is caused.

Although, now that I think about it, maybe another point of the anime is that it's meant to be obvious that the pre-crime is bullshit, because the society has the need for the Enforcer janissary class. But hey, maybe we'll see that the system is actually generally correct and the Enforcers get up to shady shit, and that giving them guns is actually a really bad idea when one clever-dick improvises a wi-fi jammer and then they all rise up and start beating the real cops to death with clubs and numerical superiority.

But that's not what the series is promising, then. An actual pre-crime analysis system should look at everyone who is willing to go "Yup, I have free societal rein to hate on this person, we've all agreed that they're the designated victim." and note that whoops, the law doesn't say you're allowed to abuse your social lessers according to a nebulous and ever-shifting social pecking order, you're all criminals. An accurate oracle will predict that you are likely to commit a crime; if they don't peg as latent criminals, either criminal harassment isn't a crime in this society (which I strongly doubt is shown), or there are explicit and complicated legal codes allowing it in specific circumstances (which again I doubt is shown)...or, what I think is likely, the show (and/or the Sibyl system) are equivocating between what the actual law is and how the actual law gets enforced. Like, is discrimination against people who once-ever pinged into latent criminal legal and mandated, or social consequence? You can't appeal the court of public opinion if you're accused of a crime in a shame-based culture...but you can go "Predliciton towards illegal discrimination and targeted harassment! The Law says that this man is now innocent and if you disagree, that means you're likely to be a criminal yourself! Jail for all of you! Oh, you think that you can get away with changing the law so we're not allowed to imprison millions of people for going along with society's most-common actual beliefs? That's evading the police, like with drugs! Death for millions of you!"

Of course, this assumes that the Sybil system is honest, which I don't think it was ever meant to be. An actual pre-crime system means that the letter of the law is vital to the story, and that we need an actual indicator of how stable crime index scores generally are and what going from sub-100 to 300 to back to sub-100 means. (Can the cops walk into City Hall and shoot at random government officials, since if the act of being mock-executed makes their crime index go up it means they were latent criminals all along? Are there even government officials any more, or is it just the will of the computer system and its physical agents?)

Hmm. I don't think that the circumstances of recent Nigerian immigrants actually do mirror, e.g., those of turn-of-the-century Chinese immigrants particularly well, but I'd be interested to hear you break down what you think the salient features of Asian immigrant waves were, and how they compare to the circumstances of the given Nigerian wave.

We've got a lot of immigrants to a lot of nations being done by a lot of ethnic groups across a lot of history for a lot of reasons under a lot of circumstances; we should be able to tell pretty quickly which of those factors (if any of them) most saliently predict outcomes.

I mean, there is such a requirement for me to engage with the show. If it is the actual case that there are no more real laws any more and the only actual crime is for the system to rate you 100+ against its own internal criteria, then yeah, it's a crime oracle, and it says nothing new or interesting.

Now, with that being said, I think that we would absolutely get some horrible "We can measure X, so let's write laws against X instead of against what is actually fair and just." Goodhart bullshit if we started implementing the early phases of Sybil, but you still need an actual code of laws to compare those to. As I said above, we've got "What if it was illegal to fail a vibe check?" at home, and it's not novel to me that you can say "The Vibe of that person is white/capitalist/monarchist/>100 arbitrary bullshit number I just made up., get 'im!"

Funny you say this, as this was, almost word for for, the reason the director of the factory gave for obstructing the investigation. One thing to keep in mind is that crime management is not the only task of the Sybil System, it manages all of society. It's not even clear how much power politicians have, and it's heavily implied they're just a human face for the system. So if it's the system that decided it needs drones from the factory, doesn't it stand to reason that it might tolerate a bit of harrasment to keep the productivity up?

Because at that point it's not an oracular system, it's an agentic one (a tyrant with extra steps). A system that lies in pursuit of a given goal is not an oracle (or at least, not a useful oracle), and if it would lie to cover its manager (just as the manager lied to cover his subordinates), then how can you trust that it's actually measuring economic output correctly?

The law is legible. The law is codified, tested with precedent, and those precedents recorded. And it is exactly because in the real world, it is frequently the case that legal systems beclown themselves and accept justifications like the manager's (for the friends of the system) while punishing those opposed to the system for rights that are written in bare ink on paper, undeniable yet denied, that I think I have so little patience for the show as presented. It's not new, original, or thought-provoking to ask "But what if the legal system ran purely on AI and vibes?", because, well, look at what we've got. Because I could predict what someone in that position would say, to put the "It's for the greater good!" blanket over themselves, since I have heard that excuse and that story many times before...and because I'm more than a little disappointed I didn't get an author willing to actually sit down and think about what a crime (and thus, a criminal) actually was.

So, I think I'm missing something. (I've not watched this show, for reference.) How do you maintain a set-up like the one in the factory when the color vibe check should reveal "Yup, you assholes are all guilty of pre-crime harassment and conspiracy to commit harassment, everyone to jail, every politician that says 'No, this is an economically-vital piece of infrastructure.' also to jail.", and so on?

Or, alternatively, can you hang around and do the same kind of murder-provoking harassment to everyone in society without having a crime index, and since you don't have a crime index, you actually doing it can't be a crime?

Conversely, if the system spits out that a latent criminal has changed their psychology enough that they are no longer likely to commit crimes (as you'd expect them to do after having good, meaningful work inside a system) doesn't that mean that they graduate pretty quickly, and everyone saying "No, they're scum, we've collectively agreed that they're the non-metaphorical underclass we agree to look down upon" is intending to commit crimes against actual-citizens and thus gets immediately vibe-checked and shot?

I feel like either there is a lot being elided here. A society as described can't be both a functional pre-crime enforcement state and a metaphor for modern society, because shooting people with criminal nature and intent regardless of their social status, connections, or cleverness in concealing their actual crimes would wipe out huge swathes of the people we consider movers and shakers in modern society.

So, I'm going to guess that there is a reveal that the Psycho-Pass is fundamentally bullshit, with vast quantities of either false positives or false negatives, because just what I've heard in the first three episode descriptions, it does not sound like the system works as described.

My opinion is that morality is not relative, but neither is it universally shared. Morality is a way for people willing and capable of positive-sum interactions to interact with each other. If you are not willing and able, you are not a moral actor, and likewise, dealing with you is a matter of pragmatism, not morality. You can have a moral war (or at least a war with moral aspects), if both sides are willing to agree on values like "Killing civilians for no significant military gain is wrong." and formalize combat to keep the fighting out of the fields and towns; when one side violates that agreement, then that is no longer a moral issue.

Again, I agree with the wolf; I agree that the wolf and those who carry water for him can and will disregard both honor and morality, and tear down every house and building to loot the rubble for themselves and their fellow-travelers.

A pretty elementary tenant of morality, or reasoning in general, is that you need to be alive to do it (or at least for other people that share your ideals to continue in your stead). If you choose to lay down and be devoured, because you feel that it's as good for the wolf to enjoy your flesh as a meal as for you to keep living in it, then that's on you. And if you hold to a morality that says that the above is the highest virtue, then that morality will end when it runs out of practitioners.

I honestly don't see this as something that can be meaningfully argued. Either you read the above comic and reach for your gun (or give fervent thanks to those around you who pick up the gun on your behalf), or you don't; if you don't, then you're not likely to share enough values with those who do to make discussing it worthwhile.

Morality is way for people who share values to coordinate and make great things. But it is only that. Absent shared values, there is only the pigs shooting every wolf, or the wolves devouring every pig.

"We'll have peace only if and when every last Palestinian renounces violence and accepts the status quo, and until then, we'll keep bombing," no, there will never be peace. At the moment, that does seem to be Israeli policy.

Wait, why not? If Israel decided to ignore optics, accept whatever level of collateral damage as was necessary, and bombed every Palestinian that didn't renounce violence, and only bombed them, then Israel would stop when only the non-irredeemable non-vengeance-monsters were left, and there'd be no more violence, yes?

I mean, given the current state of Palestinian culture, this would be at least genocide in the wholescale and eradication of their culture, and would probably end up being genocide in terms of actual real genocide, yes, but that would stop the violence.

Who has more power, Larry Summers, or some asshole with a rifle, the knowledge to use it correctly, and a willingness to die in the process of murdering Larry Summers?

Thanks and much appreciated.

Can you demonstrate this? I admit, I'm not seeing how making selection criteria harder would decrease the likelihood of cheating. I mean, in the extreme case, if you make a test that only one billionth of humanity could pass fairly, then the odds that any given person passed the test fairly (when there are great reputational and financial rewards for passing the test, no deep culture of investigating and calling out cheaters, and strong incentives to have everyone passively trust the process and not assume cheating as a default possibility) are fairly low.

Again, I'm not an expert on MIT's admission methods, but if, e.g., they hold their own proctored and blind-graded exam for all students they are considering admitting, I'd definitely update in the direction of considering MIT more reliable. But given that my default assumption about colleges (which is that they will cheerfully drop admissions standards into the ground to accept those of their favored demographics, and raise the standards on the unfavored demographics to compensate), I simply do not believe that MIT is honestly selecting students according to fair criteria.

Fair; I am operating entirely off of a few article summaries which specifically mentioned that he traded crypto at Jane Street, and if there is evidence that Jane Street wasn't trading crypto at the time, I certainly don't have either any specialist knowledge of Jane Street or notable faith in the article summaries.

The point remains, however; if my (hypothetical) investment manager was bragging to me about how much money he made with Bernie Madoff, I would seek another investment manager, even if said investment manager decried buying into the Ponzi scheme specifically and even if Bernie had other legitimate investments. It doesn't matter if they came out ahead (plus, I, being a suspicious bastard, would figure that a smart investment manager would make damn sure to conceal the fact that they lost money by not doing basic due diligence on Bernie's fund if they did lose money in it); no matter the outcome, my own trust in someone who put money down on Bernie would go inexorably down.

Did he? I honestly don't know. Has anyone done a post-fall post-mortem deep dive on SBF's time in Jane Street?

I feel like there's a point around good toupees here; it could be that I've been bamboozled by dozens of low-IQ people and just never though to check.

As for my boiling comment, I was making a joke along the lines of room-temperature IQ, in that 212 (F) and 100 (C) are both boiling depending on your measure. And, to be clear, I don't think that SBF is significantly below average, and assume he's between 107 - 115 IQ generally just based on his heritage.

But I put no faith in his words, his presentation of himself, and any evaluation by someone who would either gain by reporting him smarter or be put at risk of retaliation by reporting him dumber as indicators of his smartness. I think that his first talent is shamelessness, and his second is creativity in exploiting trust, and his third is in presentation to limit the number of people who think to check on his first two strengths, and while he could also be quite smart at the shape-rotate-y stuff (and is probably not blisteringly incompetent at it), I see at present absolutely no reason to assume that SBF is "really really good at STEM/maths".

Again, I bring up what those two factors would imply about the pool of non-immigrants; if we assume that groups are blank slates, then we should see the same demographic outcomes in non-immigrants from Nigeria, and from Israel. Do we? And again, does this assumption encode American original-colonist exceptionalism as an expected outcome, where we should assume that the best outcomes should belong to the stock of those that did the hardest initial work on arrival? Do you think there is any reversion-to-the-mean going on, and at what rate?

And yes, I'm being vague. From my perspective, I'm a guy who can watch the night sky and has an OK memory hearing astrologers confidently announce that a plague is happening because Mercury is in retrograde and that is what causes plagues, and lining up that with the other times I know that Mercury was in retrograde and there was no plague. I am not a doctor or a microbiologist or even an astronomer, but I don't need to be; all I need to do is evaluate "Does condition X, which I hear people claim as the reason for this observable event in the world Y, actually correlate with Y, or do we have cases of X not causing Y and in fact being associated with the opposite of Y?"

My own default position is vague because it's complicated. My thoughts are that sets like black Americans and Jews are a huge, confounded mass of distinct lineages and cultural influences, and that what might be true about subsets of those groups could not be true about the whole. My default position is that while knowing someone's race gives you information about their likely group outcomes, every group contains diligent sinners and callow saints and that looking at the individual in front of you and tracing their specific life outcomes to their specific choices and reactions to the events of their own specific life is the only way to get a non-statistical answer.

And so, if anyone is going to say "But for X, these groups which have wildly divergent group outcomes would have near-identical ones.", then they'd better be able to show the general principle first that groups are not distinct in the absence of X, and second that X moves the needle for a high confidence interval of groups that I can think of in the expected direction."

My own personal answer is red, for the general reasons delineated below.

For the people who choose blue: does the presence of this vigorous debate change your opinions any? I know that while my first thought was red, the fact that this has become a thing, and that there is no obvious common consensus, is more than enough to permanently cement me on Team Red. How much baseline expectation of people picking red no matter what do you need before your choice comes down to "Everyone who picks blue dies, which includes me." and "Everyone who picks blue dies, which doesn't include me?"

I do not think that you claimed that we had factual equality in good faith. I do not think that you actually believe that hate crimes committed by black people are given the same degree of attention and seriousness as hate crimes committed against black people (despite the relative numbers and severity of these two categories), and I think that you are dismissing this claim as being in bad faith in bad faith yourself.

Can you tell us what would convince you that there is not, in fact, equal treatment of black and nonblack criminal behavior? If you were given another similar incident, or another five, or another dozen? How about if you were shown a statistical gap between the amount black people were prosecuted as a group and the amount of crimes that they committed?

For me, it's just parsimony. I have no faith in colleges in general any more, and Sam has demonstrated his ability to manipulate elite institutions, and is also a goober. It could be that he is high-IQ and a goober, or that he's a goober all around and simply bamboozled MIT the same way he did a bunch of others - or it might be that he simply had a lucky interpersonal-connection 'in' to MIT that would have worked whether his boiling-hot IQ was measured in C or F.

I also think it's probably fairly unlikely that we'll get hard data either way, especially now, given the extended time that's passed from any potential objective-ish evaluations like AP exams or SATs. Also, nowadays, it would be trivial for someone of his resources to game a few slightly-harder-to-fake signals, like a ghostwritten StackOverflow profile and some boilerplate personal projects on a GitHub account.

But given the sheer stupidity and utter agnosticism towards the very idea of personal consequences he has displayed so far, I feel like I can safely say he's probably real dumb, in the classic sense, excepting for his ability to lie and manipulate people (which, in fairness, is a non-trivial skill, but also one not necessarily interacting with deep math, science, and programming geekery). I predict that no actual evidence of Sam doing anything difficult and valuable with anything not vulnerable to social pressure, where the results can be verified (meaning mostly math or programming, since those are things I feel I could verify myself) will be found. I am not super-invested in this theory, and I happily admit this is purely a balance-of-probabilities as I see it; I'd be delighted for someone to turn up, e.g., a deeper dive into Sam's LoL rankings, if nothing more objective can be found.

If historical discrimination was the primary driver of group outcomes, then we could look at two groups which had suffered similarly, and confidently predict that they have similar outcomes relative to an undiscriminated-against control group. This is not the case; you can suffer historic discrimination and be either wildly above-mean in outcomes, or distinctly below-mean in outcomes. (And, of course, you can as an individual be in the above-mean group and fail hard, or in the below-mean group and succeed hugely.)

Because discrimination does not necessarily lead to lesser outcomes and can in fact lead to greater outcomes, it cannot be the primary driver. At least, that's how I'm understanding the term; if you have a different understanding, please feel free to elaborate. Or alternately, if you want to claim that Jewish and Asian overperformance in the face of discrimination is a historic fluke specific to a place and time, reminiscent to a legless man winning a marathon due to a series of freak coincidences (while having functioning legs is still generally the primary driver of winning footraces) and that we should expect to see Asians and Jews with comparable outcomes to American blacks in other areas and times, feel free to make that case as well.

With respect, I think you're misreading me. I am making no argument as to why there are group differences; I am simply pointing out that a frequently-given explanation (prior trauma) is clearly and obviously a non-answer. Clearly, races are distinct in terms of outcome at the group-aggregate level; equally-clearly, we see that outcomes of races do not correlate directly with discrimination.

I do think that we've got a lot to untangle if we did want to claim it's all group-founder effects. If we posit that the primary determinant in group outcomes is subgroup selection and founder effects, we could also look at the outcomes of indentured servants who were shipped across to the burgeoning Americas as well; if selection effects explains all, then we should find a clear delineation in demographic destiny between the children of free colonists and indentured servants (who we would expect to be very close to the descendants of slaves). We could also dig into the histories of Irish immigrants who came over en masse in response to the Famine, as well as digging into those sentenced to transportation to, e.g., Australia.

Of course, the big issue is that if the secret sauce is selection effects and we're just getting the cream of the crop from various nations, then we could look at the pool of people who didn't immigrate from various nations and see if the world really is divided into Economic Go-Getters and Everyone Else. And now that I mention it, wouldn't the descendants of the original American colonists be the ultimate economic migrants? Shouldn't we see parity between their descendants and the others?

Again, I make no claim as to why group outcome difference exists. I just note that it does, that it's durable, and that historic discrimination doesn't account for it. As far as I'm concerned, the reason for different group outcomes is that people are different, and groups contain different people, and because I believe this, I am very skeptical of any "But for Factor X, these groups of people would have identical outcomes."

HBD is an answer to the question "Why do different races have persistent group outcomes?". I was asking a different question; specifically, "Why do people frequently claim something clearly specious (group rates of discrimination) to explain (one particular set of) group differences, when literally minutes of thought and research is enough to disprove it?"

I mean, the answer might be "Claiming anti-black discrimination explains all group outcomes is a matter of Wokist doctrine solely, and no one ever advanced the argument in good faith, and its common presence simply indicates how far public discourse has fallen.", but I figured I should at least ask people to take stabs at the argument first.

Huh. So you can make a husky act like a basset hound in terms of general laid-backness and temperament with a single gene tweak?

Respectfully, I doubt this. Can you link any sources to this effect?