In theory there's no reason they can't choose it at random.
And the fact they presumably don't even attempt to choose it at random betrays a belief in the heritability of behavior and physiology that is incompatible with the modern liberal worldview.
Uhh, do they actually have a choice? Can they say they only want kids with successful healthy parents?
My understanding was maybe they can specify gender and/or race, but otherwise it's so hard to find children to adopt *of a very young age* that they have no choice but to take what they're offered if they want to adopt.
This appears to be an attempt to smuggle in the assumption that there are, in fact, no "fair" differences in the outcomes by race, and any predictor which predicts racial disparity is biased. In fact, they specifically say that the predictive model and the protected characteristics will be independent. That leads to some very unfair predictions when the protected characteristics and the outcomes are not independent. For instance, if we're predicting defaults on a loan, and we use just credit score, we'll predict that whites default on loans less than blacks. If we produce a "counterfactual fairness" predictor based on race and credit score, then we won't predict that -- but the predictor will be blatantly unfair, since it will either overpredict white default rates, underpredict black default rates, or both.
The "counterfactual fairness" paper mentions that impossibility theorem (pp 5-6), and claims to sidestep it by being fair in neither. Which it isn't; it's straight-up equality of result, with a lot of mathematical window-dressing and some technical use of the word "causal" to make it less obvious. With their version of "causal", if black people tend to drive red cars more than white people, then blackness is (according to this paper) a cause of driving red cars.
Their notion of "counterfactual fairness" is not some general notion of counterfactual fairness, but specifically the notion that changing the protected characteristic while holding things not "causally dependent" (using causally in that technical sense) constant should not change the outcome. This is a strong version of equality of result -- stronger than demographic parity.
They give an example of a car insurance company using red cars as a proxy for aggressive driving being unfair because (hypothetically) people of a certain race prefer red cars. They describe the causal structure here as A -> X <- U -> Y, where Y is accident rate, A is race, X is red car, and U is aggressive driving.
Why would you want to rebalance average levels of police focus away from high crime neighborhoods and into low crime neighborhoods and thereby cause a lot more people to be victims of crime? Should we have lots of fire hydrants around buildings that can't catch fire, even if we have to remove them from places near old wooden buildings?
Is that an attempt to make 14% seem like a large percentage? Are you saying the preferences of the population of 7 average sized states should outweigh the preferences of the population of 42 average sized states?
How about you use actual data instead of anecdotes? I can find people who say they're scared of the crime in their high crime neighborhoods and want the police to protect them.
"even well-intentioned policing yields unfair outcomes along racial lines. I won't try to paraphrase"
No, you had better try and paraphrase it or don't bring it up at all. Making a strong claim and then then directing anyone who disagrees with you to read an entire book is totally unreasonable.
As it presently stands, arrest rates and crime victimisation survey offending rates line up extremely closely for race. Any changes to policing you support necessarily have to simply involve black criminals getting away with crime more often than they currently do. Which strikes me as an extremely marginal definition of "fair".
Alexander starts from false assumptions that the rate of criminal activity does not vary between races, and in the case of drug use, she uses a study based on *self-reported rates* of drug use. Studies that have used objective measures - such as blood or urine testing - show different rates by race, sex, and class/SES.
There is actually sparse data showing that per mile driven, traffic violations do not vary by race, and actually circumstantial evidence (accident rates, vehicle injury rates) that they do.
To be clear - none of this is indicative that any particular individual is using drugs or driving recklessly, nor that any particular cop is operating under an unwarranted negative bias. Still the effect of drug dealing and reckless driving are felt primarily on the community where those individual harms occur, piling up on the other, non-drug using, safe driving (and pedestrian) population.
In her (impressively dense and footnoted) work, Alexander focuses on the alleged unfairness to people (mostly able bodied young men) accused of those crimes and ignores the plight of those who are its victims - where children, women and the elderly are over-represented.
In a better world, those able-bodied young men would - as a group, and under the direction of law-upholding elders - be acting as protectors for the more vulnerable of the community. Unfortunately, this role has been outsourced to the police & to people outside the community, who are hampered by lack of knowledge, lack of community connection, and lack of community cooperation.
"do predictive policing but actively rebalance until average levels of police focus match across racial categories."
Okay, you're just saying that policing should not be about crime reduction if this is the line you want to take. You're claiming that differences in outcomes are necessarily "unfair", but these differences are objectively caused largely by behavioral differences.
You're ASSUMING that behavior and demographic group are independent variables, but they're most certainly not. If people belonged to a different demographic group, either their behavior would be different, or that demographic group would now have a different e.g. crime rate.
Differences in crime victimization are also a difference in "outcome". Equalizing inputs (police presence) means accepting unequal outcomes in such a case.
Related to 22, I am increasingly obsessed with Peter Turchin's concept of elite overproduction (which you've written about) and find that it explains so much of modern politics, media, and academia, and doesn't bode well for the future.
I also find it plausible, though I've never been able to find a place where he explains it in enough detail for me.
It seems like there need to be two definitions of "elite", one defining how many people think they "ought" to be elite, the other defining how many people feel satisfied that they've made it.
As far as I can tell what we're seeing is something like "many people feel like they deserve a type of upper-class creative professional job, but aren't getting it". Are there fewer upper-class creative professional jobs than before? Do more people feel like they deserve those jobs than before? Why?
I don't remember seeing an answer to any of this in the two Turchin books I read. If I had to guess, it would involve a failure of college's reputation (place that guarantees you an upper-class creative professional job) to catch up to its reality (place that half of people go to). But this seems less like an inevitable world-historic force, and more about some particular policies and cultural norms catching up to us.
I published a post about how it's really hard to make it as a writer, much worse than many young people believe, and that you should definitely have a day job and try to do something else if you can. I got two types of emails. The first was from people who had been professional or semi-professional writers for years and finally gave up because the attempt to climb the ladder was so bruising and frustrating. They were very grateful I had written it. The other was young writers trying to make it. They hated it and assured me that I only wrote it because I have not had sufficient success myself.
The rage of the latter - who, I'm sure, were all perfectly capable of getting nice corporate jobs at Nationwide Insurance or a nonprofit that works in the vague area of sustainability and having very enviable upper-middle class lives - I think there's something really powerful there. I detect such an immense frustration among youngish people about the sense that they have to occupy some sort of vaguely artistic or creative or intellectual occupation, that this is their birthright, and their continuing inability to secure a solid income by doing something that their peers recognize as enviable fills them with a profound sense of injustice.
Maybe it all comes to nothing, I don't know. But god, there's so much unfocused rage about that stuff.
I feel this. I was told I was "gifted" as a kid, and raised to get one of those elite jobs. Working for McDonalds was constantly used as a threat to me: "do your homework or you will end up working at McDonalds!" I attended one of the best, most classy public schools in the country. This climate constantly villainized common jobs. I remember my high school counselor giving a speech about how embarrassed she was that so many (only a small portion) of our students "only" ended up going to community colleges.
I was a fantastic student. But I got really fucked up when I became an adult and realized that I had been lied to all my life. Nobody in the real world gives a damn about academic performance. I had been so focused on academics that I actively shunned learning social skills, but it turns out that is exactly what you need to get an elite job.
I really like working with my hands and building stuff. Intellectually I know I would be really happy in a job like machinist, carpenter, or electrician, and in fact I have picked up a decent amount of experience in all three of those fields over the last decade doing informal work for family. But I can't actually get over the mental block required to actually apply for any kind of job like that because of how heavily the culture I was raised in told me that that would be F A I L U R E. I wish nobody had told me what to want as a kid.
So here is my take: The boomers and gen-x'ers presided over a period of profound technological and societal progress and growth, which was wonderful, but unsustainable and impermanent. This caused them to become extremely optimistic and make wild and outlandish promises to us millennials about what our lives would be like. But then that extreme progress reverted to baseline, everything stabilized to a much slower rate of growth, and while the world millennials are inheriting is in every way better than the one the boomers and x-ers were raised in it is so much worse than the world we were promised that it has led to this terrible malaise and sense of failure among so many of our generation. So let's maybe chill out a bit with the zoomers and stop stigmatizing normal jobs.
That wall you feel like you hit may be partly your non-upper class background. “Social skills” can range from not picking your nose in public, to the social style of the rich. Your post is pleasantly introspective and humble; your social skills are probably fine or close to fine. The educational sorting engine promotes upper class/upper middle class behavior. They don’t tell you when you hit that wall, but it’s real.
You’re right about the stigma about jobs though. That’s also a middle class formulation. “If you linger too close to the workers, we’ll lose all the progress we made!”
If you can do the job safely and successfully you can find permanent employment in a trade and probably more acceptance than you expect. The sense of failure affects you more if you buy in, so don’t buy in. Who has the hustle and the success? Go do what they’re doing. Educational eliteness is/can be vastly toxic in the US but the effects don’t have to be permanent. You can get over that mental block :) and on the other side you will find many like you. I escaped, mostly. Just remember money actually does matter.
Your initial point is very correct. I've had just about everything handed to me and achieved all my dreams mostly because I was raised to be able to socialize in an elite way. It's insane how important it is in having a truly excellent life.
I'm in that generation and also got the threat of McDonalds jobs if I didn't improve my grades. But McDonalds and other fast food jobs really are crap jobs. They typically pay minimum wage (and no tips), have variable and lousy hours, bosses who suck more than usual, etc. I don't think I ever got threatened with "do your homework or you'll have to be a mechanic like your uncle", or a carpenter, or an electrician. Somehow the idea of trades jobs never came up at all, possibly because I was a stereotypical '80s computer geek from the start. But back then, computer geekery only led to a regular white collar career, nothing remotely elite.
The tail end of Generation X and the vanguard of Millennials graduated into the .com boom -- and the subsequent bust. Technological progress and growth didn't stop in 2000, though. Millennials have little to complain about.
An interesting story, for me. I'm at the tail end of the boomer generation, and had a "gifted" academic trajectory -- I went to a first-rate high school, a college that routinely ranks in the top 5 internationally for science and engineer, and the best graduate school in my field. I won scholarships, and later fancy grants from the Federal agencies.
But no one in my experience ever stigmatized work qua work. I worked at least part-time from the time I was 15 onward, and that was totally expected, among my peer group and in my family. My family and teachers were totally supportive of my academic exertions, but at the same time they took it for granted (and so did I) that I would enter the workforce no later than my mid-teens, and that taking a McDonald's job after school and in the summer was not just honest labor, worthy of ample respect, but an expected thing if one were not to be thought kind of a fancy pants putz with soft hands and probably a bad attitude.
Certainly if I had ever said I thought flipping burgers or pumping gas was beneath me, as a 17-year-old, just because I scored a 5 on the AP Calculus BC exam, my mother would have slapped me and I dread the look of contempt I would have seen cross my father's face.
It's laughably ignorant to claim the world we inherited is "in every way better." That's true for me, and it may be true for you and many others, but it is very obviously not universally true.
Yeah, that probably wasn't the correct way to put it. I think I said it that way because... in the past I have found myself overly romanticizing the past, like, claiming that farmers in the 1800's were better off than today's poor, and have been thoroughly enough disabused of that notion so am to some extent over-correcting. There are certainly a lot of ways the world has been getting worse recently; I am obviously not very happy about how things are going. But I do think "better on average" is a good way to put it.
It’s partly of a really strong “not an office job” idea people have had for the last 15 years or so, combined with both increasing meritocracy and increasing belief in meritocracy - back when access to elite (in the broadest sense) jobs was basically hereditary everyone else just shrugged and went off down the widget mines, but now people think they can fight for them enough will to make it, well, a fight.
As opposed to what? It's become fashionable to blame capitalism for everything, when there are no examples of other systems doing the thing in question better, or even on a comparable level.
We are. People just don't like that idea because it means that they aren't good enough.
All data points seem to point towards people being more likely to rise or fall based on merit rather than other things today than they were historically; for instance, more people are in the upper middle and upper class, a higher percentage of billionaires didn't inherit their wealth, etc.
On top of that, there is a higher demand for credentials and past work experience, suggesting that examining people for merit is increasingly important. Indeed, a lot of the complaints about jobs having high entry requirements suggests that we require more merit to even be considered.
What exact data are you talking about here? Certainly not recent US data where social mobility has largely stalled. A fully meritocratic society would have only 20% of the top quintile stay there (while 20% would fall to the bottom quintile) and vice versa
Part of it is the existence of people like ... the two of you.
You both have jobs that involve saying your opinions on the news of the day, and getting paid enough for it to have it be your day job. I think both of you put a lot of thought into what you write, but there's a lot of people who seem to spout random nonsense on Twitter and somehow get lots of followers and probably a good living. Like I dunno Ben Shapiro.
Easy to see that and think "I'm as smart as that guy why can't *I* do that???"
I sometimes wish I was one of those guys, instead of my real, well-paying, not-particularly-enjoyable, working-right-now-instead-of-partying-with-friends job. But it reminds me of what Matt Yglesias said about NBA owners - it's not just an investment, it's a prestige position, and there's no reason ex ante to expect to get paid for it. The world is filled with people who want their opinions to hold sway. The truth is I'd pay to wake up tomorrow with 500,000 Twitter followers who respected my opinions on politics et al.
It can be true that twitter makes people dumber and also that having 500,000 followers on twitter who respect your opinions makes a person more influential than I'll ever be.
This phenomenon is noted by Eric Hoffer in "The True Believer", a book on the origin of mass movements. According to Hoffer, wealthy young people who believe their standard of living will not advance as far as the previous generation's are fertile ground for mass movements. A reservoir of rage.
I'd love to provide some quotes from Hoffer, but I can't seem to keep a copy on hand. I always find someone who needs to read it and I give them my copy.
" I detect such an immense frustration among youngish people about the sense that they have to occupy some sort of vaguely artistic or creative or intellectual occupation, that this is their birthright, and their continuing inability to secure a solid income by doing something that their peers recognize as enviable fills them with a profound sense of injustice." Well, Freddie, for what it's worth I like that sentence and think you write well, and I hope your writing makes you a good living.
Movies and self-esteem-koolaid-drinking-teachers keep telling people they can be anything if they believe hard enough and work hard enough, but that's a lie. Most people are just not born with the talent to be an upper class creative professional. Blank slatism leads to a lot of people beating themselves up for not achieving things that were never in their power to achieve.
I suspect there are two things here that people are improperly conflating:
1. Having an outlet for your creativity. Eg. writing.
2. Being paid for #1.
There are lots of jobs out there which will provide you huge amounts of free time to write "on the clock". Night watchmen was historically one of those. Hotel night auditor is great. Likewise, there are other jobs which will pay well to allow someone to make the most of their free time to engage in artistic endeavors. "Shakespear in the Park" style productions allow anybody to be a player.
But very few jobs will allow you to make a living for being artistic, and there is indeed a lot of competition for that space. And I suspect that is where the ennui comes from.
At the risk of making the system even more complicated, "deserve" in Elite_3 as you define it gives me pause.
There's a difference between "people who we might deem to ethically deserve elite jobs, and all the privileges thereof, if they can get them" and "people who have the skills such that it would be, in brute utilitarian terms, beneficial to society if they were taking up the elite jobs, regardless of whether they 'deserve' them".
For example, you can imagine a least-convenient-possible-world where, because they got better education and stuff, the children of the previous aristocratic generation are the most qualified, bu they're selfish and bigoted and corrupt (though not quite so corrupt that they're not still the most efficient at the end of the day) — while you have a class of would-be elites from poorer backgrounds who maybe wouldn't do quite as good a job but are earnest and meritorious, and morally "deserve" better than to waste away in jobs that are well beneath them while jobs that are *slightly* above them, but which they *could* do "well enough", are taken up by the oily aristocrats.
While I can see that it might suck to live in the world where the "oily aristocrats" rule while the marginally less competent but more "morally deserving" alternative have little power, I'd much rather that the past leaders of my society had been the absolute most skilled ones regardless of other concerns - 'fair' division of the pie today maters to us, but what matters to our grandkids and *their* grandkids is how fast we're growing the pie, and I think that thinking for the long-term future is what brings out the best in humanity.
P.S. I also think it'll generally be unlikely that selfish and bigoted and corrupt men are the best available leaders - while I grant for argument that the education advantage of old money can be very large, the best leaders in your toy society will most likely be that small fraction of the children of aristocracy that care about the wellbeing of others and are honourable.
Turchin does give an answer: it was the availability of mass college education and gender/racial equality opening opportunity to more people. Anyway, I really don't like Turchin but I do think there's increasing elite conflict. I just don't think it's caused by some kind of quasi-Malthusian overpopulation.
Also some speculation on my part. I never had a wage job in the 60s (because I wasn't alive then). But my impression from movies is that they had a little more freedom and discretion than wagies do now? It seems like back then, a shopkeeper/cook/factory worker was still expected to think for themselves, at least a little. Now, it's just really strict that you *follow the system*, with layers of managers and tracking systems to make sure that you do. I'm thinking of the difference between "Glenn Glarry, Glen Ross" where salesman work in a boiler room using whatever tricks they can think of to sell real estate, vs the part in "the office" where Michael (an experienced salesman) is constantly criticized by his manager for "going off script", ie saying anything at all that wasn't completely scripted.
Bottom line: not only is there more competition than ever to be an elite. But if you're not an elite, you are *acutely* conscious of your status as an underling.
That is correct in my limited experience, but I think it comes from the lopping out of many layers of lower and middle management. Your line schmo in the 60s was allowed a bit of freedom to think for himself *because* he had a foreman keeping an eye on things and backstopping his thinking, and the foreman had a shop manager watching him, and so on. There were a lot more layers of management. Many of these have been replaced by software and devices, and these are necessarily more rigid -- lacking human intelligence and flexibility -- and so the job of the upper manager becomes more precise, less telling a middle manager the general idea and more writing of scripts and programs, and the job of the line worker (or very low manager) gets extremely rigid because of the increased distance between him and the guy who makes the big decisions.
I worked in the corporate world from 1982-2000 and have supported myself as an opinion journalist since then, so I am out of touch with contemporary white collar work. But from what I hear, it sounds pretty awful.
I'm a white collar worker right now for the state.
I have an enormous degree of flexibility and personal responsibility.
This seems to be the case for most of my colleagues.
We do have various systems for trying to get stuff through (like for instance, if we create some Standard Communication With The Public, or some Standard Form for people to fill out, we run it through comms, and can't change it willy-nilly because the Spanish form and English form have to be the same), but like, in terms of personal discretion, they don't really care how we do our work, just that it gets done.
I create procedures to help people so that they can do work in a repeatable fashion, and that stuff is helpful to people - but those procedures also involve a lot of automation, so you can do your work way faster as a result, meaning that you have more time to do "the tough stuff" (which is also the interesting stuff, as it is the stuff that is NOT repetitive).
Indeed, it feels like the constant thing of being white collar is figuring out a way to automate all the repetitive work so that you can spend most of your time chasing down the things that aren't repetitive, as those, while a minority of incidents, take up the bulk of your time because they have Problems and don't fit into the system cleanly.
But those things are more interesting than the repetitive stuff, and their one-off nature means that it's more interesting.
If these "one-offs" stop being one-offs, you start looking for ways to stop them from happening so you don't have to repeatedly deal with the same issue.
What seems to makes US elite overproduction more intense (to this outsider) is that only a handful of universities, who limit their supply of admissions, are seen to produce elites.
A lot of elites (as in actually talented and educated people) outside the US move to the US for work or academia and fewer go the other way, which should worsen/lessen the problem in the US/outside.
Mancur Olson is probably the first who made this argument in any detail, in his 1982 book "The rise and decline of nations: Economic growth, stagflation, and social rigidities." Quoting from memory, Olson believes that rigidities in the form of an increasing number of "veto points" follow the growth of special interest groups. Rigidities are weakened during major wars (who tend to weaken special interests by creating a mentality that "winning the war is so important that everything else must yield"). However, the process starts again when peace is re-established. The hypothesis is that the longer the peace lasts, the more "veto points" build up, and the harder it becomes to get anything done. (Notice in this context that we now live in the second-longest period without a world war in modern history, only beaten by 1815-1914.)
Olson was a creative political scientist, and the book could be worth an ACX review.
Ok, but why then do people expect that going to college should give them a job that's not just in the top half? I think the answer is that we expect that growth means that we'll be better off than the previous generation. But even when that's true in absolute terms the issue with education is that the kind of job that an education guarantees you is below what it was in the past (certainly in terms of status and possibly even in absolute material terms).
What this suggests is that we shouldn't expand admissions to educational programs with the same name. So if you want to expand the number of people you educate in college instead create some kind of new program without the baggage of the old. But maybe that has other disadvantages.
BTW a good case study to think about this issue is the many academics who seem to feel that it's an affront to justice and fairness that it's really hard to get a tenure track academic job. From the outside that's just nuts. I mean, no one would take it for granted that just because they played basketball or studied acting for 10 years they were owed a job.
Maybe because college still trains us as "future elites" rather than slightly-above-average? You get a campus filled with luxury amenities, classes that are liberal arts and sciences rather than anything connected to a job, and it ends with a fancy graduation ceremony where you wear a robe and hear some Latin phrases. Even the generic "business" degrees seem designed for future executives and entrepreneurs, rather than the low-level office jobs they'd be more likely to get.
Turchin defines the elite as those in the upper classes, the wealthy and the powerful. In Rome, these would be in the senatorial class or its equivalent. In traditional societies it would be the upper aristocracy, the dukes, and likely the royal dukes at that. In the old USSR and modern China, it would be the inner party members, the ones who might serve in the Politburo.
Turchin argues that nothing really changes without the elite being involved, since they are the ones who run things. Being elite has nothing to do with actually being better; it is just about being closer the the center of power. Russia was taken over by the communists because the Russian elite under the tsar would no longer defend the regime.
He argues that there are times, often after a destructive war or the development of a new technology, that there is a good supply of resources versus population, and that tends to be seen as the "golden age" when "merit" was rewarded and new members were added to the elite. The elite can expand even as living standards rise. At some point, as the elite grows in size, elite competition becomes more common and eventually destructive.
Turchin's use of the term "elite" has nothing to do with the competence of its rule or some arbitrary metric of merit save for those introduced as propaganda. In our society, it isn't the banker making $500K or $1M a year, it's the people who own the bank or are living on the fortunes enabled by its founding.
Freddie: Both the tweeter and Turchin are not original:
"Capitalism, Socialism, and Democracy is a book on economics, sociology, and history by Joseph Schumpeter, arguably one of—if not his most—famous, controversial, and important works. It's also one of the most famous, controversial, and important books on social theory, social sciences, and economics—in which Schumpeter deals with capitalism, socialism, and creative destruction."
"Schumpeter's theory is that the success of capitalism will lead to a form of corporatism and a fostering of values hostile to capitalism, especially among intellectuals. ... The term "intellectuals" denotes a class of persons in a position to develop critiques of societal matters for which they are not directly responsible and able to stand up for the interests of strata to which they themselves do not belong. One of the great advantages of capitalism, he argues, is that as compared with pre-capitalist periods, when education was a privilege of the few, more and more people acquire (higher) education. The availability of fulfilling work is however limited and this, coupled with the experience of unemployment, produces discontent. The intellectual class is then able to organise protest and develop critical ideas against free markets and private property, even though these institutions are necessary for their existence. ..."
Schumpeter's theory of elites was popular with the center-right business press about 40 years ago.
Oddly, though, I don't recall people using the word "elites" much until 1999 when I went to an elite conference (e.g., a former prime minister was there). There, everybody used the word "elites."
The part of the FDA/Public health/Trump conspiracy theory that I just can’t get behind is that it proposes a counter factual where the FDA acts with enough haste to approve before the election.
Feels like a rare case of Cowen et. al. letting their vaguely anti-woke bias (I don’t say that pejoratively) overcome their extremely justified skepticism of the clowns at our public health agencies.
Please see Matthew Herper's 11/9/2020 article in Stat News in which he quotes Pfizer executive William Gruber on Pfizer's extraordinary step of shutting down lab processing in the world's most important clinical trial from late October until the day after the election:
Pfizer shut lab processing down for a very simple reason, Steve: they had informed the FDA of the proposed protocol change in late October. If they had 32 cases before the change was approved, the protocol would have required them to report the results to FDA. In addition, they may also have been required by the U.S. Securities and Exchange Commission (SEC) to make the results public. If the results of an underpowered 32-case analysis had suggested that the vaccine wasn't that effective, it would've been bad and potentially catastrophic for the company (recall that, at the time, the fear was that the vaccines would only be around 50% effective).
So, they shut down lab processing to ensure they couldn't confirm cases and therefore would avoid a protocol violation. On 3 November, FDA approved the protocol changes, and they resumed lab processing on 4 November.
Pfizer shut down lab processing in late October because they didn't want to know where they stood in the world's most important clinical trial.. They restarted lab processing the day after the Election because it was now safe for them to know.
The first sentence is correct: they didn't want to know because they didn't want to conduct a 32-case interim analysis. This was because it was less likely to lead to good results than a 62-case interim analysis, which was now possible to be done quickly because the US epidemic exploded in mid-to-late October.
If they had continued lab processing, they would have been forced to do an interim analysis at 32 cases, because their protocol change needed to be approved to prevent them from being forced to do a 32-case analysis.
If Pfizer had not shut down lab processing until the day after the election, Pfizer likely could have delivered a 62 count result before the election. But it did not want to do that, having come under strong pressure all fall from Biden, Harris, Democratic doctors, and the media to slow-walk the clinical trial until after the election to deny Trump an October surprise.
For example, see this exultant November 1, 2020 New York Times news article (not an oped) announcing Trump had failed to get a vaccine October surprise:
Welcome to November. For Trump, the October Surprise Never Came.
Trump’s hope that an economic recovery, a Covid vaccine or a Biden scandal could shake up the race faded with the last light of October.
By Shane Goldmacher and Adam Nagourney
Published Nov. 1, 2020
Updated Nov. 3, 2020
President Trump began the fall campaign rooting for, and trying to orchestrate, a last-minute surprise that would vault him ahead of Joseph R. Biden Jr.
Also, note that Pfizer kept their decision to shut down their lab processing in their clinical trial secret from late October until the day after the Election from the public, investors, and the White House until William Gruber's interview with Matthew Herper of Stat News on 11/9/2020.
At least two Pfizer stock analysts issued downbeat recommendations on Pfizer stock in late October on the grounds that if the clinical trial were going well, Bourla would have announced like he'd been promising. Indeed, Pfizer's stock drifted down when efficacy was not announced in the original time frame, in part because no explanation was given to anybody other than FDA insiders that Pfizer was not following their published protocol for commercial-political reasons.
From a normal ethical standpoint, secretly shutting down lab processing in the world's most important clinical trial is obviously dubious.
However, from the standpoint of the higher morality of denying Trump an October Surprise that his emphasis on vaccines was working, it's saintly.
That sounds like a perverse incentive from our regulations. Andrew Gelman would say something about a failure to do statistics like a Bayesian (or to do a stupid version of frequentist/Fisherian "null hypothesis testing"). He's talked about not have a set point to stop collecting data, but instead to include in your model what drives your decision to stop collecting data. And one should be able to update on receiving additional data, and our regulations should not incentivize people to stop collecting data.
I’m really struggling with the part where Pfizer decides a 32 case interim analysis is unacceptable, and in fact so unacceptable that is worth deferring the world’s most important clinical trial ever. Presumably cos these chumps are rubbish at designing these kinds of trials. And there’s no way around that allows us to progress the science like (for example) do the interim analysis as agreed but issue a joint statement with FDA saying they don’t think that’s sufficient and will wait for the 62 case milestone. Like there’s no alternative to just not knowing.
And of course we’re all just to pretend that Nov 4 is just a coincidence. And we’ve just to ignore that Topol himself made the link between the date of the trial and the Election.
It’s a complicated story you’ve got there Edward. It’s possible, but it seems there’s a more parsimonious explanation.
"recall that, at the time, the fear was that the vaccines would only be around 50% effective"
You have probably followed this more closely than I have. What is the evidence for that claim?
My interpretation of what happened is that if the vaccine turned out to be about 90% effective and the requirement was 50%, the people at Pfizer almost certainly had enough data prior to unblinding, from the earlier stages of testing, to be pretty sure it would pass. Why is that mistaken?
Prior to unblinding, how could they be sure of anything? Blinded data means not knowing whether the sick people are the ones you vaccinated or not.
And, yeah, a priori, a <50% effectiveness was plausible. Influenza vaccines made using traditional technology are often only ~50% effective, and what little we knew suggested that coronaviruses would be at least as hard if not harder to vaccinate against. Since we had no prior experience with mRNA vaccines in humans, "what if it is no better than a crappy flu vaccine?" is a reasonable concern.
It still takes a perverse incentive to deliberately postpone knowing that, even at the intermediate-result level. There may be several plausible perverse incentives to explain this one, e.g. the effect on Pfizer's stock price of an intermediate report of 40% efficacy.
Several non-mRNA vaccines have had trouble coming up with an impressive efficacy percentage. For example, mighty Merck pulled the plug on its vaccine effort on 1/25/21 because it could see from its phase one trial that it couldn't beat Pfizer's and Moderna's efficacy:
My impression is that the Pfizer and Moderna saw from their earlier clinical trials in which they measured how much antibodies their vaccines produces (a lot) that there was an excellent chance that their vaccines would smash thru the FDA's 50% efficacy barrier.
On the other hand, the media tended to be fairly gob-smacked by the >90% efficacy claimed by Merck on 11/9/20.
Pfizer CEO Albert Bourlas's public statements in August-October 2020 suggest he had good reason for optimism about his vaccine. On the other hand, he seemed slower in figuring out the politics of an announcement before election day and thus kept suggesting that his vaccine would be proven efficacious in October. He finally wised up and shut down lab processing in late October until the day after the election.
If Bourlas had delivered Trump's October Surprise, he would be, to the prestige press, history's greatest criminal. But Bourlas managed to not do that while still beating Moderna by seven days.
Attempts to convince me that delays were a result of extraordinary political considerations are directly at odds with attempts to convince me that delays were a result of institutional ossification. These are competing explanations.
I don't know what about the linked letter would move the needle specifically towards the former.
is the OPs objection is the explanations are functionally at odds? the first requires some degree of coordination and flexibility, whereas the second claims these things are missing
Methodologically, if I don't share your priors appealing to them to try and convince me is a non-starter.
NIMBYism is a hell of a lot more complicated than mere rent-seeking, and your milage will vary *wildly* about what tactics it employs in any given area.
Your opponents are not a homogenous self-reinforcing block, and if you attribute the wrong motivations to them you will lose the ability to find effective resolutions.
I disagree with NIMBY but I agree that it's unhelpful to model NIMBY as rent-seeking. For one thing, the effects are community-wide rather than individual and mediated through whatever the local political norms are.
Here's a model I do think is helpful: communities face decisions about whether/how to provide certain public goods like traffic management, cleanliness / orderliness, neighborhood continuity, nice looking streets, and so forth. They can either do that by devoting time, money, and attention to addressing issues, or by refusing to grow so the issues don't come up. The latter has a much higher cost in aggregate utility but that cost is an externality-- it's being paid in higher rents and the reduced utility of people who can no longer afford to move to the area. The cost as perceived by decision-makers (usually homeowners and/or rent-controlled renters) is much lower. This produces effects equivalent to rent-seeking without any need to model NIMBY supporters as consciously wanting to increase their property values. It's sufficient that the community faces a wide range of decisions where there's an option to push the costs onto renters and prospective residents rather than having landowners bear them via property tax (or better yet, Georgist land tax).
These are not competing explanations at all. The idea that the people in charge of things don't really care about doing their job (helping the greatest number of people be treated for diseases) is perfectly compatible with the idea that the people in charge of things are focused on short-term political enmity.
> doing their job (helping the greatest number of people be treated for diseases)
Institutional ossification in a nutshell is that these are very much different things, no matter how convenient it would be if they were better aligned.
I think he's arguing that the ossification is sufficient to explain the letter, and since we have a strong prior for ossification, the existence of the letter doesn't lend itself as evidence for the political consideration theory at all.
The conspiracy theory is likely to be false. Ignore the bad headline, but this clearly explains that the companies decided to switch from 32 to 62 cases in order to get a higher-powered analysis, which was possible as the virus had started spreading rapidly in the US again. It wasn’t at all obvious back in October 2020 that the vaccines would be as effective as they are, meaning more power was desirable.
This article you cite's headline is a pretty classic example of why Scott gets irritated by "No Evidence" headlines:
"Fact check: No evidence supports Trump's claim that COVID-19 vaccine result was suppressed to sway election"
From this article in "Science:"
"When the companies submitted their request for a protocol change, they had yet to accumulate 32 cases. If they had 32 cases before the change was approved, the protocol would have required them to report the results to FDA. In addition, if the results could impact the way investors traded company stock, they may also have been required by the U.S. Securities and Exchange Commission (SEC) to make the results public. They decided to store the nasal swabs taken from participants who had suspected SARS-CoV-2 infections: If they didn't test the swabs, they couldn't confirm cases and therefore would avoid a protocol violation."
In other words, Pfizer put the world's most important clinical trial on ice because it didn't want to know.
Please note that you and I are very close in our appraisal of what happened. We both agree that Mr. Bourlas of Pfizer was largely motivated by commercial considerations to take the extraordinary step of shutting down the lab.
My interpretation, however, is that the feverish political situation over the last weeks of the 2020 election campaign was the single biggest commercial consideration he faced. If Pfizer followed its published protocol and announced efficacy before the election, Trump would proclaim it a triumph for his vaccine-centric covid strategy and the mainstream media, being ferociously anti-Trump, would immediately amp up the already existing spreading of Fear, Uncertainty, and Doubt about Trump's vaccine.
Mr. Bourlas thus resolved, quite rationally, to take the crazy-sounding step of turning off the count in the clinical trial, which led to the Pfizer vaccine getting rapturous media coverage after Biden's victory.
Your theory appears to be that Pfizer in late October 2020 was NOT worried about the politics, they were suddenly worried again about their earlier statistical significance calculations: what if their vaccine doesn't achieve statistical significance for 50% efficacy at the 32 case checkpoint but then succeeds at a later checkpoint like 62? That would be bad from a marketing standpoint.
Well, sure, but they'd been through all that months earlier when devising their published protocol for this project that promised tens or hundreds of billions in revenue. They knew how strong the vaccine was in generating antibodies in humans from their earlier trial.
A much stronger worry along these lines no doubt was for Bourlas: what if the vaccine succeeds triumphantly at the 32 case checkpoint, and then the entire non-Fox media denounces Pfizer's statistical mumbo-jumbo that claims 32 cases are enough in order to discredit Trump's October surprise?
I you were the CEO of Pfizer, putting the trial in suspended animation until after the election must have seemed like a brilliant way to keep the prestige press from going berserk about how evil and dangerous Pfizer's Trump vaccine is.
They knew about statistics a month before, but what they couldn't predict was how many actual cases of COVID there would be. With no cases, you couldn't prove your vaccine works. More cases from a COVID surge, and you meet statistical significance sooner than you expected.
CEO Bourlas had been talking up an October announcement for some time.
But his comments agitated The Establishment since it might grant Trump an October surprise.
The CEO then found a pretty clever way to avoid getting The Establishment furious at Pfizer while still beating Moderna to the draw by a week. You can't say Bourlas didn't earn his pay.
Everyone who's not mainstream from Marxists to libertarians to reactionaries to rationalists recognizes the existence of the establishment. Entities and positions in it are high status. Entities and positions outside of it are lower status. And there a ton of social forces and organizations which devote an enormous amount of resources to making sure their projects are at least tacitly accepted by them.
(In anticipation of a criticism here; it's not an *organization* or a conspiracy or whatever. It's emergent, but it's broadly observed to exist.)
Pfizer CEO Albert Bourlas appeared to be obsessed, not unreasonably, with beating Moderna to announcing vaccine efficacy from a clinical trial first. Thus he repeatedly predicted good news in October. He appeared, however, to be slightly obtuse about the political ramifications of an October announcement in a hyper-politicized era. Perhaps he assumed, once again not unreasonably, that since Pfizer had not taken money from Trump's Operation Warp Speed, that Pfizer would be seen as above politics and its promptness in announcing the good news would be seen as admirable in humanity with covid.
Ha-ha, naive Mr. Bourlas! In the fall of 2020, everything, even matters of life and death, was political.
Eventually, Bourlas wised up. And by shutting down his lab from processing clinical trial until the day after the election, he managed to thread the eye of the needle: not announcing when it could help Trump, but still beating Moderna by one scant week.
Good argument. If it's a competition between "the company makes $billions and I get a $1.2 million Christmas bonus" and "Some ideological/cultural contest has an 0.5% chance of resolving more in the direction I like" I'm pretty confident of which motive I find more credible.
If I were writing up Pfizer CEO Albert Bourlas's decision-making process in October 2020 for a Harvard Business School case study, I'd be impressed by how he managed to navigate between Scylla (announcing in time to help Trump) and Charybdis (losing the race to Moderna).
I agree that we’re close in our appraisal of what happened. You just think it was a clever political trick, whereas I think it was a sound business decision based on the explosion of cases in the US from mid-to-late October. It is a decision I would have taken, irrespective of the political consequences, had I been in their position.
The real thing they were worried about was vaccine hesitancy. They were worried that:
* Rushing out a vaccine that wasn't proven to be safe/effective would result in low uptake.
* Rushing out a vaccine that was seen as pushed out for political reasons would result in low uptake.
* A vaccine that *actually had problems* would cripple future vaccine uptake, as it would be hyped up and then fail, and thus result in worse long-term outcomes. It could also completely ruin the company, which would be thrown under the bus by literally everyone for Lying To The Public.
All of these were very good reasons to want to be certain on their data.
How do they know the effect of any action on vaccine hesitancy? The "pause" on the J&J vaccine was supposed to establish that the authorities were taking seriously any possible problem with the vaccines and thereby reassure people, but it wound up cratering actual vaccinations.
As several Marginal Revolution commenters have already noted, other countries authorized the Pfizer vaccine on a timeline very similar to the US (including countries with a pro-Trump government), which is already pretty strong counter-evidence for the conspiracy theory.
Indeed. The UK's MHRA, which has generally been very quick throughout the pandemic, approved the Pfizer vaccine on December 2nd. The FDA approved it on December 11th.
The Trump administration asked Pfizer and Moderna to apply for a "compassionate use" license to get the shots into nursing home residents before the 2020 winter surge. These residents had 10-20% fatality rates, and were just sitting ducks in the nursing homes.
Pfizer and Moderna said no. Birx's theory was that it was election related.
But what makes this so bad is that it also prevented Israel and UK from doing the same thing (why would Pfizer allow it for UK but not US?). So elderly people outside the US died for our election.
I'll just note that the original conspiracy theory I commented on was about US health officials pushing out trial timelines to delay authorization about the vaccine, not that Pfizer leadership didn't apply to some license. It is of course possible to spawn a more convenient sister theory (which seems superficially similar but actually makes a set of completely different claims) every time an objection is encountered; such is the nature of conspiracy theories. They are generally not worth paying attention to. The original one only was because it came from Nate Silver who is usually a reasonable person.
The pressure campaign was simple: convince Pfizer and the FDA to stall until after the election. We have receipts, Eric Topol bragged about it in an Oct 2020 Technology Review article.
Garett Jones put it best, the coincidental delays of Pfizer and the FDA are an example of a Schelling point, tacit coordination in game theory.
I don't see how. The theory is that US politics influenced Pfizer to sit on the study results and delay reporting/processing till the day after the election (plus FDA goal post shifting on what the trial needed to show). Hard to see how other countries would approve without trial results...
We do know that Pfizer shut down processing of samples to avoid having to report results, that under their original protocols the trial should have reported in mid-Oct. and even under the revised protocols it should have reported in late Oct. had they not stopped testing samples. We don't definitively know why, but the fact that they resumed the day after the election and announced results a few days after that does tend to make people suspicious (this may have been triggered by FDA approval of their protocol changes the day before, but that just raises the question of why the FDA sat on it's hands till exactly then). That plus Topol's comments and letter on influencing the FDA form the basis of the charge.
In my mind it's more credible than the Wuhan lab leak theory (which I give ~35%). Way more suspicious circumstantial evidence here.
They didn’t have 32 cases in mid-October, or even in late October (when they shut lab processing down and took the decision to change their protocol). They submitted their request to change the protocol in late October, and the FDA approved it on the second working day of November.
Right. The single best guess for when Pfizer stopped the count is 31 cases, although I'd go as low as 29. My impression is that CEO Albert Bourlas really really wanted to beat Moderna so it took him a long time to realize that Pfizer would be crucified by the Establishment for delivering Trump his dreaded October Surprise.
Most of the comments focus on Pfizer, but the FDA actually moved the goalposts.
The FDA made a last minute change to the EUA standards to require a median of two months of follow up data for clinical trial participants. Do the math: it puts an EUA after the election.
"The agency [in September 2020] published new guidance for vaccine manufacturers that said they need to provide at least two months of follow-up safety data after vaccinating trial participants to apply for authorization."
Which is, not coincidentally, the standard that Eric Topol argued for as part of his pressure campaign. Pfizer delayed the announcement to avoid negative PR about the election, but the FDA had already set the timeline.
I haven't read their article, but it struck me at the time that unblinding the data just after the election was exactly what you would do if you wanted to avoid Trump getting the credit while minimizing the cost in mortality of doing so.
Similarly, I thought Covid first appearing near a research institution that was studying bat viruses was pretty strong evidence for the lab leak theory.
In both cases, a competent and unbiased expert could go beyond those simple facts to reach a different conclusion, but competent and unbiased experts are in short supply and hard for the lay observer to recognize. The Covid case was a particularly striking example where it turned out that the supposedly competent and unbiased expert was simply lying about being unbiased, being believed, for a while, by other people who shared his bias for different reasons.
Ben - I think you make a valid point. Pfizer was reacting mainly to the FDA's EUA guidance change. I don't buy the arguments from Steve Sailor that Pfizer shares a lot of the blame although I think they could have acted better (for instance by publishing interim results before the election as they had planned under their original schedule)
There are rumors, for instance from Erick Topol, that the Trump admin wanted to bypass the FDA by having the HHS Secretary rubber stamp the EUA in October when the first endpoint read-out from Pfizer's Phase III was expected to be hit. That would have been 100% legal although it would have created short term hesitancy and perception it was rushed. However I think it would have been net positive by a lot because I believe many elderly and at-risk would have gotten it and tens of thousands would have been saved. The increased hesitancy could be allayed later and offset to some degree by having more follow-up data, FDA weighing in, and observational studies. Additionally, a lot of hesitancy goes away once people see their friends and family getting the vaccine (most people are not analyzing data). So by allowing people to get vaccines earlier, the main process whereby hesitancy is reduced starts earlier as well!
I think I already posted my timeline below but it covers a lot of what happened:
One potential failure mode (especially with someone as erratic as Trump at the helm) is that the government starts vaccinating early on a "might not help but can't harm" basis, the vaccine turns out to be not very good, and the government sticks to it over better ones just to save face.
This happened in Hungary: the government started vaccinating with Sinopharm because in early 2021 they could obtain it in larger quantities than Pfizer, leaned on the health authority to skip the normal evaluation process, got a lot of criticism for it (some justified, a lot of it not), doubled down into a "West bad, China good" narrative (using inactivated viruses is old and time-tested, mRNA is still an experimental method etc. etc.) which was a nice fit for Orbán's politics in any case. Then by middle of the year, when Pfizer was available in effectively unlimited quantities, and the first reliable data on Sinopharm showed up and it turn out to be less effective, and a lot less effective in older people, and other Sinopharm-using countries started to re-vaccinate their older population, the Hungarian government didn't want to admit it was wrong and just stuck to its guns. So something like a quarter of the population ended up facing Omicron (which fortunately turned out to be not that bad) with two Sinopharm shots and maybe one Pfizer booster.
Thanks Ben, glad to hear! It took me like 8-10 hours to put together, I think. If you notice anything major I'm missing let me know (especially regarding Pfizer's timeline). I was going to try to double check and incorporate some stuff from Steve Sailor's article (https://www.takimag.com/article/the-new-normal-by-any-means-necessary/) but never did.
Yeah, again, if it had just said "unlikely", I would think they meant "compared to your lurid imagination where they're comic book villains and reoffending all the time". At "extremely unlikely" I am at least eyerolling.
Fun rationalist side project: formally define unlikely and likely as probabilities. This raises the question of whether it’s contextual (e.g. a 10% chance of death is too terrifying to be unlikely, a 10% chance of getting your hands wet is too trivial not to be unlikely), as well as if there’s a gap between “unlikely” and “likely;” they seem like a binary, but a 40% chance doesn’t intuitively seem to be either.
good find. it doesn't apply to this case, but 45 seems like a high value to call unlikely.
And are these ranges still recommended when comparing to an extremely high/low baseline? This also arises in ML for measuring performance of predicting rare events, since it is almost impossible to beat the baseline of 'is not a murderer'.
If I had a 19% chance of dying (from being hit by a bus) every time I crossed the street, would you still say that it was "very unlikely"? I mean I usually cross the street about five times a day, so that would reduce my life expectancy to about a day.
Easy to define them formally. Impossible to get anyone to stick to definitions in conversation, or any kind of discourse really. (Slight exaggeration for the sake of humor.)
I remember some of the UK's variant of concern reports had formal definitions of those terms as probability ranges. They'd say things like 'likely (>50%) that alpha was more infectious than wild-type' and have a chart at the end specifying what probability 'likely' corresponds to.
They literally said "extremely unlikely". If something with a 49% chance of happening can be considered "extremely unlikely", then these words and concepts lose all meaning/practical significance.
In the paper, it is a relative measure compare to other criminals - released non-violent criminals are more likely to be re-arrested for violent crimes than released murderers.
Which is meaningful and important if someone is telling you they favor shorter sentences for non-violent criminals and longer ones for murderers because of recidivism concerns.
But not how the tweet seemed to be phrasing it. Either the tweet misinterpreted that part of the paper, or was intentionally misrepresenting it and assuming no one would look at the paper itself (which, I'm sure 99.99% of people reading the tweet didn't, so...).
Even taking that comparison as the point, is it age-adjusted? released murderers are often like 70, at which age violent crime is much harder to do....
Middle class domestic murderers who go to prison for 10 or 20 years for killing their wives aren't that likely to commit another murder after they get out in late middle age.
On the other hand, teens who shoot somebody (not fatally) in a gang shooting and do a couple of years are not all that unlikely to wind up killing somebody in their 20s or 30s.
It's not surprising. American prisons are torture sites that turn many victims into animals. 22% is "extremely unlikely" when that context is considered.
People released for murder are monitored extremely closely. While it is certainly possible they are committing violent crimes and not being apprehended, everyone else is being watched less closely.
Seriously, you call the police because your bf hit you, maybe something gets done maybe it doesn't. Call the police because your ex-bf who's on parole for murder hit you, he's going to jail.
Yeah, I would object at "unlikely". It's much much less blatant an error - I can almost believe it's an honest misunderstanding of how people are most likely to interpret the word "unlikely" - but I still think it's very misleading.
Came here to note this. Currently the stats are (an elevated) 7.8 murders per 100,000 in the USA (annually). Assuming that each person was murdered by a separate person (which maximizes the number of murderers, but doesn't account for spree killings or murders done by groups of people) that's 0.078 percent prevalence of murderers in the population (which is very close to zero, and it's interesting how sensitive we are these days to a change in that rate.)
Again assuming that each homicide is committed by a different person, over five years we'd expect 0.39 percent murderers in the general population. Instead, the report indicates that 2% were *arrested* for murder. That's an order of magnitude difference. Given the abhorrent clearance rate for murder in African American urban populations, it seems likely that the actual rate is much higher, perhaps even twice to three times as much.
That's a lot of dead people, casually swept under the rug, because space was needed for performative heart-bleeding.
And as the report indicates, the impact on the community via non-murder violent crime is 10 times as large. Good going guys.
By this logic: If we released all violent crime offenders from state prisons tomorrow, 200k murderers (+4k) and 500k others (+5k) would result in 9k additional deaths over the next 5 years, or an annual increase in murder victims of 1.8k or 11% (from pre pandemic level of about 16.5k/year). You could argue it’s worth it but also it’s clearly not true. There would be blood in the streets if we stopped locking people up for murder. These numbers are based on the system we have now, in which people are locked up until they are old and in many cases a parole board believes they won’t kill again. You cannot plausibly claim these statistics would apply under a radically different system.
The better the convict, the sooner they will be released and vice versa. The worst convicts will literally have less time outside of jail in which to reoffend, and they will also be older and therefore less likely to reoffend. These effects almost assuredly outweigh the effects of incarceration increasing criminality (if these effects exist)
Although I'm not a subject matter expert, I would be frankly astonished if the answer wasn't yes.
For the obvious reason of criminal record = hard to get a job= poverty = crime, but also for more esoteric reasons, like:
When my cousin went to an extended stay in a hospital to treat anorexia, they wouldn't put her in wards with other anorexics for fear they would reinforce each other's behaviors/beliefs and give each other advice on how to evade detection and treatment. I imagine surrounding criminals with nothing but other criminals for years could have the same effect.
Agreed. Also, what's the average period of incarceration associated with that 22% number? Because if it's measured in decades, I don't know that this implies we can just start letting violent offenders out of prison and not suffer any public safety repercussions.
Parole violations do count towards the overall rearrest rate when they result in arrests, but they do not count as violent reoffending, so it's not relevant to the question of how many people convicted of homicide commit another violent crime after release.
It's hard to tell from the report how many were arrested for parole violations only, as opposed to parole violations in conjunction with a more serious crime. I'm no expert, but I would expect committing a crime to be a parole violation, so it's likely that many of those parole violation charges were just tacked on to whatever crime actually motivated the arrest.
Sure, but the true rate of offending is higher than the arrest rate for the non-convicted population too.
Given that people with criminal records receive far more scrutiny from the police, I would expect the discrepancy between these two numbers to be much *less* for them than for the general population.
I'm not sure this is true. I would guess that first-time offenders are more likely to commit violent crimes against people they know, making it harder to get away with it, whereas repeat offenders are more likely to commit violent crimes against strangers.
Regardless, I think it's beside the point. When we're doing cost-benefit analysis of prison sentence length, the absolute risk of reoffending is probably more important than the risk of offending relative to the general population. Suppose the true rate of violent reoffending for people convicted of homicide is 30%. What would we do with the knowledge that this is only 20 times the overall rate of violent offending, rather than 30 times?
Maybe a dumb question, but re #11, why would Xi cause offense generally? I've always pronounced it "ksai", so I'm not sure if it is the pronunciation. Is it that it is spelled the same as the family name of the president of China? But I think Chinese pronounce the spelling "she". So maybe the Chinese government would not want to see that in print? If that's the case, is there a general ban on Chinese researchers associating that Greek letter with anything negative, or using it at all? Greek letters are all over all kinds of science, so that would be a big headache. Or maybe not just something so prominent. Or is it not about Xi Jingping, but the fact that it is a common family name in China, and it's more about national sensitivity than the president's ego? So many questions!
Somewhere a clever teenager in Taipei is creating a meme involving Xi as Winnie the Pooh dying of COVID-19 that gets through the Great Firewall, China gets pissed enough to over-rattle sabers at Taiwan, a frontline observer mistakes a civilian drone for an incoming ballistic missile, leading to a shooting war in the strait, the US gets drawn in and climate change doesn't look like the biggest concern any more.
Yeah, and then think of all the stoners enjoying the daylights outta that meme, and passing it along with odd tweaks added. There's a whole ecosystem of wit out there.
In terms of how many people bear it, yeah, but in terms of how many occurrences of the surname "Mu" or "Xi" you will find in a random text, probably not ;-)
Yes. And if the next letter after Nu had coincidentally been ‘Trump” or “Biden”, it would have been skipped in exactly the same way, with a similar eye roll inducing explanation.
I don't think most English speakers would be able to pronounce a word-initial /ks/ upon their first encounter with the word. So while us more educated folk might pronounce Ξ as /ksai/ or /ksi:/, I would bet that a majority of Americans at least would pronounce it /zai/, /ʃi:/, /si:/, or /kai/.
Plus when we inevitably reach Χ, we'd get to deal with that confusion lol
I don't think the explanation requires an unusual level of sensitivity on the part of the Chinese government. Imagine that 'Biden' is the next letter in the sequence, but it's pronounced bee'doŋ. I think they still would have skipped the Biden variant.
"But I think Chinese pronounce the spelling 'she'"
Sort of. The Mandarin "x" is a consonant that doesn't exist in English. It's more like what you get if you try to say "she" with the tip of your tongue held just below the base of your lower incisors.
Is this specifically a property of some Beijing standard dialect or something? Mandarin "x" and English "sh" seem identical to me but that could be a function of geography.
They're definitely different in standard Mandarin, but apparently x, sh, and sometimes s are merged in Cantonese-accented Mandarin. It wouldn't surprise me if there were some other regional accents/dialects that merged them.
Sounds good to me Kenny. Wait, actually it doesn't, because I am currently so irritated and weirded out by the whole dangerous, irrational mess that at the moment I'm in favor of giving future variants names that as childishly disgusting as possible, names that would crack up a bunch of especially uncivilized middle school boys -- Fart, Buttne, Puke . . .
(Please excuse my crassness, if you're not amused by this kind of thing. You caught me in a crass & cranky state of mind).
There's something to be said for the idea that the WHO should contract out with roomfuls of 9 year olds around the world to come up with lists of names for disgusting things, and then work down these lists in some sort of canonical order.
Though it would be unfortunate if the puke variant turned out to cause severe buildup of methane in the colon while the fart variant turned out to cause more vomiting.
10: I've bought an audiobook in the last month based on the quality of its negative reviews. In case you're interested, it was "Money: The True Story of a Made-up Thing" by Jacob Goldstein and it's great, and I knew it would be great because the one-star reviews were all various flavors of argle bargle Murray Rothbard. The lesson being, you can learn a lot about something by listening to its enemies.
Where did you read those negative reviews? Amazon/Audible's single 1-star review is related to a physical printing issue and the handful of 2- and 3-star reviews don't reference Rothbard or thinking I'd associate with those types of takes. (And actually one of the critical reviews finds a bright spot in the book to be the author's proposition of explicitly bifurcating banking services into full-reserve and fractional-reserve entities, giving the customer choice in the matter - which sounds nearly identical to something I read from either Rothbard or one of his acolytes years ago.)
Regarding #26, I'm not a Mormon or a Utah resident, but I think your graph is cherry-picking the timescale it looks at. Based on https://gardner.utah.edu/wp-content/uploads/Fertility-FS-June2021.pdf?x71849 , the state-level Utah fertility rate was generally declining from at least 1960 (as early as they show) to 1994, recovered a tiny amount from 1994-2008, and then declined again (as your graph shows). So, I think it's fair to attribute this to a trend that's been in place since 1960. The more interesting question is what caused the 1994-2008 temporary minor recovery?
Boy, that curve looks to me like it's just basically tracking the US numbers except 19% higher, except for the anomalous peak between 1974 and 1980, where it topped out at 83% higher. (!) I'd be more interested in hearing an explanation of that peak.
I think there's a "generational" thing happening, where kids come in clumps in our population. I was part of that 1974-80 boom, and then my children were born in that 2004-2012 bump. That's one part, maybe.
On 14: Poverty is bad enough, the idea that you'd need some sort of esoteric quantitative description that maps onto neat 21st century bourgeois categories of badness in order to want to fight poverty is bizarre, and it doesn't do anything to resolve one way or the other the questions about the "best" way to fight poverty because that's a discussion about values, not statistics.
Postscript: Give a monkey a set of calipers and apparently he'll call himself the Übermensch.
I mean, the underlying question is whether poverty _can_ be solved.
Let us take it for granted that low intelligence causes poverty. If low intelligence is also caused by poverty, then this is a vicious cycle that can be broken by simple give-poor-people-money interventions in a single generation.
If, on the other hand, low intelligence is caused purely by genetics, then poverty can't be solved except through genetic interventions.
If I caused lead poisoning in a bunch of kids (or just hit them sufficiently hard in the head with a ball-peen hammer), do you think they will on average be poorer than a control group of kids I don't damage?
I didn't think we'd go STRAIGHT into 1922 eugenics but OK...so, here's the thing, we have created truly godlike levels of wealth for BILLIONS of people in the span of a couple centuries, so yeah, poverty definitely isn't genetic and maybe think twice before you launch straight into that at a dinner party.
I'm having a hard time figuring out what your point is, but I will note that steam engines improve the productivity and standard of living of all humans without changing their genome one bit.
"Given a set of technologies and opportunities" is where all the bodies are buried, and you're not invited to the dinner party on account of being two hundred years behind the conversation.
But we’ve already eliminated poverty by that definition, the homeless aside perhaps. Previous generations would be amazed that our poor were more likely to be obese.
The genetic changes already occurred. It was these genetic changes that allowed the steam engine to be developed in the first place. There's no reason to think people who never even invented the wheel (vast swathes of the globe) would have been capable of developing steam engine without significant genetic changes.
Absolute poverty has basically been eliminated in the US, Canada, and many other developed countries for decades now apart from a marginal population of people with severe mental problems/drug addiction that renders them unhousable, as well as a fraction of people who live in extremely remote locations and refuse government assistance and other things.
Gee, it's almost as if the study is question was looking at people in..."otherwise rich countries". Maybe click the link before freaking out based on your own ignorance next time?
It IS truly weird, but all definitions of poverty used with reference to the developed world at present are weird definitions. By any kind of objective or historical standard there is no poor person in any modern developed country. The biggest problems of the American poor are their abundant access to food and drugs. Poverty is *solved*. If you don't want to just shelve all further discussion of poverty (which, to be clear, would make perfect sense), you have to use a bizarre definition of it.
Calling this "1922 eugenics" is lazy and stupid. Either poverty is largely explained by heritable factors or it isn't. Calling it "1922 eugenics" does not change this and is a BS attempt as shouting down an opponent who disagrees with you.
And FWIW, the west rose out of poverty by directly producing the scientific and technological advances to do so. It needed to do this to rise out of poverty because those things didn't exist at the time.
Today, all of that hard word is already done, and the knowledge is available in seconds to anyone with an internet connection. All these impoverished countries need to do is copy what has already been discovered. Nobody is expecting them all to invent this stuff from scratch. It's literally never been easier to industrialize. If they can't industrialize on their own in 2021, there's no reason to think they will ever come close to catching up with the rest of the world.
...holy shit, this comment alone proves that you are indeed mentally stuck in 1922-era beliefs regarding genetic superiority. I hope you improve as a person this coming year.
I deleted coarse and trollish reply. Apologies for any offence caused. Let´s try again.
First, it is not correct to say that if poverty is caused strictly by low intelligence, and low intelligence is caused by genetics, then poverty cannot be solved except through genetic intervention. It could also be solved by reordering society so that intelligence would stop playing role of an income determinant. Not that I think this would be good or anything, but it is an option for solving poverty even under this unrealistically restricted assumptions.
But secondly, assumption that low intelligence causes poverty is rather reductive. I get that it is strictly speaking correct, at least in wealthy countries with lots of human capital. Intelligence is one of the causes of poverty in a sense that IQ is correlated with income. But there are many other causes of poverty. And those non-genetic causes of poverty might be amenable to other kinds of solutions than genetic engineering.
And I think it is obvious that correlation between intelligence and income is in a present society very far from being perfect. E.g. I really doubt that in median Western country someone with IQ 100 from a family in uppermost 10 % of income distribution has a higher odds of ending up in poverty than someone with IQ 120 from a family in lowest 10 % of income distribution.
Ben Franklin said that death and taxes are two things we will always have. Some rationalists think we can abolish death, but I don't think the same ones expect us to abolish taxes.
Poverty is caused by division of labor and poor bargaining power & compensation at the bottom of the wage scale. Population IQ level interventions, while likely good for other reasons, would be far, far more likely to increase elite overproduction than to solve poverty.
It would make the heritability of poverty approach zero.
And you're completely ignoring the non-income factors of poverty. Committing crimes and spending time in prison, becoming a drug addict or having a kid while you're a teenager will severely diminish your economic prospects and this has nothing to do with "bargaining power". Maybe if they grew up a wealthier family they would be less likely to do this kind of thing, but the heritability of these kinds of behaviors is certainly not insignificant (and can even explain more of the variance than environmental factors in some cases). Low IQ people have a lower savings rate even after you control for income and so would be expected to have worse financial health even if their incomes increased.
So, before there was division of labor, everyone was rich? If you wanted to blame it on agriculture, you might persuade me. We are probably not using our terms the same way.
Poverty is almost entirely caused by differential productivity levels.
People who work at WalMart stocking shelves simply produce far less value on an hourly basis than electrical engineers or linesmen do.
Indeed, if you look at companies that primarily employ poor people, like WalMart, they have very low profit margins, while companies like Microsoft, that employ top end people, have very large profit margins.
This is exactly contrary to what would be expected if it was due to "poor bargaining power and compensation" - the workers at Microsoft have more leverage, not less, but their company has a higher profit margin, not a lower one, and they get paid a lot more.
Relative poverty is insolvable because of human genetic diversity. IQ has a heritability north of 75% in adulthood, and correlates with income to 0.5 in primary wage earners.
So, you can't solve relative poverty.
Absolute poverty can be solved by increasing per capita productivity. The US has almost no one who lives in absolute poverty nowadays.
So depending on whether you are defining poverty relatively or absolutely, it's either insolvable or has already basically been solved.
There's a big difference between 'I have $5 and you have $100000000000' versus 'I have $100,000 and you have $200,000.'
We could 'solve' the massive inequality we have right now without requiring 100% identical outcomes. How much a economic system amplifies relative advantages is a contingent property of the system itself, not a natural and unchanging fact of the universe.
There is absolutely nothing "esoteric" about IQ or any other established psychometric measure. The fact that you're clearly ignorant on this stuff is a you problem.
As for fighting poverty, if you do not accurately understand the actual causes of poverty then you have no hope of ever meaningfully reducing it.
Trial and error is not so completely hopeless. Unfortunately, trial and error would require poverty fighters to try new methods and abandon failed ones.
"Cheap solar, cheap wind, and cheap storage mean that we could see the first large sustained decrease in electricity costs in over half a century. People are finally starting to realize this, and are speculating about what could be done with cheap abundant electricity. ...The reality of cheap electricity — not 30 or 50 years in the future, but in the coming decade — is thus starting to sink in. This is really happening."
Uh-huh. Meanwhile, this email from the electricity supplier I'm currently signed up with:
"We have all been impacted by the significant increases in wholesale energy prices over the past year. At [redacted] we have made a commitment to our customers to work hard to keep prices as low as possible. As a small supplier we are unable to absorb the increasing prices of wholesale electricity or to hedge against them effectively, this means that as prices increase we have to pass those increases onto our customers. In order to protect our customers from significant price rises we’ve engaged with [redacted] to offer our customers a competitive green electricity deal".
Every electricity supplier in Ireland has whacked on price increases over the past year. Although wind generation seems promising, at the same time, unusual weather patterns can mean it fails: right now, it's very very windy so the generation is high, but back in June we got a period of settled, warm weather with little wind, so no generation.
Retail customers buying power from the grid is emphatically not the point here. Industrial manufacturers building their own behind-the-meter power and building widgets with .01$/kWh power is the killer app.
If we were expecting very cheap energy in a decade, wouldn't oil companies stock value be plummeting? [1] And geologists give up on trying to find new oil and gas locations that won't be ready for over 10 years?
[1] Yes, many of them have investments in renewable energies, but there to various degrees and we should be able to notice that.
Someone who unironically used "bourgeois" pejoratively just moments ago celebrating rich companies becoming richer while consumers struggle to pay for heating. Weird.
Damn it, if I weren't too cheap/broke to subscribe to Freddie deBoer's Substack (hi, Freddie! Happy New Year!) I'd be over there whinging like a creaky gate swinging in a gale about this.
This, folks, is how you lose elections to populists instead of your nice, shiny, technocrat candidate winning.
Stand atop a heap of crystal-ball gazing, looking into the mistily undefined period of "sometime during this ten years" and declaring that cheap electricity is coming - for big industry, to reduce their manufacturing costs and make more crap. And not even the prospect of at least more jobs for people, because those industries will be relying on automation to keep their labour costs down and productivity up, so there will be a flood of (iPhones? refrigerators? widgets for fancy driverless electric cars?) but I won't be buying them, because my disposable income is reduced due to inflation and things like "no cheap electricity for you, peasant, in the very definite next year or two but instead three or more price increases during the year".
And the answer to that, from we peasants that are not getting bonuses or rises in line with inflation, is "fuck you, I prefer to dream of cheap widgets for my toys".
Thanks, guys! I'm not going to hold my breath anticipating "cheap electricity for big industrial and commercial concerns is just around the corner! like the way nuclear power would make the deserts bloom:
"for big industry, to reduce their manufacturing costs and make more crap"
The crap big industry produces includes most housing (often using energy-expensive steel), cars (again, steel, aluminum), and food (needs water--desal to the rescue!)
Meanwhile, my electricity bill is on the order of 1% of my spending and natgas *maybe* another 1%.
I would *love* cheap, clean energy for the industries that make pretty much all the things I consume.
You realise that your 1% is an anecdote not a statistic?. A quick google tells me that 14M people live in fuel poverty in the US and it’s worse in other countries.
I think it's less relevant whether people are in "fuel poverty" than it is whether that bill is high enough that doubling it would offset a reduction in food or consumer good prices.
That said, you're right, after looking into it, it likely is, unless food dropped more than I'd anticipate.
Ireland needs to build inter-connectors to both export surplus wind (of which we will have plenty when offshore is built out) and to bring in power from the continent. Unfortunately it looks like Britain is going to eat our lunch in that.
Short term price ramps aren't an issue and aren't what's meant here. Long term price falls are what's going to happen, short term volatility is only natural in a changing market.
UK electricity bills have basically been flat the last decade with energy efficiency doing the heavy lifting. Probable shift in taxation to gas as opposed to electricity will see that change.
Ireland can reach 100% renewable relatively simply, offshore wind off each coast, dense interconnection with UK, France, Iceland, Spain smooths supply profiles. Each individual country only needs a small amount of overbuild when paired with a reasonably small amount of storage - 2-24hr storage markets serviceable by batteries, longer term served probably by hydrogen/derivatives given the volume of money flowing into the market but may vary on geography.
Interconnectors are the key and really drastically help cut emissions and prices. ERCOT is the example for the US audience, with ERCOT basically screwing the US out of having a national grid which would let solar in California compliment offshore wind in New England, or whatever.
Unfortunately Ireland has no real plans for large interconnectors but Britain does, the distance is much larger of course. You are right though, yesterday I noticed that - it being windy - renewables were 70% of supply and that’s a half hearted effort of a few onshore wind farms. If we (in Ireland) double output we would have a big surplus to sell on a moderately windy day but that depends on there being a customer somewhere to take it.
re: #26, it seems too small of a change to be the whole explanation, but the source now has a link to the fact that Utah counties have been becoming less LDS over time as well as people of other faiths move in. In the addendum, he says 'there is an almost 1-to-1 ordering between "percent LDS" and TFR.'
So that explains at least part of it, people from groups with lower average fertility rates moving into the area would certainly affect the overall rate.
The percent LDS certainly explains much of the difference between counties in Utah/Idaho.
But it's also a factor that LDS folks are having smaller families, especially in places like Utah County. When I grew up here, large families were bumping the average up. Just a quick mental sample of my close friends as a kid gives me families with 11, 8, 10, and 2 kids, plus my family with 4 kids. That's not scientific, but it's not a bad sample of family size in Utah County in the 80s/90s. (One family in the neighborhood had 17 kids, same mom, no twins! Fun fact: I was friends with one of those kids, and also friends with her niece, who was a year older.)
Utah (even Utah County) is definitely less Mormon than it was, but LDS families are smaller too. As for why LDS folks are having fewer kids, I guess a few things may contribute. More women serving missions (probably a small effect), more women delaying kids while going to school, certainly the "asymmetric apostacy" mentioned above, and honestly the economics of having lots of kids. The price of a larger home and a larger car relative to income plays a role, and since the 2008 housing crisis, home prices in Utah have still been going up.
I'd have to do some digging on how much a home or a big old van costs in relative terms to when I was a lad, but my gut tells me it's more relative to income. In addition, we don't culturally cram as many people into smaller homes as we used to, nor do we overfill cars like we did when I was younger. So that's changed too.
I'm not sure how deepfakes could be used for (computer) _hacking_, but the link itself isn't about that – it's about an Instagram account being hacked and deepfake videos posted to it.
But, historically, 'social engineering' (i.e. humans compromising security via other humans) has long been recognized as an important element of practical 'hacking' so it's definitely reasonable to consider what's described in your links as 'hacking' in a broader (but still coherent and reasonable) sense.
It would be funny if, after all the panic about deep fakes, the main use case for crime was helping criminals do really good impressions of people over the phone.
So, based on item 3 from link 1, Buddha wouldn’t play Squid Game:
“ Games of marking diagrams on the floor such that the player can only walk on certain places. This is described in the Vinaya Pitaka as "having drawn a circle with various lines on the ground, there they play avoiding the line to be avoided". ”
Re: Number 9. I have read that the COVID vaccine inserts non-human DNA into your DNA, thus making you technically non-human, so therefore Jesus cannot save you.
You'd think that if He were able to turn water into wine and raise the dead, as well as being generally omnipotent and omniscient, that He would not let Himself or His works by stymied by a game of Cosmic Gotcha.
more importantly, every virus ever inserts non-human DNA or RNA into you. "You can only be saved if you've never gotten a virus" is a really aggressive way to gatekeep heaven...
Unless that retrovirus previously infected the Tree of Knowledge, then it doesn't bear original sin so arguably it's diluting the Fallen human DNA. (Why does this feel like an SMBC comic?)
I can't even begin to disentangle the bad, ignorant, and plain "that's not even wrong" understanding of theology there, so I'm going to recommend this video from "History for Atheists" - warning: very sharp language, even I thought it was a bit rough on the nominees in places, but this kind of stuff is certainly a runner for "Crappy Golden Orrery of the Year":
There was a group of engineers at Google NYC who would basically play Counterstrike all day. There were also some who barely showed up for work at all (and weren't working from home either). I don't think that's typical, though. But xkcd.com/303 isn't completely off, there's plenty of down time surfing the web, wandering the building, and having swordfights on chairs. At least until Corporate Safety shuts that last down.
"11. Why COVID variants skipped from Mu to Omicron: “In a statement, the WHO said it skipped Nu for clarity and Xi to avoid causing offense generally.” Rolling my eyes at “offense generally” and the idea of deliberately averting nominative determinism."
On the one hand, the last thing we need is to give the Nutbar-American Community more ammo by naming a virus "Xi".
On the other hand, skipping "Xi" just gives the Nutbar-American Community more ammo.
I guarantee you that the WHO was vastly more concerned about China itself (the world's largest economy!) than about "Nutbar" Americans. In fact, I'd reasonably certain that the latter group most likely didn't factor into their decision at all, nor should it have! Do you really think that global security organizations are concerned with the intricate details of American political discourse? That sort of Americentric viewpoint honestly seems bizarre to anyone who isn't living in the U.S. themselves.
Hey! This is off-topic, but I saw that you posted a comment in the classified thread a few days ago that isn't there anymore, and I just wanted to say that I'd be interested in hanging out if you're up for it! We probably wouldn't be a romantic match, but you seem cool and I'm looking for Boston friends :) My post is here: https://astralcodexten.substack.com/p/classifieds-thread-12022/comment/4731208 and you can email me at rip.my.inbox@gmail.com if you're interested!
19. 6ixBuzz acted in a way that meaningfully emboldened gang violence in Toronto that causally contributed to many deaths (mostly gang involved but innocents as well). I do appreciate that people were erroneously accusing you [Scott] of emboldening abstract bad things but I do think the cases are qualitatively different.
"doxxing is okay when it's important" is the existing rule doxxers use. Change means either reforming that rule—in effect, just changing whose opinion of 'important' is used—or being against doxxing altogether.
The latter is more in line with the 'Be nice, at least until you can coordinate meanness' principle.
Could you elaborate? Was their Instagram account actively inciting a gang war? It's hard to picture how else social media posting causally contributes to gang violence, but maybe that's just a failure of my imagination
The “public health establishment delayed the COVID vaccine” theory is false. The companies very clearly explain the reason for the delay in the article below (ignore the bad headline).
“Pfizer initially designed the trial to conduct interim analyses after 32, 62, 92, and 120 cases accrued... Fewer cases means less statistical power and therefore greater uncertainty about the vaccine's impact, so the efficacy goals to meet FDA's requirements were higher than 50% at interim analyses, ranging from 76.9% for 32 cases to 58.8% for 120.
In mid-October, the companies had yet to confirm 32 cases. But with the epidemic exploding at many of the trial's locations—which were mainly in the United States—they had second thoughts about FDA's request that their first interim analysis should have more to support an EUA request…
The math was simple: COVID-19 cases among participants were jumping from one or two per day to up to 10 or more. It became clear that the trial would accrue 62 cases shortly after hitting the 32 mark, and the higher number meant greater statistical power—and fewer debates about the meaning of the data. This 62 cutoff both lowered the efficacy bar the vaccine had to clear, and was also something of an insurance policy: If the vaccine triggered mediocre immune responses and it teetered around 50% efficacy in the trial, it could more easily have been deemed futile at 32 cases because of bad luck.
the companies' decision to seek the protocol change had nothing to do with politics… "This was our decision," Sahin says. When he read Trump's tweets, he shrugged. "This is just not true."
In late October, they informed the FDA of the protocol change. The FDA approved the protocol change in early November, which meant that the companies could start testing their samples. The results were announced on 9th November.
I think this may be a "beware isolated demands for rigor" situation. When reasonable reasons to delay the vaccine would make Trump look bad, there were a lot of them. When that stopped being a political concern, there suddenly were a lot fewer reasons being taken seriously.
While the EMA has been consistently institutionally slower at approvals than the FDA, that's not the case for the (UK) MHRA, which has been really aggressive for that sort of bureaucracy (it approved AZ first and has been much more prepared to take risks with it; it keeps complaining that Pfizer and Moderna won't apply for approval to vaccinate children).
There's no reason that MHRA would be politically motivated by what is going on with Trump (not because they don't hate Trump, but because a UK approval wouldn't have any impact on US politics). Yet they didn't approve until after the election either (and note that the first non-trial vaccination in the world was in the UK, in early December 2020).
I'm inclined to think that the fact that non-FDA agencies didn't approve until after the election suggests that this wasn't an unadulteratedly political decision.
Indeed. The timing of the *announcement*, on November 9th, was driven by sound business decisions. If you’re Pfizer, and you think you can still be the first to announce if you switch to 62 cases (thus getting a higher-powered analysis, and increasing the probability of approval) as opposed to 32 cases, why wouldn’t you do that? Many feared the vaccines would only just meet the efficacy threshold, so a higher-powered analysis was desirable. This meant that the announcement occurred after the election rather than before.
The timing of the *approval* was irrelevant to the election. It occurred, as you say, on December 2nd in the UK, and December 11th in the US. So it's just not true to say that "there suddenly were a lot fewer reasons being taken seriously" - the FDA took more than a month (!) after Pfizer had announced its vaccine was effective to actually approve it.
The decision of Pfizer was to not publish trial data until after the US election; the MHRA can't approve the vaccine without any trial data, even if they're not linked to the cause of the delay
headline: "No evidence..." wait, isn't that some color flag.
But joke aside, I also haven't seen mentioned an obvious political alternative read of the letter: They were worried the Trump administration was attempting through side channels to have the results delivered earlier, and was making it clear that should not happen.
Attempting through side channels to hasten the vaccination of the populace in a deadly pandemic, likely saving tens of thousands of lives, how diabolical!
The smoking gun evidence that Pfizer shut down lab processing in the world's most important clinical trial from late October until the day after the election is found directly in Matthew Herper's 11/9/2020 article in "Stat News" interviewing Pfizer executive William Gruber:
The story of how the data have been analyzed seems to include no small amount of drama. Pfizer, seeing an opportunity to both help battle a pandemic and demonstrate its research prowess, made decisions that were always likely to make its study the first of a Covid-19 vaccine to produce data — including its decision to have an independent group of researchers, known as a data safety and monitoring board, take an early look at the data in the 44,000-volunteer study before its completion.
The first analysis was to occur after 32 volunteers — both those who received the vaccine and those on placebo — had contracted Covid-19. If fewer than six volunteers in the group who received the vaccine had developed Covid-19, the companies would make an announcement that the vaccine appeared to be effective. The study would continue until at least 164 cases of Covid-19 — individuals with at least one symptom and a positive test result — had been reported.
That study design, as well as those of other drug makers, came under fire from experts who worried that, even if it was statistically valid, these interim analyses would not provide enough data when a vaccine could be given to billions of people.
In their announcement of the results, Pfizer and BioNTech revealed a surprise. The companies said they had decided not to conduct the 32-case analysis “after a discussion with the FDA.” Instead, they planned to conduct the analysis after 62 cases. But by the time the plan had been formalized, there had been 94 cases of Covid-19 in the study. It’s not known how many were in the vaccine arm, but it would have to be nine or fewer.
Gruber said that Pfizer and BioNTech had decided in late October that they wanted to drop the 32-case interim analysis. At that time, the companies decided to stop having their lab confirm cases of Covid-19 in the study, instead leaving samples in storage. The FDA was aware of this decision. Discussions between the agency and the companies concluded, and testing began this past Wednesday. When the samples were tested, there were 94 cases of Covid in the trial. The DSMB met on Sunday.
This means that the statistical strength of the result is likely far stronger than was initially expected. It also means that if Pfizer had held to the original plan, the data would likely have been available in October, as its CEO, Albert Bourla, had initially predicted.
Pfizer had published its clinical trial protocol calling for checkpoints when the following numbers of samples of positive covid tests had been found in its clinical trial (across the unblinded two arms: test and control):
First: 32
Second: 62
Third: 92
But Pfizer decided to shut down processing of samples in late October and only restarted its lab in the world's most important clinical trial on the day after the Election, Wednesday November 4. Pfizer wound up blowing through its first, second, and even third published checkpoints, winding up with 94 case. It made its epochal announcement of its vaccine's efficacy early on the morning of Monday, November 9, six days after the election.
Without shutting down lab processing, it seems likely Pfizer's efficacy announcement would have come no later than Monday, November 2, the day before the election, which would have given Trump his belated October Surprise that his opponents were so fearful of.
And the article I posted quite clearly explains *why* they switched from a 32-case analysis to a 62-case analysis. Pfizer wanted to be the first to announce its results, but more importantly it wanted its vaccine to be approved. An interim analysis with greater power - 62 cases - was more likely to detect a sufficient effect than a 32-case analysis. As the virus began to spread rapidly in mid-to-late October, they realized they could skip the 32-case analysis, thus getting a higher-powered analysis that was more likely to lead to approval, *and still be first to announce*.
Recall that the fear at the time was that the vaccines would only be around 50% effective (see NPR article below). The risk of failing to get approved was considered to be quite high. Pfizer therefore made the entirely sound business decision to increase the likelihood of approval, safe in the knowledge that it would still be first to announce.
They had shut down lab processing in order to prevent themselves from having a continuous case count. If they had got to 32 cases first, they would have been required to conduct an analysis on 32 cases. If they had known about 62 cases first, they would’ve been required to conduct an analysis on 62 cases. In the end, the first number they knew about after reopening lab processing was 94 cases.
It's the day after FDA approved the protocol change. But of course that raises the question why the FDA waited till exactly then. And the whole thing is just so suspicious in timing and otherwise (and given the Topol letter and comments certainly people were trying to influence the timing) that it's going to be pretty hard to disprove it. For what it's worth I think it's significantly more likely than not that delay was influenced by political considerations.
But clearly releasing a 32-count update could not have had any affect on approval, and would have been followed within a few days by the 62-count update in any case.
Don’t the Trumpy set mostly hate and fear vaccines anyway? So promoting a vaccine before the election would’ve somehow helped the anti-vaxx side win? I don’t get it.
The highest levels of vaccine skepticism seem to be found among the Black caucus, of whom ~90% vote Democrat, so I don't get what you're going for either.
The relevant timeline is otherwise this:
* in early October, Trump repeatedly says on the campaign trail that "vaccines are coming momentarily"
* in response, on October 16, Pfizer announces that it will delay its FDA applications
* in response, Trump softens his declarations to "vaccine would be coming within weeks"
* throughout this, the "neutral corporate media" are full-blast accusing Trump of hundreds of thousands of COVID deaths
In a short while it has become clear that Pfizer in October already had collected 32 RCT results (withheld instead of published, as per the last-minute protocol change), for which it is overwhelmingly certain that they *would* have crossed the 76% threshold (which would have sped up some things, "undesired political impact" aside) and for which Pfizer would not have suffered drawbacks even if they turned out below 76%.
(In a slightly longer while it has also become clear that Biden is not being full-blast accused by the same "neutral corporate media" for the ongoing hundreds of thousands of COVID deaths)
I imagine that Black Democrats are less likely to be vaccine hesitant than White Republicans (or Black Republicans) at this stage. We know that 60% of the unvaccinated are Republican, much higher than their share of the population.
Well at least your understanding that the vaccine-skeptics are 60% Rep / 40% Dem is grounded in reality, as opposed to McClain's assessment of "hur hur Trumpy set", which is grounded in the dominant propaganda of our times.
About imagining that fewer % of Black Dems are skeptical than % of white Reps, that would need to be squared with stats that Blacks are overall 41% less likely to pursue vaccination, and that given a ~10:90 voting split, we can shuffle only so many Black skeptics into the Rep group. I did not pursue googling down for polls that slice by race and party, but here's the source for the 41%:
The exact race/party split wasn't the point of my comment anyway, but how Pfizer (with FDA's backing) modified the protocol to withhold the intermediary results because of their projected impact on the 2020 elections. That the claim that the protocol was modified at the last minute because of Pfizer/BioNTech business interests - to echo the BioNTech CEO - "is just not true".
Pfizer started the RCT in July 2020 on 41 thousand candidates. Back then, when they preregistered the publication milestones of 32 / 62 / 92 cases, they *expected* -- according to their own words -- that the results would take longer to materialize than they in fact did.
In other words, that a sufficient number of results was coming back even before election day had been *unexpected*. FDA and Pfizer then did a last-minute change of that preregistered protocol to ensure that no intermediary results shall be published until after the election.
Saying that the withholding of intermediate results was a business decision done to decrease the likelihood of FDA rejection, is unadulterated horseshit. That's not how trials work, that's not how FDA approvals work. Tens of thousands of people already received the vaccines and placebos. The 62 cases *will* be arriving whether you publish or withhold the results of the first 32 cases. And the 92 and 170 cases *will* just as certainly be arriving too. And by the time you publish that your p < 0.0001 result upon the 170 cases does satisfy FDA's 52.3% threshold, it does not matter for the FDA approval whether the first 32 cases had failed their higher threshold (76.9% instead of 52.3% to account for decreased statistical significance). On the contrary, the case of intermediary results _surpassing_ that higher threshold was meant to be taken as a strong efficacy signal, encouraging protocol modification to start vaccinating placebo recipients.
Given that the 170 cases had a 95% result, just rubs it in how unlikely it would have been for the 32 cases (duly collected in October) to *not* exceed that 76.9% "higher threshold". Leaving unknowns aside and sticking to preregistered protocol - even if the 32 cases wouldn't exceed 76.9%, Pfizer wouldn't lose the FDA approval by publishing that intermediary result instead of withholding it.
In re: 26--as an LDS Utahn, I'm not sure there is a better reason generally than a vague "rise of secularism." We (American Latter-day Saints) track the broader American culture in many/most things, just with a significant delay. Economic worries about supporting a family and buying a home almost certainly play a role, both in people getting married later and in having children.
Social changes related to hanging out in groups vs. "serious dating" are fairly often referenced by the leaders of the church as an area of concern in terms of age at marriage, which I would think tracks with fewer children, but I'm not sure how strong an effect this really has.
I understand that some young married couples have been reluctant to have children based on worries about the wickedness of the world (an interesting mirroring of the "environmental concerns-->no children or small family" thing you posted about a few months ago).
The reason in my immediate family is ill health--all five of me and my siblings intended to have families more or less of that size, but chronic fatigue syndrome intervened.
There is an offset to the Mormon/Utah birth rate compared to the US birth rate. I am going to assert this is because of "culture"(a mix of social, religious, historical factors). The US birth rate has drop between 2003 and therefore so will the Mormon birth rate. The problem that tweet shows is the Mormon birth rate has dropped by a lot more than the US birth rate.
I think it is a safe conclusion to say there has been a change in culture but I don't think there is any unique, new pressure in Mormon Culture. We still talk about marriage a lot. We still have doctrines about the importance of families. What else is left besides something physical(specifically in Utah) or a more vague "Rise of secularism"?
Here is one example that I think is a small shift in norms that has come from the American culture to the Mormon Culture. It used to be(10+ years ago) considered an "acceptable" question to ask a couple who had been married several years if there were going to start a family. Or if they only had one child if they were going to have more. Now I don't think it is considered and acceptable question to ask. That is not a big shift but it is indicative of culture norm and expectations changing.
Maybe it is not weird that the offset is regressing to zero but why was there an offset in the first place? I suspect there is at least more research about that question.
I’m not Mormon (most of our friends are atheist or agnostic) but people asked us when we were going to have kids all the time, and when we had one we were regularly asked if we wanted more. I know this is sort of considered a bit of a nosy question nowadays, but I would say it’s still pretty normal. Maybe we just don’t project that we’re the sort to get bent out of shape if asked?
RE: 23, the "Effective Altruism" movement announcing that they're going all-in on supporting... Ivy League/Caltech/Swarthmore students with extra mentorship/funding/organization on campus smells a bit like reflexive elitism. Is it really not worth anyone's time to mentor students at a non-California flagship state school or seek out people who might be world-changers at schools who aren't already connected to strong pipelines to the elite? Seems like yet another way to hoard opportunities among a select group rather than seeking out new talent [and yes, they do say they will look to expand later, though I'm sure it will be to populist institutions like Cornell and Williams] and doesn't sound particularly altruistic in practice.
I am sure the "Effective Altruism" are more concerned with what will get results than what will be perceived as fair. If they feel they can double the Effective Altruism output of a student through a labor-intensive program, it makes sense to start with the people with the highest projected output.
That said, I'm skeptical their mentoring program will amount to much of a benefit at all.
I think the difference between your view (of what's more or less altruistic) and that of the folks in the linked post is that, roughly speaking, you lean towards equity while they lean towards rowing/steering, in Holden Karnofsky's schema https://www.cold-takes.com/rowing-steering-anchoring-equity-mutiny/
> Nate Silver, Tyler Cowen, and Garrett Jones come out in favor of “the public health establishment deliberately delayed the COVID vaccine by a month so it wouldn’t make Trump look good before Election Day”
Why would other countries play along with this? The UK for example only started injecting in December, well after the election.
And the UK was the first in the world with the Western vaccines (I forget when Sputnik and Sinovac/Sinopharm were first used), so there aren't any counterexamples, not even Israel.
I find this hard to answer because I don't understand why so many countries took the same amount of time to begin with - you'd think there would be lots of variation in how large/long a study they needed.
Phase III clinical trials, even the abbreviated "Warp Speed" version, are very expensive. Pfizer isn't going to do a separate one for e.g. Belgium. If Belgium says, "we're OK with 10,000 people, one months, and 20 cases", Pfizer is going to say "and you can pull that data out of the bigger study we'll be releasing to the FDA in October, er, November". If Belgium says "we need a million people, a full year, and a thousand cases", Pfizer is going to say "good luck with that". And it's unlikely the Belgian government is going to pay for it out of pocket.
The EU as a whole could have, but their standards are sufficiently similar to those of the US FDA that everybody seems to have agreed that one big trial for combined US/EU approval was the way to go, and October was too late to turn on a dime. The FDA is the 800-lb gorilla in this case, so every pharma company is incentivized to conduct their one big expensive trial to FDA standards, and every national regulatory agency is incentivized to base their approval decision on the results of an FDA-grade trial.
The theory is that political considerations influenced Pfizer to sit on the trial results (they stopped processing lab samples from mid-Oct. till the day after the election). Hard to see how other countries would approve without trial data. Pfizer just happened to be on track to be first to complete testing.
Re: 23, I agree with the thesis of the post but am not at all sure it will be possible to achieve. I was also active in Stanford EA in 2015 and I think a large amount of the effect really was just getting extraordinarily lucky in having a bunch of interested/interesting people around at the same time (many of whom never ended up directly involved in EA after graduation, to my knowledge, but were still an important component - five people does not make a sustainable campus organization). It's still worth putting effort in to attempting to make such a group coalesce, but it might simply not work if you don't get lucky with the people who happen to be around when you're trying.
Pascal says something like that: we're afraid ever to just sit alone without recreation for fear they'll think about the fact that they're going to die one day. So instead we all frantically race after constant amusements to fill our minds so no disquietingly serious thoughts about our soul and eternal fate can intrude.
Expert meditators can meditate while doing other things. If getting hit on the head with a stick can be a catalyst for enlightenment, why not getting a Yahtzee?
Check out Thi Nguyen's book on the nature of games - he claims they are the art form whose medium is human agency rather than a physical medium. He analogizes games to yoga, as a way of trying out ways of structuring one's (physical body/mental agency) that come naturally to other people but don't arise in one's own life.
Not entirely sure how this fits into Buddhism though.
2.) This reminds me of what Turkey is doing right now to deal with its own inflation. I can go into more detail but they're effectively creating financial instruments that compensate for inflation through a variety of methods I an get into. And then trying to move most normal people's accounts into those instruments. They're trying it not because the new minister knew no economics but because Erdogan has a weird ideas about economics that just happen to justify his policies. Though in this specific case his beliefs have some possibility of working. Basically, it creates pressure on the Turkish balance sheet (so the government is picking up the tab) in exchange for (theoretically) shoring up the currency's value by allowing the lira to create fake dollars/euros. This should, in principle, help restore some confidence and give the central bank (which has not been printing money) a chance to lower inflation. Effectively through wealth transfers to ordinary Turks. While this is an expensive policy that will probably put Turkey somewhat into debt it's probably worth doing that to fight high inflation.
The reason we can't do it here is that the US has been printing money and we don't have the same fiscal room with all the other spending that's been going on. But most importantly: Turkey's effectively printing Euros/Dollars through its fiscal instruments (because that's how the peg works). Brazil, iirc, did likewise, as did several American countries. But the US can't peg its currency to the dollar because it IS the dollar.
11.) I suspect it's a sign that Chinese censorship algorithms don't like homophones and someone delicately told the WHO that China would prefer their people didn't have a sneaky way to refer to Xi.
21.) There has been a lot of interesting movement in power lately. I look forward to seeing how it ends up. Though frankly, people get really politicized and polarized about power production such that I expect there will be significant barriers. I'm less qualified to comment on the rest. I also find it revealing that Fintech is not among the new technologies despite the fact we're objectively behind much of the rest of the world. Even on easily measurable things like the T3 standard.
22.) Why would this only happen in upper income countries? Plenty of poor countries strangle themselves with red tape and regulation too.
24.) I like to use the metaphor that engineers/architects/artisans are like musicians. If you give a skilled violinist a violin and sheet music and ask them to play the next day they'll come in and do a great job. If you give an unskilled violnist the same task they will produce much worse music. Yet they both spent the same time preparing the piece. If anything the skilled violinist probably spent less time. Likewise, an engineer's work product is much less a result of the specific task than their total accumulated work experience. The issue is that measuring artisanal output is often very difficult and nearly impossible outside of someone with the same skill. So corporate sets up arbitrary rules like 40 hours a week or time tracking which then get gamed. You can see the same dynamic in teaching music: the really great musicians do practice a lot but they move on from the grade school "three hours of homework" model pretty quickly once they're good/independent. (And to be fair, that kind of grinding is necessary to GET good.)
33.) I can confirm this is what YC's process is like. I think he might have a bit of hero worship going on. Don't get me wrong, they all came across as very smart and sharp. I think the maximal terms ("the smartest person I've ever met and I just met the smartest bankers in Africa") is a bit overblown. Though, to be clear, I'm not implying Sam Altman isn't highly intelligent. I agree it's cut short. There would have been pre-interviews and all that. They would have met other people. And investor stories are a favorite of mine.
34.) I have one of these. Though it's significantly less nice than that one. I also have a candle clock that tracks the time, day, and year. Though I don't actually run it that often.
#24 - I've had coworkers who do roughly an hour of work a day. I think they thought they were getting away with it, but I noticed, and it drove me crazy.
Re the Amazon article (#10) and allowing consumers to post negative reviews -- I just saw Miracle on 34th Street and the man who thinks he's Santa Claus starts telling Macy's customers about bargains at other stores. Customers are so impressed that Macy's sales go up, and all the competing department stores start doing the same thing. Learn from the classics!
Re #14: I would argue that in modern societies, especially of the welfare-state variety, it is unlikely that poverty is a major cause of cognitive problems; it's more likely the reverse. Although a third factor could cause both. Re #15: What group are we comparing to? Are we comparing to people incarcerated for non-violent crimes, or killers who weren't incarcerated, or the general population? Re #27: Maybe we need a 3-D graph.
#1: No hopscotch, then? I'm out, Buddha. That's a red line for me.
#12: The original Florida man?
#22: Elon Musk was just waxing philosophic about this with Lex Fridman about how wars help sweep away red tape and bad rules, and since we don't really have many of those anymore (not great power conflicts, anyway), we're stuck with the rent-seeking incumbents. I didn't find this portion of the conversation particularly insightful--it wasn't anything you couldn't have read in, say, Reason Magazine over the last ten years--but how to achieve similar effects without the eight figure death tolls of two world wars is certainly a topic worth thinking about. The post-war German economic miracle was a real phenomenon.
#29: eh, was successfully crowd-funding new social media sites a legit problem in recent years? I don't care much for Facebook or Twitter, but "finally, we found a way to make Gab work!" or some other right-wing alternative site seems like...not our most pressing issue. My Crypto Bear card remains valid.
#33: This left me wanting more, too, but not in the "George Costanza always leaving on a high note" sense, but rather in the "you yada yada-ed sex?" sense, where they skipped over the best part.
Robin Hanson (who had once said he was a single-issue anti-war voter) has recently been warning that a lack of military competition may cause global government (including what he thinks is a currently-existing rather informal version of it) to rot: https://www.overcomingbias.com/2021/11/will-world-government-rot.html
I am inclined to agree. Possible counterpoint, though: the Soviet Union collapsed with more of a whimper than a bang. I don't know how much their war in Afghanistan contributed to that collapse, however.
While true, I don't think it refutes my point that correct ideas don't *only* win out in non-verbal conflict. Breadlines are pretty convincing evidence that something has gone awry, for example.
Really? Lobbyists got drafted? 3 letter agencies shut down or switched mission?Which practices were abandoned? My gut says rentseekers can surf that wave. The only thing that frustrates them is when things shut down and budgets dry up.
And of course, the military-industrial complex, the god emperor of rent seekers, was born at that moment. Well, not born, it must’ve been around before… came of age?
#26, former Mormon here - it's not entirely "rise of secularism" but it rhymes.
For the past 15 years people have been leaving Mormonism in *droves*. It used to be that almost no one left, and it was incredibly taboo to even talk about.
When I tried to leave in 2008 I knew 1 person in my entire life who'd ever left, and she was 25 years older than me. When I successfully left in 2012, I probably knew a handful of people who'd exited. Today, _everyone in Mormonism_ knows someone who has left. It's still a big deal to leave, but it's much less taboo.
In trying to prevent even more people from leaving, the church has become more "big tent". Some of the defining features of the Mormon lifestyle (big families, church literally every Sunday in person, women only as homemakers) have been deemphasized, to the point where "Buffet Mormonism" is now part of the nomenclature, where you're not judged as harshly if you subscribe to certain beliefs/rules but not others.
What that all means is fewer people view "have lots of kids" as a necessary and fundamental part to "being a good Mormon".
"Net increase in membership" is not necessarily inconsistent with people leaving in droves. You can still have a net increase so long as the gross increase (conversions + births) outnumber departures.
There's also the question of how you count departures. I've heard anecdotally that the LDS Church usually continues to count ex-Mormons as current church members unless they jump through some hoops to formally "resign" their membership.
There is data to back this. Difference between church records decline (~85% in 2010 to ~82% in 2018) vs. survey data (~88% in 2010 to ~72% in 2020) is significant.
The LDS church counts everyone who has ever been baptized as a member (unless you resign or they record your death), even if they haven’t heard from you in decades (and assumes you’re alive until you’re a centenarian), so their membership number simply isn’t meaningful. Why not publish global average Sunday attendance or number of people holding temple recommends?
1) Most people who leave don't withdraw their records from the church, so they're still counted in that chart.
2) Many of the converts joining are significantly less committed than the people who are leaving. They've maybe talked to the missionaries 3 times and then they get baptized. Ask anyone who's served a mission in Latin America to attest to this. So while this does help ebb the flow of people leaving per the chart, the total level of "committedness" to the church is absolutely down
The real issue with #26 is that it isn't a chart of Mormon birthrates, but birthrates in Utah County. Many non-LDS people have moved in, and that population change accounts for the decline in birthrate.
Re the Mormon babies or lack thereof. The study uses a location sample of Utah County, UT and Bonneville County, Idaho, as a proxy for “all Mormons.” This is a flaw for a few reasons. Utah County is a sub/exurb of Salt Lake City. Bonneville County is Idaho Falls, the nearest major city to Idaho National Lab. I argue Utah Cty is subject to middle/upper middle class downward population pressure making those Mormons anomalous. Also Bonneville county is population small but has larger than average population of lab employees, also subject to the class/education population pressure. And it’s a city even w/o the lab, hence another layer of same pressure.
The conclusion may reflect reality, I don’t know, but not for the reasons given. I’m not Mormon but I lived in/near Bonneville county for a while. Something that stood out at the time was age at marriage and childbirth was lower than the west coast (anecdotal).
Getting a good sample of Mormons spanning US/world geography, urban/rural range and income range would give a better picture.
To #26 as a former Mormon, formerly Utah county resident I think I can speak to this. First younger Mormons are having fewer children for the same reason everyone else is having fewer children - they're expensive. We're just highly motivated to have them for theological/cultural reasons. Anedotally I've watched the culturally ideal family (ie the one pushed in talks, considered normative etc) drop from 5-6 (my moms generation) to 3-5 (my generation), to 2-3 (the current generation). Simply put Mormons are subject to the same pressures as everyone we're just stubborn enough to delay the effects.
You're also equating Utah County and Utah with Mormonism which is less true than it used to be 20 years ago. Lots of non-mormon childless people moving in especially to the Salt Lake Area. Utah County is of course still a Mormon bastion but even that's less true than it used to be.
Finally the last 10-15 years have seen a pretty unpresidented swing away from orthodox belief. The Church failed to publish membership growth numbers for Utah for the first time ever this past year. Anecdotally the number of people leaving the Church has increased particularly over the pandemic, and exMormon support spaces are exploding.
“ I would have been willing to let this pass if they had just said “unlikely” - somebody might honestly think 22% is unlikely compared to some hypothetical belief that it’s near-certain.”
Surely it should be compared to the likelihood of the average person killing someone. Less than 0.1% I imagine.
At any rate, we jail murderers for the crime they’ve already committed rather than worrying what they may do in the future. If some otherwise meek mannered husband kills his wife in a crime of passion, and all psychological profiles, indicate he will never kill again we jail him anyway.
That’s retributive justice for you, not to be confused with revenge which is a personal thing; it’s the state dispassionately deciding this crime is worse than that crime and applying different punishments. Long may it continue.
In retributive justice the “trigger event” (presumably you mean the crime) is the reason and the justification for the sentence. Do you really need to be told who the “we” is.
Probably not. Perhaps you actually have thought all this through, and don’t need me to get you to think about it. Maybe you think an event can provide a justification, or that a system of justice that uses punishment based on some other sort of justification should not be called “retributive.” I apologize for my inept attempt to inspire thought.
Well if you want me to think about you should do a bit better than one or two lines of drive by posts and pedantry about what “we” means. I don’t think I or anybody else knows what you think, it is only clear that you disagree with me. You have given nobody here any food for thought because you haven’t explained what you are arguing for, and very badly explained what you are arguing against.
"At any rate, we jail murderers for the crime they’ve already committed rather than worrying what they may do in the future."
Says who? If somebody commits a murder, it demonstrates a willingness to commit murder absent in the overwhelming majority of the population. Therefore we remove them from society so that they can't murder again.
Punishment and deterrence are part of the equation somehow, though, are they not? Consider that if I get convicted of stealing cars, the penalty for that in California is a maximum of three years in prison. If I've demonstrated the willingness to steal expensive stuff, three years in prison isn't likely to do much to protect society from my depredations, considering average male life expectancy is just under 79 years in the US. I could steal a lot of stuff in the other 76 years I have available.
That's not a given. 'Willingness to murder' is not some inherent unalterable property of an individual disentangled from the context of that murder.
While I understand that 'genetically cast in stone' is all the rage in some quarters, people do change, and can be changed by institutions devoted to rehabilitation rather than punishment/exploitation. (so not American prisons)
There’s plenty of examples where somebody could be genuinely rehabilitated and not a threat to society when imprisoned. Imagine a drunk driver kills two children. Before his trial he gets sober and voluntarily gives up his driving license. Psychiatrist’s agree that the remorse is genuine. This man is no longer a drunk or a driver, and can therefore never kill again as a drunk driver, so should he be imprisoned at all? Isn’t he not rehabilitated already. Is he not no longer a threat to society.
If the answer is that he should be imprisoned anyway for the crime then that is a belief in retributive justice. I’m not making this up by the way, the idea that our system is primarily retributive is a banal truism.
As for deterrence - retributive justice is part of that but deterrence can go beyond that and arrest the innocent for potential criminality or treason ie internment.
A recovering alcoholic is more likely to become an alcoholic again than a person who never was one in the first place, no matter how sincere their remorse; giving up your license removes the potential to drive legally but drunk driving is already illegal, access to a car is the thing that matters and that's very situational and much harder to prevent. Even in your toy example with no murderous intent, your remorseful repentant defendant is IMO much more likely that the average citizen to cause future deaths. I don't think that alone is reason enough to imprison, and deterrence also plays a large role, but it doesn't zero out.
Also, fwiw a repentant drunk driver might actually not go to jail for vehicular manslaughter in a lot of places, precisely because they aren't likely to hurt more people. Suspended sentences and the like are tools often used in reality and rarely discussed in these hypotheticals
I should probably stop giving examples, as it is pointless. So to generalise.
In all cases retributive justice matches the punishment to the crime regardless of other benefits to society. Or as Kant said “ “Judicial punishment can never be used merely as a means to promote some other good for the criminal himself or for civil society, but instead it must in all cases be imposed on him only on the ground that he has committed a crime; for a human being ought never to be manipulated merely as a means related to another’s purposes”
There are three reasons to jail someone for a crime, of which you have only mentioned two:
1. Retribution (you mention)
2. Removal from society (you mention and say isn't accurate for at least some criminals)
3. Deterrence (which you do not mention)
If you know that the maximum jail sentence for killing a person is 1 year (or go really absurd and make it a week), then that would genuinely change your willingness to kill someone. Maybe most people go from an extremely low likelihood to a very low likelihood and still never kill someone. Some portion of the population might go from ~0% to 5% or whatever, and dramatically increase the murder rate.
Jailing one time "crime of passion" criminals similarly to hardened career criminals is one way we get around the potential for criminals to lie about their motives. If they knew that fabricating a "crime of passion" could eliminate their jail sentence, then of course they would claim that. I would say that we knowingly jail people that are very unlikely to ever commit a crime again, as a way to maintain a system were reason #2 (keep criminals out of society so they can't offend again) can be effective even in edge cases where we are uncertain of motives. That such a system also deters potential other murders is certainly a strong plus.
Deterrence isn’t the same as retributive justice though. They are often related but retributive justice needs to prove guilt while deterrence could be achieved by rounding up people who may do something in the future, to deter them or others. Internment is an example.
I agree? Maybe you meant to respond to someone else, but I listed them as separate categories. The current system seems most interested in deterrence, followed by removal from society. Retribution, by comparison, is at best a distant third.
I still think retributive justice is the primary category. Deterrence could be achieved with more draconian laws and less concern about the innocent being affected. Any system that has higher penalties for more serious crimes is retributive.
That said the 3 strikes laws are more about taking people from society, rather than the seriousness of the crime, so in that case you are correct.
That's interesting. I would consider higher penalties for more serious crimes to also be about deterrence - we want to deter people from committing more serious crimes above those committing less serious. If all crimes have the same punishment, then you aren't sending a signal that some crimes should be avoided more than others, but we really would prefer more jay walking and petty theft and less murder and arson.
I do see your point about retribution though, and believe you are correct that it's more important to the discussion than I had previously said.
Without knowing more about the distributions of the killing-someone statistics, I don't know what would be a reasonable number. I could imagine that double the median likelihood of killing someone (should it only be criminal homicide, or all deaths, even justified ones?) would be fine if the numbers were low in absolute terms.
20 is very encouraging! If most of the important info is not in digital text, and most of the digital text is unimportant, I never have to read my backlog of AI/safety papers to contribute to the field!
(This is somewhat sarcastic, but... am I wrong tho?)
You're not wrong. Network as much as possible instead, use papers strictly as references to look up very specific things - every now and again you'll duplicate work that someone else did already but you'll probably waste less time overall.
I'd never heard before that the negative correlation between intelligence and fertility *wasn't* simply a matter of more educated and successful people having fewer children for cultural reasons--a finding that even *uneducated* intelligent people have lower fertility.
I wonder to what extent this is due to higher rates of unintended pregnancy among the less-intelligent, even at the same educational and socioeconomic level. Given that, until the last decade, c. 50% of all US pregnancies were unintended, different levels of vulnerability to unintended pregnancy could be a big factor in fertility differences.
(Recently, due to long-acting reversible contraceptives, the unintended fraction of pregnancies has declined to around 1/3. I would expect the negative correlation between intelligence and fertility when controlled for education to have significantly declined as a result.)
Contrary to popular myth, higher levels of education in developed-world women are not correlated with lower desired family size, although wealthier and more educated women are more likely to fail to achieve their fertility goals, due primarily to delayed marriage.
I suppose an innate, biological tradeoff between fertility and intelligence in humans, as seen in those fruit flies, could also be a factor, but it seems implausible that biological limits to fertility would be the main factor in a society where fertility among all groups is well below replacement--where everybody is having far fewer children than they're biologically capable of
how much does "desired family size" actually mean if people aren't trying for kids until their 30s and thus predictably not having as many kids as they wanted?
> although this is weirdly short and leaves me wanting more
lsusr (https://www.lesswrong.com/users/lsusr) writings on LessWrong are usually like that, but there's a lot of them. He's a damn good writer/thinker, highly recommended.
#28 Having complained about mainstream media misuse of “no evidence” in the past, specifically that it seemed biased against right-wing political statements, I should credit NYT: in their obituary of Harry Reid, they noted that he famously claimed “without evidence” that Mitt Romney cheated on his taxes.
A perticular claim by a particular person being without evidence is different from there being no evidence for a position. I can say '1+1=2, trust me on this' or '1+1=2 (Russell, 1910)'.
Did they only add that on because he admitted that he lied about it? Not much credit to the NYT for that statement, given Harry Reid himself openly admitted it was a lie. Helping keep Mitt Romney from getting elected is one of the things Harry is most famous for, so I can see why they would include that tidbit anyway, but it's not much in their favor to include basic historical accuracy that no one denies.
"[Claim of the first successful deepfakes based hacking.] Looking through comments elsewhere, I think [this claim falls apart](...)" - actually those two links are about two different hacks (with the first not including any kind of detail beyond hearsay, so not verifiable). It sounds somewhat implausible that two different people would make up the same deepfake story. (There was indeed a wave of extortion-based attacks against semi-famous instagrammers at the time, see e.g. https://www.vice.com/en/article/93bw9z/bitcoin-scam-hostage-videos-instagram , but it wouldn't be strange for the attackers to use multiple methods.)
Those eigenrobot tweets are pretty much just content-free shitposting and as such it's not helpful to promote them as an example of how to talk about statistics and bias on the internet, IMO. They also miss the real point of story, which IMO is that the Markup was able to access a predpol company's dataset due to sysadmin incompetence, and did some handwavey analysis (they compared predictions to actual arrest and showed that it overpredicts in black areas and underpredicts in white ones, but admitted that it doesn't really prove anything since the police tried to deter crimes from happening based on those predictions, and the mismatch between predictions and reality could simply mean they were successful at it) to whip up interest in the story and push their point that this kind of data should be available for public scrutiny and not only when someone misconfigures a firewall.
I remember that article coming up before and I don't remember the point about it having a mismatch between predictions and arrests, instead just treating different amounts in different neighborhoods by itself proof of being bad.
Sorry, re-skimming the article it seems I misremembered it. The actual claim they make (alongside various disparate-impact claims such as higher predictions for black neighborhoods and higher predictions for poor neighborhoods) is "PredPol claims that using its software is likely to lead to fewer arrests because sending officers to the company’s prediction boxes creates a deterrent effect. However, we did not observe PredPol having a measurable impact on arrest rates, in either direction." ( https://themarkup.org/show-your-work/2021/12/02/how-we-determined-crime-prediction-software-disproportionately-targeted-low-income-black-and-latino-neighborhoods#stop-arrest-and-use-of-force-analysis ) and also it's in their methodology companion article, not the main article (although in their newsletter it was more prominent).
In any case IMO it fits into the pattern of Markup articles in general: they do a lot of hard work to obtain or create a dataset about the practices of some secretive corporate entity, do some analysis to show that those practices are potentially terrible (with the "potentially" part not emphasized), and call for more public scrutiny and transparency of the data. The analysis usually has a clickbaity title and doesn't actually prove much, but calls to attention to the lack of transparency around something that has a large public impact. I think that's on the net quite useful - companies preventing access to the data which could be used to judge the harm/benefit balance of their activity is a major problem, so raising awareness about it is important; the clickbaitiness is annoying but tolerable (that's how you get donor money I suppose).
2) I’ve been waiting for a decade for someone to explain how this story is made up / not the real reason for the end of (Réal) inflation. It just sounds way too good to be true. Surely if there’s an angle that deflates the hype, someone in the ACX comments would know it. Anyone?
Inflation is caused by bad fiscal & monetary policy but can be perpetuated by people’s expectations even after policy is fixed. Switching to the real was only half the plan. It was a way to stop expectations-driven inflation, but that switch was paired with major fiscal reforms to stop the root cause.
Even that, I do not understand. Sure, it hides the inflation in your paycheck and in your grocery prices. But does nobody notice that their savings get smaller and smaller no matter how vigorously they save? Or was Brazil just so poor that nobody had any savings, or so inflation-wracked that nobody *bothered* to have savings any more?
There might be something *in general* to the notion of having a social media site that rewards first movers. But "In Praise of Ponzis" makes me angry, because the glorious outcome of the cooperation he wants to build is just "if we all cooperate, we can trick Youtube's recommendation system".
Tricking Youtube's recommendation system does not generate social good, because you're just replacing one (possibly more authentic) winner with your own artificially generated winner.
Also, Youtube can just patch the hole in their recommendation system, if they haven't already.
Re 7: If you too strongly select for intelligence (or any other trait) you will do so at the expense of other beneficial traits and overall fitness. This is because the strong pressures on intelligence will cause you to select for genes which (for example) increase IQ by 0.5 points but decrease overall fitness 0.5% (numbers made up). What you want is a weak pressure on intelligence sustained, over a long period of time, in the presence of pressure on other traits you value. This causes genes which increase intelligence but don't decrease overall fitness (or decrease it only very small amounts) to be positively selected for, while genes which increase intelligence but decrease overall fitness are rejected. Time is important as it allows new mutations to enter the gene pool and be tested by these pressures.
In the fish example, if the only factor being selected for is intelligence, then you will select for negative, neutral and positive overall genes that increase intelligence. However, negative changes (changes that increase intelligence but decrease overall fitness) will be the lowest hanging fruit because they will be barely hanging on in the gene pool and thus have the strongest potential to grow in number. This dynamic also would incorporate the homozygous/heterozygous dynamic you talk about in your 2016 article.
What you want is a situation where all selection forces for the traits you desire are in a reasonable balance. So perhaps health, symmetry and intelligence are being selected for simultaneously, and any new mutation which helps one but harms the others too much is removed.
WRT Ashkenazi Jews; in normal populations, the reproductive returns from higher IQ were probably not strong enough to overwhelm the negative effects of being more sickly or having more sickly family members. However in Ashkenazi Jewish society, for a time, the benefits of increased intelligence outweighed the reproductive harm of having more sickly family members. This would have improved the selection for positive and neutral intelligence boosting mutations, but also negative ones - which would be proportionally boosted more until a new equilibrium was reached. I'm unsure if it would be easier to develop mutations which boost IQ but harm other aspects fitness versus mutations which boost IQ and help other aspects of fitness, but initially you're going to be tapping into the harm pool.
In your 2016 article you talk about how IQ genetically correlates with many good things. Apologies if you ultimately concluded this, but (severity of) mutational load makes the most sense. 99% of impactful mutations are shit. They're still going to get passed along for a while if they aren't too major. Some people (by chance) are going to inherit more of the shittiest ones and these are most likely to just harm many things (like 10% corroding every part in a car made of a certain type of metal). For example, if you have a mutation that makes your mitochondria 10% less efficient, that will possible reduce your height, health, intelligence and cardiovascular fitness. If you're unlucky enough to have a greater proportion of your shitty mutations be the shittiest ones, then they are probably going to drag everything down.
That Ponzi scheme article (#29) is rather bad. For one thing it confuses Ponzi schemes with things that aren't. The essence of a Ponzi scheme is that the system is funded by new entrants; when you give tokens to everyone who watches your video, in the hope of recouping on ad fees, that's not a Ponzi scheme, that's just you paying people to watch ads. If that worked out economically, advertisers would just cut out the middle-man and pay people directly.
And for actual Ponzi schemes, the article conveniently ignores the reason they are forbidden in many contexts, which is that most investors will lose their money - the system doesn't produce any value, just redistributes it from late entrants to early ones. You can play the lottery of entering at what you think is an early point, and on the system level you can fundraise off of people's propensity for gambling, but it doesn't seem like a revolutionary improvement over traditional fundraising. And anything on top of a blockchain needs to be a revolutionary improvement to work, because it needs to compensate all the intrinsic disadvantages of being on top of a blockchain.
This is really interesting, and I too would like to see Lyman's Mormon preference data. LDS culture today is simply a world apart from what it was even 30 or 40 years ago. Enjoyed your post on Mormon moderation as well, it checks out from my POV.
Yeah, re:technology, I remember in like 2014 people were confidently predicting fully self-driving cars by like 2018. Now self-driving cars are not even on Noah's list. People are probably overly optimistic.
As a Mormon, and a current student at BYU-Idaho (see data for no.26), I feel like many of the other takes here in the comments are broadly correct. LDS members are not immune to broader cultural trends/pressures, though they are mitigated for a time by our community's relative insularity (like most frictional religious minorities). To the extent that lower fertility rates are a cultural issue, more than enough time has passed since the broad slowdown occured for the majority of demographic groups for cultural ideas to have permeated Mormon norms. To the extent that this effect is a product of mothers entering the workforce, etc., that has definitely happened within the Church, even though it again has lagged the same phenomenon in broader society. (I asked a few family members about what they thought the reason for this decline in the fertility rate could be and they all mentioned cost of housing, how expensive it is to rear a kid, etc. AKA the kinds of takes you would equally expect from a non-Mormon sample. A couple of them also mentioned that a large proportion of LDS friends in their age groups wanted to have kids but were unable to due to actual fertility issues.)
Mind you, this particular take seems a bit haphazard and I would quibble slightly with the methodology; I don't think that BYU-Idaho -- located in Rexburg, Madison County -- holds enormous relevance for Bonneville County. The latter is larger, and centered around Idaho Falls, which has grown tremendously in the last couple of decades and is located within a different micropolitan statistical area than Rexburg, i.e. may be a poor proxy.
This is probably why Romney is so supportive of certain aspects of BBB even if he wants them separated from the larger bill. His CTC was arguably better than Biden's for instance.
Re 2: Brazilian here, can attest that Planet Money episode was an good summary of Plano Real. I was eleven, so just old enough to remember visiting a newstand to buy comic books and checking a table posted on the side that told me how much each issue would cost that month. Spider-Man, Web of Spider-Man, and all the rest started going for "A23". One URV went from CR$647.50 to CR$2750.00 over four months and then it felt like inflation stopped.
It's still probably Brazil's greatest success story. It has become the sort of thing everyone is in favor of, like mom and apple pie and hating the NYT for doxxing bloggers, but both 2022 frontrunners were against it in 1994.
On #6, discussed many of these with my girlfriend. The one on polygenetic screening for gender took me for a spin, until I realized it was literally China’s one-child policy, then it became quite clear why it’s a form of eugenics you shouldn’t support. I don’t think the sex ratio of society is a “slider” that we can be trusted to muck around with, at least not yet.
The problem is that it is a very simple tragedy of the commons given our society. For economic value each individual family would prefer men in a one child scenario but for society we should actually prefer women in that case. This is actually pretty consistent even at a higher cap. Of course what is a 10% problem in America on this issue is a 70% problem in most Asian countries.
That was certainly true in early ag and I industrial societies but in modern, wealthy societies does that remain true? Women have a way higher chance at obtaining higher education in America and I believe Europe.
6) Quasi-Mormon here, but not in Utah. I haven’t noticed anything at all. There are 3 families with 6+ kids that go to church with me, and everyone has kids or is planning on it. Maybe it’s a trend in Utah and that’s overwhelming the small progress down south?
#26: As a Latter Day Saint who spent a decade living in Utah, I definitely don’t think it was secularism or any other cultural explanations. The explanation that I find most plausible is the high cost of child care (including the opportunity cost of having a stay at home mother). It’s increasingly difficult to sustain a family on a single income and that difficulty increases with each additional child.
It intersects in a nasty way with high housing costs. High housing costs mean you need two incomes (even if the second one is only part-time) to buy a house unless you've got a pretty high income job, and you have to move really far out west and south from both Utah and Salt Lake Valleys to find more affordable housing.
If you're living really far away from either your or your spouses' parents' houses, then you can't regularly lean on them for child care. That in turn means you either need to pay for really expensive child care, delay child-bearing until you're more financially secure, or get back into work as soon as possible after your children are school-age - all of which tends to lower the number of children you go for.
I know a few Mormon couples who are just living in one of the parents' basements with their young children while trying to save up money for a house. They don't seem to be unhappy with it, though - the grand-parents provide some child care and get lots of time with their grand-children, and the parents save money.
Agreed! The cost of housing is probably the single largest reason why single income families are so difficult to maintain these days and that definitely impacts the decisions that couples make when deciding how many children to have.
The annoying thing is that double-income families are also a leading cause of high property prices. Sixty years ago, the median house cost (hand-wavey) what the median single-income family could afford, but now the median house costs (hand-wavey) what the median double-income family can afford.
At what point did it become unusual in the US for three generations to share a house? It was more the norm than the exception at the time and place (late USSR) I was born, I think, and almost certainly when my parents were born. How long ago would it have been common in the US? Or was it never particularly common in an urban setting?
It was common for fairly poor people in NYC before WWII, at least. I suspect WWII really is the dividing line, because afterwards is when suburbia exploded.
The immediate aftermath of WWII was characterized by a substantial housing shortage in the United States, so there would certainly have been quite a few three-generation households just because there was no other place for the twenty-somethings to live. But it's certainly plausible that the postwar economy was characterized by a vastly increased *demand* for independent living accommodations for young families, as soon as the housing could be built to meet it. So, probably 1950s.
I’m aware of three politically-related delays in the mRNA clinical trials. Please note that safety, diversity, and efficacy are somewhat separate issues.
– Safety: The decision to require two months of safety data, pushing Pfizer’s first date for an application back to 11/17/2020. The campaign for this was clearly motivated in part by Democratic concerns to deny Trump his October Surprise on the vaccine front, but … safety is important. So I’m not as worked up over this as some libertarian economists are. I can't say off the top of my head what the right amount of time to collect safety information was.
– Diversity: The Trump Administration’s decision in late summer to require Moderna to delay its clinical trial by a month to recruit a more racially diverse set of volunteers to test the safety and efficacy of the vaccine on different races. I’m sympathetic toward this delay: as I may have mentioned once or twice over the years, human biodiversity can be important in a variety of settings. Ironically, however, as far as we can tell so far, HBD doesn’t matter much for vaccines: racial equality more or less reigns in terms of response to mRNA vaccines. So the one month slowdown of Moderna by the Trump Administration in the name of diversity cost Trump his October Surprise, delaying Moderna’s announcement of its high efficacy until 11/16/2020.
– Efficacy: Finally, the vastly underpublicized decision by Pfizer to shut down lab processing of clinical trial samples from late October until 11/4/2020, the day after the election. Pfizer wound up blowing through even the third checkpoint when they let their lab get back to work the day after the election. This lab shutdown strikes me as the most egregious of the three politics-related interventions. The first two called for More Data but the third led to Less Data during a critical week at the beginning of 2020’s Winter Surge.
Without the lab shutdown, my best guess is that Pfizer would have announced the very high efficacy of its vaccine on Monday, 11/2/2020, the day before the election. Trump would then have spent the last 24 hours of the campaign trumpeting the success of his vaccine strategy.
Would that have switched enough voters in Georgia, Arizona, and Nevada to cause a 269-269 tie in the Electoral College and cast the decision to the House of Representatives with one vote for each delegation, probably favoring Trump? If that looked like it would happen, would there have been a violent and/or corrupt intervention to deny Trump a second term?
Who knows?
On the other hand, I don’t think Pfizer’s lab shutdown delayed the rollout of the vaccine by much more than one week and maybe just a few days.
The government decision had already been made due to the Democrats’ anti-VAXX fear, uncertainty and doubt campaign in the fall to require, in effect, until November 17, 2020 for enough safety data to be available to begin government processing of the application. However, announcement of efficacy on 11/2 rather than 11/9 would have given states and localities an extra week to focus on the big challenge of the vaccine rollout, which most botched until about the second half of January.
So I could imagine the vaccine rollout running a few days ahead without the Pfizer lab shutdown. The number of lives lost due to the lab shutdown sounds to me like in the hundreds but probably not in the thousands.
But who knows? Perhaps if Pfizer hadn’t shut down its lab processing, Democrats would have been the anti-vaxxers in 2021?
The alternative potential timelines rapidly spin in multiple directions. A timeline in which the mainstream media spent 11/2/20 and 11/3/20 screaming in rage about how the Pfizer vaccine was a dangerous conspiracy to re-elect Trump would have been ... different.
I suspect that when Pfizer CEO Bourlas is alone with his conscience, he says to himself that, sleazy as it sounds, his ordering the lab shutdown until after the election, wound up saving lives on net.
W.r.t. Vetocracy vs. Bulldozers, it's interesting that Vitalik, given that he lives here, doesn't bring up Swiss political system, which is heavily consensus oriented (vetocracy) but then there's always a referendum hanging above any decision, which, if the consensus decision was not good enough, would bulldoze over it (referenda results are written to constitution and thus take precedence over everything else).
Per the link at the end of #27, could someone please explain Quadratic Voting to me as though I’m an idiot? How could it possibly be more equitable if billionaires can conceivably buy more votes?
In a typical quadratic voting scheme, all voters gets a fixed number of tokens with which they can "buy" votes. You typically don't buy votes with actual money. A billionaire would get the same amount of tokens as everyone else.
While direct buying of votes probably doesn't happen to a very large degree (I presume some small amount of it goes on, what with reports of mysterious bunches of blank ballot papers being found in the back seat of cars and so on), I think that a token system is just as open to fraud and bribery as any current system.
Yes, Mr. Beff Jezos only has the same number of tokens as anyone else. Like all the workers in fulfilment centres who work on contract to Amazon, and the information campaign that Amazon engaged in to tell those workers of all the bad effects getting unionised would have on them.
Why not "information campaigns" about where they should best spend their tokens? Not that this is trying to influence them to vote the way Mr. Jezos would like, not at all! Just helping their valued employees get all the information they need about this new system of voting and how to make it work.
EDIT: Of course, on second thoughts, I have been thinking too small-scale. Beff doesn't have to care that he has ten tokens to spend on votes and I have ten tokens to spend on votes. If he cares about an issue, he doesn't have to buy tokens to buy votes, he can just buy the candidate.
By being in a position to donate to national, local, and individual campaigns, Beff can make it that Bob Bobson gets selected to run instead of Jim Jimson, because Bob has sufficient funding to pay for the election campaign and Jim is trying to cajole grannies to scrape a few cents out of a jam-jar to contribute. So if Bob has the right views on an issue, which just so happen to chime in with the views Beff has, then I get to spend my tokens on either Bob or Some Loser from the other party. So long as Beff is fairly sure Bob will win, then Beff doesn't have to spend his tokens on the issue of the Lesser-Spotted Mugwumper, because once Bob is returned to office, he'll sort that one out.
I mean, what you describe isn't great, but isn't something that the proposal for quadratic voting changes at all - all of those issues are present in the status quo, so they're not an argument against quadratic voting being an improvement on the margin.
The section "Public goods, public markets" explains when you want markets ("one dollar, one vote"), when you want elections (one-person-one-vote), and when you want quadratic voting.
> ... Now, you might notice that neither of the graphs above look quite right. The first graph over-privileges people who care a lot (or are wealthy), the second graph over-privileges people who care only a little, which is also a problem. The single sheep's desire to live is more important than the two wolves' desire to have a tasty dinner.
> But what do we actually want? Ultimately, we want a scheme where how much influence you "buy" is proportional to how much you care.
> ...
> Notice that only quadratic voting has this nice property that the amount of influence you purchase is proportional to how much you care; the other two mechanisms either over-privilege concentrated interests or over-privilege diffuse interests.
The problem is even in a complex semi-direct democracy scheme where you had 10000 votes tokens and cound spend in whatever combination you thought optimizing what you cared about the wealthy can trivially buy/influence enough votes to win. Because they control the distribution of news and they control, to some degree, the funding for "science".
Quadratic voting is another in a long line of hacks by the wealthier people who have something of a conscience whereby they get to remain wealthy while sidestepping some problem related to wealth concentration and/or the bad behavior of their wealthy peers. They never work. The solution is to limit wealth concentration somehow.
If you assume that the elite control what the masses think then literally any form of government is going to be de facto controlled by the elite. In practice, the populace often has a mind of its own - dedicated propaganda is limited in bandwidth and can at best enforce the party line on a few key issues - and thus some forms of democracy do meaningfully give power to the common people.
Taking America as a very imperfect example, your ability to influence presidential candidates is very limited (though less than it used to be) because the pool of candidates to choose between is strongly filtered. Your ability too meaningfully choose the minor functionaries of you local council, however, is very real, and a surprisingly large fraction of meaningful decisions are made at very local levels.
> the fact is they made a deliberate decision to make the process take an extra month, and that some four-to-five-digit number of people died because of this decision.
I love how you can just casually drop that your side killed 50,000 people just so that Trump's approval ratings would fall, but somehow you still think they're the good guys
I think we should bring those incense clocks back.
Ferbreeze already makes an air freshener that cycles through 3 different smells so you don't get used to them and I see no reason why you couldn't expand that to 12 aromas that change on the hour. The plug-in fresheners just use a 2.4W resistor next to the wick which gets warm and promotes evaporation of the volatiles. This plus a half-watt noctua fan is well within what a USB 3 or USB-C can supply.
I'll have to find out what 3d printer filaments are compatible with air freshener juice because I've seen the solvents from them degrading plastic before.
So here I am again, to correct a mistake on vitamin D.
The study did not "find no effect". The exact quote is:
"61 (10.46%) individuals in the treated group died, compared to 386 (25.81%) in the non-treated group [odds ratio (OR): 0.597; 95% CI: 0.318-1.121; p=0.109] ... 45 (12.19%) individuals in the treated group were admitted to ICU, compared to 129 (26.27%) in the non-treated group (OR: 0.326; 95%CI: 0.149-0.712; p=0.005)."
So they found a very strong result in preventing ICU admissions. Mortality also had a reduction but it did not reach statistical significance. The correct probabilistic interpretation of that result is of course not "prevents ICU but not death", but rather "prevents ICU and will likely show significance on death as more trials are done".
So annoying. There's always lots of talk about shortage of software engineers in general - but apparently no one cares about... not even trivial inconvenience - huge ones. Maayybe there would be less of a "shortage" if there was a) a legible process to getting a job which b) wouldn't constrain candidate population to a single location on Earth.
I agree this is annoying. Nontrivial to improve, though.
The thing is, a shocking amount of software engineering productivity depends on "soft skill" activities: figuring out product requirements, socializing designs, coordinating between teams, understanding user issues, etc. Some of this can be avoided if your team is a single person or just a few people, but that sets a crippling upper limit on how much the team can build. Software businesses in a competitive environment succeed by innovating faster than their competitors, so they practically have to embrace those collaboration costs.
I mention this because of two implications: 1. the much maligned "culture fit" really is important-- effective software engineering at scale requires high levels of trust and great communication as well as technical proficiency, and 2. zoom collaboration, though it's a big improvement over previous state of the art for remote work, imposes a substantial cost in engineering effectiveness. 1) explains why your a) is hard-- software hiring remains a bit cliquey and hard to scale-- and 2) explains why your b) is hard-- software work wants to be geographically concentrated. (There are additional factors at work here too, like the typical software job being under 10 years, so people have to build their life and career plans around having jobs at several different companies.)
The role of the Bay Area as sole tech hub is overstated. The Seattle area is also a huge tech hub, with headquarters for Microsoft and Amazon and large satellite offices for a bunch of other big tech companies. And most big tech companies also have smaller satellite offices in various second or third tier tech hubs in the US: Boston, NYC, Los Angeles, Raleigh-Durham, Austin, Lehi, etc.
There are also significant tech hubs in other countries. The product team I work on (Adobe XD) is distributed between four US offices (SF, San Jose, Seattle, and Boston) and three international offices (Bucharest, Bangalore, and Mumbai). At other big tech companies and teams I've worked for, I've also worked with teams in London, Vancouver, Hyderabad, Tokyo, and Beijing.
From what I gather, it's the startup ecosystem that's most heavily concentrated in the Bay Area. My guess is that a major factor for this is that most of the VCs and potential Angel Investors are here in the Bay Area, due to the money and networks coming from people who got rich in tech a decade or two ago when the industry as a whole was much more Bay Area centric.
One of the major problems is that the companies are soul-sucking to work for. Most of the "productive" work being done is micro-optimizing sales and advertising revenue, which interests approximately zero software developers directly.
Then you get the political agitprop where the left-wing employees actively work to drive out the non-conforming (typically conservative-leaning) employees. This results in a lot of people leaving because they don't want to deal with that stress.
re: 19, even the article says their names were already out there publicly, in other articles and interviews, and they are often recognized on the street. if we lower the meaning of dox to mean "identified online but by someone we didnt want to" then sure, they were doxed"
I wish these comments were sorted by link number. Not sure if there's a platform that would support that. I guess Scott could post each link as a comment on a reddit post.
It might help, if Scott generates anchor points like "Discussion of Link # n", maybe add another one labelled "Junk, Spam & unrelated". Then it's up to the users, if they follow that structure by separating their comments to the respective threads (maybe enforced by Scott consequently deleting everything in violation of that rule)...
Actually, any early arrival at the discussion section could generate the anchor points too -- just a bunch of posts: "Disc of link n," "of link n+1" etc. If somebody does this next time around I will congratulate the daylights out of them.
A little bit further: "This has the potential not just to accomplish amazing feats (build a Mars base!), but to make space development a viable commercial activity in terms of tourism, research, and possibly even mining. It also obviously has transformative military applications."
In war, the point is to kill the other side with minimal loss of life on your own side. As Patton said, "No bastard ever won a war by dying for his country. He won it by making the other poor dumb bastard die for his country."
And of course better still is to convince the enemy that you really -can- kill his soldiers without endangering your own, as you might then not have to kill anybody.
Possibly not so much genetics as the environment; showbiz/entertainment world plus celebrity plus the blizzard of drugs in that world. Certainly having genes for addiction makes it worse, but even 'normal' people manage to fuck themselves up when they get rich and famous.
With regard to brazil and the URV how did they avoid the problem that once you price everything in inflation adjusted units you totally eliminate the ability of the central bank to affect the economy by increasing inflation or increasing interest rates. I mean it doesn't render central banks totally powerless but, one of the biggest economic dangers is that the stickiness of wages (wages tend to stay at the same level even during a downturn) can create a depression. Central banks generally respond to this in part by inflating the currency which has the effect of reducing the real value of the debts and the real value of wages but if everything is denominated in inflation adjusted units they lose that ability.
Ok, ok, this is way simplifying and if the country had a good credit rating it could still borrow and spend but surely creditors would also demand inflation adjusted units severely limiting the government's ability to engage in these kinds of interventions.
I don't think all contracts in Brazil today are made in inflation adjusted units so how did they extricate themselves from this problem?
> all contracts in Brazil today are made in inflation adjusted units.
To a large extent, they actually are -- although it takes the form of automatic inflation correction done annually by a pre-determined index (typically, a CPI measure).
I mean, if the status quo was a central bank uncontrollably printing money for decades causing massive over-inflation, then deliberately removing that capability for at least a little while was probably a key feature of the plan - no-one trusts that you won't just print more money on your say-so, but when you throw the money printer out the window suddenly money means something again.
Regarding the network heterogeneity hypothesis I'm tempted to say: well duh but does it really matter. Any number of people, including myself, have been remarking that this must, in some sense, be at play to explain the fact that the virus doesn't behave like the simple exponential models.
While I'm frustrated at the fact that it seems like all our digital gizmos mean that there should be the ability to do much more accurate population models in our epidemiological simulations I'm unconvinced it would really do much to affect how we've dealt with the pandemic. No one is really responsibly inferring substantial evidence from masks or other interventions from these hugely noisy data sets and if the total population can be thought of as a bunch of sub-populations with different reproduction rates (and occasional crossover) it's not clear to me that the standard model isn't still a pretty good guide to appropriate interventions as long as one doesn't take it too seriously (e.g. expect to be able to suppress hard for a quick win).
The article goes into detail but it's not a good guide at all. The author tested the standard model against their generated data and asked the model to estimate the effects of a non-existent lockdown.
"As you can see, according to the model and depending on when the lockdown was supposed to be in effect in each region, a lockdown that never took place and therefore had no effect whatsoever reduced the daily growth of infection by 7% to 10%. "
You can't say that a model that can be so easily fooled is a good guide.
Additionally, since the papers have been so heavily cited it does seem reasonable to assume that people are inferring substantial evidence from the data sets.
Re #21: These kinds of articles are easy to write if you don’t work in technology. There are a lot of things that look exciting that sort of work in a lab. There are a lot of things that demonstrably work but that would require significant changes in society and human behavior.
Turning something from “works in a lab” to “works in each of the 80 million iPhones I build every quarter for 3-4 years when dropped on cement every month or two” is a non-trivial problem. Many (most) technologies cannot move past this hurdle. Battery technologies in particular are extremely difficult to turn from the lab to the real world.
Autonomous cars are really exciting to people. But at least for the foreseeable future, they require very regular and structured environments to operate reliably and safely. You can do this in some places (e.g., airports, college campuses, maybe some sections of the freeway). But you can’t simply rebuild every road in America to be safe for autonomous cars. And why would you want to? There would be tremendous costs and no real benefit. So, yeah, we basically have the technology but putting it into the real world may never make any sense.
The worst example of this is nuclear energy. We literally have the means to end global warming, which I’m told by Netflix is metaphorically the same as ignoring an asteroid that’s about to smash into the earth. Nuclear has killed far fewer people (like, order of magnitude) than any other means of generating electricity (I suspect more people have fallen off of roofs installing solar panels than have been killed by nuclear). It’s been stymied from innovation by absurd regulation and unhinged people. So we literally know how to do something that will avoid the thing that people believe is an extinction event. And they still won’t get behind the technology.
This isn’t to say that we should be down on the future. We should be pushing for improved technology to improve our lives so much more than we are today. But it is to say that there are a lot of reasons you can’t just see something that’s interesting in some paper and think it will change the world. Or, rather, that the “changing the world” part is a lot more of the challenge than the figuring out the “science” part.
“works in each of the 80 million iPhones I build every quarter for 3-4 years when dropped on cement every month or two”
As we all know, dropping your iPhone on a bathroom mat, let alone cement, will muck it up beyond repair so yeah, I agree that the Holy Grail of making an iPhone tough enough to function is an impossible dream
With one exception, yes. People who like to predict technology rollouts should read some industrial history. Vaclav Smil has written many books that are useful as entry points into the literature.
The exception is solar PV. There have been big price drops for two decades, yes. We're now getting fairly close to the cost of the raw inputs in panel and system bills of material, so extrapolating the rate of decline is likely to lead you astray.
People who are excited about autonomous vehicles should ask themselves why. In nearly all cases the excitement is about reducing costs (stress; opportunity to do other things; even, bizarrely, parking fees) and increasing demand (disabled or nervous or very young potential road users). They should then ask themselves what lowering the cost of using roads and increasing the population of road users will do to traffic congestion, trip durations, and environmental harm.
Some of the excitement about autonomous vehicles is that the AI won't drink and drive - humans are imperfect drivers, and while we're still better drivers than the AI for now, at some point that won't be true and the switch to autonomous driving will save lives.
Nothing unique to explain with Mormons. They follow the same trends as the general population, just at a higher level. Fertility rates in a lot of developed countries, including US, have been collapsing in the past 10-15 years. In 2007, US fertility rate was 2.12. By this spring, CDC estimated it was just 1.64. Haven't done much research into matter, but seen a decent number of demographers suggest that it is a consequence of the modern electronic age both generally and for Mormons in particular.
20 "The Oxford Chemistry and Biology postdocs I met during bus rides to the science park (from my time at Oxford Nanopore) earned £35k at AstraZeneca 3. That's half of what someone slightly competent earns after four months of youtubing Javascript tutorials 🤷♂️."
That's nonsense if coders were getting paid that much we would never be able to hire anyone. Should we mistrust the rest of the article similarly?
I think this might be a location thing - I can genuinely believe coders in Silicon Valley earn that much straight out of boot camp, but I doubt that coders in England make that much at entry level. Nevertheless, postdocs are badly underpaid compared to the amount of training required, and as a result doing a PhD is a poor financial decision.
While I agree with this, as well as the video featured,
"Glad to see the “we should try to stop global warming for altruistic reasons, but it’s not going to destroy humanity or kill your family” perspective picking up more traction,"
I was disappointed that she then refers to "subsistence farmers and people in the third world" as if we'll still have "subsistence farmers and people in the third world" 80 years from now.
Once the Startup Cities/Charter Cities movement gets going, I don't expect there to be any subsistence farmers nor will there be any "third world" after a few decades. The very existence of mass poverty is due to bad law and governance. Most of the world's population is still stuck in regimes that make it unnecessarily difficult to create entrepreneurial value.
Rather than cite the literature on economic freedom, it might be helpful to give a concrete example. Last year Africa had its best year ever for venture capital,
"The continent’s startups raised over $4 billion this year and minted five unicorns."
Much of this was in Nigeria. What was the Nigerian government response?
"The Nigerian government is waging war on its technology industry. Within the last 12 months, President Muhammadu Buhari’s administration – through different ministries and regulatory bodies – has enacted a series of bans and operational restrictions in the country’s vibrant tech ecosystem.
Most recently, the Central Bank froze the accounts of four fintech platforms this August, claiming they were operating without a license and trading unethically sourced foreign currencies. The same month, the Nigerian Information and Technology Development Agency (NITDA) proposed introducing tax levies and licensing fees for tech companies as well as prison sentences for those who default on these payments. . . .
As economist Tunde Leye puts it: “The main thing we must understand [is] that the Nigerian elite have a consensus from the North to the East to [the] West to the Delta, that the formation of wealth independent of the political patronage system is a threat that must be exterminated.”
So called "developing nations" often have similar dynamics; Nobel laureate Douglass North, along with Wallis and Weingast, famously called this "the natural state,"
Because it is the natural condition in most societies for the oligarchs to prevent entrepreneurial capitalists from creating prosperity.
The more jurisdictions we can create where people are as free to create entrepreneurial value as they are in Singapore, Denmark, or New Zealand, the faster mass poverty will become a historical footnote. As Muhammad Yunus notes, at some point poverty will be only seen in museums, a peculiar feature of ancient human history. Children in 2100 will regard mass poverty as unfamiliar as children today regard a telephone booth or a slide rule, something that happened sometime before they were born. Their grandparents will talk about how hard life was in the old days, and thus provide "poverty within memory" for the grandchildren of today's global poor.
To shift more specifically to the climate debate: Especially when people talk about stopping fossil fuel use in Africa, these judgments should be based on empirical analyses on the relative benefits of varying rates of economic growth, on the one hand, vs. various climate impacts, on the other. There are undoubtedly scenarios in which it is more beneficial for Africans to continue to have access to affordable, reliable energy (usually fossil fuels) than to contribute marginally to lower temperatures 80 years from now. I've never seen a rigorous empirical analysis comparing such scenarios, but I wish those activists and NGOs working to end fossil fuel consumption in Africa would first undertake such scenario planning so they had an empirically informed understanding of the real human tradeoffs involved.
The "existential risk" of a population at an average GDP per capita of $80K each is very different from the "existential risk" of a population at $2K each. Climate impacts for populations at $2k would indeed be devastating. At $80K each, not so much. In 60 years, Singapore went from less than $500 to roughly $60K,
Personally, I see getting average GDP per capita up to the $80K range to be the highest moral priority. As Fred Turner notes, the goal is to "Make Everybody Rich,"
Your assumption that the Charter Cities movement will actually get up off the ground and will work as idealized are the key axioms you're overlooking here.
Special Economic Zones have exploded in recent decades with respect to sheer number, size, and diversity of models. Economists have almost completely ignored them until the last 10-15 years. Private firms with expertise in developing and evaluating zones has also expanded dramatically in recent years. There are roughly two hundred jurisdictions around the world. Many are experimenting with zone-based strategies for increasing their attractiveness to talent and capital, with job creation an urgent need for political leaders in many nations.
Charter city visionaries provide a useful set of conversations about how new zones might achieve ever greater goals. But "charter cities" are simply a category of possible zones. The scale and diversity of zone growth is a fact, not an axiom.
According to Statista the number of zones has grown globally from 79 in 1975 to 5,400 in 2018,
Many of these are banal, many are merely support for government cronies. But many nations are experimenting with new zone forms, including zones with distinctive law and governance. Some percentage of them will grow successfully.
The Adrianople Group, a private consulting firm serving the industry, just published the first map of zones around the world a few months ago,
Before that we didn't even have a map showing what was going on. Neither journalists nor academics have much sense of the zone activity currently taking place around the world.
I've been involved with this movement for almost 20 years, and know leaders who were involved since the 1970s. The transformations that have been and are taking place are stunning. Given the growth trajectory, will we have 10K, 20K, or 30K zones in another decade? My bet would be closer to 30K and maybe more. Of those it could easily be 1% that use distinctive legal frameworks to attract capital and talent, probably more.
There is evidence for growth in the number of zones around the world. There is evidence that when zones are successful in one nation, there are attempts to copy such successful zone strategies in other nations. The safest prediction is that we will continue to see more zones, more diverse zone strategies, and that successful zone strategies will be replicated. We are thus already seeing, in Patri Friedman's terminology, the beginnings of a "Cambrian Explosion in Government."
Re: 15, when you look at the paper, it looks like they are talking about low rates of recidivism as compared to *other types of convicts who get released*, rather than compared to the general public.
IE,Someone released for murder is less likely to commit a violent crime than a random criminal convicted of a nonviolent offense. This is relevant if you are talking about relative sentencing guidelines between different types of convicts, and for pointing out false dichotomies between violent and non-violent offenders. In this sense the report is saying something useful, but of more niche applicability.
The tweet doesn't convey this nuance, and seems like an attempt to lie for that reason. But that shouldn't be held against the report itself.
That would explain the difference I think. Felonies like robbery or rape or drug dealing are generally committed by people who have made a career or lifestyle decision around committing that sort of felony on a regular basis, knowing the risks and considering the rewards adequate compensation. A significant fraction of murders, and possibly a majority, are committed by people who have one specific person they really wanted to kill (e.g. the man who they caught sleeping with their wife), and now that person is dead and lightning *usually* doesn't strike twice. There are also a fair number of criminal-lifestyle murders of e.g. rival drug dealers, but including the one-and-done murders in the average would make recidivism less likely for murderers.
I've seen this credibly asserted in other places, but I'm not going to dig up references just now.
Re: 16 are we supposed to be impressed by eignenrobot here? He seems to be doing a really terrible job of the standard replyguy thing: make fun of each individual sentence out of context, never acknowledge that the criticisms of sentence 1 gets explained in sentence 5, when you do get to sentence 5 which disproves all your previous sneers, just sneer at it without making an argument and then move on as if it never happened.
Basically: yes, the fundamental issue is what your frame of reference for bias is. The algorithm is not biased relative to crime reporting and arrest rates, the journalist is claiming that crime reporting and arrest rates are themselves biased and provides several references to believe this, this is the heart of the question and eigenrobot just completely fails to address it or come to grips with it in any meaningful way.
The journalist is wrong though. Favorite example: I live in NYC. If I tell you that in 2019 the percentage of black-or-brown people in NYC is around 54%, what fraction of shooting *victims* that year would you guess were black-or-brown? Answer: 96%. And this wasn't unusual - some years it had been as high as 98%.
If we know that virtually all the people who are *getting shot* in large cities are black-or-brown and the police in those cities care about *reducing shootings* they should be spending most of their time in the neighborhoods where black-or-brown people live. That's where the victims are, that's where the perps are, that's where the shootings are happening. So that's where cops need to be.
"Biased cops" could perhaps explain some discrepancies in *arrest* rates but isn't going to explain that magnitude of discrepancy in *shooting-victim* rates unless large numbers of white people who get shot aren't reporting it to the police or somebody is consistently mischaracterizing white shooting victims as not-white. Either of those strike you as plausible? Is the idea that when victims go to the hospital or the morgue due to bullet wounds the police don't get notified just in the case of white victims out of, what, politeness? Or due to police not wanting to catch white criminals?
No, the obvious answer is the correct one: more crime happens in places where some groups tend to live than places where other groups tend to live.
UPDATE: Besides, if we look at their referenced survey the rate of non-reporting of violent crimes just isn't that different by race. It's a *little* different, but not much. In Table 5; the percent of violent victimizations not reported for "white" is 54%, "black" is 46%, "hispanic/latino" is 51%. So the crimes we're not seeing amounts to "about half of them, give or take" for the three groups we care most about. (survey is here: https://bjs.ojp.gov/content/pub/pdf/vnrp0610.pdf )
Re: 21 I'd 'get excited about' a technology that has a 10% chance of revolutionizing something I care about in the next 10 years, so I don't think 'get excited about this list of things' and 'only 10% of them are likely to happen' are in much conflict.
Re: 24 - adding my datum, same for me, couple hours of real work a week in a very high-paid job and never caught up to me.
Of course, there were some weeks with legitimately 50+ hours of real work, including weeks-long travel to factories and things, so I think a peak capacity model would probably explain a lot of why jobs are structured this way. You can't just hire 50 programmers for 2 weeks when you hit crunch, they need to already be onboarded and part of the corporate mechanism to be effective.
But it would be nice if they were only required to be in the office 20 hours a week on the weeks they only have 5 hours of work, with an understanding that they'll show up for the 50+ weeks.
Would they show up for the 50+ how weeks though? I'm an algorithm person, which isn't quite a software dev in that I usually don't have harsh deadlines, so maybe my intuitions aren't trustworthy here -- but I should think that if I structured my life around typically having 20-hour work weeks, I would have a hard time suddenly switching to putting in 50+ hours a week, through a mix of not having adequate childcare and not having adequate endurance for sitting at the office in front of a computer.
During my 3 hour flight home after xmas, I had 5 ideas that seem to me like good ideas, but maybe only 2-3 of them are actually good. This seems like a good thread for going off on disparate tangents, so here they are:
Idea 1: Biological carbon capture
Producing biomass is very cheap (90 tons +/- 30% CO2 per year per $4000 acre of farmland with minimal work using Paulownia trees or bamboo, 90% confidence). Preventing the biomass carbon from getting back into the atmosphere eventually is the hard part. This is a refinement of my prior ideas.
Step 1: Grow biomass
Step 2: Pulverize it
Step 3: Mix with water and alkaline clay (literally dirt cheap, 99% confidence)
Step 5: Almost all the resultant CO2 is trapped in solution as carbonate/bicarbonate ions due to alkalinity (99% confidence). The outgassing will be nearly pure methane (higher purity than fossil gas, 75% confidence)
Idea 2: Lyft and Uber should do just-in-time matching of drivers with passengers by physically bumping phones together with NFC. Then Ubers could pick up the same efficient way taxis pick up at the Las Vegas airport, and save everybody 10 minutes of waiting for some specific driver to come on-demand to a place that everyone already knew years in advance would have lots of passengers waiting.
Idea 3: Doing explicit Bayesian calculations in your head when you lack a good social intuition (using intuition as a synonym for what Daniel Kahneman calls "System 1"). Prior that a random girl is into me: 1:100. She starts fiddling with her hair the instant I look at her, multiply by 4:1. She offers me some gum unsolicited, multiply by 4:1. She keeps leaning over me to take pictures out the plane window, multiply by 8:1. She's with a guy but there are no PDAs, might just be a brother or platonic friend. Multiply by 1:2. She bumps my foot with hers in a way that was obviously intentional. Multiply by 4:1. Correct my numbers if they're wrong. I multiply it all out and get 64:25 Due to feminism/metoo I need pretty high confidence before making any moves (What % would be appropriate)? I should probably be reciprocating with a similar series of ambiguous gestures like she did, to gradually ratchet up our probability estimates about the other being into us, but I didn't really say or do anything. I would really appreciate a list of numbers I can use for these calculations.
Idea 4: Integrating a solar panel with a battery, a converter, and an outlet, and trying to get the entire package down below $200 like OLPC. To electrify off-the-grid places. Tesla's battery costs are around $150/kwh now, and excluding heating/cooling/cooking, 1kwh could be adequate for one household for one night. 300 watts of panel capacity could charge that in one day and have plenty left over for daytime use. Current cost of 300 watts of panel capacity is <$100. If you can get broadcast TV or internet access to remote parts of the third world (30" LCD TV = 60 watts) their ignorance will decrease, their welfare will improve, and their fertility will go down. Reducing fertility in the third world can reduce problems of hunger, longer term carbon emissions, dysgenics, and political instability.
Idea 5: I noticed there is such a thing as a "social justice quilt academy". I really don't know how to model this phenomenon of social justice people entering every hobby and making it about social justice instead of what it was originally about, except by analogy to old-timey religions that took every opportunity to proselytize. If SJW views of the causes of group performance gaps were correct, then there would be a massive utility payoff to proselytizing as much as possible. Likewise global warming. The ideas that spread will tend to be the ideas that can convince people there is a huge utility payoff for proselytizing as much as possible. The payoff need not be real. It only needs to be as convincing as religious ideas of afterlife, which is a very low bar.
For idea 1, the three questions are: who pays? If not the government, will they allow you to do it? And finally, is it better than other sequestration ideas.
For idea 2, it’s so obvious that I have to assume that Uber and Lyft executives have already discussed it and rejected it for some reason. I don’t know what those reasons are but two possibilities come to mind.
Maybe they think it’s too confusing, and want to keep the app as simple as possible - you call a ride, you get a ride.
The second is game theoretic - it allows riders to ‘steal’ rides. If you do choose to call an uber, the uber driver could cancel your ride and give one to whoever they run into. And it’s not as easy as “just make sure you’re at the pickup point in time”, because the driver could encounter another potential passenger before even getting there. This makes the service overall less reliable.
For 1: my back of the envelope math suggests it MAY be very profitable at current methane prices, but there's a lot of engineering to do and local circumstances that need to be nailed down to reduce the error bars around cost estimates. Methane produced could be up to 1/6 of the mass of CO2 removed from the atmosphere, which at current methane prices of $1.35/kg would be around $20,000/acre/year.
If the engineering gets good enough that converting wood to methane is no more difficult than converting grapes to wine, then here's a little sanity check on the pricing. One acre of grapevines produces 4 tons per year. Average grapes are worth $900/ton. (Napa/Sonoma and other prestigious regions can cost 2-10x that) So 4 tons of grapes is worth $3660. This makes about 2880 bottles of wine. If they're two buck chuck, that's $5760/acre/year. That includes some retail packaging which costs at least 25 cents per bottle. 5760-3660-2880*0.25 = $1380 estimated cost of industrially converting 4 tons of grapes to wine.
With bamboo or paulownia trees, you're dealing with a much larger biomass per acre per year, but the biomass comes in larger pieces that probably make them easier to process, and you can play fast and loose because you're not making a product for human consumption. Worst case we multiply the $1380 by (90tons/4tons) and get $31050, so that the methane sales would cover two thirds of the costs even before you start selling carbon offsets. (1/6)*(44/16) =45% of the carbon is released as methane, and the rest stays in the ground. You sell carbon offsets at $20/ton for the other 55%. 90 tons * 0.55 * 20 = $1000/year extra. So basically to break even without subsidies, industrially fermenting wood into methane needs to be 33% easier than fermenting grapes into two buck chuck.
LOTS of burdensome details here, so low confidence.
For 2: If the driver encounters another potential passenger before getting there, that's a good thing. It saves time in the aggregate. If the distance to the old passenger is X minutes, and the average pickup time is Y minutes, then the new passenger saves Y minutes, the driver saves X minutes, and the old passenger loses Y-X minutes.
Y+X > Y-X
So reserving particular drivers for particular passengers in advance will worsen aggregate time wasted. Maybe there should be another tier of service for people who are extremely sensitive to variance (catching flights, etc). But reserving-by-default seems like a huge waste of resources. The driver app gives drivers these "streaks" that function kind of like daily quests in world of warcraft. They can give drivers a quest to go to the airport, without assigning them a particular passenger yet, to precisely calibrate the number of drivers that are there in advance of pickup requests and make airport pickups a lot faster. They can do this with or without the NFC thing. Instead of being 50x slower than taxi pickups at airports, maybe uber can get it down to 5x if it stops unnecessary reservations.
I mostly don't believe in appeals to "if it was a good idea they already would have thought of it". Ever watch James Burke's Connections series? A lot of the history of science is basically: thousands of people were aware of X and Y, but nobody thought of putting them together properly, until someone did something that was totally obvious in retrospect.
I need to hurry up and finish R:A-Z (13% and 65 hours remaining on my Kindle. Sorry, I'm a slow reader, so I've developed skills of getting as much mileage as I can out of as little evidence as possible) so that I can start Inadequate Equilibria which addresses efficient markets.
I just had another idea in the same area as #1 that's 10x better. Just genetically engineer Paulownia trees to dump carbs >1ft underground (by growing unnecessarily large root systems or directly excreting carbs). Then your only marginal cost is planting them, which costs $100/acre. Those trees regrow from the root system so that they never need to be replanted. So one planting could potentially remove 30tons/acre/year of CO2 for 333 years, which comes to ~$0.01/ton for the planting and $0.2/ton to buy the acres of crappy alkali soil. The soil quality will improve due to de-alkalization and introduction of organic carbon, so the resale value of the land will be a lot higher than if it were idle. Mortgage interest rates are lower the usual appreciation in the price of land, too, so maybe it's not appropriate to count any part of the land as a cost.
Tree-parasites could ruin this plan. Monocultures are vulnerable to parasites. But the carbon that ends >1ft underground in alkalai soil is probably pretty safe from being released into the atmosphere by a plague of parasites.
On that number 24, man I've felt horribly guilty for being that guy in my career. I don't know how to feel about the idea that maybe it's fairly common.
I worked with a software engineer who could fall asleep with his fingers on the home row. Not sure how he did that but I am pretty sure he suffered from some form of narcolepsy. He had a wonderful personality though. We were partners in the company golf league . Had some great times hacking up the course together.
It's not unusual to happen to get no work done if you're working on a hard problem and everything fails. It's another because you faffing all day. I've had both happen to me. What time is it?
26. I am a member of The Church of Jesus Christ of Latter-day Saints. (We prefer to be called by the full name of the church rather than "Mormons", but I'll give you a pass this time.) I have 7 children and my wife is expecting our 8th in May, and we feel strongly about this. We love having a big family though it is countercultural in this day and age. Other commenters have pointed out that Utah is growing and the percentage of Church members is dropping, suggesting that the data doesn't actually reflect birthrates among members of the Church specifically. However, I don't think that fits the data. An outsized percent of the growth state-wide has been in Utah County, so this theory would predict that birthrates in Utah County would be falling faster than in other Utah counties - but that is not the case, at least for the counties with data shown at the link. Instead, birthrates in all of the available Utah counties are falling at a similar rate. It certainly is true that societal trends that impact everyone impact members of my church too, and pull the birthrate numbers down. But I think that trends in church teachings (softening the "multiply and replenish the earth" rhetoric) are also accelerating the trend toward lower birthrate. These changes in teaching approach are likely in response to, and out of sensitivity to, increasing numbers of unmarried and childless church members. But the softened rhetoric of course also amplifies the cycle. You can dig up old quotes from the Prophets in the 1950s and 1960s that are very straightforward about the need for and benefits of having children. Those old quotes now seem like they came from a different world. For decades now we didn't hear very much about this doctrine at all. Just in the last year or so some of the Apostles have felt a need to speak up (probably in response to the tanking birthrates we are now discussing), but that is still tentative and a marked change in approach. Even on Mothers Day at the local level, the speakers in our ward spend less time thanking mothers than they spend making sure everyone knows that those who don't have children are also ok - which is true, of course, but illustrates the point. To step back a little, and of course I am biased, but I can think of no other action that increases the overall utility of the world than bringing children into it in the first place. Existence, even in the third world, is a utility net-positive. Existing in the world vs not existing is a giant step change in utility, obviously. In our society we underestimate the value of potential existence, which almost never gets factored into our policy math. Any policy that lowers birthrates (e.g. car seat regs, or anything that hurts prosperity, or makes it harder for single-income families) has a huge unaccounted-for negative impact in overall utility via potential lives not realized as the birthrates fall as an unintended (and un-thought-about) consequence. Should be an opening for EA, but I doubt it would mood affiliate.
Well, Ecclesiastes is a holy book in all the three large Middle-East religions....and the general idea is not too different from that of Buddha either.
Besides, he writes were well. (Or all four of them. There is probably more than one author behind the text in Ecclesiastes.)
1. One line from Ecc doesn't overturn the entire rest of the texts of Judaism, Christianity, and Islam. Even Paul's exaltation of celibacy is a better root for any anti-natalist argument in Christianity.
2. Ecc and Buddhism have very different teachings. You're clinging to one vague semblance and trying to connect them. Ecc argues for a lifestyle of simplistic and passive hedonism, which is pretty close to the exact opposite of Buddhism. It is good that people are born into this world so long as Buddhism exists, because if they were not born into this world they would not be able to attain Enlightenment.
The arc of Ecclesiasties is realizing the weight of existential dread, realizing that one should be content with one's small place in existence, and in the end find peace in "fearing God."
The passage you quote is one of many from the first part of the arc, "feeling the weight of existential dread."
But you don't quote anything from part 2 or 3 of the arc. Part 2 might include something like Ecc 9:4-10,
> 4 There is hope, however, for anyone who is among the living; for even a live dog is better than a dead lion. 5For the living know that they will die, but the dead know nothing. They have no further reward, because the memory of them is forgotten. 6Their love, their hate, and their envy have already vanished, and they will never again have a share in all that is done under the sun.
> 7Go, eat your bread with joy, and drink your wine with a cheerful heart, for God has already approved your works: 8Let your garments always be white, and never spare the oil for your head.
...and of course the whole book concludes with,
> "13When all has been heard, the conclusion of the matter is this: Fear God and keep His commandments, because this is the whole duty of man."
There is nothing like a bit of theological discussion early in the morning, or late at night (depending on your time zone).
In you interpret Ecclesiastes as a narrative arch, as you do, then the final part naturally takes precedence. Since the end of a text is where an author sums things up, or extracts a final moral. In this care: Fear God and do his bidding! (12:13.)
However, the fact that the general “feel” in Ecclesiastes changes somewhat toward the end, is one of the reasons why many suspect that the text is written by different authors. Up to four authors have been suggested.
If you chose this interpretation, the first part(s) becomes as important as the last. It is just the voice of a different author.
I have always suspected that the final sentences (12:13-14) were added by a later author who recognized the beauty of the prose of the preceding authors, and added this “theologically correct” conclusion to sneak this subversive text past the eyes of those Keepers of Holy Texts who decided what to include in the Good Book.
That said, I would not claim that the “separate author” interpretation has a much stronger empirical basis than your “narrative arch” interpretation. Since the author(s) are long dead, we will unfortunately never know.
Be that as it may, any text can in any case always be interpreted in a multitude of ways. In particular a text as rich and magnificent as Ecclesiastes.
And I do not mean to belittle the late author. After all, he was the one who penned this sentence:
“There is no end to all the book writing, and too much study wearies the body.” (12:12)
I have often felt to weep with the author of Ecc 4:2-3 when listening to the stories of people who have suffered unspeakable trauma and oppression, so I understand the sentiment. But even in these cases, there is hope, forgiveness, healing, and growth possible for each and every one (through Christ, in my belief system). The Sermon on the Mount is a better reference for the downtrodden.
On the other hand, Christ does have this to say about those who abuse children: (Matthew 18:6) “But whoso shall offend one of these little ones which believe in me, it were better for him that a millstone were hanged around his neck, and that he were drowned in the depth of the sea.”
Re 2.: https://en.wikipedia.org/wiki/Brazilian_real says the Real, which replaced the URV, was worth exactly one U.S. dollar at the time it was introduced, so it seems the URV was tied to the dollar before it was converted to the Real.
The article on NRP is certainly interesting, and it explains the trick behind the URV well, but omitting the USD binding seems to give the impression that the trick's basis was purely psychological. However, binding one's currency to a stable foreign currency seems to be a pretty basic economic trick - which, of course, has a psychological side too, because the reason the inflation does not come back after it is not anymore tied to the dollar, seems to be the expectation of the people using the currency.
#30: The interviewee has an MA in political science and is a policy analyst in an organisation with a track record of minimising environmmental and social damage from resource exploitation). She does not have a track record in energy, climate, environmental, or social science.
My prior belief is that it will be full of BS, strawman opponents, misinterpretations and innumeracy.
... And my post is that the interview is full of strawmen by caricature ("children are little carbon emitting machines"), strawmanning by extremising ("stop reproducing"), and misrepresentations (a focus on birth rate, when the problem lies almost entirely with the 250 million highest-income people, who already have below-replacement birth rates; massive overclaiming, and innumeracy and ignorance of scaling timetables when it comes to technological matters), and distractions from the issue (e.g. talking about medical progress in the mid-20th century).
There are also outright falsehoods being brought out: the idea that we need technological breakthroughs to "solve" climate change (and therefore need more people to have those ideas). We have all the needed technology now. Ideas, from now on, have negative value. Climate change *is* solved...in theory. All that's left is the politics.
She *is* right that our collective non-climate-affecting depradations are pretty bad too, and those chickens will also come home to roost this century. That doesn't help her argument.
> But here’s a story about someone selecting guppies for intelligence (successfully) and finding that they had smaller guts and lower fertility.
Aren't humans descended from animals that had larger guts (due to less meat in their diet, and not cooking it)? Also, while no-one knows how fertile extinct hominins were, the probability for pregnancy per copulation is probably lower from humans than most mammals.
Humans invented cooking 250,000 years ago (and it probably didn't take too long for their guts to shrink after that). But they didn't get smart enough to invent a new tool more than once a millennium until the mesolithic ca 20k years ago. So if someone is suggesting that humans got small guts *because* they got big brains, the timeline doesn't support that.
"Glad to see the “we should try to stop global warming for altruistic reasons, but it’s not going to destroy humanity or kill your family” perspective picking up more traction"
For the longest time, this was the *default* position among progressives and environmentalists. In Al Gore's documentary "An Inconvenient Truth," the big threat posed by climate change was an increase in the amount and severity of flooding, hurricanes, and other natural disasters due to the melting of the polar ice caps. In Kim Stanley Robinson's book "New York 2140" - which was praised by environmental activists as being a starkly realistic vision of the devastation that climate change may cause - Manhattan is partly submerged and has become a poor and squalid place due to flooding and infrastructure damage, with Denver (a landlocked, high-altitude mountain city) taking its place as the economic capital of the nation. These are certainly grim predictions that present a rather bleak future, but they don't come anywhere close to the claim that humans will literally go extinct or suffer an apocalyptic catastrophe that kills billions and destroys modern civilization entirely.
But somehow, over just the past few years, the idea of climate change as an *existential threat* to humanity has become a mainstream idea among online progressives and leftists. I still don't think it's the majority opinion among left-leaning folks, but it's definitely a popular one. A recent poll I saw in a progressive Facebook group was about evenly split between "climate change is an existential threat that could cause human extinction" and "climate change is really bad, but it's not an extinction risk."
Of course, the extinction narrative has virtually no support from actual climate scientists. Only a vanishingly small minority of climatologists actually believe it will cause the literal extinction of humanity, and they tend to be viewed as crackpots by the rest of the scientific community, just like the denialists on the other end of the spectrum. Roger Hallam's claim that "climate change will kill 6 billion people by 2100" was extensively debunked by scientists, and the fringe theory of a "runaway greenhouse effect" that would turn Earth into an uninhabitable wasteland was dismissed by the IPCC as a literal impossibility. Yet that hasn't kept the extinction narrative from gaining traction among the activist crowd since the mid-2010s. Any ideas why this might be the case?
I think a lot of public discussion of things tends to flatting them into either "this will kill everyone" or "this is great for everyone" or "this is completely irrelevant". This is probably related to the "halo effect", where anything that you like for one reason tends to get assumed to be good for other reasons as well.
Per #26, I'm an active Mormon who lived in Utah for 17 years (10 of them in Utah County). I don't think the birthrate decline has much to do with secularism or a reduced percentage of Mormons in the state. It's much more simple than that: Utah is booming!! Utah County especially is growing like mad. What was once a collection of small towns throughout the Provo-SLC-Ogden corridor is not an unbroken sea of increasingly dense suburbs. Lehi is becoming a tech hub of its own (Silicon Slopes!). Utah feels more and more like California every year.
The result of this is that Utah is much less isolated than it used to be. There are far more people with far more connections to the outside world, and it has given the state a more metropolitan feel. The state is still dominated by Mormons. In Utah County especially, you can still assume that everyone you see at the grocery store either is or used to be an active Mormon. But with a more metropolitan culture, people just aren't having as many kids. My dad was one of eight kids, I was one of eight kids, but my wife and I only have three. That's a common story. There's a lot more to life now that there are things to do.
The older generations still feel very strongly that women should stay home with their kids (my wife and I get this talk from both our parents), but more and more women are working now. Managing a huge family is very difficult to do when you also have a job. Religious changes are mostly a red herring. It's just simple affluence.
#6 Pleasantly surprised that lots of people are supporting some obvious utility gains, and not getting derailed by the worst argument in the world (P1: nazi death camps killed people partly for eugenics. P2: nazi death camps were super evil. P3: eugenics means any organized attempt to improve the gene pool P4: therefore any organized attempt to improve the gene pool is super evil. To see how ridiculous this is, compare: P1: nazis invaded eastern europe partly for agriculture P2: the nazi invasion of eastern europe was super evil. P3: agriculture means any organized attempt to improve the food supply. P4: therefore any organized attempt to improve the food supply is super evil.)
#7 the article says they were selecting for bigger brains *relative* to body size. That implies they weren't just selecting for bigger brains -- they were selecting for bigger brains and smaller rest-of-body. In that context the result of smaller guts is less surprising and I should update much less towards intelligence genes being tradeoffs.
#8 seems very plausible but in its current form it's too vague to constrain anticipation. The south Korean government had everyone's cell phone GPS data in real time for test and trace purposes. They could use that data to characterize population structure and get better at modeling interventions. Models definitely don't need any more free variables that researchers can tweak to arrive at whatever conclusion they already wanted to arrive at.
#9 is probably a joke (90% confidence that the author doesn't actually believe that). But there's no parody disclaimer in the profile. It's probably unvirtuous to make an ambiguously-parodic twitter account that tweets straw men which people might mistake for the genuine article and update towards negative karma for various tribes that never espoused that straw man. (generalizing from fictional evidence, ethnic tension)
#15 The subset of murderers who ever get released is a very nonrandom sample of murderers, plus they've been in jail so long that they've aged out of their peak violent crime years. So low-ish recidivism of murderers doesn't say much about what the recidivism rates would be in the context of some radical anti-incarceration reform, and it's probably a mistake to use the former to argue for the latter.
#16 The NVCS shows that racial disparities in arrest rates correspond to disparities in actually-committing-crime, not police bias, so assuming neighborhood disproportionality = bias is not just an unwarranted assumption, it's actively undermined by the available evidence. Black and Latino neighborhoods also have a younger average age, and youth commit several times more crime than old folks. (having a not-yet-fully-developed brain is a bit like having a low IQ and correlates with many of the same things: impulsivity, aggression, criminality, low productivity)
#21 My most pessimistic take on this analogizes big tech to land speculators who got in early, and are now just collecting rent on their natural monopoly. But worse than actual land speculators, because they use all that money to hire the smartest people to figure out how to make people click on ads 1% more instead of actually innovating. So there's a brain drain from everything else into figuring out how to make people click ads 1% more. Related Neal Stephenson speech which is great: https://www.youtube.com/watch?v=TE0n_5qPmRM "I saw the best minds of my generation writing spam filters"
Noah blaming energy costs for the stagnation seems dubious. Total US energy spending is only 5.7% of GDP (https://css.umich.edu/factsheets/us-energy-system-factsheet). US total government spending (federal+state+local) was 45% of GDP in 2021. A new government program that wastes an extra 6% of GDP could have a bigger impact than doubling energy prices. It's plausible that affirmative action or uncertainties about inflation or regulations could make the economy 5% less efficient too. But Noah singles out energy as the scapegoat for declining innovation in the physical realm, without considering any of the other possible causes.
Sanity check: If energy is really going to become a lot cheaper over the coming decade, CME oil futures for 2033 should be a lot cheaper than oil prices today, because people will substitute other energy sources for oil. But actually, the futures prices decline from $75/barrel to $60/barrel by 2025 (which is basically just reversion to the mean) and then approximately stay there until 2033. There's no longer term trend towards vastly cheaper energy visible in this futures curve. (CME also has futures for electricity prices, which show a similar pattern of quick reversion to the mean followed by stasis, but they're a lot less liquid than oil futures so I don't trust the accuracy of the quotes as much. Zero volume on a typical day. https://www.cmegroup.com/markets/energy/electricity/pjm-western-hub-peak-calendar-month-real-time-lmp.quotes.html)
Brain computer interfaces are still slow and lossy. They're inevitably lossy, because external electrodes can't perfectly read the state of the neurons. They'll probably not in my lifetime be as fast and accurate as a mouse, but they might enable something more compact which is useful for non-paraplegics (e.g., google glass style computers)
The vision implant is super cool but it sounds like it still has a very long way to go to get to Geordi on Star Trek TNG. They basically upgraded a blind person to a person who can just barely see large shapes.
The fine print about the pig-to-human kidney transplant was that it only ran for a couple of days, and it was external to the body. (h/t NSQ or freakonomics podcasts) They're still nowhere close to what people usually mean by the words "organ transplant".
This Comment Is Too Damn Long so I'm going to write a separate one if I have any thoughts on 22-34
#22 Nomad capitalists have observed that as soon as a country gets developed enough for its passports to have visa-free access to the US, it usually stops having a very favorable tax and regulatory regime. The intersection of countries with visa free access to the US (https://en.wikipedia.org/wiki/Visa_Waiver_Program) and tax havens (https://howmuch.net/articles/world-map-of-personal-income-tax-havens) is empty except for Poland, which has low-ish but not impressively low taxes.
#23 When I worked for a startup in LA, I worked a lot harder than my usual laziness. I put in about 35 hours of high quality effort every week. (because I actually liked the job, I had a substantial equity stake, my boss was great at motivating me, and I was overcompensating after getting fired from a previous job for extreme laziness and insubordination) A year after I quit that job my boss told me that he had to get 3 people to do what I used to do. There are huge differences in productivity between programmers, so it's not surprising that some programmer can get away with working 5-10 hours a week.
#27 the bulldozer-vetocratic axis is really cool.
#33 Sam Altman was in my accelerated math and science classes for three years in middle school. I went over to his house once to make a 1m parabolic reflector out of mylar in 8th grade. He didn't seem like the smartest in the class, but he was very polite, scrupulous, and hard working. I had no idea he was gay until I read about it on Wikipedia decades later. He did seem annoyed that I was talking about girls while I was at his house. Those were the days when I was widely bullied for being gay, and wasn't actually gay, but people might have thought so. And a couple years later I got a 9.6 on hotornot, so maybe he was into me and I was oblivious as usual.
The reality is that even if they had approved the vaccine sooner, it would not have saved many, if any lives. They started ramping production even before they got approval, and we were short for many months on vaccines. We would have at best saved a tiny number of lives - and probably not that.
Moreover, they were concerned about rushing through a vaccine that either didn't work or which caused damage to people, as it would increase vaccine hesitancy - which, as we know, has killed far more people than any delay would have.
So the argument that it killed lots of people is false.
This seems to be a common cognitive glitch amongst a certain brand of libertarian - you constantly and consistently disregard this stuff.
To be fair to the people at Toronto Life, if you wore a bejeweled SSC pendant around your neck and had a STAFF, the NYT would NOT have been acting inappropriately.
It's literally their jobs, and they have corporations involved. This ain't the same thing.
This isn't an open thread.
I think it was Razib Khan who noted that the latter form of eugenics is extremely common.
Really?
I have literally never heard of anybody outside of arranged-marriage cultures using a family history of longevity as as a factor in choosing a spouse.
I meant generally choosing a person with desirable characteristics & from a good family.
In theory there's no reason they can't choose it at random.
And the fact they presumably don't even attempt to choose it at random betrays a belief in the heritability of behavior and physiology that is incompatible with the modern liberal worldview.
Does the fact that so many couples choose to adopt children essentially "at random" tell against this?
Uhh, do they actually have a choice? Can they say they only want kids with successful healthy parents?
My understanding was maybe they can specify gender and/or race, but otherwise it's so hard to find children to adopt *of a very young age* that they have no choice but to take what they're offered if they want to adopt.
The demand for babies is very high so prospective adoptive parents will accept a lot.
But they still won't adopt the kid who will need a wheelchair and feeding tube for the next 12 years of his life (when he dies).
This appears to be an attempt to smuggle in the assumption that there are, in fact, no "fair" differences in the outcomes by race, and any predictor which predicts racial disparity is biased. In fact, they specifically say that the predictive model and the protected characteristics will be independent. That leads to some very unfair predictions when the protected characteristics and the outcomes are not independent. For instance, if we're predicting defaults on a loan, and we use just credit score, we'll predict that whites default on loans less than blacks. If we produce a "counterfactual fairness" predictor based on race and credit score, then we won't predict that -- but the predictor will be blatantly unfair, since it will either overpredict white default rates, underpredict black default rates, or both.
The "counterfactual fairness" paper mentions that impossibility theorem (pp 5-6), and claims to sidestep it by being fair in neither. Which it isn't; it's straight-up equality of result, with a lot of mathematical window-dressing and some technical use of the word "causal" to make it less obvious. With their version of "causal", if black people tend to drive red cars more than white people, then blackness is (according to this paper) a cause of driving red cars.
Their notion of "counterfactual fairness" is not some general notion of counterfactual fairness, but specifically the notion that changing the protected characteristic while holding things not "causally dependent" (using causally in that technical sense) constant should not change the outcome. This is a strong version of equality of result -- stronger than demographic parity.
They give an example of a car insurance company using red cars as a proxy for aggressive driving being unfair because (hypothetically) people of a certain race prefer red cars. They describe the causal structure here as A -> X <- U -> Y, where Y is accident rate, A is race, X is red car, and U is aggressive driving.
Why would you want to rebalance average levels of police focus away from high crime neighborhoods and into low crime neighborhoods and thereby cause a lot more people to be victims of crime? Should we have lots of fire hydrants around buildings that can't catch fire, even if we have to remove them from places near old wooden buildings?
How many is "many"? https://reason.com/2020/08/06/81-percent-of-black-americans-want-the-same-level-or-more-of-police-presence-gallup/
That poll suggests 14%, so as many people as the population of 7 average-sized US states.
Is that an attempt to make 14% seem like a large percentage? Are you saying the preferences of the population of 7 average sized states should outweigh the preferences of the population of 42 average sized states?
No, it's an attempt to counter the implication in posting that poll that 'many' can only mean a statement of proportion and not one of number.
That is, "many" is not the same as "most" and should not be expected to be.
What percentage of that 14% are criminals or the families of criminals?
How about you use actual data instead of anecdotes? I can find people who say they're scared of the crime in their high crime neighborhoods and want the police to protect them.
"even well-intentioned policing yields unfair outcomes along racial lines. I won't try to paraphrase"
No, you had better try and paraphrase it or don't bring it up at all. Making a strong claim and then then directing anyone who disagrees with you to read an entire book is totally unreasonable.
As it presently stands, arrest rates and crime victimisation survey offending rates line up extremely closely for race. Any changes to policing you support necessarily have to simply involve black criminals getting away with crime more often than they currently do. Which strikes me as an extremely marginal definition of "fair".
Alexander starts from false assumptions that the rate of criminal activity does not vary between races, and in the case of drug use, she uses a study based on *self-reported rates* of drug use. Studies that have used objective measures - such as blood or urine testing - show different rates by race, sex, and class/SES.
There is actually sparse data showing that per mile driven, traffic violations do not vary by race, and actually circumstantial evidence (accident rates, vehicle injury rates) that they do.
To be clear - none of this is indicative that any particular individual is using drugs or driving recklessly, nor that any particular cop is operating under an unwarranted negative bias. Still the effect of drug dealing and reckless driving are felt primarily on the community where those individual harms occur, piling up on the other, non-drug using, safe driving (and pedestrian) population.
In her (impressively dense and footnoted) work, Alexander focuses on the alleged unfairness to people (mostly able bodied young men) accused of those crimes and ignores the plight of those who are its victims - where children, women and the elderly are over-represented.
In a better world, those able-bodied young men would - as a group, and under the direction of law-upholding elders - be acting as protectors for the more vulnerable of the community. Unfortunately, this role has been outsourced to the police & to people outside the community, who are hampered by lack of knowledge, lack of community connection, and lack of community cooperation.
"do predictive policing but actively rebalance until average levels of police focus match across racial categories."
Okay, you're just saying that policing should not be about crime reduction if this is the line you want to take. You're claiming that differences in outcomes are necessarily "unfair", but these differences are objectively caused largely by behavioral differences.
You're ASSUMING that behavior and demographic group are independent variables, but they're most certainly not. If people belonged to a different demographic group, either their behavior would be different, or that demographic group would now have a different e.g. crime rate.
Differences in crime victimization are also a difference in "outcome". Equalizing inputs (police presence) means accepting unequal outcomes in such a case.
If these people were thinking along those lines, then these comments would never have been made in the first place.
It's a simple observation but the "disparity" in how we police men vs women illustrates your point.
There are some different geographic concentrations of the sexes, but not quite on the level of ethnic neighborhoods.
Related to 22, I am increasingly obsessed with Peter Turchin's concept of elite overproduction (which you've written about) and find that it explains so much of modern politics, media, and academia, and doesn't bode well for the future.
Which thinker should we credit instead?
I also find it plausible, though I've never been able to find a place where he explains it in enough detail for me.
It seems like there need to be two definitions of "elite", one defining how many people think they "ought" to be elite, the other defining how many people feel satisfied that they've made it.
As far as I can tell what we're seeing is something like "many people feel like they deserve a type of upper-class creative professional job, but aren't getting it". Are there fewer upper-class creative professional jobs than before? Do more people feel like they deserve those jobs than before? Why?
I don't remember seeing an answer to any of this in the two Turchin books I read. If I had to guess, it would involve a failure of college's reputation (place that guarantees you an upper-class creative professional job) to catch up to its reality (place that half of people go to). But this seems less like an inevitable world-historic force, and more about some particular policies and cultural norms catching up to us.
Oblomov is a favorite
You possibly mean Goncharev, the author of "Oblomov"?
No, he referenced books, and I mentioned Oblomov
I published a post about how it's really hard to make it as a writer, much worse than many young people believe, and that you should definitely have a day job and try to do something else if you can. I got two types of emails. The first was from people who had been professional or semi-professional writers for years and finally gave up because the attempt to climb the ladder was so bruising and frustrating. They were very grateful I had written it. The other was young writers trying to make it. They hated it and assured me that I only wrote it because I have not had sufficient success myself.
The rage of the latter - who, I'm sure, were all perfectly capable of getting nice corporate jobs at Nationwide Insurance or a nonprofit that works in the vague area of sustainability and having very enviable upper-middle class lives - I think there's something really powerful there. I detect such an immense frustration among youngish people about the sense that they have to occupy some sort of vaguely artistic or creative or intellectual occupation, that this is their birthright, and their continuing inability to secure a solid income by doing something that their peers recognize as enviable fills them with a profound sense of injustice.
Maybe it all comes to nothing, I don't know. But god, there's so much unfocused rage about that stuff.
"My grandfather was a warrior, so that my dad could be a merchant, so where the hell is my NEA Literature Fellowship?!"
I feel this. I was told I was "gifted" as a kid, and raised to get one of those elite jobs. Working for McDonalds was constantly used as a threat to me: "do your homework or you will end up working at McDonalds!" I attended one of the best, most classy public schools in the country. This climate constantly villainized common jobs. I remember my high school counselor giving a speech about how embarrassed she was that so many (only a small portion) of our students "only" ended up going to community colleges.
I was a fantastic student. But I got really fucked up when I became an adult and realized that I had been lied to all my life. Nobody in the real world gives a damn about academic performance. I had been so focused on academics that I actively shunned learning social skills, but it turns out that is exactly what you need to get an elite job.
I really like working with my hands and building stuff. Intellectually I know I would be really happy in a job like machinist, carpenter, or electrician, and in fact I have picked up a decent amount of experience in all three of those fields over the last decade doing informal work for family. But I can't actually get over the mental block required to actually apply for any kind of job like that because of how heavily the culture I was raised in told me that that would be F A I L U R E. I wish nobody had told me what to want as a kid.
So here is my take: The boomers and gen-x'ers presided over a period of profound technological and societal progress and growth, which was wonderful, but unsustainable and impermanent. This caused them to become extremely optimistic and make wild and outlandish promises to us millennials about what our lives would be like. But then that extreme progress reverted to baseline, everything stabilized to a much slower rate of growth, and while the world millennials are inheriting is in every way better than the one the boomers and x-ers were raised in it is so much worse than the world we were promised that it has led to this terrible malaise and sense of failure among so many of our generation. So let's maybe chill out a bit with the zoomers and stop stigmatizing normal jobs.
That wall you feel like you hit may be partly your non-upper class background. “Social skills” can range from not picking your nose in public, to the social style of the rich. Your post is pleasantly introspective and humble; your social skills are probably fine or close to fine. The educational sorting engine promotes upper class/upper middle class behavior. They don’t tell you when you hit that wall, but it’s real.
You’re right about the stigma about jobs though. That’s also a middle class formulation. “If you linger too close to the workers, we’ll lose all the progress we made!”
If you can do the job safely and successfully you can find permanent employment in a trade and probably more acceptance than you expect. The sense of failure affects you more if you buy in, so don’t buy in. Who has the hustle and the success? Go do what they’re doing. Educational eliteness is/can be vastly toxic in the US but the effects don’t have to be permanent. You can get over that mental block :) and on the other side you will find many like you. I escaped, mostly. Just remember money actually does matter.
Your initial point is very correct. I've had just about everything handed to me and achieved all my dreams mostly because I was raised to be able to socialize in an elite way. It's insane how important it is in having a truly excellent life.
Generation X was never optimistic about anything.
I'm in that generation and also got the threat of McDonalds jobs if I didn't improve my grades. But McDonalds and other fast food jobs really are crap jobs. They typically pay minimum wage (and no tips), have variable and lousy hours, bosses who suck more than usual, etc. I don't think I ever got threatened with "do your homework or you'll have to be a mechanic like your uncle", or a carpenter, or an electrician. Somehow the idea of trades jobs never came up at all, possibly because I was a stereotypical '80s computer geek from the start. But back then, computer geekery only led to a regular white collar career, nothing remotely elite.
The tail end of Generation X and the vanguard of Millennials graduated into the .com boom -- and the subsequent bust. Technological progress and growth didn't stop in 2000, though. Millennials have little to complain about.
An interesting story, for me. I'm at the tail end of the boomer generation, and had a "gifted" academic trajectory -- I went to a first-rate high school, a college that routinely ranks in the top 5 internationally for science and engineer, and the best graduate school in my field. I won scholarships, and later fancy grants from the Federal agencies.
But no one in my experience ever stigmatized work qua work. I worked at least part-time from the time I was 15 onward, and that was totally expected, among my peer group and in my family. My family and teachers were totally supportive of my academic exertions, but at the same time they took it for granted (and so did I) that I would enter the workforce no later than my mid-teens, and that taking a McDonald's job after school and in the summer was not just honest labor, worthy of ample respect, but an expected thing if one were not to be thought kind of a fancy pants putz with soft hands and probably a bad attitude.
Certainly if I had ever said I thought flipping burgers or pumping gas was beneath me, as a 17-year-old, just because I scored a 5 on the AP Calculus BC exam, my mother would have slapped me and I dread the look of contempt I would have seen cross my father's face.
Yeah, how true, after college nobody's gonna give a shit about your Shakespeare paper.
It's laughably ignorant to claim the world we inherited is "in every way better." That's true for me, and it may be true for you and many others, but it is very obviously not universally true.
Yeah, maybe "better on average" would be more accurate, based on statistics about disease and poverty.
Yeah, that probably wasn't the correct way to put it. I think I said it that way because... in the past I have found myself overly romanticizing the past, like, claiming that farmers in the 1800's were better off than today's poor, and have been thoroughly enough disabused of that notion so am to some extent over-correcting. There are certainly a lot of ways the world has been getting worse recently; I am obviously not very happy about how things are going. But I do think "better on average" is a good way to put it.
It’s partly of a really strong “not an office job” idea people have had for the last 15 years or so, combined with both increasing meritocracy and increasing belief in meritocracy - back when access to elite (in the broadest sense) jobs was basically hereditary everyone else just shrugged and went off down the widget mines, but now people think they can fight for them enough will to make it, well, a fight.
I’m not buying we are in a time of increased meritocracy.
Yeah meritocracy under capitalism is a myth.
As opposed to what? It's become fashionable to blame capitalism for everything, when there are no examples of other systems doing the thing in question better, or even on a comparable level.
Capitalism encourages meritocracy by its very nature.
We are. People just don't like that idea because it means that they aren't good enough.
All data points seem to point towards people being more likely to rise or fall based on merit rather than other things today than they were historically; for instance, more people are in the upper middle and upper class, a higher percentage of billionaires didn't inherit their wealth, etc.
On top of that, there is a higher demand for credentials and past work experience, suggesting that examining people for merit is increasingly important. Indeed, a lot of the complaints about jobs having high entry requirements suggests that we require more merit to even be considered.
All those are signs of increasing meritocracy.
What exact data are you talking about here? Certainly not recent US data where social mobility has largely stalled. A fully meritocratic society would have only 20% of the top quintile stay there (while 20% would fall to the bottom quintile) and vice versa
https://www.oecd.org/social/action-needed-to-tackle-stalled-social-mobility.htm
Part of it is the existence of people like ... the two of you.
You both have jobs that involve saying your opinions on the news of the day, and getting paid enough for it to have it be your day job. I think both of you put a lot of thought into what you write, but there's a lot of people who seem to spout random nonsense on Twitter and somehow get lots of followers and probably a good living. Like I dunno Ben Shapiro.
Easy to see that and think "I'm as smart as that guy why can't *I* do that???"
I sometimes wish I was one of those guys, instead of my real, well-paying, not-particularly-enjoyable, working-right-now-instead-of-partying-with-friends job. But it reminds me of what Matt Yglesias said about NBA owners - it's not just an investment, it's a prestige position, and there's no reason ex ante to expect to get paid for it. The world is filled with people who want their opinions to hold sway. The truth is I'd pay to wake up tomorrow with 500,000 Twitter followers who respected my opinions on politics et al.
I'm sorry you feel this way because twitter is for losers and makes its daily users dumber without exception.
This post has my nomination for ACX's "Best of 2021" category.
It can be true that twitter makes people dumber and also that having 500,000 followers on twitter who respect your opinions makes a person more influential than I'll ever be.
Whereas info-free snark is good for all our souls?
This phenomenon is noted by Eric Hoffer in "The True Believer", a book on the origin of mass movements. According to Hoffer, wealthy young people who believe their standard of living will not advance as far as the previous generation's are fertile ground for mass movements. A reservoir of rage.
I'd love to provide some quotes from Hoffer, but I can't seem to keep a copy on hand. I always find someone who needs to read it and I give them my copy.
Fortunately, we have the internet: https://www.academia.edu/21464682/The_True_Believer_Eric_Hoffer
" I detect such an immense frustration among youngish people about the sense that they have to occupy some sort of vaguely artistic or creative or intellectual occupation, that this is their birthright, and their continuing inability to secure a solid income by doing something that their peers recognize as enviable fills them with a profound sense of injustice." Well, Freddie, for what it's worth I like that sentence and think you write well, and I hope your writing makes you a good living.
Movies and self-esteem-koolaid-drinking-teachers keep telling people they can be anything if they believe hard enough and work hard enough, but that's a lie. Most people are just not born with the talent to be an upper class creative professional. Blank slatism leads to a lot of people beating themselves up for not achieving things that were never in their power to achieve.
I suspect there are two things here that people are improperly conflating:
1. Having an outlet for your creativity. Eg. writing.
2. Being paid for #1.
There are lots of jobs out there which will provide you huge amounts of free time to write "on the clock". Night watchmen was historically one of those. Hotel night auditor is great. Likewise, there are other jobs which will pay well to allow someone to make the most of their free time to engage in artistic endeavors. "Shakespear in the Park" style productions allow anybody to be a player.
But very few jobs will allow you to make a living for being artistic, and there is indeed a lot of competition for that space. And I suspect that is where the ennui comes from.
This isn't anything new. Karl Marx was the same way. Probably why he appeals to such narcissistic types.
Let's have three definitions of "elite":
Elite_1. People who think they deserve high-status positions
Elite_2. People who actually have high-status positions
Elite_3. People who actually deserve high-status positions
By this standard, we're massively overproducing Elite_1s, somewhat overproducing Elite_2s, and massively underproducing Elite_3s .
At the risk of making the system even more complicated, "deserve" in Elite_3 as you define it gives me pause.
There's a difference between "people who we might deem to ethically deserve elite jobs, and all the privileges thereof, if they can get them" and "people who have the skills such that it would be, in brute utilitarian terms, beneficial to society if they were taking up the elite jobs, regardless of whether they 'deserve' them".
For example, you can imagine a least-convenient-possible-world where, because they got better education and stuff, the children of the previous aristocratic generation are the most qualified, bu they're selfish and bigoted and corrupt (though not quite so corrupt that they're not still the most efficient at the end of the day) — while you have a class of would-be elites from poorer backgrounds who maybe wouldn't do quite as good a job but are earnest and meritorious, and morally "deserve" better than to waste away in jobs that are well beneath them while jobs that are *slightly* above them, but which they *could* do "well enough", are taken up by the oily aristocrats.
While I can see that it might suck to live in the world where the "oily aristocrats" rule while the marginally less competent but more "morally deserving" alternative have little power, I'd much rather that the past leaders of my society had been the absolute most skilled ones regardless of other concerns - 'fair' division of the pie today maters to us, but what matters to our grandkids and *their* grandkids is how fast we're growing the pie, and I think that thinking for the long-term future is what brings out the best in humanity.
P.S. I also think it'll generally be unlikely that selfish and bigoted and corrupt men are the best available leaders - while I grant for argument that the education advantage of old money can be very large, the best leaders in your toy society will most likely be that small fraction of the children of aristocracy that care about the wellbeing of others and are honourable.
Turchin does give an answer: it was the availability of mass college education and gender/racial equality opening opportunity to more people. Anyway, I really don't like Turchin but I do think there's increasing elite conflict. I just don't think it's caused by some kind of quasi-Malthusian overpopulation.
Isn't it obviously just the massive increase in education? We've gone from 7.7% college grads to almost 40% now: https://www.statista.com/statistics/184260/educational-attainment-in-the-us/
Also some speculation on my part. I never had a wage job in the 60s (because I wasn't alive then). But my impression from movies is that they had a little more freedom and discretion than wagies do now? It seems like back then, a shopkeeper/cook/factory worker was still expected to think for themselves, at least a little. Now, it's just really strict that you *follow the system*, with layers of managers and tracking systems to make sure that you do. I'm thinking of the difference between "Glenn Glarry, Glen Ross" where salesman work in a boiler room using whatever tricks they can think of to sell real estate, vs the part in "the office" where Michael (an experienced salesman) is constantly criticized by his manager for "going off script", ie saying anything at all that wasn't completely scripted.
Bottom line: not only is there more competition than ever to be an elite. But if you're not an elite, you are *acutely* conscious of your status as an underling.
That is correct in my limited experience, but I think it comes from the lopping out of many layers of lower and middle management. Your line schmo in the 60s was allowed a bit of freedom to think for himself *because* he had a foreman keeping an eye on things and backstopping his thinking, and the foreman had a shop manager watching him, and so on. There were a lot more layers of management. Many of these have been replaced by software and devices, and these are necessarily more rigid -- lacking human intelligence and flexibility -- and so the job of the upper manager becomes more precise, less telling a middle manager the general idea and more writing of scripts and programs, and the job of the line worker (or very low manager) gets extremely rigid because of the increased distance between him and the guy who makes the big decisions.
I worked in the corporate world from 1982-2000 and have supported myself as an opinion journalist since then, so I am out of touch with contemporary white collar work. But from what I hear, it sounds pretty awful.
I'm a white collar worker right now for the state.
I have an enormous degree of flexibility and personal responsibility.
This seems to be the case for most of my colleagues.
We do have various systems for trying to get stuff through (like for instance, if we create some Standard Communication With The Public, or some Standard Form for people to fill out, we run it through comms, and can't change it willy-nilly because the Spanish form and English form have to be the same), but like, in terms of personal discretion, they don't really care how we do our work, just that it gets done.
I create procedures to help people so that they can do work in a repeatable fashion, and that stuff is helpful to people - but those procedures also involve a lot of automation, so you can do your work way faster as a result, meaning that you have more time to do "the tough stuff" (which is also the interesting stuff, as it is the stuff that is NOT repetitive).
Indeed, it feels like the constant thing of being white collar is figuring out a way to automate all the repetitive work so that you can spend most of your time chasing down the things that aren't repetitive, as those, while a minority of incidents, take up the bulk of your time because they have Problems and don't fit into the system cleanly.
But those things are more interesting than the repetitive stuff, and their one-off nature means that it's more interesting.
If these "one-offs" stop being one-offs, you start looking for ways to stop them from happening so you don't have to repeatedly deal with the same issue.
If its massive increase in education, is this elite overproduction higher in Canada or Israel?
https://en.wikipedia.org/wiki/List_of_countries_by_tertiary_education_attainment
What seems to makes US elite overproduction more intense (to this outsider) is that only a handful of universities, who limit their supply of admissions, are seen to produce elites.
A lot of elites (as in actually talented and educated people) outside the US move to the US for work or academia and fewer go the other way, which should worsen/lessen the problem in the US/outside.
Mancur Olson is probably the first who made this argument in any detail, in his 1982 book "The rise and decline of nations: Economic growth, stagflation, and social rigidities." Quoting from memory, Olson believes that rigidities in the form of an increasing number of "veto points" follow the growth of special interest groups. Rigidities are weakened during major wars (who tend to weaken special interests by creating a mentality that "winning the war is so important that everything else must yield"). However, the process starts again when peace is re-established. The hypothesis is that the longer the peace lasts, the more "veto points" build up, and the harder it becomes to get anything done. (Notice in this context that we now live in the second-longest period without a world war in modern history, only beaten by 1815-1914.)
Olson was a creative political scientist, and the book could be worth an ACX review.
Ok, but why then do people expect that going to college should give them a job that's not just in the top half? I think the answer is that we expect that growth means that we'll be better off than the previous generation. But even when that's true in absolute terms the issue with education is that the kind of job that an education guarantees you is below what it was in the past (certainly in terms of status and possibly even in absolute material terms).
What this suggests is that we shouldn't expand admissions to educational programs with the same name. So if you want to expand the number of people you educate in college instead create some kind of new program without the baggage of the old. But maybe that has other disadvantages.
BTW a good case study to think about this issue is the many academics who seem to feel that it's an affront to justice and fairness that it's really hard to get a tenure track academic job. From the outside that's just nuts. I mean, no one would take it for granted that just because they played basketball or studied acting for 10 years they were owed a job.
Maybe because college still trains us as "future elites" rather than slightly-above-average? You get a campus filled with luxury amenities, classes that are liberal arts and sciences rather than anything connected to a job, and it ends with a fancy graduation ceremony where you wear a robe and hear some Latin phrases. Even the generic "business" degrees seem designed for future executives and entrepreneurs, rather than the low-level office jobs they'd be more likely to get.
Turchin defines the elite as those in the upper classes, the wealthy and the powerful. In Rome, these would be in the senatorial class or its equivalent. In traditional societies it would be the upper aristocracy, the dukes, and likely the royal dukes at that. In the old USSR and modern China, it would be the inner party members, the ones who might serve in the Politburo.
Turchin argues that nothing really changes without the elite being involved, since they are the ones who run things. Being elite has nothing to do with actually being better; it is just about being closer the the center of power. Russia was taken over by the communists because the Russian elite under the tsar would no longer defend the regime.
He argues that there are times, often after a destructive war or the development of a new technology, that there is a good supply of resources versus population, and that tends to be seen as the "golden age" when "merit" was rewarded and new members were added to the elite. The elite can expand even as living standards rise. At some point, as the elite grows in size, elite competition becomes more common and eventually destructive.
Turchin's use of the term "elite" has nothing to do with the competence of its rule or some arbitrary metric of merit save for those introduced as propaganda. In our society, it isn't the banker making $500K or $1M a year, it's the people who own the bank or are living on the fortunes enabled by its founding.
Freddie: Both the tweeter and Turchin are not original:
"Capitalism, Socialism, and Democracy is a book on economics, sociology, and history by Joseph Schumpeter, arguably one of—if not his most—famous, controversial, and important works. It's also one of the most famous, controversial, and important books on social theory, social sciences, and economics—in which Schumpeter deals with capitalism, socialism, and creative destruction."
"Schumpeter's theory is that the success of capitalism will lead to a form of corporatism and a fostering of values hostile to capitalism, especially among intellectuals. ... The term "intellectuals" denotes a class of persons in a position to develop critiques of societal matters for which they are not directly responsible and able to stand up for the interests of strata to which they themselves do not belong. One of the great advantages of capitalism, he argues, is that as compared with pre-capitalist periods, when education was a privilege of the few, more and more people acquire (higher) education. The availability of fulfilling work is however limited and this, coupled with the experience of unemployment, produces discontent. The intellectual class is then able to organise protest and develop critical ideas against free markets and private property, even though these institutions are necessary for their existence. ..."
https://en.wikipedia.org/wiki/Capitalism,_Socialism_and_Democracy
Schumpeter's theory of elites was popular with the center-right business press about 40 years ago.
Oddly, though, I don't recall people using the word "elites" much until 1999 when I went to an elite conference (e.g., a former prime minister was there). There, everybody used the word "elites."
Related(?)
https://www.conradbastable.com/essays/elite-underproduction-why-we-cant-solve-hard-problems-anymore
You linked to the list of games Buddha would not play on a previous Link post, back in April of 2016.
Not that there's anything wrong with that, it's still interesting.
https://slatestarcodex.com/2016/04/15/links-416-they-cant-link-our-dick/
Darn.
I wasn’t sure if I was misremembering, so nice to have confirmation.
https://xkcd.com/978/
Citogenesis, but the cycle goes between SSC Linkposts and Ratblr
The part of the FDA/Public health/Trump conspiracy theory that I just can’t get behind is that it proposes a counter factual where the FDA acts with enough haste to approve before the election.
Feels like a rare case of Cowen et. al. letting their vaguely anti-woke bias (I don’t say that pejoratively) overcome their extremely justified skepticism of the clowns at our public health agencies.
Please see Matthew Herper's 11/9/2020 article in Stat News in which he quotes Pfizer executive William Gruber on Pfizer's extraordinary step of shutting down lab processing in the world's most important clinical trial from late October until the day after the election:
https://www.statnews.com/2020/11/09/covid-19-vaccine-from-pfizer-and-biontech-is-strongly-effective-early-data-from-large-trial-indicate/
Pfizer shut lab processing down for a very simple reason, Steve: they had informed the FDA of the proposed protocol change in late October. If they had 32 cases before the change was approved, the protocol would have required them to report the results to FDA. In addition, they may also have been required by the U.S. Securities and Exchange Commission (SEC) to make the results public. If the results of an underpowered 32-case analysis had suggested that the vaccine wasn't that effective, it would've been bad and potentially catastrophic for the company (recall that, at the time, the fear was that the vaccines would only be around 50% effective).
So, they shut down lab processing to ensure they couldn't confirm cases and therefore would avoid a protocol violation. On 3 November, FDA approved the protocol changes, and they resumed lab processing on 4 November.
Pfizer shut down lab processing in late October because they didn't want to know where they stood in the world's most important clinical trial.. They restarted lab processing the day after the Election because it was now safe for them to know.
The first sentence is correct: they didn't want to know because they didn't want to conduct a 32-case interim analysis. This was because it was less likely to lead to good results than a 62-case interim analysis, which was now possible to be done quickly because the US epidemic exploded in mid-to-late October.
If they had continued lab processing, they would have been forced to do an interim analysis at 32 cases, because their protocol change needed to be approved to prevent them from being forced to do a 32-case analysis.
If Pfizer had not shut down lab processing until the day after the election, Pfizer likely could have delivered a 62 count result before the election. But it did not want to do that, having come under strong pressure all fall from Biden, Harris, Democratic doctors, and the media to slow-walk the clinical trial until after the election to deny Trump an October surprise.
For example, see this exultant November 1, 2020 New York Times news article (not an oped) announcing Trump had failed to get a vaccine October surprise:
Welcome to November. For Trump, the October Surprise Never Came.
Trump’s hope that an economic recovery, a Covid vaccine or a Biden scandal could shake up the race faded with the last light of October.
By Shane Goldmacher and Adam Nagourney
Published Nov. 1, 2020
Updated Nov. 3, 2020
President Trump began the fall campaign rooting for, and trying to orchestrate, a last-minute surprise that would vault him ahead of Joseph R. Biden Jr.
A coronavirus vaccine. ...
https://www.nytimes.com/2020/11/01/us/politics/trump-october.html
Also, note that Pfizer kept their decision to shut down their lab processing in their clinical trial secret from late October until the day after the Election from the public, investors, and the White House until William Gruber's interview with Matthew Herper of Stat News on 11/9/2020.
At least two Pfizer stock analysts issued downbeat recommendations on Pfizer stock in late October on the grounds that if the clinical trial were going well, Bourla would have announced like he'd been promising. Indeed, Pfizer's stock drifted down when efficacy was not announced in the original time frame, in part because no explanation was given to anybody other than FDA insiders that Pfizer was not following their published protocol for commercial-political reasons.
From a normal ethical standpoint, secretly shutting down lab processing in the world's most important clinical trial is obviously dubious.
However, from the standpoint of the higher morality of denying Trump an October Surprise that his emphasis on vaccines was working, it's saintly.
That sounds like a perverse incentive from our regulations. Andrew Gelman would say something about a failure to do statistics like a Bayesian (or to do a stupid version of frequentist/Fisherian "null hypothesis testing"). He's talked about not have a set point to stop collecting data, but instead to include in your model what drives your decision to stop collecting data. And one should be able to update on receiving additional data, and our regulations should not incentivize people to stop collecting data.
Steve seems to be 100% correct here.
I’m really struggling with the part where Pfizer decides a 32 case interim analysis is unacceptable, and in fact so unacceptable that is worth deferring the world’s most important clinical trial ever. Presumably cos these chumps are rubbish at designing these kinds of trials. And there’s no way around that allows us to progress the science like (for example) do the interim analysis as agreed but issue a joint statement with FDA saying they don’t think that’s sufficient and will wait for the 62 case milestone. Like there’s no alternative to just not knowing.
And of course we’re all just to pretend that Nov 4 is just a coincidence. And we’ve just to ignore that Topol himself made the link between the date of the trial and the Election.
It’s a complicated story you’ve got there Edward. It’s possible, but it seems there’s a more parsimonious explanation.
"recall that, at the time, the fear was that the vaccines would only be around 50% effective"
You have probably followed this more closely than I have. What is the evidence for that claim?
My interpretation of what happened is that if the vaccine turned out to be about 90% effective and the requirement was 50%, the people at Pfizer almost certainly had enough data prior to unblinding, from the earlier stages of testing, to be pretty sure it would pass. Why is that mistaken?
Prior to unblinding, how could they be sure of anything? Blinded data means not knowing whether the sick people are the ones you vaccinated or not.
And, yeah, a priori, a <50% effectiveness was plausible. Influenza vaccines made using traditional technology are often only ~50% effective, and what little we knew suggested that coronaviruses would be at least as hard if not harder to vaccinate against. Since we had no prior experience with mRNA vaccines in humans, "what if it is no better than a crappy flu vaccine?" is a reasonable concern.
It still takes a perverse incentive to deliberately postpone knowing that, even at the intermediate-result level. There may be several plausible perverse incentives to explain this one, e.g. the effect on Pfizer's stock price of an intermediate report of 40% efficacy.
Several non-mRNA vaccines have had trouble coming up with an impressive efficacy percentage. For example, mighty Merck pulled the plug on its vaccine effort on 1/25/21 because it could see from its phase one trial that it couldn't beat Pfizer's and Moderna's efficacy:
https://www.merck.com/news/merck-discontinues-development-of-sars-cov-2-covid-19-vaccine-candidates-continues-development-of-two-investigational-therapeutic-candidates/
My impression is that the Pfizer and Moderna saw from their earlier clinical trials in which they measured how much antibodies their vaccines produces (a lot) that there was an excellent chance that their vaccines would smash thru the FDA's 50% efficacy barrier.
On the other hand, the media tended to be fairly gob-smacked by the >90% efficacy claimed by Merck on 11/9/20.
Pfizer CEO Albert Bourlas's public statements in August-October 2020 suggest he had good reason for optimism about his vaccine. On the other hand, he seemed slower in figuring out the politics of an announcement before election day and thus kept suggesting that his vaccine would be proven efficacious in October. He finally wised up and shut down lab processing in late October until the day after the election.
If Bourlas had delivered Trump's October Surprise, he would be, to the prestige press, history's greatest criminal. But Bourlas managed to not do that while still beating Moderna by seven days.
Now that's what I call management!
Attempts to convince me that delays were a result of extraordinary political considerations are directly at odds with attempts to convince me that delays were a result of institutional ossification. These are competing explanations.
I don't know what about the linked letter would move the needle specifically towards the former.
is the OPs objection is the explanations are functionally at odds? the first requires some degree of coordination and flexibility, whereas the second claims these things are missing
Am I to find "we all know" persuasive?
Methodologically, if I don't share your priors appealing to them to try and convince me is a non-starter.
NIMBYism is a hell of a lot more complicated than mere rent-seeking, and your milage will vary *wildly* about what tactics it employs in any given area.
Your opponents are not a homogenous self-reinforcing block, and if you attribute the wrong motivations to them you will lose the ability to find effective resolutions.
I disagree with NIMBY but I agree that it's unhelpful to model NIMBY as rent-seeking. For one thing, the effects are community-wide rather than individual and mediated through whatever the local political norms are.
Here's a model I do think is helpful: communities face decisions about whether/how to provide certain public goods like traffic management, cleanliness / orderliness, neighborhood continuity, nice looking streets, and so forth. They can either do that by devoting time, money, and attention to addressing issues, or by refusing to grow so the issues don't come up. The latter has a much higher cost in aggregate utility but that cost is an externality-- it's being paid in higher rents and the reduced utility of people who can no longer afford to move to the area. The cost as perceived by decision-makers (usually homeowners and/or rent-controlled renters) is much lower. This produces effects equivalent to rent-seeking without any need to model NIMBY supporters as consciously wanting to increase their property values. It's sufficient that the community faces a wide range of decisions where there's an option to push the costs onto renters and prospective residents rather than having landowners bear them via property tax (or better yet, Georgist land tax).
These are not competing explanations at all. The idea that the people in charge of things don't really care about doing their job (helping the greatest number of people be treated for diseases) is perfectly compatible with the idea that the people in charge of things are focused on short-term political enmity.
> doing their job (helping the greatest number of people be treated for diseases)
Institutional ossification in a nutshell is that these are very much different things, no matter how convenient it would be if they were better aligned.
I think he's arguing that the ossification is sufficient to explain the letter, and since we have a strong prior for ossification, the existence of the letter doesn't lend itself as evidence for the political consideration theory at all.
It's very easy to pour sand into a machine, even easier if it's old and ossified.
The conspiracy theory is likely to be false. Ignore the bad headline, but this clearly explains that the companies decided to switch from 32 to 62 cases in order to get a higher-powered analysis, which was possible as the virus had started spreading rapidly in the US again. It wasn’t at all obvious back in October 2020 that the vaccines would be as effective as they are, meaning more power was desirable.
https://www.science.org/content/article/fact-check-no-evidence-supports-trump-s-claim-covid-19-vaccine-result-was-suppressed
This article you cite's headline is a pretty classic example of why Scott gets irritated by "No Evidence" headlines:
"Fact check: No evidence supports Trump's claim that COVID-19 vaccine result was suppressed to sway election"
From this article in "Science:"
"When the companies submitted their request for a protocol change, they had yet to accumulate 32 cases. If they had 32 cases before the change was approved, the protocol would have required them to report the results to FDA. In addition, if the results could impact the way investors traded company stock, they may also have been required by the U.S. Securities and Exchange Commission (SEC) to make the results public. They decided to store the nasal swabs taken from participants who had suspected SARS-CoV-2 infections: If they didn't test the swabs, they couldn't confirm cases and therefore would avoid a protocol violation."
In other words, Pfizer put the world's most important clinical trial on ice because it didn't want to know.
Please note that you and I are very close in our appraisal of what happened. We both agree that Mr. Bourlas of Pfizer was largely motivated by commercial considerations to take the extraordinary step of shutting down the lab.
My interpretation, however, is that the feverish political situation over the last weeks of the 2020 election campaign was the single biggest commercial consideration he faced. If Pfizer followed its published protocol and announced efficacy before the election, Trump would proclaim it a triumph for his vaccine-centric covid strategy and the mainstream media, being ferociously anti-Trump, would immediately amp up the already existing spreading of Fear, Uncertainty, and Doubt about Trump's vaccine.
Mr. Bourlas thus resolved, quite rationally, to take the crazy-sounding step of turning off the count in the clinical trial, which led to the Pfizer vaccine getting rapturous media coverage after Biden's victory.
Your theory appears to be that Pfizer in late October 2020 was NOT worried about the politics, they were suddenly worried again about their earlier statistical significance calculations: what if their vaccine doesn't achieve statistical significance for 50% efficacy at the 32 case checkpoint but then succeeds at a later checkpoint like 62? That would be bad from a marketing standpoint.
Well, sure, but they'd been through all that months earlier when devising their published protocol for this project that promised tens or hundreds of billions in revenue. They knew how strong the vaccine was in generating antibodies in humans from their earlier trial.
A much stronger worry along these lines no doubt was for Bourlas: what if the vaccine succeeds triumphantly at the 32 case checkpoint, and then the entire non-Fox media denounces Pfizer's statistical mumbo-jumbo that claims 32 cases are enough in order to discredit Trump's October surprise?
I you were the CEO of Pfizer, putting the trial in suspended animation until after the election must have seemed like a brilliant way to keep the prestige press from going berserk about how evil and dangerous Pfizer's Trump vaccine is.
And that worked.
They knew about statistics a month before, but what they couldn't predict was how many actual cases of COVID there would be. With no cases, you couldn't prove your vaccine works. More cases from a COVID surge, and you meet statistical significance sooner than you expected.
CEO Bourlas had been talking up an October announcement for some time.
But his comments agitated The Establishment since it might grant Trump an October surprise.
The CEO then found a pretty clever way to avoid getting The Establishment furious at Pfizer while still beating Moderna to the draw by a week. You can't say Bourlas didn't earn his pay.
Why would the CEO of Pfizer care about agitating an entity that exists only in your imagination?
Everyone who's not mainstream from Marxists to libertarians to reactionaries to rationalists recognizes the existence of the establishment. Entities and positions in it are high status. Entities and positions outside of it are lower status. And there a ton of social forces and organizations which devote an enormous amount of resources to making sure their projects are at least tacitly accepted by them.
(In anticipation of a criticism here; it's not an *organization* or a conspiracy or whatever. It's emergent, but it's broadly observed to exist.)
Pfizer CEO Albert Bourlas appeared to be obsessed, not unreasonably, with beating Moderna to announcing vaccine efficacy from a clinical trial first. Thus he repeatedly predicted good news in October. He appeared, however, to be slightly obtuse about the political ramifications of an October announcement in a hyper-politicized era. Perhaps he assumed, once again not unreasonably, that since Pfizer had not taken money from Trump's Operation Warp Speed, that Pfizer would be seen as above politics and its promptness in announcing the good news would be seen as admirable in humanity with covid.
Ha-ha, naive Mr. Bourlas! In the fall of 2020, everything, even matters of life and death, was political.
Eventually, Bourlas wised up. And by shutting down his lab from processing clinical trial until the day after the election, he managed to thread the eye of the needle: not announcing when it could help Trump, but still beating Moderna by one scant week.
I'm impressed.
Good argument. If it's a competition between "the company makes $billions and I get a $1.2 million Christmas bonus" and "Some ideological/cultural contest has an 0.5% chance of resolving more in the direction I like" I'm pretty confident of which motive I find more credible.
If I were writing up Pfizer CEO Albert Bourlas's decision-making process in October 2020 for a Harvard Business School case study, I'd be impressed by how he managed to navigate between Scylla (announcing in time to help Trump) and Charybdis (losing the race to Moderna).
I agree that we’re close in our appraisal of what happened. You just think it was a clever political trick, whereas I think it was a sound business decision based on the explosion of cases in the US from mid-to-late October. It is a decision I would have taken, irrespective of the political consequences, had I been in their position.
The real thing they were worried about was vaccine hesitancy. They were worried that:
* Rushing out a vaccine that wasn't proven to be safe/effective would result in low uptake.
* Rushing out a vaccine that was seen as pushed out for political reasons would result in low uptake.
* A vaccine that *actually had problems* would cripple future vaccine uptake, as it would be hyped up and then fail, and thus result in worse long-term outcomes. It could also completely ruin the company, which would be thrown under the bus by literally everyone for Lying To The Public.
All of these were very good reasons to want to be certain on their data.
Their worries were well founded, I think.
How do they know the effect of any action on vaccine hesitancy? The "pause" on the J&J vaccine was supposed to establish that the authorities were taking seriously any possible problem with the vaccines and thereby reassure people, but it wound up cratering actual vaccinations.
And it turns out in the end that they aren't very effective at all. If they'd spent more time they'd have figured that out.
🦃
As several Marginal Revolution commenters have already noted, other countries authorized the Pfizer vaccine on a timeline very similar to the US (including countries with a pro-Trump government), which is already pretty strong counter-evidence for the conspiracy theory.
Indeed. The UK's MHRA, which has generally been very quick throughout the pandemic, approved the Pfizer vaccine on December 2nd. The FDA approved it on December 11th.
And that's even worse ...
The Trump administration asked Pfizer and Moderna to apply for a "compassionate use" license to get the shots into nursing home residents before the 2020 winter surge. These residents had 10-20% fatality rates, and were just sitting ducks in the nursing homes.
Pfizer and Moderna said no. Birx's theory was that it was election related.
But what makes this so bad is that it also prevented Israel and UK from doing the same thing (why would Pfizer allow it for UK but not US?). So elderly people outside the US died for our election.
I'll just note that the original conspiracy theory I commented on was about US health officials pushing out trial timelines to delay authorization about the vaccine, not that Pfizer leadership didn't apply to some license. It is of course possible to spawn a more convenient sister theory (which seems superficially similar but actually makes a set of completely different claims) every time an objection is encountered; such is the nature of conspiracy theories. They are generally not worth paying attention to. The original one only was because it came from Nate Silver who is usually a reasonable person.
The pressure campaign was simple: convince Pfizer and the FDA to stall until after the election. We have receipts, Eric Topol bragged about it in an Oct 2020 Technology Review article.
Garett Jones put it best, the coincidental delays of Pfizer and the FDA are an example of a Schelling point, tacit coordination in game theory.
I don't see how. The theory is that US politics influenced Pfizer to sit on the study results and delay reporting/processing till the day after the election (plus FDA goal post shifting on what the trial needed to show). Hard to see how other countries would approve without trial results...
We do know that Pfizer shut down processing of samples to avoid having to report results, that under their original protocols the trial should have reported in mid-Oct. and even under the revised protocols it should have reported in late Oct. had they not stopped testing samples. We don't definitively know why, but the fact that they resumed the day after the election and announced results a few days after that does tend to make people suspicious (this may have been triggered by FDA approval of their protocol changes the day before, but that just raises the question of why the FDA sat on it's hands till exactly then). That plus Topol's comments and letter on influencing the FDA form the basis of the charge.
In my mind it's more credible than the Wuhan lab leak theory (which I give ~35%). Way more suspicious circumstantial evidence here.
They didn’t have 32 cases in mid-October, or even in late October (when they shut lab processing down and took the decision to change their protocol). They submitted their request to change the protocol in late October, and the FDA approved it on the second working day of November.
Of course they had 32 cases in late October. They had 94 cases by some time between Nov 4 and Nov 9. They probably shut down at 31.
Right. The single best guess for when Pfizer stopped the count is 31 cases, although I'd go as low as 29. My impression is that CEO Albert Bourlas really really wanted to beat Moderna so it took him a long time to realize that Pfizer would be crucified by the Establishment for delivering Trump his dreaded October Surprise.
Most of the comments focus on Pfizer, but the FDA actually moved the goalposts.
The FDA made a last minute change to the EUA standards to require a median of two months of follow up data for clinical trial participants. Do the math: it puts an EUA after the election.
https://www.cnbc.com/2020/10/07/trump-says-no-presidents-ever-pushed-the-fda-like-him-vaccine-coming-very-shortly.html
"The agency [in September 2020] published new guidance for vaccine manufacturers that said they need to provide at least two months of follow-up safety data after vaccinating trial participants to apply for authorization."
Which is, not coincidentally, the standard that Eric Topol argued for as part of his pressure campaign. Pfizer delayed the announcement to avoid negative PR about the election, but the FDA had already set the timeline.
Pfizer already had 2 months for 37,000 participants, though.
I haven't read their article, but it struck me at the time that unblinding the data just after the election was exactly what you would do if you wanted to avoid Trump getting the credit while minimizing the cost in mortality of doing so.
Similarly, I thought Covid first appearing near a research institution that was studying bat viruses was pretty strong evidence for the lab leak theory.
In both cases, a competent and unbiased expert could go beyond those simple facts to reach a different conclusion, but competent and unbiased experts are in short supply and hard for the lay observer to recognize. The Covid case was a particularly striking example where it turned out that the supposedly competent and unbiased expert was simply lying about being unbiased, being believed, for a while, by other people who shared his bias for different reasons.
Ben - I think you make a valid point. Pfizer was reacting mainly to the FDA's EUA guidance change. I don't buy the arguments from Steve Sailor that Pfizer shares a lot of the blame although I think they could have acted better (for instance by publishing interim results before the election as they had planned under their original schedule)
There are rumors, for instance from Erick Topol, that the Trump admin wanted to bypass the FDA by having the HHS Secretary rubber stamp the EUA in October when the first endpoint read-out from Pfizer's Phase III was expected to be hit. That would have been 100% legal although it would have created short term hesitancy and perception it was rushed. However I think it would have been net positive by a lot because I believe many elderly and at-risk would have gotten it and tens of thousands would have been saved. The increased hesitancy could be allayed later and offset to some degree by having more follow-up data, FDA weighing in, and observational studies. Additionally, a lot of hesitancy goes away once people see their friends and family getting the vaccine (most people are not analyzing data). So by allowing people to get vaccines earlier, the main process whereby hesitancy is reduced starts earlier as well!
I think I already posted my timeline below but it covers a lot of what happened:
https://moreisdifferent.substack.com/p/what-happened-a-timeline-on-the-development
One potential failure mode (especially with someone as erratic as Trump at the helm) is that the government starts vaccinating early on a "might not help but can't harm" basis, the vaccine turns out to be not very good, and the government sticks to it over better ones just to save face.
This happened in Hungary: the government started vaccinating with Sinopharm because in early 2021 they could obtain it in larger quantities than Pfizer, leaned on the health authority to skip the normal evaluation process, got a lot of criticism for it (some justified, a lot of it not), doubled down into a "West bad, China good" narrative (using inactivated viruses is old and time-tested, mRNA is still an experimental method etc. etc.) which was a nice fit for Orbán's politics in any case. Then by middle of the year, when Pfizer was available in effectively unlimited quantities, and the first reliable data on Sinopharm showed up and it turn out to be less effective, and a lot less effective in older people, and other Sinopharm-using countries started to re-vaccinate their older population, the Hungarian government didn't want to admit it was wrong and just stuck to its guns. So something like a quarter of the population ended up facing Omicron (which fortunately turned out to be not that bad) with two Sinopharm shots and maybe one Pfizer booster.
I found this timeline tremendously helpful and would recommend everyone here read it.
Thanks Ben, glad to hear! It took me like 8-10 hours to put together, I think. If you notice anything major I'm missing let me know (especially regarding Pfizer's timeline). I was going to try to double check and incorporate some stuff from Steve Sailor's article (https://www.takimag.com/article/the-new-normal-by-any-means-necessary/) but never did.
>18: Does “Moore’s law of genome sequencing” still hold? If not, who we should blame? Here’s a Twitter discussion.
Yeah, even before reading the tweet, it's Illumina and their monopoly. We'll just have to wait until their patents start expiring in the mid-2020s.
22 percent is a LOT higher than the base rate for committing violent crimes.
Yeah, again, if it had just said "unlikely", I would think they meant "compared to your lurid imagination where they're comic book villains and reoffending all the time". At "extremely unlikely" I am at least eyerolling.
are they using unlikely to just mean <50% ?
Fun rationalist side project: formally define unlikely and likely as probabilities. This raises the question of whether it’s contextual (e.g. a 10% chance of death is too terrifying to be unlikely, a 10% chance of getting your hands wet is too trivial not to be unlikely), as well as if there’s a gap between “unlikely” and “likely;” they seem like a binary, but a 40% chance doesn’t intuitively seem to be either.
Apparently "unlikely" is 20-45% according to CIA guidelines: https://www.edge.org/3rd_culture/Master%20Class%202015/Slide033.jpg / https://www.edge.org/conversation/philip_tetlock-edge-master-class-2015-a-short-course-in-superforecasting-class-ii
"very unlikely" would be <20% so applying it to 22% wouldn't be that far off. No guidelines for "extremely unlikely" though.
good find. it doesn't apply to this case, but 45 seems like a high value to call unlikely.
And are these ranges still recommended when comparing to an extremely high/low baseline? This also arises in ML for measuring performance of predicting rare events, since it is almost impossible to beat the baseline of 'is not a murderer'.
If I had a 19% chance of dying (from being hit by a bus) every time I crossed the street, would you still say that it was "very unlikely"? I mean I usually cross the street about five times a day, so that would reduce my life expectancy to about a day.
Easy to define them formally. Impossible to get anyone to stick to definitions in conversation, or any kind of discourse really. (Slight exaggeration for the sake of humor.)
Yeah, let's just say it's very unlikely you can get anyone to stick to definitions.
I remember some of the UK's variant of concern reports had formal definitions of those terms as probability ranges. They'd say things like 'likely (>50%) that alpha was more infectious than wild-type' and have a chart at the end specifying what probability 'likely' corresponds to.
They literally said "extremely unlikely". If something with a 49% chance of happening can be considered "extremely unlikely", then these words and concepts lose all meaning/practical significance.
oh, i'm with you... I was just trying to think of what justification was in their minds.
In the paper, it is a relative measure compare to other criminals - released non-violent criminals are more likely to be re-arrested for violent crimes than released murderers.
Which is meaningful and important if someone is telling you they favor shorter sentences for non-violent criminals and longer ones for murderers because of recidivism concerns.
But not how the tweet seemed to be phrasing it. Either the tweet misinterpreted that part of the paper, or was intentionally misrepresenting it and assuming no one would look at the paper itself (which, I'm sure 99.99% of people reading the tweet didn't, so...).
So.. would that support *increased* sentences for non-violent crimes?
It would support lower sentences for violent crimes if your only sentencing parameter is risk of violent recidivism.
Even taking that comparison as the point, is it age-adjusted? released murderers are often like 70, at which age violent crime is much harder to do....
Middle class domestic murderers who go to prison for 10 or 20 years for killing their wives aren't that likely to commit another murder after they get out in late middle age.
On the other hand, teens who shoot somebody (not fatally) in a gang shooting and do a couple of years are not all that unlikely to wind up killing somebody in their 20s or 30s.
It's not surprising. American prisons are torture sites that turn many victims into animals. 22% is "extremely unlikely" when that context is considered.
Also, it's probably much higher than 22% if you consider the violent crimes they weren't apprehended for
People released for murder are monitored extremely closely. While it is certainly possible they are committing violent crimes and not being apprehended, everyone else is being watched less closely.
Seriously, you call the police because your bf hit you, maybe something gets done maybe it doesn't. Call the police because your ex-bf who's on parole for murder hit you, he's going to jail.
Yeah, I would object at "unlikely". It's much much less blatant an error - I can almost believe it's an honest misunderstanding of how people are most likely to interpret the word "unlikely" - but I still think it's very misleading.
Came here to note this. Currently the stats are (an elevated) 7.8 murders per 100,000 in the USA (annually). Assuming that each person was murdered by a separate person (which maximizes the number of murderers, but doesn't account for spree killings or murders done by groups of people) that's 0.078 percent prevalence of murderers in the population (which is very close to zero, and it's interesting how sensitive we are these days to a change in that rate.)
Again assuming that each homicide is committed by a different person, over five years we'd expect 0.39 percent murderers in the general population. Instead, the report indicates that 2% were *arrested* for murder. That's an order of magnitude difference. Given the abhorrent clearance rate for murder in African American urban populations, it seems likely that the actual rate is much higher, perhaps even twice to three times as much.
That's a lot of dead people, casually swept under the rug, because space was needed for performative heart-bleeding.
And as the report indicates, the impact on the community via non-murder violent crime is 10 times as large. Good going guys.
I think you are off by a factor of 10. 7.8 per 100,000 is 0.0078%, or 0.039% over 5 years.
Good catch, that's what I get for doing math without double checking.
By this logic: If we released all violent crime offenders from state prisons tomorrow, 200k murderers (+4k) and 500k others (+5k) would result in 9k additional deaths over the next 5 years, or an annual increase in murder victims of 1.8k or 11% (from pre pandemic level of about 16.5k/year). You could argue it’s worth it but also it’s clearly not true. There would be blood in the streets if we stopped locking people up for murder. These numbers are based on the system we have now, in which people are locked up until they are old and in many cases a parole board believes they won’t kill again. You cannot plausibly claim these statistics would apply under a radically different system.
The arrow of causality is tangled here. Does a prison sentence actually make people MORE likely to commit violent crime? (Some studies say yes...)
It matters greatly when your sentence ends, because people age out of crime.
Wouldn't that be an independent factor you could control for?
You could, but the stats being responded to don't as far as I can tell.
The better the convict, the sooner they will be released and vice versa. The worst convicts will literally have less time outside of jail in which to reoffend, and they will also be older and therefore less likely to reoffend. These effects almost assuredly outweigh the effects of incarceration increasing criminality (if these effects exist)
Although I'm not a subject matter expert, I would be frankly astonished if the answer wasn't yes.
For the obvious reason of criminal record = hard to get a job= poverty = crime, but also for more esoteric reasons, like:
When my cousin went to an extended stay in a hospital to treat anorexia, they wouldn't put her in wards with other anorexics for fear they would reinforce each other's behaviors/beliefs and give each other advice on how to evade detection and treatment. I imagine surrounding criminals with nothing but other criminals for years could have the same effect.
Agreed. Also, what's the average period of incarceration associated with that 22% number? Because if it's measured in decades, I don't know that this implies we can just start letting violent offenders out of prison and not suffer any public safety repercussions.
Median time served for homicide before release is on the order of 15 years.
Note that this is just rearrest rate. True rate of reoffending is likely higher, although I have no idea how to find out how much higher.
Would getting caught breaking probation rules (they're very tight) lead to being rearrested?
The report they used as a data source is here:
https://bjs.ojp.gov/content/pub/pdf/rprts05p0510.pdf
Parole violations do count towards the overall rearrest rate when they result in arrests, but they do not count as violent reoffending, so it's not relevant to the question of how many people convicted of homicide commit another violent crime after release.
It's hard to tell from the report how many were arrested for parole violations only, as opposed to parole violations in conjunction with a more serious crime. I'm no expert, but I would expect committing a crime to be a parole violation, so it's likely that many of those parole violation charges were just tacked on to whatever crime actually motivated the arrest.
Sure, but the true rate of offending is higher than the arrest rate for the non-convicted population too.
Given that people with criminal records receive far more scrutiny from the police, I would expect the discrepancy between these two numbers to be much *less* for them than for the general population.
I'm not sure this is true. I would guess that first-time offenders are more likely to commit violent crimes against people they know, making it harder to get away with it, whereas repeat offenders are more likely to commit violent crimes against strangers.
Regardless, I think it's beside the point. When we're doing cost-benefit analysis of prison sentence length, the absolute risk of reoffending is probably more important than the risk of offending relative to the general population. Suppose the true rate of violent reoffending for people convicted of homicide is 30%. What would we do with the knowledge that this is only 20 times the overall rate of violent offending, rather than 30 times?
Maybe a dumb question, but re #11, why would Xi cause offense generally? I've always pronounced it "ksai", so I'm not sure if it is the pronunciation. Is it that it is spelled the same as the family name of the president of China? But I think Chinese pronounce the spelling "she". So maybe the Chinese government would not want to see that in print? If that's the case, is there a general ban on Chinese researchers associating that Greek letter with anything negative, or using it at all? Greek letters are all over all kinds of science, so that would be a big headache. Or maybe not just something so prominent. Or is it not about Xi Jingping, but the fact that it is a common family name in China, and it's more about national sensitivity than the president's ego? So many questions!
I think it would mostly cause trouble because China would believe foreigners would associate it with Xi Jinping, or take it as a deliberate snub
Somewhere a clever teenager in Taipei is creating a meme involving Xi as Winnie the Pooh dying of COVID-19 that gets through the Great Firewall, China gets pissed enough to over-rattle sabers at Taiwan, a frontline observer mistakes a civilian drone for an incoming ballistic missile, leading to a shooting war in the strait, the US gets drawn in and climate change doesn't look like the biggest concern any more.
Yeah, and then think of all the stoners enjoying the daylights outta that meme, and passing it along with odd tweaks added. There's a whole ecosystem of wit out there.
I think you are correct, and the here is a reason to eye-roll even harder:
The original article states “… and ‘Xi’ was not used because it is a common last name”.
However, it looks like “Mu” (the last named variant) is a more common last name than “Xi”.
Strictly speaking “Xi” is also actually a first name…
In terms of how many people bear it, yeah, but in terms of how many occurrences of the surname "Mu" or "Xi" you will find in a random text, probably not ;-)
We know the reason that Xi wasn’t picked because that decision of the WHO was taped. Indeed they didn’t want to snub Xi.
https://www.instagram.com/tv/CW-RYqyq80o/?utm_medium=copy_link
If you like FAH, you might also like Viva la Dirt League, especially if you are into computer and/or role-playing gaming at all.
Oh. That was a comedy skit?
Yes. And if the next letter after Nu had coincidentally been ‘Trump” or “Biden”, it would have been skipped in exactly the same way, with a similar eye roll inducing explanation.
I don't think most English speakers would be able to pronounce a word-initial /ks/ upon their first encounter with the word. So while us more educated folk might pronounce Ξ as /ksai/ or /ksi:/, I would bet that a majority of Americans at least would pronounce it /zai/, /ʃi:/, /si:/, or /kai/.
Plus when we inevitably reach Χ, we'd get to deal with that confusion lol
A Chinese might pronounce Xi and Xi the same. Not everyone in the world will pronounce Xi in the English way. Or Xi, for that matter.
By the way, the correct (i.e, Greek) way to pronounce the letter is "ksee".
I don't think the explanation requires an unusual level of sensitivity on the part of the Chinese government. Imagine that 'Biden' is the next letter in the sequence, but it's pronounced bee'doŋ. I think they still would have skipped the Biden variant.
Yep. It seems pretty reasonable to me especially considering China as the origin of the virus.
"But I think Chinese pronounce the spelling 'she'"
Sort of. The Mandarin "x" is a consonant that doesn't exist in English. It's more like what you get if you try to say "she" with the tip of your tongue held just below the base of your lower incisors.
Is this specifically a property of some Beijing standard dialect or something? Mandarin "x" and English "sh" seem identical to me but that could be a function of geography.
They're definitely different in standard Mandarin, but apparently x, sh, and sometimes s are merged in Cantonese-accented Mandarin. It wouldn't surprise me if there were some other regional accents/dialects that merged them.
What region are you from?
I enjoyed the twitter take:
"Technically they're all Xi variants"
I will be glad when we get to the end of the Greek alphabet and start identifying variants using first names. Covid Doreen, Covid Dwayne . . .
We're doing Chinese characters next. That should last for a while.
The hurricane people used to move from human names to Greek letters, so maybe the variant people can move from Greek letters to human names?
Sounds good to me Kenny. Wait, actually it doesn't, because I am currently so irritated and weirded out by the whole dangerous, irrational mess that at the moment I'm in favor of giving future variants names that as childishly disgusting as possible, names that would crack up a bunch of especially uncivilized middle school boys -- Fart, Buttne, Puke . . .
(Please excuse my crassness, if you're not amused by this kind of thing. You caught me in a crass & cranky state of mind).
There's something to be said for the idea that the WHO should contract out with roomfuls of 9 year olds around the world to come up with lists of names for disgusting things, and then work down these lists in some sort of canonical order.
Though it would be unfortunate if the puke variant turned out to cause severe buildup of methane in the colon while the fart variant turned out to cause more vomiting.
Damn, you're right. It appears that my proposal for scatologizing ourselves out of this mess has a major flaw.
10: I've bought an audiobook in the last month based on the quality of its negative reviews. In case you're interested, it was "Money: The True Story of a Made-up Thing" by Jacob Goldstein and it's great, and I knew it would be great because the one-star reviews were all various flavors of argle bargle Murray Rothbard. The lesson being, you can learn a lot about something by listening to its enemies.
Where did you read those negative reviews? Amazon/Audible's single 1-star review is related to a physical printing issue and the handful of 2- and 3-star reviews don't reference Rothbard or thinking I'd associate with those types of takes. (And actually one of the critical reviews finds a bright spot in the book to be the author's proposition of explicitly bifurcating banking services into full-reserve and fractional-reserve entities, giving the customer choice in the matter - which sounds nearly identical to something I read from either Rothbard or one of his acolytes years ago.)
Regarding #26, I'm not a Mormon or a Utah resident, but I think your graph is cherry-picking the timescale it looks at. Based on https://gardner.utah.edu/wp-content/uploads/Fertility-FS-June2021.pdf?x71849 , the state-level Utah fertility rate was generally declining from at least 1960 (as early as they show) to 1994, recovered a tiny amount from 1994-2008, and then declined again (as your graph shows). So, I think it's fair to attribute this to a trend that's been in place since 1960. The more interesting question is what caused the 1994-2008 temporary minor recovery?
Boy, that curve looks to me like it's just basically tracking the US numbers except 19% higher, except for the anomalous peak between 1974 and 1980, where it topped out at 83% higher. (!) I'd be more interested in hearing an explanation of that peak.
I think there's a "generational" thing happening, where kids come in clumps in our population. I was part of that 1974-80 boom, and then my children were born in that 2004-2012 bump. That's one part, maybe.
On 14: Poverty is bad enough, the idea that you'd need some sort of esoteric quantitative description that maps onto neat 21st century bourgeois categories of badness in order to want to fight poverty is bizarre, and it doesn't do anything to resolve one way or the other the questions about the "best" way to fight poverty because that's a discussion about values, not statistics.
Postscript: Give a monkey a set of calipers and apparently he'll call himself the Übermensch.
I mean, the underlying question is whether poverty _can_ be solved.
Let us take it for granted that low intelligence causes poverty. If low intelligence is also caused by poverty, then this is a vicious cycle that can be broken by simple give-poor-people-money interventions in a single generation.
If, on the other hand, low intelligence is caused purely by genetics, then poverty can't be solved except through genetic interventions.
If I caused lead poisoning in a bunch of kids (or just hit them sufficiently hard in the head with a ball-peen hammer), do you think they will on average be poorer than a control group of kids I don't damage?
It also sounds more doable than changing culture
I didn't think we'd go STRAIGHT into 1922 eugenics but OK...so, here's the thing, we have created truly godlike levels of wealth for BILLIONS of people in the span of a couple centuries, so yeah, poverty definitely isn't genetic and maybe think twice before you launch straight into that at a dinner party.
I'm having a hard time figuring out what your point is, but I will note that steam engines improve the productivity and standard of living of all humans without changing their genome one bit.
"Given a set of technologies and opportunities" is where all the bodies are buried, and you're not invited to the dinner party on account of being two hundred years behind the conversation.
But we’ve already eliminated poverty by that definition, the homeless aside perhaps. Previous generations would be amazed that our poor were more likely to be obese.
Franco's inability to post his laughable takes without being a condescending jackass must be genetic too.
The genetic changes already occurred. It was these genetic changes that allowed the steam engine to be developed in the first place. There's no reason to think people who never even invented the wheel (vast swathes of the globe) would have been capable of developing steam engine without significant genetic changes.
I believe there where wheels in the New World... but only used for toys.
They are talking about relative poverty here.
Absolute poverty has basically been eliminated in the US, Canada, and many other developed countries for decades now apart from a marginal population of people with severe mental problems/drug addiction that renders them unhousable, as well as a fraction of people who live in extremely remote locations and refuse government assistance and other things.
I should have clarified, I was talking specifically about poverty in otherwise-rich countries.
That's a truly weird definition of poverty, given that it excludes basically all poverty that has ever existed in all of human history.
I agree it's weird, but also I can see why someone would say it off-handedly:
I know lots of people in poverty and every single one of them is in the USA.
It's not weird at all. The article was looking at people in "rich countries". They weren't going into Chad and the DRC to study this stuff.
Gee, it's almost as if the study is question was looking at people in..."otherwise rich countries". Maybe click the link before freaking out based on your own ignorance next time?
It IS truly weird, but all definitions of poverty used with reference to the developed world at present are weird definitions. By any kind of objective or historical standard there is no poor person in any modern developed country. The biggest problems of the American poor are their abundant access to food and drugs. Poverty is *solved*. If you don't want to just shelve all further discussion of poverty (which, to be clear, would make perfect sense), you have to use a bizarre definition of it.
This is a mind-numbingly ignorant take.
Calling this "1922 eugenics" is lazy and stupid. Either poverty is largely explained by heritable factors or it isn't. Calling it "1922 eugenics" does not change this and is a BS attempt as shouting down an opponent who disagrees with you.
And FWIW, the west rose out of poverty by directly producing the scientific and technological advances to do so. It needed to do this to rise out of poverty because those things didn't exist at the time.
Today, all of that hard word is already done, and the knowledge is available in seconds to anyone with an internet connection. All these impoverished countries need to do is copy what has already been discovered. Nobody is expecting them all to invent this stuff from scratch. It's literally never been easier to industrialize. If they can't industrialize on their own in 2021, there's no reason to think they will ever come close to catching up with the rest of the world.
...holy shit, this comment alone proves that you are indeed mentally stuck in 1922-era beliefs regarding genetic superiority. I hope you improve as a person this coming year.
You have not made one substantive post in this whole thread, just insults. Impressive!
I deleted coarse and trollish reply. Apologies for any offence caused. Let´s try again.
First, it is not correct to say that if poverty is caused strictly by low intelligence, and low intelligence is caused by genetics, then poverty cannot be solved except through genetic intervention. It could also be solved by reordering society so that intelligence would stop playing role of an income determinant. Not that I think this would be good or anything, but it is an option for solving poverty even under this unrealistically restricted assumptions.
But secondly, assumption that low intelligence causes poverty is rather reductive. I get that it is strictly speaking correct, at least in wealthy countries with lots of human capital. Intelligence is one of the causes of poverty in a sense that IQ is correlated with income. But there are many other causes of poverty. And those non-genetic causes of poverty might be amenable to other kinds of solutions than genetic engineering.
And I think it is obvious that correlation between intelligence and income is in a present society very far from being perfect. E.g. I really doubt that in median Western country someone with IQ 100 from a family in uppermost 10 % of income distribution has a higher odds of ending up in poverty than someone with IQ 120 from a family in lowest 10 % of income distribution.
> could also be solved by reordering society so that intelligence would stop playing role of an income determinant.
Yes, impoverishing everyone would break any correlation between poverty and intelligence.
We can (and should) tax people based on height. http://www.overcomingbias.com/2007/12/tax-the-tall.html In the future, we'll have even better PRS scores for IQ, and can tax based on that.
Who needs taxes? Aren’t taxes racist?
Ben Franklin said that death and taxes are two things we will always have. Some rationalists think we can abolish death, but I don't think the same ones expect us to abolish taxes.
You don't need a "perfect" correlation for something to be extremely predictive/explanatory.
Poverty is caused by division of labor and poor bargaining power & compensation at the bottom of the wage scale. Population IQ level interventions, while likely good for other reasons, would be far, far more likely to increase elite overproduction than to solve poverty.
It would make the heritability of poverty approach zero.
And you're completely ignoring the non-income factors of poverty. Committing crimes and spending time in prison, becoming a drug addict or having a kid while you're a teenager will severely diminish your economic prospects and this has nothing to do with "bargaining power". Maybe if they grew up a wealthier family they would be less likely to do this kind of thing, but the heritability of these kinds of behaviors is certainly not insignificant (and can even explain more of the variance than environmental factors in some cases). Low IQ people have a lower savings rate even after you control for income and so would be expected to have worse financial health even if their incomes increased.
So, before there was division of labor, everyone was rich? If you wanted to blame it on agriculture, you might persuade me. We are probably not using our terms the same way.
Incorrect.
Poverty is almost entirely caused by differential productivity levels.
People who work at WalMart stocking shelves simply produce far less value on an hourly basis than electrical engineers or linesmen do.
Indeed, if you look at companies that primarily employ poor people, like WalMart, they have very low profit margins, while companies like Microsoft, that employ top end people, have very large profit margins.
This is exactly contrary to what would be expected if it was due to "poor bargaining power and compensation" - the workers at Microsoft have more leverage, not less, but their company has a higher profit margin, not a lower one, and they get paid a lot more.
Or decoupling poverty and intelligence... I still like UBI, and a miner definitely works harder than I do at my desk.
This is the best idea.
I think there's two layers to this.
Relative poverty is insolvable because of human genetic diversity. IQ has a heritability north of 75% in adulthood, and correlates with income to 0.5 in primary wage earners.
So, you can't solve relative poverty.
Absolute poverty can be solved by increasing per capita productivity. The US has almost no one who lives in absolute poverty nowadays.
So depending on whether you are defining poverty relatively or absolutely, it's either insolvable or has already basically been solved.
There's a big difference between 'I have $5 and you have $100000000000' versus 'I have $100,000 and you have $200,000.'
We could 'solve' the massive inequality we have right now without requiring 100% identical outcomes. How much a economic system amplifies relative advantages is a contingent property of the system itself, not a natural and unchanging fact of the universe.
There is absolutely nothing "esoteric" about IQ or any other established psychometric measure. The fact that you're clearly ignorant on this stuff is a you problem.
As for fighting poverty, if you do not accurately understand the actual causes of poverty then you have no hope of ever meaningfully reducing it.
Trial and error is not so completely hopeless. Unfortunately, trial and error would require poverty fighters to try new methods and abandon failed ones.
"Cheap solar, cheap wind, and cheap storage mean that we could see the first large sustained decrease in electricity costs in over half a century. People are finally starting to realize this, and are speculating about what could be done with cheap abundant electricity. ...The reality of cheap electricity — not 30 or 50 years in the future, but in the coming decade — is thus starting to sink in. This is really happening."
Uh-huh. Meanwhile, this email from the electricity supplier I'm currently signed up with:
"We have all been impacted by the significant increases in wholesale energy prices over the past year. At [redacted] we have made a commitment to our customers to work hard to keep prices as low as possible. As a small supplier we are unable to absorb the increasing prices of wholesale electricity or to hedge against them effectively, this means that as prices increase we have to pass those increases onto our customers. In order to protect our customers from significant price rises we’ve engaged with [redacted] to offer our customers a competitive green electricity deal".
Every electricity supplier in Ireland has whacked on price increases over the past year. Although wind generation seems promising, at the same time, unusual weather patterns can mean it fails: right now, it's very very windy so the generation is high, but back in June we got a period of settled, warm weather with little wind, so no generation.
https://www.farmersjournal.ie/ireland-s-wind-energy-generation-plummets-during-heatwave-636463
So I'm not expecting 100% renewables electricity and I'm very much 'wait and see' if we do get cheap electricity sometime within the next nine years.
Retail customers buying power from the grid is emphatically not the point here. Industrial manufacturers building their own behind-the-meter power and building widgets with .01$/kWh power is the killer app.
It’s emphatically the point to retail consumers.
If we were expecting very cheap energy in a decade, wouldn't oil companies stock value be plummeting? [1] And geologists give up on trying to find new oil and gas locations that won't be ready for over 10 years?
[1] Yes, many of them have investments in renewable energies, but there to various degrees and we should be able to notice that.
Someone who unironically used "bourgeois" pejoratively just moments ago celebrating rich companies becoming richer while consumers struggle to pay for heating. Weird.
It's always "jam tomorrow", isn't it?
Damn it, if I weren't too cheap/broke to subscribe to Freddie deBoer's Substack (hi, Freddie! Happy New Year!) I'd be over there whinging like a creaky gate swinging in a gale about this.
This, folks, is how you lose elections to populists instead of your nice, shiny, technocrat candidate winning.
Stand atop a heap of crystal-ball gazing, looking into the mistily undefined period of "sometime during this ten years" and declaring that cheap electricity is coming - for big industry, to reduce their manufacturing costs and make more crap. And not even the prospect of at least more jobs for people, because those industries will be relying on automation to keep their labour costs down and productivity up, so there will be a flood of (iPhones? refrigerators? widgets for fancy driverless electric cars?) but I won't be buying them, because my disposable income is reduced due to inflation and things like "no cheap electricity for you, peasant, in the very definite next year or two but instead three or more price increases during the year".
And the answer to that, from we peasants that are not getting bonuses or rises in line with inflation, is "fuck you, I prefer to dream of cheap widgets for my toys".
Thanks, guys! I'm not going to hold my breath anticipating "cheap electricity for big industrial and commercial concerns is just around the corner! like the way nuclear power would make the deserts bloom:
https://www.environmentandsociety.org/sites/default/files/thumbnails/image/seagram2.jpg
I don’t notice any difference in my bills when the wind blows vs when it doesn’t. The only carbon tax applied is to all electricity in Ireland.
This is an awesome comment.
"for big industry, to reduce their manufacturing costs and make more crap"
The crap big industry produces includes most housing (often using energy-expensive steel), cars (again, steel, aluminum), and food (needs water--desal to the rescue!)
Meanwhile, my electricity bill is on the order of 1% of my spending and natgas *maybe* another 1%.
I would *love* cheap, clean energy for the industries that make pretty much all the things I consume.
You realise that your 1% is an anecdote not a statistic?. A quick google tells me that 14M people live in fuel poverty in the US and it’s worse in other countries.
I think it's less relevant whether people are in "fuel poverty" than it is whether that bill is high enough that doubling it would offset a reduction in food or consumer good prices.
That said, you're right, after looking into it, it likely is, unless food dropped more than I'd anticipate.
Ireland needs to build inter-connectors to both export surplus wind (of which we will have plenty when offshore is built out) and to bring in power from the continent. Unfortunately it looks like Britain is going to eat our lunch in that.
They probably mean when fusion power comes on line, which is Just Ten Years Away™.
Short term price ramps aren't an issue and aren't what's meant here. Long term price falls are what's going to happen, short term volatility is only natural in a changing market.
UK electricity bills have basically been flat the last decade with energy efficiency doing the heavy lifting. Probable shift in taxation to gas as opposed to electricity will see that change.
https://twitter.com/drsimevans/status/1387064025654116355
Ireland can reach 100% renewable relatively simply, offshore wind off each coast, dense interconnection with UK, France, Iceland, Spain smooths supply profiles. Each individual country only needs a small amount of overbuild when paired with a reasonably small amount of storage - 2-24hr storage markets serviceable by batteries, longer term served probably by hydrogen/derivatives given the volume of money flowing into the market but may vary on geography.
Interconnectors are the key and really drastically help cut emissions and prices. ERCOT is the example for the US audience, with ERCOT basically screwing the US out of having a national grid which would let solar in California compliment offshore wind in New England, or whatever.
https://www.volts.wtf/p/lessons-from-the-texas-mess
https://www.volts.wtf/p/transmission-month-everything-in
https://ec.europa.eu/energy/infrastructure/transparency_platform/map-viewer/main.html
Unfortunately Ireland has no real plans for large interconnectors but Britain does, the distance is much larger of course. You are right though, yesterday I noticed that - it being windy - renewables were 70% of supply and that’s a half hearted effort of a few onshore wind farms. If we (in Ireland) double output we would have a big surplus to sell on a moderately windy day but that depends on there being a customer somewhere to take it.
ERCOT is just Texas being stupid. Why does it prevent California and New England joining together?
26. As a childless white nonmormon young professional who moved to salt lake city recently, myself and quite a few people like me are the problem.
I have also read a bit about mormon women having trouble finding mormon men to marry due to asymmetric apostasy but don't have any detailed sources.
"I have also read a bit about mormon women having trouble finding mormon men to marry due to asymmetric apostasy but don't have any detailed sources."
Well, traditionally, the Mormons do have a solution they can fall back on if that's the case....
re: #26, it seems too small of a change to be the whole explanation, but the source now has a link to the fact that Utah counties have been becoming less LDS over time as well as people of other faiths move in. In the addendum, he says 'there is an almost 1-to-1 ordering between "percent LDS" and TFR.'
So that explains at least part of it, people from groups with lower average fertility rates moving into the area would certainly affect the overall rate.
The percent LDS certainly explains much of the difference between counties in Utah/Idaho.
This is the correct answer. "Lives in Utah" is no longer a good proxy for "is LDS".
But it's also a factor that LDS folks are having smaller families, especially in places like Utah County. When I grew up here, large families were bumping the average up. Just a quick mental sample of my close friends as a kid gives me families with 11, 8, 10, and 2 kids, plus my family with 4 kids. That's not scientific, but it's not a bad sample of family size in Utah County in the 80s/90s. (One family in the neighborhood had 17 kids, same mom, no twins! Fun fact: I was friends with one of those kids, and also friends with her niece, who was a year older.)
Utah (even Utah County) is definitely less Mormon than it was, but LDS families are smaller too. As for why LDS folks are having fewer kids, I guess a few things may contribute. More women serving missions (probably a small effect), more women delaying kids while going to school, certainly the "asymmetric apostacy" mentioned above, and honestly the economics of having lots of kids. The price of a larger home and a larger car relative to income plays a role, and since the 2008 housing crisis, home prices in Utah have still been going up.
I'd have to do some digging on how much a home or a big old van costs in relative terms to when I was a lad, but my gut tells me it's more relative to income. In addition, we don't culturally cram as many people into smaller homes as we used to, nor do we overfill cars like we did when I was younger. So that's changed too.
Not to mention women getting more educated correlates strongly with lower fertility.
This isn’t exactly hacking, but deepfake audio has been allegedly used twice in plots to steal money by mimicking the voices of company executives.
https://www.wsj.com/articles/fraudsters-use-ai-to-mimic-ceos-voice-in-unusual-cybercrime-case-11567157402
https://www.forbes.com/sites/thomasbrewster/2021/10/14/huge-bank-fraud-uses-deep-fake-voice-tech-to-steal-millions/?sh=687a06d47559
I'm not sure how deepfakes could be used for (computer) _hacking_, but the link itself isn't about that – it's about an Instagram account being hacked and deepfake videos posted to it.
But, historically, 'social engineering' (i.e. humans compromising security via other humans) has long been recognized as an important element of practical 'hacking' so it's definitely reasonable to consider what's described in your links as 'hacking' in a broader (but still coherent and reasonable) sense.
It would be funny if, after all the panic about deep fakes, the main use case for crime was helping criminals do really good impressions of people over the phone.
So, based on item 3 from link 1, Buddha wouldn’t play Squid Game:
“ Games of marking diagrams on the floor such that the player can only walk on certain places. This is described in the Vinaya Pitaka as "having drawn a circle with various lines on the ground, there they play avoiding the line to be avoided". ”
Re: Number 9. I have read that the COVID vaccine inserts non-human DNA into your DNA, thus making you technically non-human, so therefore Jesus cannot save you.
You'd think that if He were able to turn water into wine and raise the dead, as well as being generally omnipotent and omniscient, that He would not let Himself or His works by stymied by a game of Cosmic Gotcha.
more importantly, every virus ever inserts non-human DNA or RNA into you. "You can only be saved if you've never gotten a virus" is a really aggressive way to gatekeep heaven...
Don't confuse folks with facts.
How would you look at a piece of DNA and say whether it is human or non-human?
By that definition, if it's not a sequence present in the genome of Adam or Eve, it's not human.
Unless that retrovirus previously infected the Tree of Knowledge, then it doesn't bear original sin so arguably it's diluting the Fallen human DNA. (Why does this feel like an SMBC comic?)
"This comic book brought to you by the Jack Chick Institute of Biological Sciences!"
Hey, nobody's perfect.
I can't even begin to disentangle the bad, ignorant, and plain "that's not even wrong" understanding of theology there, so I'm going to recommend this video from "History for Atheists" - warning: very sharp language, even I thought it was a bit rough on the nominees in places, but this kind of stuff is certainly a runner for "Crappy Golden Orrery of the Year":
https://www.youtube.com/watch?v=WWfa3PKqXuI
There was a group of engineers at Google NYC who would basically play Counterstrike all day. There were also some who barely showed up for work at all (and weren't working from home either). I don't think that's typical, though. But xkcd.com/303 isn't completely off, there's plenty of down time surfing the web, wandering the building, and having swordfights on chairs. At least until Corporate Safety shuts that last down.
"11. Why COVID variants skipped from Mu to Omicron: “In a statement, the WHO said it skipped Nu for clarity and Xi to avoid causing offense generally.” Rolling my eyes at “offense generally” and the idea of deliberately averting nominative determinism."
On the one hand, the last thing we need is to give the Nutbar-American Community more ammo by naming a virus "Xi".
On the other hand, skipping "Xi" just gives the Nutbar-American Community more ammo.
I guarantee you that the WHO was vastly more concerned about China itself (the world's largest economy!) than about "Nutbar" Americans. In fact, I'd reasonably certain that the latter group most likely didn't factor into their decision at all, nor should it have! Do you really think that global security organizations are concerned with the intricate details of American political discourse? That sort of Americentric viewpoint honestly seems bizarre to anyone who isn't living in the U.S. themselves.
Hey! This is off-topic, but I saw that you posted a comment in the classified thread a few days ago that isn't there anymore, and I just wanted to say that I'd be interested in hanging out if you're up for it! We probably wouldn't be a romantic match, but you seem cool and I'm looking for Boston friends :) My post is here: https://astralcodexten.substack.com/p/classifieds-thread-12022/comment/4731208 and you can email me at rip.my.inbox@gmail.com if you're interested!
19. 6ixBuzz acted in a way that meaningfully emboldened gang violence in Toronto that causally contributed to many deaths (mostly gang involved but innocents as well). I do appreciate that people were erroneously accusing you [Scott] of emboldening abstract bad things but I do think the cases are qualitatively different.
"doxxing is okay when it's important" is the existing rule doxxers use. Change means either reforming that rule—in effect, just changing whose opinion of 'important' is used—or being against doxxing altogether.
The latter is more in line with the 'Be nice, at least until you can coordinate meanness' principle.
Could you elaborate? Was their Instagram account actively inciting a gang war? It's hard to picture how else social media posting causally contributes to gang violence, but maybe that's just a failure of my imagination
The “public health establishment delayed the COVID vaccine” theory is false. The companies very clearly explain the reason for the delay in the article below (ignore the bad headline).
“Pfizer initially designed the trial to conduct interim analyses after 32, 62, 92, and 120 cases accrued... Fewer cases means less statistical power and therefore greater uncertainty about the vaccine's impact, so the efficacy goals to meet FDA's requirements were higher than 50% at interim analyses, ranging from 76.9% for 32 cases to 58.8% for 120.
In mid-October, the companies had yet to confirm 32 cases. But with the epidemic exploding at many of the trial's locations—which were mainly in the United States—they had second thoughts about FDA's request that their first interim analysis should have more to support an EUA request…
The math was simple: COVID-19 cases among participants were jumping from one or two per day to up to 10 or more. It became clear that the trial would accrue 62 cases shortly after hitting the 32 mark, and the higher number meant greater statistical power—and fewer debates about the meaning of the data. This 62 cutoff both lowered the efficacy bar the vaccine had to clear, and was also something of an insurance policy: If the vaccine triggered mediocre immune responses and it teetered around 50% efficacy in the trial, it could more easily have been deemed futile at 32 cases because of bad luck.
the companies' decision to seek the protocol change had nothing to do with politics… "This was our decision," Sahin says. When he read Trump's tweets, he shrugged. "This is just not true."
In late October, they informed the FDA of the protocol change. The FDA approved the protocol change in early November, which meant that the companies could start testing their samples. The results were announced on 9th November.
https://www.science.org/content/article/fact-check-no-evidence-supports-trump-s-claim-covid-19-vaccine-result-was-suppressed
I think this may be a "beware isolated demands for rigor" situation. When reasonable reasons to delay the vaccine would make Trump look bad, there were a lot of them. When that stopped being a political concern, there suddenly were a lot fewer reasons being taken seriously.
While the EMA has been consistently institutionally slower at approvals than the FDA, that's not the case for the (UK) MHRA, which has been really aggressive for that sort of bureaucracy (it approved AZ first and has been much more prepared to take risks with it; it keeps complaining that Pfizer and Moderna won't apply for approval to vaccinate children).
There's no reason that MHRA would be politically motivated by what is going on with Trump (not because they don't hate Trump, but because a UK approval wouldn't have any impact on US politics). Yet they didn't approve until after the election either (and note that the first non-trial vaccination in the world was in the UK, in early December 2020).
I'm inclined to think that the fact that non-FDA agencies didn't approve until after the election suggests that this wasn't an unadulteratedly political decision.
Indeed. The timing of the *announcement*, on November 9th, was driven by sound business decisions. If you’re Pfizer, and you think you can still be the first to announce if you switch to 62 cases (thus getting a higher-powered analysis, and increasing the probability of approval) as opposed to 32 cases, why wouldn’t you do that? Many feared the vaccines would only just meet the efficacy threshold, so a higher-powered analysis was desirable. This meant that the announcement occurred after the election rather than before.
The timing of the *approval* was irrelevant to the election. It occurred, as you say, on December 2nd in the UK, and December 11th in the US. So it's just not true to say that "there suddenly were a lot fewer reasons being taken seriously" - the FDA took more than a month (!) after Pfizer had announced its vaccine was effective to actually approve it.
Sound business decisions, political chicanery...equally bad in my book! \s
The decision of Pfizer was to not publish trial data until after the US election; the MHRA can't approve the vaccine without any trial data, even if they're not linked to the cause of the delay
headline: "No evidence..." wait, isn't that some color flag.
But joke aside, I also haven't seen mentioned an obvious political alternative read of the letter: They were worried the Trump administration was attempting through side channels to have the results delivered earlier, and was making it clear that should not happen.
Attempting through side channels to hasten the vaccination of the populace in a deadly pandemic, likely saving tens of thousands of lives, how diabolical!
The smoking gun evidence that Pfizer shut down lab processing in the world's most important clinical trial from late October until the day after the election is found directly in Matthew Herper's 11/9/2020 article in "Stat News" interviewing Pfizer executive William Gruber:
The story of how the data have been analyzed seems to include no small amount of drama. Pfizer, seeing an opportunity to both help battle a pandemic and demonstrate its research prowess, made decisions that were always likely to make its study the first of a Covid-19 vaccine to produce data — including its decision to have an independent group of researchers, known as a data safety and monitoring board, take an early look at the data in the 44,000-volunteer study before its completion.
The first analysis was to occur after 32 volunteers — both those who received the vaccine and those on placebo — had contracted Covid-19. If fewer than six volunteers in the group who received the vaccine had developed Covid-19, the companies would make an announcement that the vaccine appeared to be effective. The study would continue until at least 164 cases of Covid-19 — individuals with at least one symptom and a positive test result — had been reported.
That study design, as well as those of other drug makers, came under fire from experts who worried that, even if it was statistically valid, these interim analyses would not provide enough data when a vaccine could be given to billions of people.
In their announcement of the results, Pfizer and BioNTech revealed a surprise. The companies said they had decided not to conduct the 32-case analysis “after a discussion with the FDA.” Instead, they planned to conduct the analysis after 62 cases. But by the time the plan had been formalized, there had been 94 cases of Covid-19 in the study. It’s not known how many were in the vaccine arm, but it would have to be nine or fewer.
Gruber said that Pfizer and BioNTech had decided in late October that they wanted to drop the 32-case interim analysis. At that time, the companies decided to stop having their lab confirm cases of Covid-19 in the study, instead leaving samples in storage. The FDA was aware of this decision. Discussions between the agency and the companies concluded, and testing began this past Wednesday. When the samples were tested, there were 94 cases of Covid in the trial. The DSMB met on Sunday.
This means that the statistical strength of the result is likely far stronger than was initially expected. It also means that if Pfizer had held to the original plan, the data would likely have been available in October, as its CEO, Albert Bourla, had initially predicted.
Pfizer had published its clinical trial protocol calling for checkpoints when the following numbers of samples of positive covid tests had been found in its clinical trial (across the unblinded two arms: test and control):
First: 32
Second: 62
Third: 92
But Pfizer decided to shut down processing of samples in late October and only restarted its lab in the world's most important clinical trial on the day after the Election, Wednesday November 4. Pfizer wound up blowing through its first, second, and even third published checkpoints, winding up with 94 case. It made its epochal announcement of its vaccine's efficacy early on the morning of Monday, November 9, six days after the election.
Without shutting down lab processing, it seems likely Pfizer's efficacy announcement would have come no later than Monday, November 2, the day before the election, which would have given Trump his belated October Surprise that his opponents were so fearful of.
And the article I posted quite clearly explains *why* they switched from a 32-case analysis to a 62-case analysis. Pfizer wanted to be the first to announce its results, but more importantly it wanted its vaccine to be approved. An interim analysis with greater power - 62 cases - was more likely to detect a sufficient effect than a 32-case analysis. As the virus began to spread rapidly in mid-to-late October, they realized they could skip the 32-case analysis, thus getting a higher-powered analysis that was more likely to lead to approval, *and still be first to announce*.
Recall that the fear at the time was that the vaccines would only be around 50% effective (see NPR article below). The risk of failing to get approved was considered to be quite high. Pfizer therefore made the entirely sound business decision to increase the likelihood of approval, safe in the knowledge that it would still be first to announce.
https://www.npr.org/sections/health-shots/2020/09/12/911987987/a-covid-19-vaccine-may-be-only-50-effective-is-that-good-enough
You've explained a shift from 32 to 62, but not from 62 to 94.
They had shut down lab processing in order to prevent themselves from having a continuous case count. If they had got to 32 cases first, they would have been required to conduct an analysis on 32 cases. If they had known about 62 cases first, they would’ve been required to conduct an analysis on 62 cases. In the end, the first number they knew about after reopening lab processing was 94 cases.
Then how do they determine when to reopen lab processing if it wasn't based on the election?
It's the day after FDA approved the protocol change. But of course that raises the question why the FDA waited till exactly then. And the whole thing is just so suspicious in timing and otherwise (and given the Topol letter and comments certainly people were trying to influence the timing) that it's going to be pretty hard to disprove it. For what it's worth I think it's significantly more likely than not that delay was influenced by political considerations.
But clearly releasing a 32-count update could not have had any affect on approval, and would have been followed within a few days by the 62-count update in any case.
Don’t the Trumpy set mostly hate and fear vaccines anyway? So promoting a vaccine before the election would’ve somehow helped the anti-vaxx side win? I don’t get it.
(I think) the thinking is that the 'pro-vaxx' sides might have been reversed (to some degree anyways) in the relevant counterfactual world.
The highest levels of vaccine skepticism seem to be found among the Black caucus, of whom ~90% vote Democrat, so I don't get what you're going for either.
The relevant timeline is otherwise this:
* in early October, Trump repeatedly says on the campaign trail that "vaccines are coming momentarily"
* in response, on October 16, Pfizer announces that it will delay its FDA applications
* in response, Trump softens his declarations to "vaccine would be coming within weeks"
* throughout this, the "neutral corporate media" are full-blast accusing Trump of hundreds of thousands of COVID deaths
In a short while it has become clear that Pfizer in October already had collected 32 RCT results (withheld instead of published, as per the last-minute protocol change), for which it is overwhelmingly certain that they *would* have crossed the 76% threshold (which would have sped up some things, "undesired political impact" aside) and for which Pfizer would not have suffered drawbacks even if they turned out below 76%.
(In a slightly longer while it has also become clear that Biden is not being full-blast accused by the same "neutral corporate media" for the ongoing hundreds of thousands of COVID deaths)
A recap of that October segment is in https://news.yahoo.com/trumps-vaccine-promises-meet-reality-184543949.html
Um, ok, I guess? And, therefore…?
I imagine that Black Democrats are less likely to be vaccine hesitant than White Republicans (or Black Republicans) at this stage. We know that 60% of the unvaccinated are Republican, much higher than their share of the population.
Well at least your understanding that the vaccine-skeptics are 60% Rep / 40% Dem is grounded in reality, as opposed to McClain's assessment of "hur hur Trumpy set", which is grounded in the dominant propaganda of our times.
About imagining that fewer % of Black Dems are skeptical than % of white Reps, that would need to be squared with stats that Blacks are overall 41% less likely to pursue vaccination, and that given a ~10:90 voting split, we can shuffle only so many Black skeptics into the Rep group. I did not pursue googling down for polls that slice by race and party, but here's the source for the 41%:
https://www.medicalnewstoday.com/articles/which-us-demographics-are-more-likely-to-refuse-a-covid-19-vaccine
The exact race/party split wasn't the point of my comment anyway, but how Pfizer (with FDA's backing) modified the protocol to withhold the intermediary results because of their projected impact on the 2020 elections. That the claim that the protocol was modified at the last minute because of Pfizer/BioNTech business interests - to echo the BioNTech CEO - "is just not true".
Pfizer started the RCT in July 2020 on 41 thousand candidates. Back then, when they preregistered the publication milestones of 32 / 62 / 92 cases, they *expected* -- according to their own words -- that the results would take longer to materialize than they in fact did.
In other words, that a sufficient number of results was coming back even before election day had been *unexpected*. FDA and Pfizer then did a last-minute change of that preregistered protocol to ensure that no intermediary results shall be published until after the election.
Saying that the withholding of intermediate results was a business decision done to decrease the likelihood of FDA rejection, is unadulterated horseshit. That's not how trials work, that's not how FDA approvals work. Tens of thousands of people already received the vaccines and placebos. The 62 cases *will* be arriving whether you publish or withhold the results of the first 32 cases. And the 92 and 170 cases *will* just as certainly be arriving too. And by the time you publish that your p < 0.0001 result upon the 170 cases does satisfy FDA's 52.3% threshold, it does not matter for the FDA approval whether the first 32 cases had failed their higher threshold (76.9% instead of 52.3% to account for decreased statistical significance). On the contrary, the case of intermediary results _surpassing_ that higher threshold was meant to be taken as a strong efficacy signal, encouraging protocol modification to start vaccinating placebo recipients.
Given that the 170 cases had a 95% result, just rubs it in how unlikely it would have been for the 32 cases (duly collected in October) to *not* exceed that 76.9% "higher threshold". Leaving unknowns aside and sticking to preregistered protocol - even if the 32 cases wouldn't exceed 76.9%, Pfizer wouldn't lose the FDA approval by publishing that intermediary result instead of withholding it.
In re: 26--as an LDS Utahn, I'm not sure there is a better reason generally than a vague "rise of secularism." We (American Latter-day Saints) track the broader American culture in many/most things, just with a significant delay. Economic worries about supporting a family and buying a home almost certainly play a role, both in people getting married later and in having children.
Social changes related to hanging out in groups vs. "serious dating" are fairly often referenced by the leaders of the church as an area of concern in terms of age at marriage, which I would think tracks with fewer children, but I'm not sure how strong an effect this really has.
I understand that some young married couples have been reluctant to have children based on worries about the wickedness of the world (an interesting mirroring of the "environmental concerns-->no children or small family" thing you posted about a few months ago).
The reason in my immediate family is ill health--all five of me and my siblings intended to have families more or less of that size, but chronic fatigue syndrome intervened.
Just adding onto this(as another LDS Utahn).
There is an offset to the Mormon/Utah birth rate compared to the US birth rate. I am going to assert this is because of "culture"(a mix of social, religious, historical factors). The US birth rate has drop between 2003 and therefore so will the Mormon birth rate. The problem that tweet shows is the Mormon birth rate has dropped by a lot more than the US birth rate.
I think it is a safe conclusion to say there has been a change in culture but I don't think there is any unique, new pressure in Mormon Culture. We still talk about marriage a lot. We still have doctrines about the importance of families. What else is left besides something physical(specifically in Utah) or a more vague "Rise of secularism"?
Here is one example that I think is a small shift in norms that has come from the American culture to the Mormon Culture. It used to be(10+ years ago) considered an "acceptable" question to ask a couple who had been married several years if there were going to start a family. Or if they only had one child if they were going to have more. Now I don't think it is considered and acceptable question to ask. That is not a big shift but it is indicative of culture norm and expectations changing.
Maybe it is not weird that the offset is regressing to zero but why was there an offset in the first place? I suspect there is at least more research about that question.
The real problem is that the graph doesn't show the Mormon birth rate, but the Utah birth rate.
I’m not Mormon (most of our friends are atheist or agnostic) but people asked us when we were going to have kids all the time, and when we had one we were regularly asked if we wanted more. I know this is sort of considered a bit of a nosy question nowadays, but I would say it’s still pretty normal. Maybe we just don’t project that we’re the sort to get bent out of shape if asked?
You *all* have Chronic Fatigue Syndrome? I'd heard there was a genetic component but even so that seems like truly rotten luck. My condolences
RE: 23, the "Effective Altruism" movement announcing that they're going all-in on supporting... Ivy League/Caltech/Swarthmore students with extra mentorship/funding/organization on campus smells a bit like reflexive elitism. Is it really not worth anyone's time to mentor students at a non-California flagship state school or seek out people who might be world-changers at schools who aren't already connected to strong pipelines to the elite? Seems like yet another way to hoard opportunities among a select group rather than seeking out new talent [and yes, they do say they will look to expand later, though I'm sure it will be to populist institutions like Cornell and Williams] and doesn't sound particularly altruistic in practice.
I am sure the "Effective Altruism" are more concerned with what will get results than what will be perceived as fair. If they feel they can double the Effective Altruism output of a student through a labor-intensive program, it makes sense to start with the people with the highest projected output.
That said, I'm skeptical their mentoring program will amount to much of a benefit at all.
I think the difference between your view (of what's more or less altruistic) and that of the folks in the linked post is that, roughly speaking, you lean towards equity while they lean towards rowing/steering, in Holden Karnofsky's schema https://www.cold-takes.com/rowing-steering-anchoring-equity-mutiny/
> Nate Silver, Tyler Cowen, and Garrett Jones come out in favor of “the public health establishment deliberately delayed the COVID vaccine by a month so it wouldn’t make Trump look good before Election Day”
Why would other countries play along with this? The UK for example only started injecting in December, well after the election.
And the UK was the first in the world with the Western vaccines (I forget when Sputnik and Sinovac/Sinopharm were first used), so there aren't any counterexamples, not even Israel.
Was the UK not dependent on Pfizer’s trial results? If that’s where the delay occurred, would it not affect the UK’s approval timeline?
I find this hard to answer because I don't understand why so many countries took the same amount of time to begin with - you'd think there would be lots of variation in how large/long a study they needed.
Phase III clinical trials, even the abbreviated "Warp Speed" version, are very expensive. Pfizer isn't going to do a separate one for e.g. Belgium. If Belgium says, "we're OK with 10,000 people, one months, and 20 cases", Pfizer is going to say "and you can pull that data out of the bigger study we'll be releasing to the FDA in October, er, November". If Belgium says "we need a million people, a full year, and a thousand cases", Pfizer is going to say "good luck with that". And it's unlikely the Belgian government is going to pay for it out of pocket.
The EU as a whole could have, but their standards are sufficiently similar to those of the US FDA that everybody seems to have agreed that one big trial for combined US/EU approval was the way to go, and October was too late to turn on a dime. The FDA is the 800-lb gorilla in this case, so every pharma company is incentivized to conduct their one big expensive trial to FDA standards, and every national regulatory agency is incentivized to base their approval decision on the results of an FDA-grade trial.
The theory is that political considerations influenced Pfizer to sit on the trial results (they stopped processing lab samples from mid-Oct. till the day after the election). Hard to see how other countries would approve without trial data. Pfizer just happened to be on track to be first to complete testing.
If Pfizer themselves chose to delay it, then presumably the health bodies of other countries wouldn't have a choice.
Re: 23, I agree with the thesis of the post but am not at all sure it will be possible to achieve. I was also active in Stanford EA in 2015 and I think a large amount of the effect really was just getting extraordinarily lucky in having a bunch of interested/interesting people around at the same time (many of whom never ended up directly involved in EA after graduation, to my knowledge, but were still an important component - five people does not make a sustainable campus organization). It's still worth putting effort in to attempting to make such a group coalesce, but it might simply not work if you don't get lucky with the people who happen to be around when you're trying.
Are there any games Buddha would have approved of, or at least tolerated?
Hypothesis: The purpose of games is to distract people from both daily life and any risk of enlightenment while playing.
Pascal says something like that: we're afraid ever to just sit alone without recreation for fear they'll think about the fact that they're going to die one day. So instead we all frantically race after constant amusements to fill our minds so no disquietingly serious thoughts about our soul and eternal fate can intrude.
Expert meditators can meditate while doing other things. If getting hit on the head with a stick can be a catalyst for enlightenment, why not getting a Yahtzee?
One-hand clapping contests?
Check out Thi Nguyen's book on the nature of games - he claims they are the art form whose medium is human agency rather than a physical medium. He analogizes games to yoga, as a way of trying out ways of structuring one's (physical body/mental agency) that come naturally to other people but don't arise in one's own life.
Not entirely sure how this fits into Buddhism though.
https://objectionable.net/games-agency-as-art/
2.) This reminds me of what Turkey is doing right now to deal with its own inflation. I can go into more detail but they're effectively creating financial instruments that compensate for inflation through a variety of methods I an get into. And then trying to move most normal people's accounts into those instruments. They're trying it not because the new minister knew no economics but because Erdogan has a weird ideas about economics that just happen to justify his policies. Though in this specific case his beliefs have some possibility of working. Basically, it creates pressure on the Turkish balance sheet (so the government is picking up the tab) in exchange for (theoretically) shoring up the currency's value by allowing the lira to create fake dollars/euros. This should, in principle, help restore some confidence and give the central bank (which has not been printing money) a chance to lower inflation. Effectively through wealth transfers to ordinary Turks. While this is an expensive policy that will probably put Turkey somewhat into debt it's probably worth doing that to fight high inflation.
The reason we can't do it here is that the US has been printing money and we don't have the same fiscal room with all the other spending that's been going on. But most importantly: Turkey's effectively printing Euros/Dollars through its fiscal instruments (because that's how the peg works). Brazil, iirc, did likewise, as did several American countries. But the US can't peg its currency to the dollar because it IS the dollar.
11.) I suspect it's a sign that Chinese censorship algorithms don't like homophones and someone delicately told the WHO that China would prefer their people didn't have a sneaky way to refer to Xi.
21.) There has been a lot of interesting movement in power lately. I look forward to seeing how it ends up. Though frankly, people get really politicized and polarized about power production such that I expect there will be significant barriers. I'm less qualified to comment on the rest. I also find it revealing that Fintech is not among the new technologies despite the fact we're objectively behind much of the rest of the world. Even on easily measurable things like the T3 standard.
22.) Why would this only happen in upper income countries? Plenty of poor countries strangle themselves with red tape and regulation too.
24.) I like to use the metaphor that engineers/architects/artisans are like musicians. If you give a skilled violinist a violin and sheet music and ask them to play the next day they'll come in and do a great job. If you give an unskilled violnist the same task they will produce much worse music. Yet they both spent the same time preparing the piece. If anything the skilled violinist probably spent less time. Likewise, an engineer's work product is much less a result of the specific task than their total accumulated work experience. The issue is that measuring artisanal output is often very difficult and nearly impossible outside of someone with the same skill. So corporate sets up arbitrary rules like 40 hours a week or time tracking which then get gamed. You can see the same dynamic in teaching music: the really great musicians do practice a lot but they move on from the grade school "three hours of homework" model pretty quickly once they're good/independent. (And to be fair, that kind of grinding is necessary to GET good.)
33.) I can confirm this is what YC's process is like. I think he might have a bit of hero worship going on. Don't get me wrong, they all came across as very smart and sharp. I think the maximal terms ("the smartest person I've ever met and I just met the smartest bankers in Africa") is a bit overblown. Though, to be clear, I'm not implying Sam Altman isn't highly intelligent. I agree it's cut short. There would have been pre-interviews and all that. They would have met other people. And investor stories are a favorite of mine.
34.) I have one of these. Though it's significantly less nice than that one. I also have a candle clock that tracks the time, day, and year. Though I don't actually run it that often.
#24 - I've had coworkers who do roughly an hour of work a day. I think they thought they were getting away with it, but I noticed, and it drove me crazy.
Why would you care? They milk the corp and the corp needs to be milked. Treat it as their contribution to fighting the rise of unfriendly AI.
A better question is why you're not doing it yourself, as you've seen it's easy, and using the leftover time for something actually useful.
Those people are a burden on everyone else. Solution is to fire them and let them starve.
They aren't contributing to society anyway, we can use those resources for more useful things.
I see you haven't been employed for long if you think of most of office work as "contributing to society".
Re the Amazon article (#10) and allowing consumers to post negative reviews -- I just saw Miracle on 34th Street and the man who thinks he's Santa Claus starts telling Macy's customers about bargains at other stores. Customers are so impressed that Macy's sales go up, and all the competing department stores start doing the same thing. Learn from the classics!
It's been a thing since at least 1999 https://en.wikipedia.org/wiki/Office_Space
Re #14: I would argue that in modern societies, especially of the welfare-state variety, it is unlikely that poverty is a major cause of cognitive problems; it's more likely the reverse. Although a third factor could cause both. Re #15: What group are we comparing to? Are we comparing to people incarcerated for non-violent crimes, or killers who weren't incarcerated, or the general population? Re #27: Maybe we need a 3-D graph.
Minor error: The Indian Jews converted to Judaism in the 1950s and 1970s. "19th century" is when they converted to Christianity.
Thank you. I was confused about this even after reading the Wikipedia summary
#1: No hopscotch, then? I'm out, Buddha. That's a red line for me.
#12: The original Florida man?
#22: Elon Musk was just waxing philosophic about this with Lex Fridman about how wars help sweep away red tape and bad rules, and since we don't really have many of those anymore (not great power conflicts, anyway), we're stuck with the rent-seeking incumbents. I didn't find this portion of the conversation particularly insightful--it wasn't anything you couldn't have read in, say, Reason Magazine over the last ten years--but how to achieve similar effects without the eight figure death tolls of two world wars is certainly a topic worth thinking about. The post-war German economic miracle was a real phenomenon.
#29: eh, was successfully crowd-funding new social media sites a legit problem in recent years? I don't care much for Facebook or Twitter, but "finally, we found a way to make Gab work!" or some other right-wing alternative site seems like...not our most pressing issue. My Crypto Bear card remains valid.
#33: This left me wanting more, too, but not in the "George Costanza always leaving on a high note" sense, but rather in the "you yada yada-ed sex?" sense, where they skipped over the best part.
Robin Hanson (who had once said he was a single-issue anti-war voter) has recently been warning that a lack of military competition may cause global government (including what he thinks is a currently-existing rather informal version of it) to rot: https://www.overcomingbias.com/2021/11/will-world-government-rot.html
Greg Cochran wrote that war keeps us sane, because correct ideas win out in non-verbal conflict: https://twitter.com/gcochran99/status/1127995103585267712
I am inclined to agree. Possible counterpoint, though: the Soviet Union collapsed with more of a whimper than a bang. I don't know how much their war in Afghanistan contributed to that collapse, however.
North Korea still exists.
While true, I don't think it refutes my point that correct ideas don't *only* win out in non-verbal conflict. Breadlines are pretty convincing evidence that something has gone awry, for example.
Chernobyl was probably the single most important contributor, which was enough of a bang.
It also doesn’t get rid of rent-seeking incumbents for the winner.
depends how desperate the war is - the allies got rid of much rent-seeking in switching to a total war footing in WW2
Really? Lobbyists got drafted? 3 letter agencies shut down or switched mission?Which practices were abandoned? My gut says rentseekers can surf that wave. The only thing that frustrates them is when things shut down and budgets dry up.
And of course, the military-industrial complex, the god emperor of rent seekers, was born at that moment. Well, not born, it must’ve been around before… came of age?
#26, former Mormon here - it's not entirely "rise of secularism" but it rhymes.
For the past 15 years people have been leaving Mormonism in *droves*. It used to be that almost no one left, and it was incredibly taboo to even talk about.
When I tried to leave in 2008 I knew 1 person in my entire life who'd ever left, and she was 25 years older than me. When I successfully left in 2012, I probably knew a handful of people who'd exited. Today, _everyone in Mormonism_ knows someone who has left. It's still a big deal to leave, but it's much less taboo.
In trying to prevent even more people from leaving, the church has become more "big tent". Some of the defining features of the Mormon lifestyle (big families, church literally every Sunday in person, women only as homemakers) have been deemphasized, to the point where "Buffet Mormonism" is now part of the nomenclature, where you're not judged as harshly if you subscribe to certain beliefs/rules but not others.
What that all means is fewer people view "have lots of kids" as a necessary and fundamental part to "being a good Mormon".
I like how you put droves in asterisk quotes. Up to 2019 there has still been a net increase in membership. https://newsroom.churchofjesuschrist.org/facts-and-statistics
Droves can be negative, per Amazon’s experience.
"Net increase in membership" is not necessarily inconsistent with people leaving in droves. You can still have a net increase so long as the gross increase (conversions + births) outnumber departures.
There's also the question of how you count departures. I've heard anecdotally that the LDS Church usually continues to count ex-Mormons as current church members unless they jump through some hoops to formally "resign" their membership.
There is data to back this. Difference between church records decline (~85% in 2010 to ~82% in 2018) vs. survey data (~88% in 2010 to ~72% in 2020) is significant.
https://twitter.com/ageofinfovores/status/1473884304723697667
https://twitter.com/ageofinfovores/status/1473832538917388290
The LDS church counts everyone who has ever been baptized as a member (unless you resign or they record your death), even if they haven’t heard from you in decades (and assumes you’re alive until you’re a centenarian), so their membership number simply isn’t meaningful. Why not publish global average Sunday attendance or number of people holding temple recommends?
1) Most people who leave don't withdraw their records from the church, so they're still counted in that chart.
2) Many of the converts joining are significantly less committed than the people who are leaving. They've maybe talked to the missionaries 3 times and then they get baptized. Ask anyone who's served a mission in Latin America to attest to this. So while this does help ebb the flow of people leaving per the chart, the total level of "committedness" to the church is absolutely down
The real issue with #26 is that it isn't a chart of Mormon birthrates, but birthrates in Utah County. Many non-LDS people have moved in, and that population change accounts for the decline in birthrate.
Re the Mormon babies or lack thereof. The study uses a location sample of Utah County, UT and Bonneville County, Idaho, as a proxy for “all Mormons.” This is a flaw for a few reasons. Utah County is a sub/exurb of Salt Lake City. Bonneville County is Idaho Falls, the nearest major city to Idaho National Lab. I argue Utah Cty is subject to middle/upper middle class downward population pressure making those Mormons anomalous. Also Bonneville county is population small but has larger than average population of lab employees, also subject to the class/education population pressure. And it’s a city even w/o the lab, hence another layer of same pressure.
The conclusion may reflect reality, I don’t know, but not for the reasons given. I’m not Mormon but I lived in/near Bonneville county for a while. Something that stood out at the time was age at marriage and childbirth was lower than the west coast (anecdotal).
Getting a good sample of Mormons spanning US/world geography, urban/rural range and income range would give a better picture.
To #26 as a former Mormon, formerly Utah county resident I think I can speak to this. First younger Mormons are having fewer children for the same reason everyone else is having fewer children - they're expensive. We're just highly motivated to have them for theological/cultural reasons. Anedotally I've watched the culturally ideal family (ie the one pushed in talks, considered normative etc) drop from 5-6 (my moms generation) to 3-5 (my generation), to 2-3 (the current generation). Simply put Mormons are subject to the same pressures as everyone we're just stubborn enough to delay the effects.
You're also equating Utah County and Utah with Mormonism which is less true than it used to be 20 years ago. Lots of non-mormon childless people moving in especially to the Salt Lake Area. Utah County is of course still a Mormon bastion but even that's less true than it used to be.
Finally the last 10-15 years have seen a pretty unpresidented swing away from orthodox belief. The Church failed to publish membership growth numbers for Utah for the first time ever this past year. Anecdotally the number of people leaving the Church has increased particularly over the pandemic, and exMormon support spaces are exploding.
Out of curiosity, why can't you link to Money Stuff on Bloomberg?
Maybe behind a paywall? But could say the same for WSJ so ¯\_(ツ)_/¯
Yeah, Bloomberg is soft paywalled, but it's easily circumvented with, for example, archive.is.
“ I would have been willing to let this pass if they had just said “unlikely” - somebody might honestly think 22% is unlikely compared to some hypothetical belief that it’s near-certain.”
Surely it should be compared to the likelihood of the average person killing someone. Less than 0.1% I imagine.
At any rate, we jail murderers for the crime they’ve already committed rather than worrying what they may do in the future. If some otherwise meek mannered husband kills his wife in a crime of passion, and all psychological profiles, indicate he will never kill again we jail him anyway.
That’s retributive justice for you, not to be confused with revenge which is a personal thing; it’s the state dispassionately deciding this crime is worse than that crime and applying different punishments. Long may it continue.
> we jail murderers for the crime they’ve already committed
That is the trigger event, but not necessarily the justification. And who is this “we”?
In retributive justice the “trigger event” (presumably you mean the crime) is the reason and the justification for the sentence. Do you really need to be told who the “we” is.
No, it's not. Yes, I do.
I don’t think engaging with you is going to be productive.
Probably not. Perhaps you actually have thought all this through, and don’t need me to get you to think about it. Maybe you think an event can provide a justification, or that a system of justice that uses punishment based on some other sort of justification should not be called “retributive.” I apologize for my inept attempt to inspire thought.
Well if you want me to think about you should do a bit better than one or two lines of drive by posts and pedantry about what “we” means. I don’t think I or anybody else knows what you think, it is only clear that you disagree with me. You have given nobody here any food for thought because you haven’t explained what you are arguing for, and very badly explained what you are arguing against.
"At any rate, we jail murderers for the crime they’ve already committed rather than worrying what they may do in the future."
Says who? If somebody commits a murder, it demonstrates a willingness to commit murder absent in the overwhelming majority of the population. Therefore we remove them from society so that they can't murder again.
Punishment and deterrence are part of the equation somehow, though, are they not? Consider that if I get convicted of stealing cars, the penalty for that in California is a maximum of three years in prison. If I've demonstrated the willingness to steal expensive stuff, three years in prison isn't likely to do much to protect society from my depredations, considering average male life expectancy is just under 79 years in the US. I could steal a lot of stuff in the other 76 years I have available.
That's not a given. 'Willingness to murder' is not some inherent unalterable property of an individual disentangled from the context of that murder.
While I understand that 'genetically cast in stone' is all the rage in some quarters, people do change, and can be changed by institutions devoted to rehabilitation rather than punishment/exploitation. (so not American prisons)
There’s plenty of examples where somebody could be genuinely rehabilitated and not a threat to society when imprisoned. Imagine a drunk driver kills two children. Before his trial he gets sober and voluntarily gives up his driving license. Psychiatrist’s agree that the remorse is genuine. This man is no longer a drunk or a driver, and can therefore never kill again as a drunk driver, so should he be imprisoned at all? Isn’t he not rehabilitated already. Is he not no longer a threat to society.
If the answer is that he should be imprisoned anyway for the crime then that is a belief in retributive justice. I’m not making this up by the way, the idea that our system is primarily retributive is a banal truism.
As for deterrence - retributive justice is part of that but deterrence can go beyond that and arrest the innocent for potential criminality or treason ie internment.
https://www.britannica.com/topic/retributive-justice/History-of-retribution
A recovering alcoholic is more likely to become an alcoholic again than a person who never was one in the first place, no matter how sincere their remorse; giving up your license removes the potential to drive legally but drunk driving is already illegal, access to a car is the thing that matters and that's very situational and much harder to prevent. Even in your toy example with no murderous intent, your remorseful repentant defendant is IMO much more likely that the average citizen to cause future deaths. I don't think that alone is reason enough to imprison, and deterrence also plays a large role, but it doesn't zero out.
Also, fwiw a repentant drunk driver might actually not go to jail for vehicular manslaughter in a lot of places, precisely because they aren't likely to hurt more people. Suspended sentences and the like are tools often used in reality and rarely discussed in these hypotheticals
I should probably stop giving examples, as it is pointless. So to generalise.
In all cases retributive justice matches the punishment to the crime regardless of other benefits to society. Or as Kant said “ “Judicial punishment can never be used merely as a means to promote some other good for the criminal himself or for civil society, but instead it must in all cases be imposed on him only on the ground that he has committed a crime; for a human being ought never to be manipulated merely as a means related to another’s purposes”
There are three reasons to jail someone for a crime, of which you have only mentioned two:
1. Retribution (you mention)
2. Removal from society (you mention and say isn't accurate for at least some criminals)
3. Deterrence (which you do not mention)
If you know that the maximum jail sentence for killing a person is 1 year (or go really absurd and make it a week), then that would genuinely change your willingness to kill someone. Maybe most people go from an extremely low likelihood to a very low likelihood and still never kill someone. Some portion of the population might go from ~0% to 5% or whatever, and dramatically increase the murder rate.
Jailing one time "crime of passion" criminals similarly to hardened career criminals is one way we get around the potential for criminals to lie about their motives. If they knew that fabricating a "crime of passion" could eliminate their jail sentence, then of course they would claim that. I would say that we knowingly jail people that are very unlikely to ever commit a crime again, as a way to maintain a system were reason #2 (keep criminals out of society so they can't offend again) can be effective even in edge cases where we are uncertain of motives. That such a system also deters potential other murders is certainly a strong plus.
Deterrence isn’t the same as retributive justice though. They are often related but retributive justice needs to prove guilt while deterrence could be achieved by rounding up people who may do something in the future, to deter them or others. Internment is an example.
I agree? Maybe you meant to respond to someone else, but I listed them as separate categories. The current system seems most interested in deterrence, followed by removal from society. Retribution, by comparison, is at best a distant third.
I still think retributive justice is the primary category. Deterrence could be achieved with more draconian laws and less concern about the innocent being affected. Any system that has higher penalties for more serious crimes is retributive.
That said the 3 strikes laws are more about taking people from society, rather than the seriousness of the crime, so in that case you are correct.
That's interesting. I would consider higher penalties for more serious crimes to also be about deterrence - we want to deter people from committing more serious crimes above those committing less serious. If all crimes have the same punishment, then you aren't sending a signal that some crimes should be avoided more than others, but we really would prefer more jay walking and petty theft and less murder and arson.
I do see your point about retribution though, and believe you are correct that it's more important to the discussion than I had previously said.
Without knowing more about the distributions of the killing-someone statistics, I don't know what would be a reasonable number. I could imagine that double the median likelihood of killing someone (should it only be criminal homicide, or all deaths, even justified ones?) would be fine if the numbers were low in absolute terms.
Do you think the average person has a 22% chance of killing someone?
“ but the will of God is really easy to thwart,”
Hey, nobody’s perfect.
20 is very encouraging! If most of the important info is not in digital text, and most of the digital text is unimportant, I never have to read my backlog of AI/safety papers to contribute to the field!
(This is somewhat sarcastic, but... am I wrong tho?)
You're not wrong. Network as much as possible instead, use papers strictly as references to look up very specific things - every now and again you'll duplicate work that someone else did already but you'll probably waste less time overall.
#7 somehow feels reminiscent of this Icelandic study:
https://www.theguardian.com/science/2017/jan/16/natural-selection-making-education-genes-rarer-says-icelandic-study
As for guppies, so for yuppies?
That's quite interesting.
I'd never heard before that the negative correlation between intelligence and fertility *wasn't* simply a matter of more educated and successful people having fewer children for cultural reasons--a finding that even *uneducated* intelligent people have lower fertility.
I wonder to what extent this is due to higher rates of unintended pregnancy among the less-intelligent, even at the same educational and socioeconomic level. Given that, until the last decade, c. 50% of all US pregnancies were unintended, different levels of vulnerability to unintended pregnancy could be a big factor in fertility differences.
(Recently, due to long-acting reversible contraceptives, the unintended fraction of pregnancies has declined to around 1/3. I would expect the negative correlation between intelligence and fertility when controlled for education to have significantly declined as a result.)
Contrary to popular myth, higher levels of education in developed-world women are not correlated with lower desired family size, although wealthier and more educated women are more likely to fail to achieve their fertility goals, due primarily to delayed marriage.
I suppose an innate, biological tradeoff between fertility and intelligence in humans, as seen in those fruit flies, could also be a factor, but it seems implausible that biological limits to fertility would be the main factor in a society where fertility among all groups is well below replacement--where everybody is having far fewer children than they're biologically capable of
This is a first whack, but biology is weirdly entangled. Why would reproductive hormones be related to the ability to regulate body temperature?
how much does "desired family size" actually mean if people aren't trying for kids until their 30s and thus predictably not having as many kids as they wanted?
> although this is weirdly short and leaves me wanting more
lsusr (https://www.lesswrong.com/users/lsusr) writings on LessWrong are usually like that, but there's a lot of them. He's a damn good writer/thinker, highly recommended.
That link didn't quite work but thanks for the pointer anyhow
ah, thanks for pointing that out! (no comment markdown on substack?)
#28 Having complained about mainstream media misuse of “no evidence” in the past, specifically that it seemed biased against right-wing political statements, I should credit NYT: in their obituary of Harry Reid, they noted that he famously claimed “without evidence” that Mitt Romney cheated on his taxes.
A perticular claim by a particular person being without evidence is different from there being no evidence for a position. I can say '1+1=2, trust me on this' or '1+1=2 (Russell, 1910)'.
Did they only add that on because he admitted that he lied about it? Not much credit to the NYT for that statement, given Harry Reid himself openly admitted it was a lie. Helping keep Mitt Romney from getting elected is one of the things Harry is most famous for, so I can see why they would include that tidbit anyway, but it's not much in their favor to include basic historical accuracy that no one denies.
"[Claim of the first successful deepfakes based hacking.] Looking through comments elsewhere, I think [this claim falls apart](...)" - actually those two links are about two different hacks (with the first not including any kind of detail beyond hearsay, so not verifiable). It sounds somewhat implausible that two different people would make up the same deepfake story. (There was indeed a wave of extortion-based attacks against semi-famous instagrammers at the time, see e.g. https://www.vice.com/en/article/93bw9z/bitcoin-scam-hostage-videos-instagram , but it wouldn't be strange for the attackers to use multiple methods.)
In any case, wouldn't be the first time deepfakes are used in a crime, see e.g. https://timesofindia.indiatimes.com/city/ahmedabad/deepfakes-replace-women-on-sextortion-calls/articleshow/86020397.cms
Those eigenrobot tweets are pretty much just content-free shitposting and as such it's not helpful to promote them as an example of how to talk about statistics and bias on the internet, IMO. They also miss the real point of story, which IMO is that the Markup was able to access a predpol company's dataset due to sysadmin incompetence, and did some handwavey analysis (they compared predictions to actual arrest and showed that it overpredicts in black areas and underpredicts in white ones, but admitted that it doesn't really prove anything since the police tried to deter crimes from happening based on those predictions, and the mismatch between predictions and reality could simply mean they were successful at it) to whip up interest in the story and push their point that this kind of data should be available for public scrutiny and not only when someone misconfigures a firewall.
I remember that article coming up before and I don't remember the point about it having a mismatch between predictions and arrests, instead just treating different amounts in different neighborhoods by itself proof of being bad.
Sorry, re-skimming the article it seems I misremembered it. The actual claim they make (alongside various disparate-impact claims such as higher predictions for black neighborhoods and higher predictions for poor neighborhoods) is "PredPol claims that using its software is likely to lead to fewer arrests because sending officers to the company’s prediction boxes creates a deterrent effect. However, we did not observe PredPol having a measurable impact on arrest rates, in either direction." ( https://themarkup.org/show-your-work/2021/12/02/how-we-determined-crime-prediction-software-disproportionately-targeted-low-income-black-and-latino-neighborhoods#stop-arrest-and-use-of-force-analysis ) and also it's in their methodology companion article, not the main article (although in their newsletter it was more prominent).
In any case IMO it fits into the pattern of Markup articles in general: they do a lot of hard work to obtain or create a dataset about the practices of some secretive corporate entity, do some analysis to show that those practices are potentially terrible (with the "potentially" part not emphasized), and call for more public scrutiny and transparency of the data. The analysis usually has a clickbaity title and doesn't actually prove much, but calls to attention to the lack of transparency around something that has a large public impact. I think that's on the net quite useful - companies preventing access to the data which could be used to judge the harm/benefit balance of their activity is a major problem, so raising awareness about it is important; the clickbaitiness is annoying but tolerable (that's how you get donor money I suppose).
So eigenrobot was not merely engaged in "shitposting" about the main article, rather than the companion article?
Funny you follow up the story about Amazon allowing customer reviews being fundamental to their success with a story about not offending Xi: Amazon wouldn't allow negative reviews of a book of President Xi Jinping’s speeches and writings on its Chinese website. https://www.reuters.com/world/china/amazon-partnered-with-china-propaganda-arm-win-beijings-favor-document-shows-2021-12-17/
2) I’ve been waiting for a decade for someone to explain how this story is made up / not the real reason for the end of (Réal) inflation. It just sounds way too good to be true. Surely if there’s an angle that deflates the hype, someone in the ACX comments would know it. Anyone?
Inflation is caused by bad fiscal & monetary policy but can be perpetuated by people’s expectations even after policy is fixed. Switching to the real was only half the plan. It was a way to stop expectations-driven inflation, but that switch was paired with major fiscal reforms to stop the root cause.
Even that, I do not understand. Sure, it hides the inflation in your paycheck and in your grocery prices. But does nobody notice that their savings get smaller and smaller no matter how vigorously they save? Or was Brazil just so poor that nobody had any savings, or so inflation-wracked that nobody *bothered* to have savings any more?
There might be something *in general* to the notion of having a social media site that rewards first movers. But "In Praise of Ponzis" makes me angry, because the glorious outcome of the cooperation he wants to build is just "if we all cooperate, we can trick Youtube's recommendation system".
Tricking Youtube's recommendation system does not generate social good, because you're just replacing one (possibly more authentic) winner with your own artificially generated winner.
Also, Youtube can just patch the hole in their recommendation system, if they haven't already.
Re 7: If you too strongly select for intelligence (or any other trait) you will do so at the expense of other beneficial traits and overall fitness. This is because the strong pressures on intelligence will cause you to select for genes which (for example) increase IQ by 0.5 points but decrease overall fitness 0.5% (numbers made up). What you want is a weak pressure on intelligence sustained, over a long period of time, in the presence of pressure on other traits you value. This causes genes which increase intelligence but don't decrease overall fitness (or decrease it only very small amounts) to be positively selected for, while genes which increase intelligence but decrease overall fitness are rejected. Time is important as it allows new mutations to enter the gene pool and be tested by these pressures.
In the fish example, if the only factor being selected for is intelligence, then you will select for negative, neutral and positive overall genes that increase intelligence. However, negative changes (changes that increase intelligence but decrease overall fitness) will be the lowest hanging fruit because they will be barely hanging on in the gene pool and thus have the strongest potential to grow in number. This dynamic also would incorporate the homozygous/heterozygous dynamic you talk about in your 2016 article.
What you want is a situation where all selection forces for the traits you desire are in a reasonable balance. So perhaps health, symmetry and intelligence are being selected for simultaneously, and any new mutation which helps one but harms the others too much is removed.
WRT Ashkenazi Jews; in normal populations, the reproductive returns from higher IQ were probably not strong enough to overwhelm the negative effects of being more sickly or having more sickly family members. However in Ashkenazi Jewish society, for a time, the benefits of increased intelligence outweighed the reproductive harm of having more sickly family members. This would have improved the selection for positive and neutral intelligence boosting mutations, but also negative ones - which would be proportionally boosted more until a new equilibrium was reached. I'm unsure if it would be easier to develop mutations which boost IQ but harm other aspects fitness versus mutations which boost IQ and help other aspects of fitness, but initially you're going to be tapping into the harm pool.
In your 2016 article you talk about how IQ genetically correlates with many good things. Apologies if you ultimately concluded this, but (severity of) mutational load makes the most sense. 99% of impactful mutations are shit. They're still going to get passed along for a while if they aren't too major. Some people (by chance) are going to inherit more of the shittiest ones and these are most likely to just harm many things (like 10% corroding every part in a car made of a certain type of metal). For example, if you have a mutation that makes your mitochondria 10% less efficient, that will possible reduce your height, health, intelligence and cardiovascular fitness. If you're unlucky enough to have a greater proportion of your shitty mutations be the shittiest ones, then they are probably going to drag everything down.
That Ponzi scheme article (#29) is rather bad. For one thing it confuses Ponzi schemes with things that aren't. The essence of a Ponzi scheme is that the system is funded by new entrants; when you give tokens to everyone who watches your video, in the hope of recouping on ad fees, that's not a Ponzi scheme, that's just you paying people to watch ads. If that worked out economically, advertisers would just cut out the middle-man and pay people directly.
And for actual Ponzi schemes, the article conveniently ignores the reason they are forbidden in many contexts, which is that most investors will lose their money - the system doesn't produce any value, just redistributes it from late entrants to early ones. You can play the lottery of entering at what you think is an early point, and on the system level you can fundraise off of people's propensity for gambling, but it doesn't seem like a revolutionary improvement over traditional fundraising. And anything on top of a blockchain needs to be a revolutionary improvement to work, because it needs to compensate all the intrinsic disadvantages of being on top of a blockchain.
26. I wrote a short thread on this, still some loose ends to tie up tho.
https://mobile.twitter.com/ageofinfovores/status/1473811731419779072
This is really interesting, and I too would like to see Lyman's Mormon preference data. LDS culture today is simply a world apart from what it was even 30 or 40 years ago. Enjoyed your post on Mormon moderation as well, it checks out from my POV.
Yeah, re:technology, I remember in like 2014 people were confidently predicting fully self-driving cars by like 2018. Now self-driving cars are not even on Noah's list. People are probably overly optimistic.
As a Mormon, and a current student at BYU-Idaho (see data for no.26), I feel like many of the other takes here in the comments are broadly correct. LDS members are not immune to broader cultural trends/pressures, though they are mitigated for a time by our community's relative insularity (like most frictional religious minorities). To the extent that lower fertility rates are a cultural issue, more than enough time has passed since the broad slowdown occured for the majority of demographic groups for cultural ideas to have permeated Mormon norms. To the extent that this effect is a product of mothers entering the workforce, etc., that has definitely happened within the Church, even though it again has lagged the same phenomenon in broader society. (I asked a few family members about what they thought the reason for this decline in the fertility rate could be and they all mentioned cost of housing, how expensive it is to rear a kid, etc. AKA the kinds of takes you would equally expect from a non-Mormon sample. A couple of them also mentioned that a large proportion of LDS friends in their age groups wanted to have kids but were unable to due to actual fertility issues.)
Mind you, this particular take seems a bit haphazard and I would quibble slightly with the methodology; I don't think that BYU-Idaho -- located in Rexburg, Madison County -- holds enormous relevance for Bonneville County. The latter is larger, and centered around Idaho Falls, which has grown tremendously in the last couple of decades and is located within a different micropolitan statistical area than Rexburg, i.e. may be a poor proxy.
This is probably why Romney is so supportive of certain aspects of BBB even if he wants them separated from the larger bill. His CTC was arguably better than Biden's for instance.
Re 2: Brazilian here, can attest that Planet Money episode was an good summary of Plano Real. I was eleven, so just old enough to remember visiting a newstand to buy comic books and checking a table posted on the side that told me how much each issue would cost that month. Spider-Man, Web of Spider-Man, and all the rest started going for "A23". One URV went from CR$647.50 to CR$2750.00 over four months and then it felt like inflation stopped.
It's still probably Brazil's greatest success story. It has become the sort of thing everyone is in favor of, like mom and apple pie and hating the NYT for doxxing bloggers, but both 2022 frontrunners were against it in 1994.
Ironic for politicians to be against a confidence trick.
On #6, discussed many of these with my girlfriend. The one on polygenetic screening for gender took me for a spin, until I realized it was literally China’s one-child policy, then it became quite clear why it’s a form of eugenics you shouldn’t support. I don’t think the sex ratio of society is a “slider” that we can be trusted to muck around with, at least not yet.
The problem is that it is a very simple tragedy of the commons given our society. For economic value each individual family would prefer men in a one child scenario but for society we should actually prefer women in that case. This is actually pretty consistent even at a higher cap. Of course what is a 10% problem in America on this issue is a 70% problem in most Asian countries.
That was certainly true in early ag and I industrial societies but in modern, wealthy societies does that remain true? Women have a way higher chance at obtaining higher education in America and I believe Europe.
Wouldn't women have outsized power in the dating market if there are way fewer of them?
6) Quasi-Mormon here, but not in Utah. I haven’t noticed anything at all. There are 3 families with 6+ kids that go to church with me, and everyone has kids or is planning on it. Maybe it’s a trend in Utah and that’s overwhelming the small progress down south?
#26: As a Latter Day Saint who spent a decade living in Utah, I definitely don’t think it was secularism or any other cultural explanations. The explanation that I find most plausible is the high cost of child care (including the opportunity cost of having a stay at home mother). It’s increasingly difficult to sustain a family on a single income and that difficulty increases with each additional child.
It intersects in a nasty way with high housing costs. High housing costs mean you need two incomes (even if the second one is only part-time) to buy a house unless you've got a pretty high income job, and you have to move really far out west and south from both Utah and Salt Lake Valleys to find more affordable housing.
If you're living really far away from either your or your spouses' parents' houses, then you can't regularly lean on them for child care. That in turn means you either need to pay for really expensive child care, delay child-bearing until you're more financially secure, or get back into work as soon as possible after your children are school-age - all of which tends to lower the number of children you go for.
I know a few Mormon couples who are just living in one of the parents' basements with their young children while trying to save up money for a house. They don't seem to be unhappy with it, though - the grand-parents provide some child care and get lots of time with their grand-children, and the parents save money.
Agreed! The cost of housing is probably the single largest reason why single income families are so difficult to maintain these days and that definitely impacts the decisions that couples make when deciding how many children to have.
The annoying thing is that double-income families are also a leading cause of high property prices. Sixty years ago, the median house cost (hand-wavey) what the median single-income family could afford, but now the median house costs (hand-wavey) what the median double-income family can afford.
At what point did it become unusual in the US for three generations to share a house? It was more the norm than the exception at the time and place (late USSR) I was born, I think, and almost certainly when my parents were born. How long ago would it have been common in the US? Or was it never particularly common in an urban setting?
It was common for fairly poor people in NYC before WWII, at least. I suspect WWII really is the dividing line, because afterwards is when suburbia exploded.
The immediate aftermath of WWII was characterized by a substantial housing shortage in the United States, so there would certainly have been quite a few three-generation households just because there was no other place for the twenty-somethings to live. But it's certainly plausible that the postwar economy was characterized by a vastly increased *demand* for independent living accommodations for young families, as soon as the housing could be built to meet it. So, probably 1950s.
I’m aware of three politically-related delays in the mRNA clinical trials. Please note that safety, diversity, and efficacy are somewhat separate issues.
– Safety: The decision to require two months of safety data, pushing Pfizer’s first date for an application back to 11/17/2020. The campaign for this was clearly motivated in part by Democratic concerns to deny Trump his October Surprise on the vaccine front, but … safety is important. So I’m not as worked up over this as some libertarian economists are. I can't say off the top of my head what the right amount of time to collect safety information was.
– Diversity: The Trump Administration’s decision in late summer to require Moderna to delay its clinical trial by a month to recruit a more racially diverse set of volunteers to test the safety and efficacy of the vaccine on different races. I’m sympathetic toward this delay: as I may have mentioned once or twice over the years, human biodiversity can be important in a variety of settings. Ironically, however, as far as we can tell so far, HBD doesn’t matter much for vaccines: racial equality more or less reigns in terms of response to mRNA vaccines. So the one month slowdown of Moderna by the Trump Administration in the name of diversity cost Trump his October Surprise, delaying Moderna’s announcement of its high efficacy until 11/16/2020.
– Efficacy: Finally, the vastly underpublicized decision by Pfizer to shut down lab processing of clinical trial samples from late October until 11/4/2020, the day after the election. Pfizer wound up blowing through even the third checkpoint when they let their lab get back to work the day after the election. This lab shutdown strikes me as the most egregious of the three politics-related interventions. The first two called for More Data but the third led to Less Data during a critical week at the beginning of 2020’s Winter Surge.
Without the lab shutdown, my best guess is that Pfizer would have announced the very high efficacy of its vaccine on Monday, 11/2/2020, the day before the election. Trump would then have spent the last 24 hours of the campaign trumpeting the success of his vaccine strategy.
Would that have switched enough voters in Georgia, Arizona, and Nevada to cause a 269-269 tie in the Electoral College and cast the decision to the House of Representatives with one vote for each delegation, probably favoring Trump? If that looked like it would happen, would there have been a violent and/or corrupt intervention to deny Trump a second term?
Who knows?
On the other hand, I don’t think Pfizer’s lab shutdown delayed the rollout of the vaccine by much more than one week and maybe just a few days.
The government decision had already been made due to the Democrats’ anti-VAXX fear, uncertainty and doubt campaign in the fall to require, in effect, until November 17, 2020 for enough safety data to be available to begin government processing of the application. However, announcement of efficacy on 11/2 rather than 11/9 would have given states and localities an extra week to focus on the big challenge of the vaccine rollout, which most botched until about the second half of January.
So I could imagine the vaccine rollout running a few days ahead without the Pfizer lab shutdown. The number of lives lost due to the lab shutdown sounds to me like in the hundreds but probably not in the thousands.
But who knows? Perhaps if Pfizer hadn’t shut down its lab processing, Democrats would have been the anti-vaxxers in 2021?
The alternative potential timelines rapidly spin in multiple directions. A timeline in which the mainstream media spent 11/2/20 and 11/3/20 screaming in rage about how the Pfizer vaccine was a dangerous conspiracy to re-elect Trump would have been ... different.
I suspect that when Pfizer CEO Bourlas is alone with his conscience, he says to himself that, sleazy as it sounds, his ordering the lab shutdown until after the election, wound up saving lives on net.
He may be right.
W.r.t. Vetocracy vs. Bulldozers, it's interesting that Vitalik, given that he lives here, doesn't bring up Swiss political system, which is heavily consensus oriented (vetocracy) but then there's always a referendum hanging above any decision, which, if the consensus decision was not good enough, would bulldoze over it (referenda results are written to constitution and thus take precedence over everything else).
Per the link at the end of #27, could someone please explain Quadratic Voting to me as though I’m an idiot? How could it possibly be more equitable if billionaires can conceivably buy more votes?
In a typical quadratic voting scheme, all voters gets a fixed number of tokens with which they can "buy" votes. You typically don't buy votes with actual money. A billionaire would get the same amount of tokens as everyone else.
While direct buying of votes probably doesn't happen to a very large degree (I presume some small amount of it goes on, what with reports of mysterious bunches of blank ballot papers being found in the back seat of cars and so on), I think that a token system is just as open to fraud and bribery as any current system.
Yes, Mr. Beff Jezos only has the same number of tokens as anyone else. Like all the workers in fulfilment centres who work on contract to Amazon, and the information campaign that Amazon engaged in to tell those workers of all the bad effects getting unionised would have on them.
Why not "information campaigns" about where they should best spend their tokens? Not that this is trying to influence them to vote the way Mr. Jezos would like, not at all! Just helping their valued employees get all the information they need about this new system of voting and how to make it work.
EDIT: Of course, on second thoughts, I have been thinking too small-scale. Beff doesn't have to care that he has ten tokens to spend on votes and I have ten tokens to spend on votes. If he cares about an issue, he doesn't have to buy tokens to buy votes, he can just buy the candidate.
By being in a position to donate to national, local, and individual campaigns, Beff can make it that Bob Bobson gets selected to run instead of Jim Jimson, because Bob has sufficient funding to pay for the election campaign and Jim is trying to cajole grannies to scrape a few cents out of a jam-jar to contribute. So if Bob has the right views on an issue, which just so happen to chime in with the views Beff has, then I get to spend my tokens on either Bob or Some Loser from the other party. So long as Beff is fairly sure Bob will win, then Beff doesn't have to spend his tokens on the issue of the Lesser-Spotted Mugwumper, because once Bob is returned to office, he'll sort that one out.
I mean, what you describe isn't great, but isn't something that the proposal for quadratic voting changes at all - all of those issues are present in the status quo, so they're not an argument against quadratic voting being an improvement on the margin.
You might be interested in Vitalik's longer write-up on quadratic funding and quadratic voting: https://vitalik.ca/general/2019/12/07/quadratic.html
The section "Public goods, public markets" explains when you want markets ("one dollar, one vote"), when you want elections (one-person-one-vote), and when you want quadratic voting.
> ... Now, you might notice that neither of the graphs above look quite right. The first graph over-privileges people who care a lot (or are wealthy), the second graph over-privileges people who care only a little, which is also a problem. The single sheep's desire to live is more important than the two wolves' desire to have a tasty dinner.
> But what do we actually want? Ultimately, we want a scheme where how much influence you "buy" is proportional to how much you care.
> ...
> Notice that only quadratic voting has this nice property that the amount of influence you purchase is proportional to how much you care; the other two mechanisms either over-privilege concentrated interests or over-privilege diffuse interests.
The problem is even in a complex semi-direct democracy scheme where you had 10000 votes tokens and cound spend in whatever combination you thought optimizing what you cared about the wealthy can trivially buy/influence enough votes to win. Because they control the distribution of news and they control, to some degree, the funding for "science".
Quadratic voting is another in a long line of hacks by the wealthier people who have something of a conscience whereby they get to remain wealthy while sidestepping some problem related to wealth concentration and/or the bad behavior of their wealthy peers. They never work. The solution is to limit wealth concentration somehow.
If you assume that the elite control what the masses think then literally any form of government is going to be de facto controlled by the elite. In practice, the populace often has a mind of its own - dedicated propaganda is limited in bandwidth and can at best enforce the party line on a few key issues - and thus some forms of democracy do meaningfully give power to the common people.
Taking America as a very imperfect example, your ability to influence presidential candidates is very limited (though less than it used to be) because the pool of candidates to choose between is strongly filtered. Your ability too meaningfully choose the minor functionaries of you local council, however, is very real, and a surprisingly large fraction of meaningful decisions are made at very local levels.
Well, count me in as the idiot standing, jaw agape, beside Antioch because this sounds to me like re-inventing rotten boroughs.
https://www.youtube.com/watch?v=Tkb9SIe4WWo
https://www.youtube.com/watch?v=EOEMRXI3sRs
https://www.youtube.com/watch?v=h6mJw50OdZ4
> the fact is they made a deliberate decision to make the process take an extra month, and that some four-to-five-digit number of people died because of this decision.
I love how you can just casually drop that your side killed 50,000 people just so that Trump's approval ratings would fall, but somehow you still think they're the good guys
Haha how many "brilliant AI journal extraction ideas" did you get?
I think we should bring those incense clocks back.
Ferbreeze already makes an air freshener that cycles through 3 different smells so you don't get used to them and I see no reason why you couldn't expand that to 12 aromas that change on the hour. The plug-in fresheners just use a 2.4W resistor next to the wick which gets warm and promotes evaporation of the volatiles. This plus a half-watt noctua fan is well within what a USB 3 or USB-C can supply.
I'll have to find out what 3d printer filaments are compatible with air freshener juice because I've seen the solvents from them degrading plastic before.
The pivot that juicero missed
So here I am again, to correct a mistake on vitamin D.
The study did not "find no effect". The exact quote is:
"61 (10.46%) individuals in the treated group died, compared to 386 (25.81%) in the non-treated group [odds ratio (OR): 0.597; 95% CI: 0.318-1.121; p=0.109] ... 45 (12.19%) individuals in the treated group were admitted to ICU, compared to 129 (26.27%) in the non-treated group (OR: 0.326; 95%CI: 0.149-0.712; p=0.005)."
So they found a very strong result in preventing ICU admissions. Mortality also had a reduction but it did not reach statistical significance. The correct probabilistic interpretation of that result is of course not "prevents ICU but not death", but rather "prevents ICU and will likely show significance on death as more trials are done".
"Related: AI Safety Needs Great Engineers."
* Who happen to live in Bay Area.
So annoying. There's always lots of talk about shortage of software engineers in general - but apparently no one cares about... not even trivial inconvenience - huge ones. Maayybe there would be less of a "shortage" if there was a) a legible process to getting a job which b) wouldn't constrain candidate population to a single location on Earth.
I agree this is annoying. Nontrivial to improve, though.
The thing is, a shocking amount of software engineering productivity depends on "soft skill" activities: figuring out product requirements, socializing designs, coordinating between teams, understanding user issues, etc. Some of this can be avoided if your team is a single person or just a few people, but that sets a crippling upper limit on how much the team can build. Software businesses in a competitive environment succeed by innovating faster than their competitors, so they practically have to embrace those collaboration costs.
I mention this because of two implications: 1. the much maligned "culture fit" really is important-- effective software engineering at scale requires high levels of trust and great communication as well as technical proficiency, and 2. zoom collaboration, though it's a big improvement over previous state of the art for remote work, imposes a substantial cost in engineering effectiveness. 1) explains why your a) is hard-- software hiring remains a bit cliquey and hard to scale-- and 2) explains why your b) is hard-- software work wants to be geographically concentrated. (There are additional factors at work here too, like the typical software job being under 10 years, so people have to build their life and career plans around having jobs at several different companies.)
The role of the Bay Area as sole tech hub is overstated. The Seattle area is also a huge tech hub, with headquarters for Microsoft and Amazon and large satellite offices for a bunch of other big tech companies. And most big tech companies also have smaller satellite offices in various second or third tier tech hubs in the US: Boston, NYC, Los Angeles, Raleigh-Durham, Austin, Lehi, etc.
There are also significant tech hubs in other countries. The product team I work on (Adobe XD) is distributed between four US offices (SF, San Jose, Seattle, and Boston) and three international offices (Bucharest, Bangalore, and Mumbai). At other big tech companies and teams I've worked for, I've also worked with teams in London, Vancouver, Hyderabad, Tokyo, and Beijing.
From what I gather, it's the startup ecosystem that's most heavily concentrated in the Bay Area. My guess is that a major factor for this is that most of the VCs and potential Angel Investors are here in the Bay Area, due to the money and networks coming from people who got rich in tech a decade or two ago when the industry as a whole was much more Bay Area centric.
Cohere is advertising jobs in Toronto (https://jobs.lever.co/cohere?location=Toronto) and a lot of Deep Mind employees work in London. So it isn't *all* in the bay area.
One of the major problems is that the companies are soul-sucking to work for. Most of the "productive" work being done is micro-optimizing sales and advertising revenue, which interests approximately zero software developers directly.
Then you get the political agitprop where the left-wing employees actively work to drive out the non-conforming (typically conservative-leaning) employees. This results in a lot of people leaving because they don't want to deal with that stress.
re: 19, even the article says their names were already out there publicly, in other articles and interviews, and they are often recognized on the street. if we lower the meaning of dox to mean "identified online but by someone we didnt want to" then sure, they were doxed"
Which, to be fair, is also what happened to Scott.
I wish these comments were sorted by link number. Not sure if there's a platform that would support that. I guess Scott could post each link as a comment on a reddit post.
Substack could do this trivially if they wanted to. Question is would enough people use it besides Scott?
The most trivial is for Scott to post 40 comments of the numbers 1-40 and people respond in the right place.
It might help, if Scott generates anchor points like "Discussion of Link # n", maybe add another one labelled "Junk, Spam & unrelated". Then it's up to the users, if they follow that structure by separating their comments to the respective threads (maybe enforced by Scott consequently deleting everything in violation of that rule)...
Actually, any early arrival at the discussion section could generate the anchor points too -- just a bunch of posts: "Disc of link n," "of link n+1" etc. If somebody does this next time around I will congratulate the daylights out of them.
One line in Noah's article on techno-optimism is particularly striking to me:
"Drones are the most obvious application; small battery-powered quadcopters may change the face of warfare and offer all kinds of delivery services."
Killing people will be much easier but so will be getting a late night MacDonald delivery.
Seems worth it.
A little bit further: "This has the potential not just to accomplish amazing feats (build a Mars base!), but to make space development a viable commercial activity in terms of tourism, research, and possibly even mining. It also obviously has transformative military applications."
And I've not yet read the part on AI...
Weird to talk like that's the trade-off.
In war, the point is to kill the other side with minimal loss of life on your own side. As Patton said, "No bastard ever won a war by dying for his country. He won it by making the other poor dumb bastard die for his country."
And of course better still is to convince the enemy that you really -can- kill his soldiers without endangering your own, as you might then not have to kill anybody.
Not sure where McDonald's enters into it.
#6 Melissa Etheridge had a son using sperm donated by David Crosby, a musically talented man with a long history of addiction.
The son went on to die at 21 of a drug overdose. Eugenics? Too real world tragic for me to play at a thought experiment.
Possibly not so much genetics as the environment; showbiz/entertainment world plus celebrity plus the blizzard of drugs in that world. Certainly having genes for addiction makes it worse, but even 'normal' people manage to fuck themselves up when they get rich and famous.
With regard to brazil and the URV how did they avoid the problem that once you price everything in inflation adjusted units you totally eliminate the ability of the central bank to affect the economy by increasing inflation or increasing interest rates. I mean it doesn't render central banks totally powerless but, one of the biggest economic dangers is that the stickiness of wages (wages tend to stay at the same level even during a downturn) can create a depression. Central banks generally respond to this in part by inflating the currency which has the effect of reducing the real value of the debts and the real value of wages but if everything is denominated in inflation adjusted units they lose that ability.
Ok, ok, this is way simplifying and if the country had a good credit rating it could still borrow and spend but surely creditors would also demand inflation adjusted units severely limiting the government's ability to engage in these kinds of interventions.
I don't think all contracts in Brazil today are made in inflation adjusted units so how did they extricate themselves from this problem?
> all contracts in Brazil today are made in inflation adjusted units.
To a large extent, they actually are -- although it takes the form of automatic inflation correction done annually by a pre-determined index (typically, a CPI measure).
I mean, if the status quo was a central bank uncontrollably printing money for decades causing massive over-inflation, then deliberately removing that capability for at least a little while was probably a key feature of the plan - no-one trusts that you won't just print more money on your say-so, but when you throw the money printer out the window suddenly money means something again.
Regarding the network heterogeneity hypothesis I'm tempted to say: well duh but does it really matter. Any number of people, including myself, have been remarking that this must, in some sense, be at play to explain the fact that the virus doesn't behave like the simple exponential models.
While I'm frustrated at the fact that it seems like all our digital gizmos mean that there should be the ability to do much more accurate population models in our epidemiological simulations I'm unconvinced it would really do much to affect how we've dealt with the pandemic. No one is really responsibly inferring substantial evidence from masks or other interventions from these hugely noisy data sets and if the total population can be thought of as a bunch of sub-populations with different reproduction rates (and occasional crossover) it's not clear to me that the standard model isn't still a pretty good guide to appropriate interventions as long as one doesn't take it too seriously (e.g. expect to be able to suppress hard for a quick win).
The article goes into detail but it's not a good guide at all. The author tested the standard model against their generated data and asked the model to estimate the effects of a non-existent lockdown.
"As you can see, according to the model and depending on when the lockdown was supposed to be in effect in each region, a lockdown that never took place and therefore had no effect whatsoever reduced the daily growth of infection by 7% to 10%. "
You can't say that a model that can be so easily fooled is a good guide.
Additionally, since the papers have been so heavily cited it does seem reasonable to assume that people are inferring substantial evidence from the data sets.
Re #21: These kinds of articles are easy to write if you don’t work in technology. There are a lot of things that look exciting that sort of work in a lab. There are a lot of things that demonstrably work but that would require significant changes in society and human behavior.
Turning something from “works in a lab” to “works in each of the 80 million iPhones I build every quarter for 3-4 years when dropped on cement every month or two” is a non-trivial problem. Many (most) technologies cannot move past this hurdle. Battery technologies in particular are extremely difficult to turn from the lab to the real world.
Autonomous cars are really exciting to people. But at least for the foreseeable future, they require very regular and structured environments to operate reliably and safely. You can do this in some places (e.g., airports, college campuses, maybe some sections of the freeway). But you can’t simply rebuild every road in America to be safe for autonomous cars. And why would you want to? There would be tremendous costs and no real benefit. So, yeah, we basically have the technology but putting it into the real world may never make any sense.
The worst example of this is nuclear energy. We literally have the means to end global warming, which I’m told by Netflix is metaphorically the same as ignoring an asteroid that’s about to smash into the earth. Nuclear has killed far fewer people (like, order of magnitude) than any other means of generating electricity (I suspect more people have fallen off of roofs installing solar panels than have been killed by nuclear). It’s been stymied from innovation by absurd regulation and unhinged people. So we literally know how to do something that will avoid the thing that people believe is an extinction event. And they still won’t get behind the technology.
This isn’t to say that we should be down on the future. We should be pushing for improved technology to improve our lives so much more than we are today. But it is to say that there are a lot of reasons you can’t just see something that’s interesting in some paper and think it will change the world. Or, rather, that the “changing the world” part is a lot more of the challenge than the figuring out the “science” part.
“works in each of the 80 million iPhones I build every quarter for 3-4 years when dropped on cement every month or two”
As we all know, dropping your iPhone on a bathroom mat, let alone cement, will muck it up beyond repair so yeah, I agree that the Holy Grail of making an iPhone tough enough to function is an impossible dream
With one exception, yes. People who like to predict technology rollouts should read some industrial history. Vaclav Smil has written many books that are useful as entry points into the literature.
The exception is solar PV. There have been big price drops for two decades, yes. We're now getting fairly close to the cost of the raw inputs in panel and system bills of material, so extrapolating the rate of decline is likely to lead you astray.
People who are excited about autonomous vehicles should ask themselves why. In nearly all cases the excitement is about reducing costs (stress; opportunity to do other things; even, bizarrely, parking fees) and increasing demand (disabled or nervous or very young potential road users). They should then ask themselves what lowering the cost of using roads and increasing the population of road users will do to traffic congestion, trip durations, and environmental harm.
Some of the excitement about autonomous vehicles is that the AI won't drink and drive - humans are imperfect drivers, and while we're still better drivers than the AI for now, at some point that won't be true and the switch to autonomous driving will save lives.
Reducing parking *fees* isn't the exciting thing - reducing the amount of *land* in urban areas dedicated to parking is exciting.
Re #1. I don't get the point of avoiding games 1 through 15. Avoiding 16 thought, sounds like a very good advice. Happy new year to all.
Nothing unique to explain with Mormons. They follow the same trends as the general population, just at a higher level. Fertility rates in a lot of developed countries, including US, have been collapsing in the past 10-15 years. In 2007, US fertility rate was 2.12. By this spring, CDC estimated it was just 1.64. Haven't done much research into matter, but seen a decent number of demographers suggest that it is a consequence of the modern electronic age both generally and for Mormons in particular.
20 "The Oxford Chemistry and Biology postdocs I met during bus rides to the science park (from my time at Oxford Nanopore) earned £35k at AstraZeneca 3. That's half of what someone slightly competent earns after four months of youtubing Javascript tutorials 🤷♂️."
That's nonsense if coders were getting paid that much we would never be able to hire anyone. Should we mistrust the rest of the article similarly?
I would also like to know where I can make 70 grand as an entry level webdev.
Definitely vaguely hoping I'm wrong about this and can find a much better paid job...
I think this might be a location thing - I can genuinely believe coders in Silicon Valley earn that much straight out of boot camp, but I doubt that coders in England make that much at entry level. Nevertheless, postdocs are badly underpaid compared to the amount of training required, and as a result doing a PhD is a poor financial decision.
While I agree with this, as well as the video featured,
"Glad to see the “we should try to stop global warming for altruistic reasons, but it’s not going to destroy humanity or kill your family” perspective picking up more traction,"
I was disappointed that she then refers to "subsistence farmers and people in the third world" as if we'll still have "subsistence farmers and people in the third world" 80 years from now.
Once the Startup Cities/Charter Cities movement gets going, I don't expect there to be any subsistence farmers nor will there be any "third world" after a few decades. The very existence of mass poverty is due to bad law and governance. Most of the world's population is still stuck in regimes that make it unnecessarily difficult to create entrepreneurial value.
Rather than cite the literature on economic freedom, it might be helpful to give a concrete example. Last year Africa had its best year ever for venture capital,
"The continent’s startups raised over $4 billion this year and minted five unicorns."
https://techcrunch.com/2021/12/30/african-tech-took-center-stage-in-2021/
Much of this was in Nigeria. What was the Nigerian government response?
"The Nigerian government is waging war on its technology industry. Within the last 12 months, President Muhammadu Buhari’s administration – through different ministries and regulatory bodies – has enacted a series of bans and operational restrictions in the country’s vibrant tech ecosystem.
Most recently, the Central Bank froze the accounts of four fintech platforms this August, claiming they were operating without a license and trading unethically sourced foreign currencies. The same month, the Nigerian Information and Technology Development Agency (NITDA) proposed introducing tax levies and licensing fees for tech companies as well as prison sentences for those who default on these payments. . . .
As economist Tunde Leye puts it: “The main thing we must understand [is] that the Nigerian elite have a consensus from the North to the East to [the] West to the Delta, that the formation of wealth independent of the political patronage system is a threat that must be exterminated.”
https://africanarguments.org/2021/10/nigerias-war-on-tech/
So called "developing nations" often have similar dynamics; Nobel laureate Douglass North, along with Wallis and Weingast, famously called this "the natural state,"
http://www.fnf.org.ph/downloadables/Douglass%20North%20Article.pdf
Because it is the natural condition in most societies for the oligarchs to prevent entrepreneurial capitalists from creating prosperity.
The more jurisdictions we can create where people are as free to create entrepreneurial value as they are in Singapore, Denmark, or New Zealand, the faster mass poverty will become a historical footnote. As Muhammad Yunus notes, at some point poverty will be only seen in museums, a peculiar feature of ancient human history. Children in 2100 will regard mass poverty as unfamiliar as children today regard a telephone booth or a slide rule, something that happened sometime before they were born. Their grandparents will talk about how hard life was in the old days, and thus provide "poverty within memory" for the grandchildren of today's global poor.
To shift more specifically to the climate debate: Especially when people talk about stopping fossil fuel use in Africa, these judgments should be based on empirical analyses on the relative benefits of varying rates of economic growth, on the one hand, vs. various climate impacts, on the other. There are undoubtedly scenarios in which it is more beneficial for Africans to continue to have access to affordable, reliable energy (usually fossil fuels) than to contribute marginally to lower temperatures 80 years from now. I've never seen a rigorous empirical analysis comparing such scenarios, but I wish those activists and NGOs working to end fossil fuel consumption in Africa would first undertake such scenario planning so they had an empirically informed understanding of the real human tradeoffs involved.
The "existential risk" of a population at an average GDP per capita of $80K each is very different from the "existential risk" of a population at $2K each. Climate impacts for populations at $2k would indeed be devastating. At $80K each, not so much. In 60 years, Singapore went from less than $500 to roughly $60K,
https://www.macrotrends.net/countries/SGP/singapore/gdp-per-capita
Personally, I see getting average GDP per capita up to the $80K range to be the highest moral priority. As Fred Turner notes, the goal is to "Make Everybody Rich,"
https://www.jstor.org/stable/24562532
Your assumption that the Charter Cities movement will actually get up off the ground and will work as idealized are the key axioms you're overlooking here.
Special Economic Zones have exploded in recent decades with respect to sheer number, size, and diversity of models. Economists have almost completely ignored them until the last 10-15 years. Private firms with expertise in developing and evaluating zones has also expanded dramatically in recent years. There are roughly two hundred jurisdictions around the world. Many are experimenting with zone-based strategies for increasing their attractiveness to talent and capital, with job creation an urgent need for political leaders in many nations.
Charter city visionaries provide a useful set of conversations about how new zones might achieve ever greater goals. But "charter cities" are simply a category of possible zones. The scale and diversity of zone growth is a fact, not an axiom.
According to Statista the number of zones has grown globally from 79 in 1975 to 5,400 in 2018,
https://www.statista.com/statistics/1203457/worldwide-number-of-special-economic-zones/
Many of these are banal, many are merely support for government cronies. But many nations are experimenting with new zone forms, including zones with distinctive law and governance. Some percentage of them will grow successfully.
The Adrianople Group, a private consulting firm serving the industry, just published the first map of zones around the world a few months ago,
https://www.openzonemap.com/
Before that we didn't even have a map showing what was going on. Neither journalists nor academics have much sense of the zone activity currently taking place around the world.
I've been involved with this movement for almost 20 years, and know leaders who were involved since the 1970s. The transformations that have been and are taking place are stunning. Given the growth trajectory, will we have 10K, 20K, or 30K zones in another decade? My bet would be closer to 30K and maybe more. Of those it could easily be 1% that use distinctive legal frameworks to attract capital and talent, probably more.
There is evidence for growth in the number of zones around the world. There is evidence that when zones are successful in one nation, there are attempts to copy such successful zone strategies in other nations. The safest prediction is that we will continue to see more zones, more diverse zone strategies, and that successful zone strategies will be replicated. We are thus already seeing, in Patri Friedman's terminology, the beginnings of a "Cambrian Explosion in Government."
Re: 15, when you look at the paper, it looks like they are talking about low rates of recidivism as compared to *other types of convicts who get released*, rather than compared to the general public.
IE,Someone released for murder is less likely to commit a violent crime than a random criminal convicted of a nonviolent offense. This is relevant if you are talking about relative sentencing guidelines between different types of convicts, and for pointing out false dichotomies between violent and non-violent offenders. In this sense the report is saying something useful, but of more niche applicability.
The tweet doesn't convey this nuance, and seems like an attempt to lie for that reason. But that shouldn't be held against the report itself.
That would explain the difference I think. Felonies like robbery or rape or drug dealing are generally committed by people who have made a career or lifestyle decision around committing that sort of felony on a regular basis, knowing the risks and considering the rewards adequate compensation. A significant fraction of murders, and possibly a majority, are committed by people who have one specific person they really wanted to kill (e.g. the man who they caught sleeping with their wife), and now that person is dead and lightning *usually* doesn't strike twice. There are also a fair number of criminal-lifestyle murders of e.g. rival drug dealers, but including the one-and-done murders in the average would make recidivism less likely for murderers.
I've seen this credibly asserted in other places, but I'm not going to dig up references just now.
Is that normalised by base rates of murder vs other crimes?
I think it's per capita
Re: 16 are we supposed to be impressed by eignenrobot here? He seems to be doing a really terrible job of the standard replyguy thing: make fun of each individual sentence out of context, never acknowledge that the criticisms of sentence 1 gets explained in sentence 5, when you do get to sentence 5 which disproves all your previous sneers, just sneer at it without making an argument and then move on as if it never happened.
Basically: yes, the fundamental issue is what your frame of reference for bias is. The algorithm is not biased relative to crime reporting and arrest rates, the journalist is claiming that crime reporting and arrest rates are themselves biased and provides several references to believe this, this is the heart of the question and eigenrobot just completely fails to address it or come to grips with it in any meaningful way.
The journalist is wrong though. Favorite example: I live in NYC. If I tell you that in 2019 the percentage of black-or-brown people in NYC is around 54%, what fraction of shooting *victims* that year would you guess were black-or-brown? Answer: 96%. And this wasn't unusual - some years it had been as high as 98%.
If we know that virtually all the people who are *getting shot* in large cities are black-or-brown and the police in those cities care about *reducing shootings* they should be spending most of their time in the neighborhoods where black-or-brown people live. That's where the victims are, that's where the perps are, that's where the shootings are happening. So that's where cops need to be.
"Biased cops" could perhaps explain some discrepancies in *arrest* rates but isn't going to explain that magnitude of discrepancy in *shooting-victim* rates unless large numbers of white people who get shot aren't reporting it to the police or somebody is consistently mischaracterizing white shooting victims as not-white. Either of those strike you as plausible? Is the idea that when victims go to the hospital or the morgue due to bullet wounds the police don't get notified just in the case of white victims out of, what, politeness? Or due to police not wanting to catch white criminals?
No, the obvious answer is the correct one: more crime happens in places where some groups tend to live than places where other groups tend to live.
UPDATE: Besides, if we look at their referenced survey the rate of non-reporting of violent crimes just isn't that different by race. It's a *little* different, but not much. In Table 5; the percent of violent victimizations not reported for "white" is 54%, "black" is 46%, "hispanic/latino" is 51%. So the crimes we're not seeing amounts to "about half of them, give or take" for the three groups we care most about. (survey is here: https://bjs.ojp.gov/content/pub/pdf/vnrp0610.pdf )
Re: 21 I'd 'get excited about' a technology that has a 10% chance of revolutionizing something I care about in the next 10 years, so I don't think 'get excited about this list of things' and 'only 10% of them are likely to happen' are in much conflict.
Re: 24 - adding my datum, same for me, couple hours of real work a week in a very high-paid job and never caught up to me.
Of course, there were some weeks with legitimately 50+ hours of real work, including weeks-long travel to factories and things, so I think a peak capacity model would probably explain a lot of why jobs are structured this way. You can't just hire 50 programmers for 2 weeks when you hit crunch, they need to already be onboarded and part of the corporate mechanism to be effective.
But it would be nice if they were only required to be in the office 20 hours a week on the weeks they only have 5 hours of work, with an understanding that they'll show up for the 50+ weeks.
Would they show up for the 50+ how weeks though? I'm an algorithm person, which isn't quite a software dev in that I usually don't have harsh deadlines, so maybe my intuitions aren't trustworthy here -- but I should think that if I structured my life around typically having 20-hour work weeks, I would have a hard time suddenly switching to putting in 50+ hours a week, through a mix of not having adequate childcare and not having adequate endurance for sitting at the office in front of a computer.
My take on the predictive policing article:
https://www.facebook.com/michael.wiebe.10/posts/10228172975016519
Stating the obvious: Scott needs a sidegig substack called "Wait, what?"
During my 3 hour flight home after xmas, I had 5 ideas that seem to me like good ideas, but maybe only 2-3 of them are actually good. This seems like a good thread for going off on disparate tangents, so here they are:
Idea 1: Biological carbon capture
Producing biomass is very cheap (90 tons +/- 30% CO2 per year per $4000 acre of farmland with minimal work using Paulownia trees or bamboo, 90% confidence). Preventing the biomass carbon from getting back into the atmosphere eventually is the hard part. This is a refinement of my prior ideas.
Step 1: Grow biomass
Step 2: Pulverize it
Step 3: Mix with water and alkaline clay (literally dirt cheap, 99% confidence)
Step 4: Anaerobically produce methane from this slurry via either microbes or synthetic enzymes. (related study: https://www.sciencedirect.com/science/article/abs/pii/S072320208680011X)
Step 5: Almost all the resultant CO2 is trapped in solution as carbonate/bicarbonate ions due to alkalinity (99% confidence). The outgassing will be nearly pure methane (higher purity than fossil gas, 75% confidence)
Step 6: The remnant may be marketable as topsoil because you've reduced its pH and added lots of humus to the clay. (30% confidence... actually after reading this I updated to 50%: https://forum.gardenersworld.com/discussion/1020379/i-have-alkaline-clay-soil(
Idea 2: Lyft and Uber should do just-in-time matching of drivers with passengers by physically bumping phones together with NFC. Then Ubers could pick up the same efficient way taxis pick up at the Las Vegas airport, and save everybody 10 minutes of waiting for some specific driver to come on-demand to a place that everyone already knew years in advance would have lots of passengers waiting.
Idea 3: Doing explicit Bayesian calculations in your head when you lack a good social intuition (using intuition as a synonym for what Daniel Kahneman calls "System 1"). Prior that a random girl is into me: 1:100. She starts fiddling with her hair the instant I look at her, multiply by 4:1. She offers me some gum unsolicited, multiply by 4:1. She keeps leaning over me to take pictures out the plane window, multiply by 8:1. She's with a guy but there are no PDAs, might just be a brother or platonic friend. Multiply by 1:2. She bumps my foot with hers in a way that was obviously intentional. Multiply by 4:1. Correct my numbers if they're wrong. I multiply it all out and get 64:25 Due to feminism/metoo I need pretty high confidence before making any moves (What % would be appropriate)? I should probably be reciprocating with a similar series of ambiguous gestures like she did, to gradually ratchet up our probability estimates about the other being into us, but I didn't really say or do anything. I would really appreciate a list of numbers I can use for these calculations.
Idea 4: Integrating a solar panel with a battery, a converter, and an outlet, and trying to get the entire package down below $200 like OLPC. To electrify off-the-grid places. Tesla's battery costs are around $150/kwh now, and excluding heating/cooling/cooking, 1kwh could be adequate for one household for one night. 300 watts of panel capacity could charge that in one day and have plenty left over for daytime use. Current cost of 300 watts of panel capacity is <$100. If you can get broadcast TV or internet access to remote parts of the third world (30" LCD TV = 60 watts) their ignorance will decrease, their welfare will improve, and their fertility will go down. Reducing fertility in the third world can reduce problems of hunger, longer term carbon emissions, dysgenics, and political instability.
Idea 5: I noticed there is such a thing as a "social justice quilt academy". I really don't know how to model this phenomenon of social justice people entering every hobby and making it about social justice instead of what it was originally about, except by analogy to old-timey religions that took every opportunity to proselytize. If SJW views of the causes of group performance gaps were correct, then there would be a massive utility payoff to proselytizing as much as possible. Likewise global warming. The ideas that spread will tend to be the ideas that can convince people there is a huge utility payoff for proselytizing as much as possible. The payoff need not be real. It only needs to be as convincing as religious ideas of afterlife, which is a very low bar.
For idea 1, the three questions are: who pays? If not the government, will they allow you to do it? And finally, is it better than other sequestration ideas.
For idea 2, it’s so obvious that I have to assume that Uber and Lyft executives have already discussed it and rejected it for some reason. I don’t know what those reasons are but two possibilities come to mind.
Maybe they think it’s too confusing, and want to keep the app as simple as possible - you call a ride, you get a ride.
The second is game theoretic - it allows riders to ‘steal’ rides. If you do choose to call an uber, the uber driver could cancel your ride and give one to whoever they run into. And it’s not as easy as “just make sure you’re at the pickup point in time”, because the driver could encounter another potential passenger before even getting there. This makes the service overall less reliable.
For 1: my back of the envelope math suggests it MAY be very profitable at current methane prices, but there's a lot of engineering to do and local circumstances that need to be nailed down to reduce the error bars around cost estimates. Methane produced could be up to 1/6 of the mass of CO2 removed from the atmosphere, which at current methane prices of $1.35/kg would be around $20,000/acre/year.
If the engineering gets good enough that converting wood to methane is no more difficult than converting grapes to wine, then here's a little sanity check on the pricing. One acre of grapevines produces 4 tons per year. Average grapes are worth $900/ton. (Napa/Sonoma and other prestigious regions can cost 2-10x that) So 4 tons of grapes is worth $3660. This makes about 2880 bottles of wine. If they're two buck chuck, that's $5760/acre/year. That includes some retail packaging which costs at least 25 cents per bottle. 5760-3660-2880*0.25 = $1380 estimated cost of industrially converting 4 tons of grapes to wine.
With bamboo or paulownia trees, you're dealing with a much larger biomass per acre per year, but the biomass comes in larger pieces that probably make them easier to process, and you can play fast and loose because you're not making a product for human consumption. Worst case we multiply the $1380 by (90tons/4tons) and get $31050, so that the methane sales would cover two thirds of the costs even before you start selling carbon offsets. (1/6)*(44/16) =45% of the carbon is released as methane, and the rest stays in the ground. You sell carbon offsets at $20/ton for the other 55%. 90 tons * 0.55 * 20 = $1000/year extra. So basically to break even without subsidies, industrially fermenting wood into methane needs to be 33% easier than fermenting grapes into two buck chuck.
LOTS of burdensome details here, so low confidence.
For 2: If the driver encounters another potential passenger before getting there, that's a good thing. It saves time in the aggregate. If the distance to the old passenger is X minutes, and the average pickup time is Y minutes, then the new passenger saves Y minutes, the driver saves X minutes, and the old passenger loses Y-X minutes.
Y+X > Y-X
So reserving particular drivers for particular passengers in advance will worsen aggregate time wasted. Maybe there should be another tier of service for people who are extremely sensitive to variance (catching flights, etc). But reserving-by-default seems like a huge waste of resources. The driver app gives drivers these "streaks" that function kind of like daily quests in world of warcraft. They can give drivers a quest to go to the airport, without assigning them a particular passenger yet, to precisely calibrate the number of drivers that are there in advance of pickup requests and make airport pickups a lot faster. They can do this with or without the NFC thing. Instead of being 50x slower than taxi pickups at airports, maybe uber can get it down to 5x if it stops unnecessary reservations.
I mostly don't believe in appeals to "if it was a good idea they already would have thought of it". Ever watch James Burke's Connections series? A lot of the history of science is basically: thousands of people were aware of X and Y, but nobody thought of putting them together properly, until someone did something that was totally obvious in retrospect.
I need to hurry up and finish R:A-Z (13% and 65 hours remaining on my Kindle. Sorry, I'm a slow reader, so I've developed skills of getting as much mileage as I can out of as little evidence as possible) so that I can start Inadequate Equilibria which addresses efficient markets.
I just had another idea in the same area as #1 that's 10x better. Just genetically engineer Paulownia trees to dump carbs >1ft underground (by growing unnecessarily large root systems or directly excreting carbs). Then your only marginal cost is planting them, which costs $100/acre. Those trees regrow from the root system so that they never need to be replanted. So one planting could potentially remove 30tons/acre/year of CO2 for 333 years, which comes to ~$0.01/ton for the planting and $0.2/ton to buy the acres of crappy alkali soil. The soil quality will improve due to de-alkalization and introduction of organic carbon, so the resale value of the land will be a lot higher than if it were idle. Mortgage interest rates are lower the usual appreciation in the price of land, too, so maybe it's not appropriate to count any part of the land as a cost.
Tree-parasites could ruin this plan. Monocultures are vulnerable to parasites. But the carbon that ends >1ft underground in alkalai soil is probably pretty safe from being released into the atmosphere by a plague of parasites.
What's NFC? What's OLPC?
NFC is Near-Field Contact. Two phones exchange data via a high-bandwidth but low-range transfer. Requires them to essentially be touching.
OLPC is One Laptop Per Child.
Does anyone know what the recidivism rates are for fraud?
https://www.ussc.gov/sites/default/files/pdf/research-and-publications/research-publications/2016/recidivism_overview.pdf
"Although the specific numerical offense severity determined under the
sentencing guidelines appears to not be correlated with recidivism, the type of
crime the offender committed does have some correlation with the risk of
future crime. Offenders whose federal offense involved firearms were most
likely to be rearrested (68.3%), followed by those arrested for robbery (67.3%),
immigration (55.7%), drug trafficking (49.9%), larceny (44.4%), other (42.0%),
and fraud (34.2%)."
Happy new year!
so.. extremely unlikely ;)
On that number 24, man I've felt horribly guilty for being that guy in my career. I don't know how to feel about the idea that maybe it's fairly common.
I've been that guy due to health issues before. Brain fog and headaches from narcoelpsy type 2. Feels bad
I worked with a software engineer who could fall asleep with his fingers on the home row. Not sure how he did that but I am pretty sure he suffered from some form of narcolepsy. He had a wonderful personality though. We were partners in the company golf league . Had some great times hacking up the course together.
It's not unusual to happen to get no work done if you're working on a hard problem and everything fails. It's another because you faffing all day. I've had both happen to me. What time is it?
Re Mormon birthdates: According to https://religionnews.com/2019/06/15/the-incredible-shrinking-mormon-american-family/, Mormon leaders used to be extremely opposed to birth control, and they gradually stopped talking about it in 1980-1990's. Is it possible it took about a decade or two for their influence on the population to fade?
26. I am a member of The Church of Jesus Christ of Latter-day Saints. (We prefer to be called by the full name of the church rather than "Mormons", but I'll give you a pass this time.) I have 7 children and my wife is expecting our 8th in May, and we feel strongly about this. We love having a big family though it is countercultural in this day and age. Other commenters have pointed out that Utah is growing and the percentage of Church members is dropping, suggesting that the data doesn't actually reflect birthrates among members of the Church specifically. However, I don't think that fits the data. An outsized percent of the growth state-wide has been in Utah County, so this theory would predict that birthrates in Utah County would be falling faster than in other Utah counties - but that is not the case, at least for the counties with data shown at the link. Instead, birthrates in all of the available Utah counties are falling at a similar rate. It certainly is true that societal trends that impact everyone impact members of my church too, and pull the birthrate numbers down. But I think that trends in church teachings (softening the "multiply and replenish the earth" rhetoric) are also accelerating the trend toward lower birthrate. These changes in teaching approach are likely in response to, and out of sensitivity to, increasing numbers of unmarried and childless church members. But the softened rhetoric of course also amplifies the cycle. You can dig up old quotes from the Prophets in the 1950s and 1960s that are very straightforward about the need for and benefits of having children. Those old quotes now seem like they came from a different world. For decades now we didn't hear very much about this doctrine at all. Just in the last year or so some of the Apostles have felt a need to speak up (probably in response to the tanking birthrates we are now discussing), but that is still tentative and a marked change in approach. Even on Mothers Day at the local level, the speakers in our ward spend less time thanking mothers than they spend making sure everyone knows that those who don't have children are also ok - which is true, of course, but illustrates the point. To step back a little, and of course I am biased, but I can think of no other action that increases the overall utility of the world than bringing children into it in the first place. Existence, even in the third world, is a utility net-positive. Existing in the world vs not existing is a giant step change in utility, obviously. In our society we underestimate the value of potential existence, which almost never gets factored into our policy math. Any policy that lowers birthrates (e.g. car seat regs, or anything that hurts prosperity, or makes it harder for single-income families) has a huge unaccounted-for negative impact in overall utility via potential lives not realized as the birthrates fall as an unintended (and un-thought-about) consequence. Should be an opening for EA, but I doubt it would mood affiliate.
Interesting life story, thanks for sharing. But concerning utility, do not forget Ecclesiastes 4:2-3:
2 Then I praised the dead who are already dead more than the living who are yet alive;
3 and more fortunate than both is he who has not yet been, who has not seen the evil work that is done under the sun.
Ecc by itself is a very poor argument for antinatalism, if that's what you're trying to get at.
Well, Ecclesiastes is a holy book in all the three large Middle-East religions....and the general idea is not too different from that of Buddha either.
Besides, he writes were well. (Or all four of them. There is probably more than one author behind the text in Ecclesiastes.)
1. One line from Ecc doesn't overturn the entire rest of the texts of Judaism, Christianity, and Islam. Even Paul's exaltation of celibacy is a better root for any anti-natalist argument in Christianity.
2. Ecc and Buddhism have very different teachings. You're clinging to one vague semblance and trying to connect them. Ecc argues for a lifestyle of simplistic and passive hedonism, which is pretty close to the exact opposite of Buddhism. It is good that people are born into this world so long as Buddhism exists, because if they were not born into this world they would not be able to attain Enlightenment.
The arc of Ecclesiasties is realizing the weight of existential dread, realizing that one should be content with one's small place in existence, and in the end find peace in "fearing God."
The passage you quote is one of many from the first part of the arc, "feeling the weight of existential dread."
But you don't quote anything from part 2 or 3 of the arc. Part 2 might include something like Ecc 9:4-10,
> 4 There is hope, however, for anyone who is among the living; for even a live dog is better than a dead lion. 5For the living know that they will die, but the dead know nothing. They have no further reward, because the memory of them is forgotten. 6Their love, their hate, and their envy have already vanished, and they will never again have a share in all that is done under the sun.
> 7Go, eat your bread with joy, and drink your wine with a cheerful heart, for God has already approved your works: 8Let your garments always be white, and never spare the oil for your head.
...and of course the whole book concludes with,
> "13When all has been heard, the conclusion of the matter is this: Fear God and keep His commandments, because this is the whole duty of man."
There is nothing like a bit of theological discussion early in the morning, or late at night (depending on your time zone).
In you interpret Ecclesiastes as a narrative arch, as you do, then the final part naturally takes precedence. Since the end of a text is where an author sums things up, or extracts a final moral. In this care: Fear God and do his bidding! (12:13.)
However, the fact that the general “feel” in Ecclesiastes changes somewhat toward the end, is one of the reasons why many suspect that the text is written by different authors. Up to four authors have been suggested.
If you chose this interpretation, the first part(s) becomes as important as the last. It is just the voice of a different author.
I have always suspected that the final sentences (12:13-14) were added by a later author who recognized the beauty of the prose of the preceding authors, and added this “theologically correct” conclusion to sneak this subversive text past the eyes of those Keepers of Holy Texts who decided what to include in the Good Book.
That said, I would not claim that the “separate author” interpretation has a much stronger empirical basis than your “narrative arch” interpretation. Since the author(s) are long dead, we will unfortunately never know.
Be that as it may, any text can in any case always be interpreted in a multitude of ways. In particular a text as rich and magnificent as Ecclesiastes.
And I do not mean to belittle the late author. After all, he was the one who penned this sentence:
“There is no end to all the book writing, and too much study wearies the body.” (12:12)
I have often felt to weep with the author of Ecc 4:2-3 when listening to the stories of people who have suffered unspeakable trauma and oppression, so I understand the sentiment. But even in these cases, there is hope, forgiveness, healing, and growth possible for each and every one (through Christ, in my belief system). The Sermon on the Mount is a better reference for the downtrodden.
On the other hand, Christ does have this to say about those who abuse children: (Matthew 18:6) “But whoso shall offend one of these little ones which believe in me, it were better for him that a millstone were hanged around his neck, and that he were drowned in the depth of the sea.”
Re 2.: https://en.wikipedia.org/wiki/Brazilian_real says the Real, which replaced the URV, was worth exactly one U.S. dollar at the time it was introduced, so it seems the URV was tied to the dollar before it was converted to the Real.
The article on NRP is certainly interesting, and it explains the trick behind the URV well, but omitting the USD binding seems to give the impression that the trick's basis was purely psychological. However, binding one's currency to a stable foreign currency seems to be a pretty basic economic trick - which, of course, has a psychological side too, because the reason the inflation does not come back after it is not anymore tied to the dollar, seems to be the expectation of the people using the currency.
#30: The interviewee has an MA in political science and is a policy analyst in an organisation with a track record of minimising environmmental and social damage from resource exploitation). She does not have a track record in energy, climate, environmental, or social science.
My prior belief is that it will be full of BS, strawman opponents, misinterpretations and innumeracy.
... And my post is that the interview is full of strawmen by caricature ("children are little carbon emitting machines"), strawmanning by extremising ("stop reproducing"), and misrepresentations (a focus on birth rate, when the problem lies almost entirely with the 250 million highest-income people, who already have below-replacement birth rates; massive overclaiming, and innumeracy and ignorance of scaling timetables when it comes to technological matters), and distractions from the issue (e.g. talking about medical progress in the mid-20th century).
There are also outright falsehoods being brought out: the idea that we need technological breakthroughs to "solve" climate change (and therefore need more people to have those ideas). We have all the needed technology now. Ideas, from now on, have negative value. Climate change *is* solved...in theory. All that's left is the politics.
She *is* right that our collective non-climate-affecting depradations are pretty bad too, and those chickens will also come home to roost this century. That doesn't help her argument.
> But here’s a story about someone selecting guppies for intelligence (successfully) and finding that they had smaller guts and lower fertility.
Aren't humans descended from animals that had larger guts (due to less meat in their diet, and not cooking it)? Also, while no-one knows how fertile extinct hominins were, the probability for pregnancy per copulation is probably lower from humans than most mammals.
Humans invented cooking 250,000 years ago (and it probably didn't take too long for their guts to shrink after that). But they didn't get smart enough to invent a new tool more than once a millennium until the mesolithic ca 20k years ago. So if someone is suggesting that humans got small guts *because* they got big brains, the timeline doesn't support that.
A new tool once a millennium is still a lot more than the vast majority of species out there...
"Glad to see the “we should try to stop global warming for altruistic reasons, but it’s not going to destroy humanity or kill your family” perspective picking up more traction"
For the longest time, this was the *default* position among progressives and environmentalists. In Al Gore's documentary "An Inconvenient Truth," the big threat posed by climate change was an increase in the amount and severity of flooding, hurricanes, and other natural disasters due to the melting of the polar ice caps. In Kim Stanley Robinson's book "New York 2140" - which was praised by environmental activists as being a starkly realistic vision of the devastation that climate change may cause - Manhattan is partly submerged and has become a poor and squalid place due to flooding and infrastructure damage, with Denver (a landlocked, high-altitude mountain city) taking its place as the economic capital of the nation. These are certainly grim predictions that present a rather bleak future, but they don't come anywhere close to the claim that humans will literally go extinct or suffer an apocalyptic catastrophe that kills billions and destroys modern civilization entirely.
But somehow, over just the past few years, the idea of climate change as an *existential threat* to humanity has become a mainstream idea among online progressives and leftists. I still don't think it's the majority opinion among left-leaning folks, but it's definitely a popular one. A recent poll I saw in a progressive Facebook group was about evenly split between "climate change is an existential threat that could cause human extinction" and "climate change is really bad, but it's not an extinction risk."
Of course, the extinction narrative has virtually no support from actual climate scientists. Only a vanishingly small minority of climatologists actually believe it will cause the literal extinction of humanity, and they tend to be viewed as crackpots by the rest of the scientific community, just like the denialists on the other end of the spectrum. Roger Hallam's claim that "climate change will kill 6 billion people by 2100" was extensively debunked by scientists, and the fringe theory of a "runaway greenhouse effect" that would turn Earth into an uninhabitable wasteland was dismissed by the IPCC as a literal impossibility. Yet that hasn't kept the extinction narrative from gaining traction among the activist crowd since the mid-2010s. Any ideas why this might be the case?
I think a lot of public discussion of things tends to flatting them into either "this will kill everyone" or "this is great for everyone" or "this is completely irrelevant". This is probably related to the "halo effect", where anything that you like for one reason tends to get assumed to be good for other reasons as well.
Per #26, I'm an active Mormon who lived in Utah for 17 years (10 of them in Utah County). I don't think the birthrate decline has much to do with secularism or a reduced percentage of Mormons in the state. It's much more simple than that: Utah is booming!! Utah County especially is growing like mad. What was once a collection of small towns throughout the Provo-SLC-Ogden corridor is not an unbroken sea of increasingly dense suburbs. Lehi is becoming a tech hub of its own (Silicon Slopes!). Utah feels more and more like California every year.
The result of this is that Utah is much less isolated than it used to be. There are far more people with far more connections to the outside world, and it has given the state a more metropolitan feel. The state is still dominated by Mormons. In Utah County especially, you can still assume that everyone you see at the grocery store either is or used to be an active Mormon. But with a more metropolitan culture, people just aren't having as many kids. My dad was one of eight kids, I was one of eight kids, but my wife and I only have three. That's a common story. There's a lot more to life now that there are things to do.
The older generations still feel very strongly that women should stay home with their kids (my wife and I get this talk from both our parents), but more and more women are working now. Managing a huge family is very difficult to do when you also have a job. Religious changes are mostly a red herring. It's just simple affluence.
#6 Pleasantly surprised that lots of people are supporting some obvious utility gains, and not getting derailed by the worst argument in the world (P1: nazi death camps killed people partly for eugenics. P2: nazi death camps were super evil. P3: eugenics means any organized attempt to improve the gene pool P4: therefore any organized attempt to improve the gene pool is super evil. To see how ridiculous this is, compare: P1: nazis invaded eastern europe partly for agriculture P2: the nazi invasion of eastern europe was super evil. P3: agriculture means any organized attempt to improve the food supply. P4: therefore any organized attempt to improve the food supply is super evil.)
#7 the article says they were selecting for bigger brains *relative* to body size. That implies they weren't just selecting for bigger brains -- they were selecting for bigger brains and smaller rest-of-body. In that context the result of smaller guts is less surprising and I should update much less towards intelligence genes being tradeoffs.
#8 seems very plausible but in its current form it's too vague to constrain anticipation. The south Korean government had everyone's cell phone GPS data in real time for test and trace purposes. They could use that data to characterize population structure and get better at modeling interventions. Models definitely don't need any more free variables that researchers can tweak to arrive at whatever conclusion they already wanted to arrive at.
#9 is probably a joke (90% confidence that the author doesn't actually believe that). But there's no parody disclaimer in the profile. It's probably unvirtuous to make an ambiguously-parodic twitter account that tweets straw men which people might mistake for the genuine article and update towards negative karma for various tribes that never espoused that straw man. (generalizing from fictional evidence, ethnic tension)
#12 Bowles' life story makes me want to take the Bran pill so I can observe history in first person. That'd be better than all the other pills (https://slatestarcodex.com/2015/06/02/and-i-show-you-how-deep-the-rabbit-hole-goes/) except Orange.
#15 The subset of murderers who ever get released is a very nonrandom sample of murderers, plus they've been in jail so long that they've aged out of their peak violent crime years. So low-ish recidivism of murderers doesn't say much about what the recidivism rates would be in the context of some radical anti-incarceration reform, and it's probably a mistake to use the former to argue for the latter.
#16 The NVCS shows that racial disparities in arrest rates correspond to disparities in actually-committing-crime, not police bias, so assuming neighborhood disproportionality = bias is not just an unwarranted assumption, it's actively undermined by the available evidence. Black and Latino neighborhoods also have a younger average age, and youth commit several times more crime than old folks. (having a not-yet-fully-developed brain is a bit like having a low IQ and correlates with many of the same things: impulsivity, aggression, criminality, low productivity)
#21 My most pessimistic take on this analogizes big tech to land speculators who got in early, and are now just collecting rent on their natural monopoly. But worse than actual land speculators, because they use all that money to hire the smartest people to figure out how to make people click on ads 1% more instead of actually innovating. So there's a brain drain from everything else into figuring out how to make people click ads 1% more. Related Neal Stephenson speech which is great: https://www.youtube.com/watch?v=TE0n_5qPmRM "I saw the best minds of my generation writing spam filters"
Noah blaming energy costs for the stagnation seems dubious. Total US energy spending is only 5.7% of GDP (https://css.umich.edu/factsheets/us-energy-system-factsheet). US total government spending (federal+state+local) was 45% of GDP in 2021. A new government program that wastes an extra 6% of GDP could have a bigger impact than doubling energy prices. It's plausible that affirmative action or uncertainties about inflation or regulations could make the economy 5% less efficient too. But Noah singles out energy as the scapegoat for declining innovation in the physical realm, without considering any of the other possible causes.
Sanity check: If energy is really going to become a lot cheaper over the coming decade, CME oil futures for 2033 should be a lot cheaper than oil prices today, because people will substitute other energy sources for oil. But actually, the futures prices decline from $75/barrel to $60/barrel by 2025 (which is basically just reversion to the mean) and then approximately stay there until 2033. There's no longer term trend towards vastly cheaper energy visible in this futures curve. (CME also has futures for electricity prices, which show a similar pattern of quick reversion to the mean followed by stasis, but they're a lot less liquid than oil futures so I don't trust the accuracy of the quotes as much. Zero volume on a typical day. https://www.cmegroup.com/markets/energy/electricity/pjm-western-hub-peak-calendar-month-real-time-lmp.quotes.html)
Brain computer interfaces are still slow and lossy. They're inevitably lossy, because external electrodes can't perfectly read the state of the neurons. They'll probably not in my lifetime be as fast and accurate as a mouse, but they might enable something more compact which is useful for non-paraplegics (e.g., google glass style computers)
The vision implant is super cool but it sounds like it still has a very long way to go to get to Geordi on Star Trek TNG. They basically upgraded a blind person to a person who can just barely see large shapes.
The fine print about the pig-to-human kidney transplant was that it only ran for a couple of days, and it was external to the body. (h/t NSQ or freakonomics podcasts) They're still nowhere close to what people usually mean by the words "organ transplant".
This Comment Is Too Damn Long so I'm going to write a separate one if I have any thoughts on 22-34
#22 Nomad capitalists have observed that as soon as a country gets developed enough for its passports to have visa-free access to the US, it usually stops having a very favorable tax and regulatory regime. The intersection of countries with visa free access to the US (https://en.wikipedia.org/wiki/Visa_Waiver_Program) and tax havens (https://howmuch.net/articles/world-map-of-personal-income-tax-havens) is empty except for Poland, which has low-ish but not impressively low taxes.
#23 When I worked for a startup in LA, I worked a lot harder than my usual laziness. I put in about 35 hours of high quality effort every week. (because I actually liked the job, I had a substantial equity stake, my boss was great at motivating me, and I was overcompensating after getting fired from a previous job for extreme laziness and insubordination) A year after I quit that job my boss told me that he had to get 3 people to do what I used to do. There are huge differences in productivity between programmers, so it's not surprising that some programmer can get away with working 5-10 hours a week.
#27 the bulldozer-vetocratic axis is really cool.
#33 Sam Altman was in my accelerated math and science classes for three years in middle school. I went over to his house once to make a 1m parabolic reflector out of mylar in 8th grade. He didn't seem like the smartest in the class, but he was very polite, scrupulous, and hard working. I had no idea he was gay until I read about it on Wikipedia decades later. He did seem annoyed that I was talking about girls while I was at his house. Those were the days when I was widely bullied for being gay, and wasn't actually gay, but people might have thought so. And a couple years later I got a 9.6 on hotornot, so maybe he was into me and I was oblivious as usual.
The reality is that even if they had approved the vaccine sooner, it would not have saved many, if any lives. They started ramping production even before they got approval, and we were short for many months on vaccines. We would have at best saved a tiny number of lives - and probably not that.
Moreover, they were concerned about rushing through a vaccine that either didn't work or which caused damage to people, as it would increase vaccine hesitancy - which, as we know, has killed far more people than any delay would have.
So the argument that it killed lots of people is false.
This seems to be a common cognitive glitch amongst a certain brand of libertarian - you constantly and consistently disregard this stuff.
I just realized that link posts would make very good twitter threads - how do you feel about making them or me making them?
To be fair to the people at Toronto Life, if you wore a bejeweled SSC pendant around your neck and had a STAFF, the NYT would NOT have been acting inappropriately.
It's literally their jobs, and they have corporations involved. This ain't the same thing.