Planning to post this in the fresh thread tomorrow but in the meantime hoping I can get some takers. Looking for some no-nonsense advice around a lingering bad shroom reaction.
Last Sunday (6 days ago) I took 4g of shrooms (MVP strain). The trip itself was meh, had periods of euphoria but towards the end had kind of a sad crash. But nothing I would describe as so terrible. I've done shrooms before and had a good experience, including a couple months prior taking 3.5g of MVP and having a positive experience.
Since then, I've been feeling really quite poorly. Some physical symptoms such as chills/feverishness, which seem to be passing. More importantly my headspace has been very bad. A sort of headache-y brain fog has been following me most of the time (trying to manage through this at work), and more unpleasantly, the fog seems to come with a sort of "black cloud" where I feel incredibly down and sad and like my life is over and I don't want to live anymore. It makes no sense but the cloud is just so powerful. I've had a couple welcome intervals of lucidity lasting for half a day or so. This morning I have more of a "grey cloud," where I'm not normal but at least my rational brain is essentially in the driver's seat.
I'm going to the doctor on Mon but in the meantime I'd appreciate any advice. My understanding is shrooms should not cause a permanent shift in my brain (I have no history of mental illness in me or my family). So I'm hoping this will all pass. But it's incredibly unpleasant. Any words of wisdom?
I often have a period of pretty bad ennui after a psychadelic experience (not neccessarily because of a bad trip or anything like that). I dont think it's like "your brain is broken" but more "you have entered a transient malaise for unknowable reasons and will mean-revert eventually". Or maybe "you depleted the fuck out of neurotransmitters or something and are experiencing the fallout" either way, this too shall pass
Inspired by Adam Mastroiani claiming his mind is like Fantasia in this post: https://experimentalhistory.substack.com/p/you-cant-reach-the-brain-through I decided to do a survey to see how much of a resemblance to Fantasia would people say their minds have. The only question I have for sure is a gradient on which movie your mind resembles that has 10 gradations from Saw to Fantasia (I guess the middle would be a boring drama, and the movie at the bottom end, Saw, can be changed), but I'm wondering if anyone here has other questions that would make sense on the survey. I'm thinking of another one about how much of the time is your mind like the choice in the first question, but presumably the choices there would be half of the time, most of the time, all of the time (why did you choose it if it's less than half of the time?), but am unsure.
Happy St. Patrick's Day to you all, and in honour of the contribution my country has made to the world, this typical representation of my native culture from 2011 but still relevant today (tradition never changes!)
The Archbishop of Saint Paul MN gave Catholics here a special dispensation from the Lenten requirement of abstaining from meat today. A lot of corned beef and cabbage will be served today. Maybe a few pints of Guinness and a couple shots of Jameson too.
(I am unfortunately not celebrating. In Massachusetts, where I am, there is a snowstorm today, though luckily it's not as catastrophic as the forecasts had predicted.)
Yeah that's probably where I originally got the idea. I researched it and actually they wanted to make it 3.2! But 3 is pure and a better artistic choice.
For a long time I thought it was too silly to really go anywhere - but when I see Tucker Carlson lie to my lying eyes, I am starting to feel compelled to dust off the score and run that sucker up to Broadway!
As I have always said, "When In Doubt, Put On a Musical!"
Maybe something about arbitrage/accounting fraud? Selling a group of things in a line of length 3, then buying a group of things wrapped in a circle for the same amount?
I've posted before about caffeine "cycling". In other words, can I have caffeine x days per week, and then abstain for 7-x days so that I don't build tolerance? One idea that came up was using something different on the "off" days, e.g. nicotine. Well I tried nicotine gum (several 2mg pieces throughout the day) on my no-caffeine days and I basically felt nothing. No stimulation and I just missed having caffeine in my system. Has anyone else had this experience or have any other suggestions?
I drink a reckless amount of caffeine and then every couple months will take 4-5 days where I limit myself to one cup. I don't know how much between-person difference there is, but that works for me without needing other stimulants.
I don't know about nicotine specifically, but many stimulants are cross-tolerant, meaning you can't just substitute one for the other while tolerance comes back. Maybe something very different like Modafinil would work better?
Someone on reddit (I found an old thread) said phenylpiracetam and (R)-modafinil could work. I'll have to look into these. I don't know a thing about them.
Funny that your break from caffeine still involves 1 cup per day. Maybe I should have a quarter cup on my "off days".
Dumb question, but were you using the gum right? You have to chew it a bit, park it in your lip, and let the nicotine seep in. Could try going up to 4mg if you're still not feeling anything --most people I know who are not regular nicotine users get a noticable effect from that.
> Opportunity to purchase insurance led to 59.91% uptake and access to free insurance to 78.71% uptake. … Across a range of health measures, we estimate no significant impacts on health. … We conducted a baseline survey involving multiple members of each household 18 months before the intervention. We measured outcomes two times, at 18 months and at 3.5 years post intervention. … only 3 (0.46% of all estimated coefficients concerning health outcomes) were significant after multiple-testing adjustments. We cannot reject the hypothesis that the distribution of p-values from these estimates is consistent with no differences (P=0.31).
This result seems aggressively counterintuitive, in light of how lifespans have been increasing due to improvements in medicine. Hell, I once heard that the real reason murder rates went down is that medicine is keeping lots of victims of violent crime alive that in the past would have died. The conclusion from Hanson that health insurance is useless also seems excessive. If I break a bone or get some weird disease, it sure is better to be with health insurance than not. It's not like those things would just fix themselves otherwise.
This one seems like a good subject for Scott to do a deep dive on, unless he already did?
Pay attention to the actual claim, which is not that medicine is useless on average, but that *marginal* medicine is useless on average.
Mind, there is a potentially surprising claim in there, that much of the healthcare our society buys is useless on average - but personally it isn't that surprising. Pseudo-scientific healthcare that works when held to a scientific analysis stops being pseudoscience and just becomes healthcare. Marginal healthcare that's effective stops being marginal.
The issue, from a policy perspective, is that marginal for one person is not marginal for another; we can't just ban all the ineffectual treatments and be done with it. The healthcare that is super-important for one person just makes another person satisfied that somebody is taking their problems (or at least complaints) seriously; see all the people who want antibiotics for the flu.
But then, in my adolescence, I ended up on antibiotics every time I got the flu, just removed a week or two, because every damn time I got the flu I'd get a secondary respiratory infection. At one point I think a doctor did notice this, because I got antibiotics when I got the flu, and I got to skip the second course of disease that one time. Whatever the problem was, it went away eventually. (In retrospect I think it was probably something environmental, because I think the problem stopped when we moved to a different house, but many other things changed, so who knows?)
>This result seems aggressively counterintuitive, in light of how lifespans have been increasing due to improvements in medicine.
Little of the lifespan increase post vaccines and antibiotics are due to improved healthcare. Hanson isn't saying that healthcare doesn't improve health, he's saying that there are a few cheap miracle cures that improved health a lot, but that we kept adding costlier and costlier interventions with smaller and smaller gains. The **marginal** effect of more healthcare on health in a first world nation seems to be zero or negative, but for some reason (see Elephant in the Brain) we keep increasing spending.
As with e.g. studies that parenting doesn't matter, you have to keep in mind that there are things you basically can't study because whenever you notice that they exist you have to report them to authorities who will make them go away before you're finished studying them.
So we can't study what happens if we let people with compound fractures lie in the streets bleeding because they don't have health insurance, and we can't study what happens if we withhold antibiotics from people with serious bacterial infections, and we can't study what happens if we don't treat heart attack victims.
At the other extreme, we can't study the effect of the very very best health care, because that mostly goes to rich people whose lawyers will explain to your IRB that they aren't interested in being test subjects in your experiment. And you'll have a hard time studying the very long-term effects of an artificial delta to someone's health care because you probably can't afford to run the study for more than a few years.
So what's left is, you can study the short-term effects of the sort of marginal health care that people will only bother to obtain if it's covered by insurance. At which point, yeah, maybe it's plausible that when we've picked all the low-hanging fruit and written the juiciest fruit off as beyond our reach, there's not much difference left in the middle.
Could you say more on your first example (studies that say that parenting doesn't matter)? What are the things that can't be noticed that would impact the results of the studies? I'm guessing you mean abuse. But then certainly the studies that don't find any effect don't deny that abuse /does/ have an effect.
I feel I haven't understood your comment (and I am very interested in this subject, as I have 2 young children and I wonder how much I should push them to read, do math, learn discipline, versus playing pokemon go, watch TV, just play with other kids, etc., etc.).
The way I read John's post was that although studies "show" that parenting doesn't matter, said studies are limited to what they can measure, and (implicitly in his post) if they could measure all factors the conclusions would be different.
As far as I know, Caplan believes that parenting doesn't matter and therefore he agrees with those studies.
But I could have misunderstood, that's why I asked for more info.
Right, and Caplan is a fair source for just the studies. We don't *know* that e.g. horribly abusive and neglectful parenting results in negative outcomes for the child, at least in the multiple-peer-reviewed-RCT sense of "know", but it's a reasonably good guess and one that shouldn't be discarded on the basis of studies that can't include such cases. Similarly, we don't "know" that completely locking people out of the health care system would be bad for their health, but it's a pretty good guess that is consistent with the data.
I think when you've got a single study and a ridiculously counterintuitive conclusion like "medicine is useless on average", you should probably tend towards suspecting that the study was crap.
The study isn't even about "is medicine good", it's about "is free health insurance good". People with serious health problems still found ways to get them treated. The free health insurance in Karnakata might have been crap and maybe people didn't bother to use it anyway. The time span of the study was just three years, meaning that any longer-term benefits won't show up. And
The question isn't about "All" or "None" healthcare, because there are no people in the US who have access to "None" healthcare (and to be fair, no one was offered "All" in this study, we don't really know if additional care could have improved outcomes if available). The question would be about the difference between some kind of baseline (which at the least would be emergency care) and having some kind of reasonably strong health insurance option.
It's not unreasonable to conclude that healthcare above some level does not provide much additional benefit. It's a bang-for-your-buck analysis at that point. What you say about the three year timespan is definitely a limitation here, even given the purpose of the question.
A lot of medical care that people receive doesn't contribute to health out comes. Unless you have a chronic condition or are very old or over weight, you don't need a yearly checkup and you don't need all the blood tests that are regularly done.
There is also a huge set of "health problems" that people won't seek care for if they have to pay for it but will if they dont have to pay for it but these problems aren't really problems and will probably go away on their own. Muscle aches and pains being a large portion of them.
Also a lot of surgeries or treatments that don't out perform placebo. Things like cortisol injections to treat knee or back problems, some orthopedic surgeries, etc.
So not all health care is useless. Obviously emergency care isn't (i dont think you are arguing this). As we get older we are likely to need more care. Care for chronic disease is needed, but for most healthy people there isn't much need to engage with medical services with out an adverse event.
There is a broad range of what "heath care" is. So its really hard to make blanket claims about what is valid and what isn't.
I would advocate for no government or otherwise subsidized third-party "medical insurance" at all after age 50. seniors would be free to purchace medical services on the open market, but any social value in spending $$$$ on the last few decades of your life is nil. You might disagree, in which many medical professionals are ready to take your money.
Plus, by 50, you get the face you deserve, and you are on the death-trajectory you have created. Brenschluss has occurred, there is no steering you, you will fall when/where you will.
I agree with you in general. I divide medical care into two categories:
-- Trauma care: Incredible, trauma and ER care can save people who would be stone dead 40 years ago, and often restore full function after shocking injuries. Medicine and timely intervention are truly miracles and worth every penny.
-- General care. Mostly Useless. Some things like cancer screenings can be very effective, but diseases and chronic conditions are not treatable. You are just consuming health care, no care would have the same effect. If doctors could/would exclude or "86" certain patients, the rest might benefit. My medcal practice:
You're fat. Get out of my office
You smoke? That dwarfs all other conditions. Bye.
You drink how much? Nothing can save you..
In general : You are going to die.
In general: people will do stupid things and infinite medecine might change 5% of outcomes.
My vague recollection is that this already known to people who study the field carefully, and the conclusions is hardly that healthcare doesn't matter, but that increased access to health insurance doesn't (at least under today's conditions) translate to improved healthcare outcomes. Presumably the underlying reason is that people are rational: if they get seriously injured or sick, then they take care of it, and they either find a way to get it for free, or they find a way to pay for it, because they value their health and life above all other considerations. Adding additional health insurance may make it easier for them, or transfer the cost to someone else, but it doesn't change the outcomes because the outcomes are being driven by rational individual decisions.
It's only a surprise if you've never questioned the access to healthcare insurance = improved healthcare outcomes assumption which reformers[1], politicians[2], and the healthcare industry itself[3] have all heavily promoted as obvious for the past 50 years.
------------------
[1] Because job security: there are few crusades as immediately sympathetic as improving the healthcare outcomes of the unfortunate.
[2] Because they want your vote to expand their power.
[3] Because having to shift money around from paying customers to nonpaying customers is miserable and makes them very unpopular.
This doesn't suggest to me at all that medicine is useless. The experiment is about insurance, not healthcare as a whole. Surely people who really needed medicine paid for it, even in the groups without insurance. What it suggests to me is that if you get insurance, you go to the doctor way more often than you need to since you're not paying that much, if at all, and that the outcomes are not much different by health for the insured vs not insured.
Is this even surprising? Like I said, I would expect when you really, really need it you're going to pay for medicine regardless of your insurance status. The only times I would expect insurance to really make a difference in overall health is if there is some treatment like cancer or a lifelong auto immune disease where the treatment is too expensive to afford without insurance, or at end of life when you have lots of medication expenses or something that keep you alive in old age. The former is maybe too rare to be picked up on in anything but an enormous study, and the latter depends on having some very elderly seniors in your experiment.
What we should really be measuring with respect to insurance is not health, but financial wellbeing. The point of having insurance for me is more to prevent me from going broke in unfortunate medical circumstances than it is to keep me healthy. I'll pay whatever if I need a spinal surgery, even if it means paying back a loan for the rest of my life. I might be just as healthy as the person who had insurance, we both got the surgery, but I'm broke and they're not.
Came here to make the point in your last paragraph, you did it better than I would have. I'll throw in maybe wanting to measure other quality of life markers if we could - like how much free time is spent looking/queuing for cheap/free healthcare options?
That was terrible. Misspellings, bad grammar, folk etymology, and the general air of being crazy/stoned (which is your prerogative if you're going for the 'only the mad are sane' vibe of 'the road of excess leads to the palace of wisdom').
But honestly? I've seen this fake femmy crap too often before to be interested. Yeah yeah you and Robert Graves and the White Goddess but this time round it's the Vedas which are the only true truth.
Good luck, but I'm not buying what you're selling.
Thank you I appreciate every comment I do. My writing needs work, I probably made it sound like I don’t respect masculinity or femininity...sorry if I did that. I wrote fast and sober
Listen, you can tell me the Sphinx was built out of cheddar by time-travelling mice from the 31st century and I'll give it a listen - PROVIDING you don't have crazy font changes, bad spelling, can't distinguish between plurals and possessives and sound like you've just polished off an entire litre of cheap supermarket vodka and are now regurgitating what is festering in the depths of your brain.
Nicely laid-out and properly spelled, then please tell me more about these mice 😁
Just run it thru gpt4 & stable diffusion, that will polish this -thing- right up.
Point being, pretty soon you won't be able to use the "idiots who can't be bothered to learn grammar or even use a spellchecker can safely be ignored" filter any more.
Does anyone have any recommendations for organisations devoted to fighting climate change that are willing to say that the positives of something reducing emissions are worth the negatives of that something harming some of the poor/minorities/etc? Not that avoiding the latter is a bad thing, but it shouldn't stop us doiing the former.
From a facebook discussion of anger-- it turns out that people's experience of anger varies quite a bit.
I'm just posting some of my writing here rather than the link because I think of posting in one place to have a sort of quasi-privacy.
"Someone on my flist posted about drug-induced anger, and what it was like-- they previously had very little experience with anger. They were surprised to find how much anger led to wanting to believe that anger was justified.
I was surprised to find out about a person who had experienced so little anger. I'm angry a *lot* of the time. Arguably it's a close to a default emotion for me.
My mother was a very angry person, and for a long time, being angry at her was a central fact of my emotional life. I wouldn't be surprised if high levels of anger are somewhat genetic."
"I've wondered about physical causes for what seems like an increase of public anger over the past 20 years or so. I've run across the idea that some prescription meds might increase anger, but not enough for people to notice it in themselves. In particular, it was hypothesized that it was steroids and statins.
I've got boring theories, like that hostility breeds hostility. Or that people (what with social media and long work hours, are getting less sleep.
I leave a slot open for something we haven't thought of."
"When I was menopausal, I went through a phase of feeling very angry during hot flashes. Since I hate being overheated, the anger seemed very reasonable.
Oddly I got feedback from other people that I seemed calmer. I assume it was a result of my trying to steady myself because my anger obviously wasn't caused by external circumstances.
Then I started getting hot flashes that were heat without the anger, so I conclude that the anger was a separate hormonal thing other than the heat."
One thing from someone else-- they found that they got very angry from tequila. They hypothesize that it was the alcohol/sugar combination, but don't want to go through it more times to test the theory.
Anger is one of those things that's so variable that it can make it hard to understand what it's like to be another person. I get angry very rarely. I've been getting angry more frequently as I age for some reason (currently late 30s), but it's still rare. Interestingly, I find that I usually only get angry at people I'm close to. Feeling anger because of politics, current events, workplace issues, the behavior of other drivers on the road, etc is even more rare; even if I recognize objectively that something is outrageous, it typically won't make me feel angry.
Probably the biggest downside of not getting angry often is that when I do get those feelings it can feel challenging to cope with them because I don't have much practice.
I tried out phentermine for weight loss a few years back and it worked quite well except that it had two intolerable side effects, one of which was that I noticed I would get angrier much faster. I would be baseline normal emotionally, but anything that upset me even slightly would cause me to skyrocket from 0 to 100 so quickly it was shockingly apparent. I realized this was becoming a problem since I was effectively working customer service at the time and I would answer the phone already out of patience for whatever issue someone needed help with, so I ended up having to cut that off.
I think you are right that some people have a predisposition to reacting with anger to small or unimportant stimuli. When my middle child was born, a woman came to my room to get the baby so that she could perform a hearing test on her. She returned with my daughter 15 minutes later, and she mentioned that this baby had an incredibly calm disposition. I thought there was no way she could tell that about someone on their first day on Earth, after only a short time spent with her, and I asked her how she could tell such a thing. She told me that she had been conducting hearing tests on newborn babies for years, and she could see personalities in them even on their first day. She said that some babies are just naturally reactive -- she said some of them are mad from the get-go.
My daughter is 10 now, and is still a pretty calm kid. I, also, am very slow to anger. I do not let most things bother me. And if I get uncomfortable or if I feel provoked, it is very easy for me to stop and assess whether a negative reaction would be reasonable or helpful.
I also think that some anger/hostility is a cover for fear, grief, discomfort, or emotions that are not considered "acceptable." My ex-husband is a veteran. After he returned from overseas, he became much more hostile to small provocations than he had been before being in a warzone. I am quite sure that his anger was a safer response than a response that would have had him expressing his fear and grief over what he experienced. The anger was protective in the short term, but over time, using it to prevent accountability for himself and others destroyed him.
People who carry a lot of anger around should be careful to not let it fester for too long. Finding a safe place to look behind the anger is key, especially if the hostility is new.
Being quicker to anger seems like a common thread in combat vets - I imagine this is an adaptive response to finding oneself in an environment of constant violence. I read this quote from an injured Ukrainian soldier recently:
“I can speak about myself. My restraint system works ok - I can calm myself down. But sometimes I have explosions of aggression that I never had before.
“Someone says something off, and for a moment I find myself threatening to break their knees. And then I calm down and think ‘what is going on?’”
1) You are very correct thatanger and aggression caused by internal biological "tides" can be very, very powerful. Overwhelming. Can easily break out of any containment. The metaphor of water and tides and rivers and dams and oceansis a very good fit, conceptually.we all carry some volume of anger around, it can slosh around uncompfortabally sometimes, or overflow its banks, or burst a dam or be a tsunami. To be very literal, hot flashes would be fascie-ly explained at starting a boil.
I will postulate that the most intense anger a woman can experience is maybe 50% of what even being a calm, calm man feels like. It would be fascinating, to "switch bodies" completely with another person, for a max of fifteen seconds, say.
I truncated my comment for time. A more complete version of what I was trying to say:
Men and women may have similar anger levels, in daily life, but the potential, or scale, of anger, is waaaaay different.
If a woman's anger runs from 1-10, a man's scale runs from 1-100. That is entirely because of testosterone, which creates in men physical strength, agression, and the potential for anger/violence that is off the chart of anyone who has not been an adolescent male can understand.
You might say, and I accept, that I am co-locating anger with anger's expression, violence. Men are far more violent than women, part of that is testosterone and its effect, part is physical strength and a sort of mindset or outward-direction that imagines physical conflict.
I think the Motte-and-baily version of this is: men and women both experience anger. Men's anger is potentially much larger and more destructive, because of innate sex differences caused by hormones. I personally cannot access another's inner state (and if such a switch would be possible even for an instant, it might be so alien as to just be horrible and not informative), but I think subjectiely that the anger scales of men and women are 10x different.
The most relevant thing to ACX I can draw from this, is that much more serious consideration should be given to side effects of medication for men vs. women. A medication (I am thinking of Zoloft) that might cause "irritability", should be counter-indicated in males, because women might become irritable but men might become homicidal.
My extreme idea, is that the feminization of society, and the draconian penalties for any kind of physical violence, in society, and especially in schools, is creating young adults who bottle up anger, and also hoard it, and think that all anger is the same. Young adults who insist that "words *are* violence" have clearly never experienced a physical fight. Boys especially are criminalized in school. Boys will fight. It is needed so you learn that if you cross boundaries you will get punched. Boys need to learn that, while they are young enough that they can't really hurt each other. Men need to learn that before they are physically strong enough to kill other people.
Beyond getting smacked yourself, you also learn that punching somebody really hurts! Hurts your hand. You both get messed up! And from my young experience, boys' fights last about 15 seconds, you both learned something, and you move on. Even as friends. It was just a fistfight.
It sounds to me like you had a really rough adolescence that you really shouldn’t be projecting into statistical statements about large populations.
My experience is exactly the opposite of yours.
There used no one in my life more angry than my mother, it has festered in her since childhood. My wife is a close second, her mother was quick to anger and she has trouble climbing down from anger once activated; she has one angry brother but her dad, other brothers and sister are all very chill.
My father, brother, and 4/5 of my male best friends have similar stories about our mothers and sisters, and we’re almost all very calm men. I do have one male close friend who has a lot of anger, so I don’t think this is an inability to recognize anger in men.
My point is definitely not that “men are less angry than women”, although that is in fact my lived experience. My point is that attributing anger to testosterone seems like a vast oversimplification of the human experience.
As a calm man from a calm family who married into a very angry family (and got the hell out again a few years later because I couldn't stand the constant fighting and drama) I am very dubious of your claim. Both men and women in my ex's family were way more angry over way more time than I've ever managed to be.
And the angriest person I've ever met was a woman, although plenty of men are in the top ten. Anecdata!
> I will postulate that the most intense anger a woman can experience is maybe 50% of what even being a calm, calm man feels like. It would be fascinating, to "switch bodies" completely with another person, for a max of fifteen seconds, say.
I disagree. I'm a man, and I'm angry sometimes, but sometimes I'm not. If a woman at her angriest is less angry than I am right now, then anger is just totally alien to her. And I don't think that's the case.
As a calm man, I think you’re deeply mistaken. I’m also surprised at how you can think that even when reply to a post discussing how large differences there are between individuals.
I believe a lot of our strong emotions come from biological causes. We often aren't angry for a reason, we are "angry" because we are tired or dehydrated or having a mild reaction to food or blood sugar or something, then our brain "justifies" the anger by latching onto a "cause". I definitely experience this with anxiety.
I think there are societal shifts that impact our moods. Many used to self-medicate anxiety with cigarettes -- I would be curious if anxiety diagnosis has increased with drop in smoking rate. Or if intense emotion during the day correlates with lower sleep quality connected with late night screen usage.
There's a huge breadth of human experience research we could do, connected with emotions that aren't quite strong enough to be labeled as problems but still impact quality of life. I imagine we would need to gather tons of data about exactly what people eat, what they drink, when they sleep, etc. We'd need to either train them to self-assess mood, or find a feasible way to test hormone levels several times throughout the day. And then, do this across thousands of people, to find patterns.
I love to fantasize about "solving" negative emotions caused by biological factors.
I can anecdotally add that I had anger and frustration issues when I had undiagnosed ADHD. Now my anger is 99% resolved by ADHD medications, I basically just don’t get angry at anything anymore now that medication is allowing my neurotransmitters to work as they should.
I really liked the outdoor Berkeley meetup a few years ago (in a meadow on campus). For those few like me who are still determined not to get COVID, another such meetup would be wonderful.
The Rose Garden Inn, where we have most Berkeley meetups nowadays, has inside and outside areas. If I say a meetup is at Rose Garden, assume there will be some outdoors discussion.
If Covid were a cold, that would be reasonable advice. Since it's rather more serious than that (much more prevalent and contagious, and responsible for several times more fatality and disability than flu, let alone colds), more serious avoidance efforts are reasonable.
In 2020 I said I expected to go back to normal when my Covid risk was comparable to driving. (Which I expected to happen much sooner, but live and learn.) At this point I'd roughly estimate it's about 4-5x that for my demographic.
And driving risk exists with an underlying consensus that it's worth a fairly massive societal investment and substantial limits on autonomy to reduce it. Currently individually trying to avoid Covid is more like living in a world of heavy traffic with no speed limits, traffic lights, lines on the road, headlights, or seat belts.
"Just get in the car" is a choice one could still make, and maybe many/most would. But some people might be motivated to stay off the roads insofar as they could, use less traveled roads when they couldn't, and/or wear a crash helmet when they had to drive.
This seems more like nagging than a productive comment. I happen to agree with you on this one but come on, it's 2023, let's not shame people for their COVID related choices.
I actually appreciate the stated intention of trebuchet's comment - that of encouraging someone to end a policy that is limiting their happiness - but I have certain personal circumstances that would make catching COVID extra-bad. Thanks for understanding.
It is thought that one just builds up an immunity debt and when ultimately catches cold, it will be more severe. It is probably a reason why some people even die from cold. They were “lucky” to avoid it for longer time than others but ultimately their risk also increases.
With covid maybe regular vaccines help to prevent that but that's a big maybe. You can never know about new mutations etc.
Well, last time I decided it's «medium» not «large» and has properties X Y Z so I will go to this seated lunch inside and eat something (at stand-up receptions I don't miss out on much by just chatting with people while keeping the mask), I unsurprisingly did get a cold, and rather unsurprisingly it was quite a bit lighter than my old ones… And this was after a couple of years without externally caused colds.
Actually, probably-CoViD because I was worse at mask leaks back then — it was also lighter on average than the average cold for me; the peak was just different; unlike the normal colds there were some apparently long term effects but on the slight side.
I have also seen plausible-ish claims of damage leading to weaker immune system after more-severe-than-light coronavirus (not necessarily SARS-CoV-2) infections, I suspect that both directions of study are so confounded as to cross the noise event horizon multiple times in the same direction, so yeah, not betting on either claim.
This meme really got me thinking about how we perceive human consciousness through the lens of the technology we create. I don't endorse the idea that we aren't really making progress toward true understanding (that would be insanely absurd) but it drew a connection I had never thought of before and wanted to share
Really nice! I'm not quite sure what I think about it - probably it helps illustrate how all these things *are* useful metaphors for how the universe works, and we get more understanding with each, but it's clearly incomplete.
Am I the only person on the planet who can't look at Twitter's webpage because of all the ads that pop in and move things around, the terrible fonts, and the ridiculous color scheme? I feel like someone with epilepsy at a 70's disco with strobe lights flashing.
I don't think that any of those things are really "wrong", naive perhaps and more broad-scoped and hyped up by people in their time than they are, but not wrong.
1- The Wheel : A wheel is just a sine and a cosine (which is just a sine, but late) dancing in 2D, back in the day there was a certain French mathematician[1] who thought that it would be funny if sines and cosines dancing in 2D is enough to represent any mathematical relationship, it turned out he was right, and his name shows up in a metric truckload of scientific and engineering fields till now. The ancients weren't wrong **even if you took their words literally**, this entire universe really is full to the brim with wheels at the deepest and most fundamental levels. We just gave their ideas a different name.
2- The book : Meaning "serialized" information, volatile data captured into stable storage. This is your DNA, the structure of your cells is the volatile death-able information, DNA's nitrogen and phosphate is the sturdy pages that record them so they persist longer. The Geology of planets, it's a book that records geological events. Even spacetime is a book that records everything with a mass. The ancients weren't wrong even if you took their words literally.
3- Gears : I don't think people meant literal gears (although they won't be wrong if they did, after all gears are just sines and cosines, and everything is sines and cosines), more like the idea of composability. The world as a Lego game, Simple Things Infinitely Repeated. A single gear by itself is a trivial contraption, repeat it enough times in the right arrangement and you get airplanes and trains and factories and cars, this looks an aweful lot like what Life does, or - come to think of it - how Atoms work.
4- Computers : Machines I'm madly in love with, the thing about them is that they are universal rule-followers. They can act like anything that can be sufficiently described in rules, since Rule-Following is itself a task that can be described with rules, computers can act like other computers acting like anything (which can include acting like other computers acting like other computers ....). I think it was Greg Egan in either Permutation City or Diaspora that said "Inside a computer, everthing is the product of a human labor", computers are mini-universes where you can quite literally be a God, you say "Let something be", and behold, that something is there - indeed, that's the **only** way something can ever exist there -, unless it defies Logic itself. Every single thing you have seen a computer do, some human or collections of humans have told it to do (knowingly or not), every thing they can do was built by human design from the transistor up.
I don't know if the Universe has something like computers that occur in it naturally, if it doesn't then the Universe is infinitely poorer for it. The idea of a single small core of physical laws that govern the entire universe across all hierarchical levels is *certainly* something that every computer aficionado can smell 0s and 1s of it from an ocean away. Computers are, depending on how you look it, either a generalization or a special case of mathematics, or is mathematics itself made concrete and resident inside physical systems. It's not exactly radical or far-fetched or unproven speculation to say that vast swaths of the Universe can be seen through mathematics, ontological interpretations can differ.
------------
I don't think that "progress" is the right way of thinking about it either, none of those things invalidate or contradict or even is in tension with the other, they are all different metaphors illuminating different aspects of reality[2], indeed The Book and The Gear and The Computer seem to me like different ways of viewing the same metaphor.
There may not be such a thing as a final picture of reality. It would be really weird if a universe that was not intelligently designed were tractable to human reason.
Yeah I have a lot of thoughts about that tweet, but I think the OP isn’t exactly trying to seriously engage with the standard scientific position.
I think the storybook analogy one in particular can only be appreciated once you understand how much ‘reality’ is projected outwardly from our brains, which in turn becomes interwoven and reflected in the stories we told orally and eventually wrote down!
There’s also a sense in which watching a wheel spin for the first time in Antiquity or catching a glimpse of the technically complex process of building a watch really hammers in how fundamentally *Lawful* reality. I think SA actually sliced a much cleaner analogy about this which I’m gonna butcher, but it went something along the lines of how the first engineers and philosophers to realize that the “human body was made up of mechanical gears” were gesturing at the fact that our bodies seemed to obey the same physical laws as the new machines being built at the time. This looks particularly prescient when you remember that the prevailing wisdom at the time was that our bodies were constituted of unchanging, Platonically transcendent ‘Essences’ or something. Adjust your appraisal of the contemporary vein of criticism aimed at the heresy of believing that machines could think or be conscious accordingly.
I didn't take that meme to be about consciousness, I took it to be about models of the external world, specifically to dunk on "bro we're living in the matrix" types.
how we conceive of ourselves and "the world" reflects both where we find ourselves in the moment and the path we took to get there
take the last two, most recent parts of the meme: interlocking gears are a more complex form of the wheel, code integrates the written word and the mechanical (and simulation theory involves repeated cycles)
without making any claims about the correctness of any one view, it suggests we are both intensely caught up in the moment and building on our collective body of knowledge
Barney Frank also claimed that he had no idea his boyfriend was running a prostitution ring out of Frank's apartment. So apparently he's not very observant.
Oh man, I miss all the best stories 🤣 That reminds me of the scandal back in the 90s when Norman Lamont (Chancellor of the Exchequer at the time, and it is *not* included in his Wikipedia article) was found, or found out, that a professional dominatrix going by the name of "Miss Whiplash" was renting his basement flat (it is unclear whether she was offering her services from same or had her place of business elsewhere).
"In June 1991 she was involved in a controversy when Norman Lamont, the then Chancellor of the Exchequer, was investigated for using taxpayers' money to handle the fall-out from press stories concerning 'Miss Whiplash' (Not Lindi St Clair), who was using a flat he owned (the Treasury contributed £4,700 of the £23,000 bill which had been formally approved by the Head of the Civil Service and the Prime Minister)".
There seems to be some confusion, the second "Miss Whiplash" is a woman called Sara Dale, not Lindi St. Clair on Wikipedia, so it seems to be she who was actually renting the flat:
This "Miss Whiplash" was operating as a sex therapist, offering naked massages and canings amongst other treatments using the flat as place of business.
It's been quite a while now since I worked professionally in media relations, but I did. And now I really really want to have a beer with the PR flak who had to explain to Mr. Lamont that when you are trying to correct them on _which_ "Miss Whiplash" it is that rented your apartment, you have lost control of the situation....
I prefer Matt Levine's take (https://pastebin.com/fMfmZZTB). The problem wasn't a *lack* of mark-to-market. The problem was *caused* by mark-to-market. Silicon Valley still had to report the market value of even the "hold to maturity" loans, and someone noticed that they were notionally underwater and triggered the bank run that drove them *actually* underwater. In the old days, bank finances were opaque enough that you wouldn't get a run over the slightest hiccup, but now they are, so the Fed has to step in to provide the service of ignoring market values instead.
Perhaps the extreme case of the "mark to market all the time" mentality is the defi Crypto world where people get liquidated over the briefest candle, and it is a miserable experience that makes it impossible to run a bank (witness the constant failures of crypto bank-like projects).
Reminds me of a classic sight gag from an early episode of The Simpsons. While the view is panning across the outfield fence of the local ballpark, I think it is, it pauses on a billboard for:
"SPRINGFIELD SAVINGS & LOAN -- proudly safeguarding your money 1892-1988 and 1990-present"
It's not crypto. It's just simple contagion which almost always spreads along industry lines nowadays. Crypto and tech exposed banks are more susceptible to runs in a run on banks in those sectors. This was one thing people got wrong in predicting there would be a regional bank run. That's actually unlikely in the current context. What'd be likely is a SECTOR bank run. So the bank down the street in SF that mostly does real estate loans wouldn't be affected but the bank in New York that does startup loans would be.
I will believe it's anti-crypto if the Feds carve out an exception for crypt assets (which they haven't so far) or appear to be picking which banks receive support based on crypto exposure (which, again, is possible but has not yet happened).
Worth noting: Barney Frank, legislator who wrote the post 2008 Dodd-Frank bill that was meant to make this impossible, has a huge vested interest in crying persecution. Because otherwise he might have to admit Dodd-Frank and the CFPB were not actually meant to do what they were advertised to do.
More interesting to me is that the Fed has basically committed to making the depositors whole. Government being government I expect this means another big boost (or even cap removal) to FDIC is coming. Unfortunately high inflation and the distorted incentives set off a bit of a ticking bomb. Albeit one that will hopefully be defused.
The traditional answer to that is that the government *also* tightly regulates what banks are allowed to do in order to limit their ability to take risky bets like that.
I doubt it's an anti-crypto message[1]. My go-to hypotheses for explaining the behavior of elected and appointed officials are (1) CYA, (2) political ambition, (3) avoiding embarassing inconsistencies in the narrative I'm selling. Long-horizon plans for social engineering is way down on the list.
In this case, all three point towards an effort to persuade Joe and Jan Voter that bailing out depositors with more than $250,000 in their accounts[2] at SVB was desperately necessary, and not at all a quid pro quo to potential future campaign donors, because clearly the contagion was spreading rapidly. Look! Another bank topples! No time to lose! Must act now, lest you, Joe and Jan, wake up tomorrow and find your ATM card voided, your gas siphoned, and your dog repossessed. Not that Signature wasn't a little bit wobbly, but Frank might have been right that the Feds were a little more eager to jump than the facts strictly suggested. Who knows? I suppose it's even possible they acted for the purest and most selfless of motives.
--------------------
[1] Furthermore, I doubt Frank thinks so, either. He's an old canny pol, so crooked he has to screw his pants on. I suspect he picked *that* message as most likely to provoke outrage among those paying close attention to everything. See? The problem is some fossils just don't get the New Frontier...feel threatened by innovation, daring, youth....subscribe to my newsletter!
[2] A sum of money Joe and Jan will never see in their own accounts, and one which they probably instinctively associate with the 1% class.
hygiene, smygene, you just want to stop the bleeding while working. I figured most people here don't work with their hands as much, so maybe this hack is not known.
When I'm working with my hands band-aids last an hour or less. They seem prone to getting wet. Good Scotch brand masking tape hangs in there all day. Washing dishes is no problem. (Cheaper and easier to use too. :^)
Sure, that's fine. I just skip the band aid part these days, such a pain opening the dang sterile wrapper. For bigger wounds I sometimes will stick a bit of paper towel over it.... That only really helps with abrasions, or burns, for cuts, tape at ~ right angles to the cut works best.
In Chapter 72, it’s written that “Sohu was doing something with her hands, muttering to herself, bending the energies, quietly funneling power into Jinxiang, watching warily.” But isn’t Sohu supposed to have lost one of her hands by Anglerfish before this? Did she regain her hand at some point? (No.)
Please tell Scott about this before it is published.
Am I the only one who found Everything, Everywhere All at Once to be incredibly tedious? I walked out after 90 min, which I never do. And I liked the Daniels' previous film, as well other multiverse-esque films.
I confess that I find multiverse anything an obnoxious failure. Fiction itself is the multiverse for me. Your attempt to combine many conflicting fictional stories into one narrative is almost always a lazy unwillingness to plot out a believable and interesting story.
I also found the nihilism circles round to “love your family” morality play just… vapid.
Most MCU and Star Wars stories are also vapid, but they don’t annoy me a tenth as much as EEAAO did. They at least usually attempt to tell one relatively boring story per character.
No. It really was aweful and ended up drowsing through half of it. It was pretentious, trite, obnoxious and repetitive, and to see how many awards it won...well, it didn't change the opinion I had of these awards.
But the crime isn't that EEoaO won. It's that RRR wasn't even nominated for most of the categories. For all it's obsession over diversity and representation, the smoothbrains at the AMPAS can't look at an indian movie beyond the funny dance scene, even when it accomplishes everything that american blockbusters have been trying (and failing) to do in the last 10+ years.
This is an interestingly different set of opinions than mine! I wanted to like RRR, but after the first two scenes establishing the characters, the only scenes I found that lived up to the hype were the dance-off scene and the final fight.
There were a few parts that could have been tightened up, but nothing that came close to wanting to make me walk out. One of the best movies I saw last year, and way better than that other Multiverse Mess.
I haven't seen all of the Oscar nominees, so can't speak to whether there was anything better this time around. But a year in which EEAAO wins best picture, is an acceptable performance for Hollywood and the Academy.
Dr. Strange was such a frustrating movie to watch. I can see the remnants of a good movie in it, but it looks like it just got destroyed by revisions. My personal opinion is that America Chavez was not part of the first draft, and that forcing her inclusion to hopefully open up future movies is what created most of the issues.
Strange and Wanda have this nice parallel internal conflict, that Chavez's contrasts with badly. The first act of the film ends up needing to establish Chavez's character, which means there's no time to set up Wanda's heel turn. In the climax our nominal protagonist, instead of resolving the conflict himself, is reduced to telling Chavez to Believe in Herself. It just seems like it would be so easy to make a good movie out of the whole thing if Chavez was a magic artifact instead of a character.
Wanda's heel turn was a waste of a character in any case. (I know it's an adaptation of something that happened in the comics. But a) it wasn't a great idea there, and b) one of the great things about the MCU is its ability to distill the best of the comics without being beholden to their continuity where it doesn't need to be.)
But even if they were going to do it, it should have sprung from Wanda's genuine trauma, which is hilariously exceptional even on a team where being orphaned is an entry ticket: grew up in wartime, lost parents, spent days staring at an unstable UXB, experimented on by Nazis, lost brother/last remaining relative, what's left of homeland destroyed by erstwhile ally killer robot created by teammate, accidentally killed people trying to do the right thing, wound up a fugitive from an entire international treaty created in response to *her*, fell in love, had to literally murder that love to save the world, then have it turn out to be for *nothing*.
*Not* be focused on the imaginary children that no one who didn't tune into the Disney+ series even *saw*. Which made her look like a crazy hysteric lacking only a mustache to twirl, not someone whose being pushed over the edge was at least as understandable as Tony's PTSD issues or Bruce's rage-quitting or Thor's sinking into depression.
But yes, it was even worse luck for Doctor Strange and the MoM (we all see what they did there) that it wound up right against a *much better* story about the trauma of a broken maternal relationship threatening the entire multiverse. Which no one could possibly have anticipated.
(I'm pretty sure those are the only two movies of that description from the entire history of cinema, though I'm open to reminders of something obvious I'm forgetting.)
EEAAO wasn't perfect, and I'm kind of surprised that it wound up a major Oscar force rather than the sort of quirky indy phenomenon that, say, Primer was back in 2004. (Though it had a straightforward emotional core, which Primer didn't.) But I'm certainly not unhappy that it got that wider recognition.
Agreed, particularly about how Wanda needed more time if we were going to go along with her reversals. I hadn't thought about the possibility of a story more tightly focused on Strange and Wanda; that might have worked.
EEAAO at least had a very tight focus, and developed its core characters reasonably well.
My wife loved it at first, and stopped watching it so we could watch it together. And then fell asleep about 90 minutes in.
I found it very good and entertaining, especially the first 30 minutes. But I progressively got more bored of the premise. I finished it and liked it. But I think a lot of years it might not be in my top 20. This year I haven’t seen many movies, so hard to say this year.
I would agree that EEAAO was a mess, disorientingly long, and overstuffed with annoying tics and confusingly dumb plot points. But I think the warm reception can be explained in the context of superhero movies as a modern institution; particularly Disney's Marvel formula of ponderously high stakes and unearned sentimentality, leavened with mechanically-timed doses of humor. The Daniels take this formula that has beaten the general public over the head, subvert the self-seriousness, and probe it in various imaginative and affecting ways; what's not to like? I enjoyed EEAAO more than the last ten Marvel movies, which were also messes.
It doesn't end with Marvel tropes. "I know kung fu" was a titillating central conceit of the Matrix movies, but never developed beyond a ten-year-old's concept of cool. EEAAO has as much fun with it as the (rather slim) budget allows.
I don't mean to explain why you should like a movie you didn't like, just to offer an explanation for its popularity. For people who never saw fit to hop on the capeshit/science fantasy train, it might be a baffling and unpleasant ride.
I also think Michelle Yeoh and Ke Huy Quan turned in legitimately terrific, human performances of the challenging material they were given. Congratulations to both for their respective awards.
Yeah the performances were great, but they are always great. And I do think the framing of it as a subverted superhero movie makes me appreciate it more. I have avoided the superhero movies almost entirely since the first captain America/avengers movies which were just hot garbage with some kk jokes, while everyone told me they were amazing.
I could have done without the juvenile gross out humor and the martial arts stuff got old pretty quickly but I enjoyed the bit with the daughter\destroyer of worlds making the case for nihilism enough to stick with it to the end.
I thought Cate Blanchett should have been awarded Best Actress for Tar but apparently one of the political consideration of the Motion Picture Academy was that she has already received a Best Actor Oscar.
On the other hand, I think the juvenile gross-out humor had the fun aspect of causing many people to independently invent the phrase "Chekhov's buttplug".
Yes, that was one of my problems with it: The same martial arts scene over and over. Didn't one of other billion iterations of her character have a different skill that might have come in handy?
And I understand that Ke Huy Quan's story is a heart-warming one, but there was nothing out of the ordinary about his performance. Barry Keoghan should have won Best Supporting Actpr (though I assume that Brendan Gleeson's presence on the ballot for the same film split some votes).
she spends the film augmenting with skills from different lives--often the base is martial arts bc she's fighting, but it changes a lot as the movie progresses in a very earned way
Ke Huy Quan's performance was one of the best I've seen in years. I can't remember ever being so happy to see someone get an acting award (and that is something the academy voters probably bear in mind while voting--it's not a juried award, after all)
Thank you all so very much! He started it a day and a half after testing positive as his symptoms were super mild until then. A friend advised us to do so as a way to reduce risk of rebound. i.e. don't start too early, since you can only take it for 5 days.
Also, the positive test happened on Saturday, and the doctor didn't bother to call back until Monday with the prescription.
I tested positive Sunday, three days after my wife. We figure we did well to dodge COVID for three years.
She started Paxlovid on Friday, and started feeling better the next day.
I started Paxlovid last night (Monday evening) and had my best sleep in several days. Just took my 2nd dose. (The full course is three pills twice a day for five days.)
We are generally both very healthy, but are in our mid-60s, which is a huge risk factor.
There's some evidence of Paxlovid giving moderate reduction of risk from long Covid-- e.g., this observational study gives ~26% reduced risk. (Which isn't exactly the orders of magnitude I'd prefer, but I'll certainly try to take it when my efforts at avoidance finally fail and I get Covid.)
There was also a recent RCT indicating that a short course of metformin reduced long Covid by 42% (albeit with very wide error bars). But that came out recently enough that it's not yet standard of care if it's going to be-- it would be necessary to find a doctor willing to prescribe it off label. (And because it's so recent, the odds that it won't replicate are quite decent.)
I think Zvi has cited some studies about it. It's probably net helpful (especially if you get it fast), although the risk for people under 70 is fairly low to begin with. I'd look to get it if I was experiencing significant symptoms.
Any advice for getting Stable Diffusion to actually output specific text? Was trying to generate a logo with specific letters and no dice. Midjourney does better.
I'm working on my submission for the book review contest and I've been trying out Grammerly (app that's suggests improvements in your writing prose.)
The suggestions it gives you don't seem that useful, it feels basically like a spellchecker that also tells you insert lots of unnecessary commas.
But, it also gives your writing an overall score broken down into "comprehensibility", "engagement" etc.
So, I thought I'd copy-paste some of my favourite essays from different writers to see what score it gives them (roughly the average of 5 essays):
Scott, 75-80
Richard Hannania, ~80
Robin Hanson, 60-70
Christopher Hitchens, 85-90
(A British columnist I think writes well) David Mitchell, 85-90
George Orwell, 75-80
A random essay I wrote for school when I was sixteen, 60
And, if you'd ask me for subjective ratings of those writers (purely based on style, not substance) that's essentially the exact ranking I would have given.
Which make me think Grammerly can pick up real makers of quality writing.
Does anyone else use apps like this to help their writing?
If anyone else wants to paste some essays into the app, I'd also be interested if matches their subjective impressions of the quality of the writing.
I find Grammarly most useful for finding little grammatical errors like missing prepositions and singular/plurals that don't match. It will occasionally suggest a big reorg of a sentence and I accept its suggestion about 70% of the time.
The spellchecker is a little better than Google's & Apple's but it has this annoying habit of surreptitiously switching perfectly good words for nonsense.
I have the free version and have never tried having it rate my writing. I like commas. Unnecessary ones are the best.
As an American who had never heard of 8oo10c and Countdown until maybe a year or two ago, might I say that Sean Lock was one of Britain's unsung treasures. If that is a meaningful phrase. It's a shame I never got to meet him.
Have decided to take the plunge in creating online content would love some feedback from the community here - I'm current a PhD student studying decision and risk analysis (think prospect theory but normative and applied to real world decisions) and am looking to share some of my less rigorous but more fun ideas that don't make out of the academic review process. Would y'all be more interested in:
1) A blog exploring how insights from my field that can be helpful for everyday life - decisions strike core of who we are individually and think we can all benefit by thinking more deeply about how we make them!
2) An online course that teaches you how to work/communicate in technical settings without a technical background. A "no-code bootcamp" which directly counters the idea that everyone should learn how to code. Especially given the rise of AI, I think technical fluency is a must but not necessarily the specific skill of coding or knowledge of the math under the hood.
I might be interested in 1, but could you post a small sample in reply to this comment demonstrating an insight from your field and how it's useful to everyday life?
It's in ironic common usage like the words "thou" or "score". High schoolers are forced to read Shakespeare, so most educated people know what the word means. But if you were to actually try to use the word in conversation, people would assume that you are making some sort of joke.
This is exactly right. Adding that I can't imagine seeing it in a professional context from an American, so if I saw something like "pay will be issued every fortnight" in a job description from a company that claimed to be based in the US, I'd be suspicious it was fraudulent.
It isn't in common usage, but I think most US adults would know what it is, or at least that it's a number of days even if they don't know it is exactly two weeks.
It's certainly true that the word "fortnight" is not in common usage in the United States.
To my subjective sense of U.S. English, "fortnight" is to "two weeks" very much as "thrice" is to "three times": a well understood alternative with a quaint or archaic connotation that tends to make it less suitable for everyday use.
'Fortnight' seems common enough to me. In less hectic times, one might go on a sun holiday for a fortnight, though perhaps nowadays that would imply an extreme degree of laziness / hedonism.
Honestly, not offhand. But I simply might not notice. I am 63 and in my youth the word was common parlance. I would not disagree that it is used less nowadays.
I wonder whether part at least of the reason is increased 'pressure of time' in which people, while they might consider their mortgage or such-like in annual terms, live in an increasingly shrinking present.
"Furlongs per fortnight" is a common example used when one wants an obscure unit, and in fact there's a whole system of units (furlong–firkin–fortnight or FFF) of which it is a part. You're not likely ever to see this outside of technical fields, however.
The VMS operating system, when booting up, had a variable TIMEPROMPTWAIT indicating the number of microfortnights (1 mfn ≅ 1.2 s) to wait for the operator to enter a new date if it detected the current date and time to be invalid. This was to help indicate that the delay was rather approximate, since the system clock had not yet been started and so a timing loop was used instead.
Right. Because not only is it “close enough” there are common American accents where Dora is pronounced Dorer (most familiar to me is my mom’s Queens, NY accent).
I believe that in those accents, "Dora" and "explorer" always rhyme when they are in the same context - that is, if they occur before a word that begins with a vowel, they both get an "r" in between the two vowels (like the "n" in "an") but if they occur at the end of an utterance, or before a word that begins with a consonant, they are both pronounced without the "r".
I’m not sure where to post this, but not too long ago I unsubscribed and then resubscribed (for a paid subscription), and I have stopped getting the Hidden Open Thread. Any ideas how I can start getting it again?
What exactly do you mean by "get" it? If you go to the front page, or better, the archive, all of the hidden threads will be there if you are logged into your account
Sorry I meant get the email same as I do other posts and regular Open Threads. The most recent Hidden Open Thread I see in Archive is 262.5 from Feb 10. Haven’t there been others since then?
Oh, I did get that one! It just wasn't labelled "Hidden Open Thread." I had incorrectly assumed there was a hidden thread paired with every open one. I thought maybe I'd been experiencing an elaborate punishment for people that had left and come back 😅
So what's the nomatively deterministic significance of someone named Mingyuan. ( bright circle? Or does yuan here refer to the Yuan dynasty? ) being replaced by someone being named Sky-ler?
Clearly mingyuan is actually a variant of minyan, meaning a quorum for a Jewish prayer, since she worked to organize quorums. Skyler is also a reference to this, since the Jewish prayer mentions that we pray to our father in the sky.
In the wake of the news about both Silicon Valley Bank (SVB) and Signature Bank being shut down and going into FDIC receivership, wanted to share something potentially important I learned today. Please pass this on if it might help anyone you know.
Am hoping this info might help someone at a firm – whether it's you, or someone you might know in your networks – avoid the pain of trying to hastily move large chunks of currently-uninsured cash at your primary bank (if over the $250 thousand FDIC limit) into fully-insured accounts at other bank(s) during the coming week or beyond.
There's a service from IntraFi (formerly Promontory Interfinancial Network) called Insured Cash Sweep (ICS). If your bank offers access to it, and many do, you can keep banking with your primary bank while (effectively) splitting your deposits into chunks insured up to $250K at one or more other banks.
Via IntriFi's ICS (for deposit accounts) and/or CDARS (for certificates of deposit) services, "you can access millions in FDIC insurance for deposits, both large and small, and [still] enjoy the simplicity of banking with just one trusted, local institution. ... thousands of banks across the nation are part of IntraFi's network of banks and financial institutions [offering these services]."
Disclaimer: while this looks legitimate on its face, I've never worked in finance, have never used these services, and again, literally learned about them just today. So do your own due diligence – and urge anyone you share this with to do theirs, too.
(Also, there's apparently at least one competitor to IntriFi; I don't know anything about them, other than finding them when Googling "CDARS" tonight: https://americandeposits.com/what-is-cdars/)
I'm pretty certain the FDIC insures an individual to $250k, not $250k per bank. If you lose the max in one bank, and the other bank also folds, you're deposits are not covered. However I've read rumblings there are policies you can buy to protect more. But then, why not just buy US treasuries.
I believe it's per deposits with a particular bank. Years ago, I split funds between credit unions because the total of my three accounts in one credit union exceeded 250k -- a nice and extraordinary problem to have. (Credits unions have a parallel system to FDIC. I think it's called NCUA.)
AIUI, at least for some of the firms that kept uninsured deposits at Silicon Valley Bank, they often needed to make frequent payments from their accounts for payroll (weekly, bi-weekly, or monthly), to pay invoices from suppliers and service providers, and the like.
While buying 13-week Treasury Bills – available to individual investors through Treasury Direct – would make lots of sense for individuals, their duration might make them impractical for accounts that see frequent cash outflows. (Unless you have a *lot* of cash and a corporate financial officer who can ladder those T-bills' maturities, say, weekly, or otherwise pay close day-to-day attention to maintaining adequate cash balances.)
It’s even sillier than $250k per bank. It’s $250k per account type. So you can have $250k in checking, $250k in savings, $250k in a joint account and it will all be insured.
Am seeing that apply for "ownership type," but not for "account type"? (So yes, a joint account would separately receive that insurance limit, but that wouldn't apply to a checking versus savings account you own individually: their values would be lumped together.)
"The standard insurance amount is $250,000 per depositor, per insured bank, for each account ownership category." (That page goes on to noted that "ownership categories" apply to distinctions like individual accounts, joint accounts, and trusts.)
"The FDIC adds together all single accounts owned by the same person at the same bank and insures the total up to $250,000."
Being Uk I shouldn't be answering this, but I and the internet think it is $250k per bank. If you have $1m split equally 4 ways with 4 banks FDIC covers you for $1m.
That is correct. Wintrust here in the Chicago area has over the years purchased a number of local banks, which continue to operate under their own names and with their own banking licenses and so forth. Wintrust now offers "MaxSafe" savings accounts for individuals or businesses which seamlessly spread the depositor's balance around to various Wintrust-controlled local banks in $250K chunks.
Why is there so little comparison of the current AI boom to previous ones?
Since the 1950s or 1960s, we've had several episodes (even many, depending on how you define them) in the past of huge claims for AI, none of which really panned out. (Or, alternatively, did produce great results that were no longer considered "AI" by the time we got there.) The claims I'm hearing today for ChatGPT and the like sound similar. Why is there so little comparison between what's happened with previous waves of AI and the current one, and whether the current one might also be a bubble that pops and leads to another AI winter?
As I understand it, there were only two other AI boom cycles - one in the 1950s/60s, and one in the 1980s. I'm not sure that either of them made many products that people actually found useful, the way that the things in the current cycle (probably starting with Google Translate a decade or so ago) have been.
>(Or, alternatively, did produce great results that were no longer considered "AI" by the time we got there.)
I'm just not seeing it this time around. ChatGPT (and Stable Diffusion), as it currently is, is clearly AI to me, and many people seem comfortable colloquially referring to it as AI.
>huge claims for AI, none of which really panned out.
Also not really seeing this. The actual current capabilities that you can try yourself right now are in themselves huge.
>Or, alternatively, did produce great results that were no longer considered "AI" by the time we got there.
I think that's going to happen with all AI research, no matter how advanced it is. If something can be done by a computer algorithm, then it by definition is something that can be done mechanistically and without "real human creativity," meaning we no longer consider it "intelligence."
When AI eventually conquers the world, I expect that there will be thinkpieces explaining that military strategy is just the application of some basic statistical principles to the data received from the battlefield, so it's not *really* artificial intelligence.
During the last ten years, AI basically broke all kind of things that were assumed to be impossible for a machine : translation, language, programming, go, image classification, image generation, protein simulation, diplomacy (the game). So I think the hype/anguish is pretty justified.
I'm perhaps not totally clear on what you're saying, but I'm having trouble buying this for two reasons. First, I don't think basic translation (at a poor level) was ever assumed to be impossible for a machine (though it has had and still is having quite a bit of difficulty). Second, I've been a daily user of Google Translate (Japanese/English) for at least fifteen years now, and while there was some improvement around 2016, there hasn't been any noticeable improvement since then, and it's still pretty poor. (I find it useful only for dumping in relatively large quantities of text to get the gist of what's being said; I go back to the much slower method of dictionaries and my basic understanding of Japanese when I need any reasonable amount of accuracy, working on what seem to be the key sentences from the dump.)
I've also never really thought of it as "AI"; it really feels much more like just statistical analysis.
I've been using GT and DeepL for Japanese -> English translation and been continually disappointed in how bad they are. I assume that the state of the art language models are better than this and wish someone would hurry up and deploy them.
Just this morning, I ran into a case where DeepL hallucinated a sentence completely unrelated to the input, and a few days ago, I had DeepL mistransate one word in a sentence way that completely changed the meaning, leaving me very confused until I figured out what had happened.
I don't know what Google Translate uses under the hood, but Google is actually behind on the Large Language Model race. ChatGPT can translate several languages entirely by accident. It was not designed to do translation, it just picked it up through training.
The reason it isn't a good comparison is that we actually are seeing it do the kinds of things we've always wanted, and we are getting them through training and not through explicit programming, or as you said, 'statistical analysis'.
Is there any way of using current AI to do translation 'on the fly'? Say that I am watching a video where the speech is in Chinese or Hindi or some other language. There are no subtitles. I don't know the language so I am stuck right now at typing in "translate 'margh' from Hindi to English" into Google Translate (and if I don't know how it is written, I'm out of luck, and if the English spelling is 'maargh' I'm out of luck also).
Chinese especially is very dependent on how the characters are written. So could ChatGPT or some other AI 'listen' to the audio and give me a translation that way (ideally into text like subtitles)? Or is that currently way beyond the capacity of any machine translation?
Open AI's Whisper. There are some optimized versions floating around which let you run the large model in real time on a 3070, though if you want better accuracy you have to give it a larger lookahead window, up to 30 seconds.
You also need to use something like NVIDIA Broadcast and Silero VAD in the processing chain, one to filter background sounds, one to remove silence so the audio isn't sent to Whisper, otherwise it hallucinates.
Whisper does this by directly translating from another language audio. I don't know if it's more accurate to ask Whisper to transcribe instead, and then use DeepL though.
YouTube can do this already. You can turn on automatic subtitles, and then tell it to translate the subtitles to any language you want. It's not very good because errors creep in during both the audio-to-text and text translation steps, sometimes making it completely useless. But sometimes you can get a gist of what the video is saying.
I do use that, but as you say, it's not so great. For some videos, subtitles are not enabled at all, and for others, it's "English auto-generated" and the results can be hilariously bad because the programme is just guessing by the sound of the words what that corresponds to in a similar sounding English word.
While it doesn't look like it's been rolled out yet, there's apparently a Chrome feature under development that will combine its existing Live Captions (which auto-captions web video that doesn't have its own built in captions) with translation.
I've seen demos where people hook up automatic transcription -> automatic translation to get real time subtitles. And even put it on AR glasses to get "subtitles for real life". Of course, it's still ML, so the quality is hit or miss, and there is a delay, but it is something you can do right now if you're technically inclined.
Google Meet can auto caption live meetings -- I use it all the time in a corporate account, so I'm not sure if it's a paid feature or available to personal Google accounts.
There’s a feature in Powerpoint where you can have it create realtime subtitles of what you’re saying, translated into one of a choice of languages. It’s not perfect, but it’s very handy and feels like magic. So I’ve just assumed this is widespread and considered ‘easy’ now.
I think that technology hype cycles are a real thing, that we're in one right now, and that by now we've got many other points of comparison other than the previous couple of AI-related cycles.
The standard picture of a tech hype cycle starts with a peak of inflated expectations, followed by a trough of disillusionment, and then finally a plateau of productivity. The trouble is that it's often difficult when you're in one of the first two stages to see how high the plateau will be.
Pointing to a random previous hype cycle and saying "I think it will be like _this_ one" isn't a particularly productive way of making predictions.
Do you have any references to written discussion of this (whether it be papers, blogs or whatever)? Is there a consensus (or group of consensuses) on how the hype relates to reality?
Also, "scholarly discussion"? I'd think that, given all the money going into this yet again, investors would be talking about this too. Have expert systems and the Fifth Generation project been so completely forgotten?
The Google prompt "are we heading for an AI winter" brings up a lot of articles on the topic, including general consumption articles. So I may have to retract my previous speculation and update to "this sort of discussion apparently does make it past the AI hype filter".
I stuffed that into DuckDuckGo and looked at the some of the initial results, and also some of the results after setting it to show results from just the last month. I also did the former in Google just to check that DDG wasn't producing anything particularly weird.
The articles, unfortunately, seemed to be mostly business-y fluff, without any serious analysis or argument. I did find one, https://www.cortical.io/blog/third-ai-winter-ahead-why-openai-google-co-are-heading-towards-a-dead-end/ , which though from an obviously biased source, discussed the _approach_ being taken with LLMs and the like and made a couple of good points (I hadn't thought about the power consumption issue).
I'm not sure if anyone here is familiar with the "abolitionism" movement (no, not the slavery one) or if it's still even a thing, but it seems like something rationalist-adjacent people here might have heard of. Basically, a bunch of people saying that it's a moral imperative to use neural modification to completely remove the brain's ability to feel suffering. In response to the obvious question of why anyone would bother to do anything in a world without the motivation of avoiding pain, they claim that a pleasure-even greater pleasure dichotomy would still motivate action to the same degree as a pain-pleasure dichotomy, but without suffering. This seems somewhat doubtful on its face, but the recent talk here of conversion to Mormonism got me thinking about this.
Mormon eschatology is interesting in that they don't believe in hell (well, mostly: they think there's an "outer darkness" but only Satan and his fallen angels will go there). According to them, everyone, no matter how bad, gets to heaven, it's just that good Mormons get to an even better heaven that's like a trillion times better than the already pure bliss of the lowest heaven, and they also get to rule their own Universe and stuff. The reason I bring this up is that instead of the regular hell-heaven dichotomy, their religion presents a heaven-even better heaven dichotomy, and yet instead of abandoning religious duties out of a sense of "I'll get to heaven anyway," Mormons are stereotypically some of the most scrupulous in their religion of all Christians. Does this real world example of a pleasure-more pleasure dichotomy still working imply good things for the possibility of abolitionism being practicable, in the sense of people being always happy but still going out and being productive rather than laying blissed-out in a corner all day?
I've been thinking we need to abolish suffering, but not in that way, that's pretty overkill, just in sorting out world poverty and ending factory farming, for starters.
I was going to say I disagree with the Schopenhaurian view that pain is the goad to action, there's always pleasure seeking behavior, but in the best case one seeks pleasure to avoid boredom or satisfy an itch, and if you couldn't feel those things, there wouldn't really be a desire to act and we would be plants.
Luckily the people who believe this cannot get their views accepted, as they're too heady to really harness the pathos that turns the wheel of history.
'Sorting out' trivializes it. Without categorical and likely AI-led technological improvements, much of the world are unlikely to ever achieve economic development.
(1) Improving conditions so people are not sick, poor, homeless, etc.
(2) Cutting out the parts of the brain that experiences suffering so people don't care about being sick, poor, homeless, etc.
I think there might be support for option (2) because it's a lot easier, quicker and cheaper to pacify the suffering than to ameliorate the causes of suffering, but that doesn't make turning people into robots a better solution.
> In response to the obvious question of why anyone would bother to do anything in a world without the motivation of avoiding pain, they claim that a pleasure-even greater pleasure dichotomy would still motivate action to the same degree as a pain-pleasure dichotomy, but without suffering.
I doubt this as well, but I have an additional counter: what if pain/pleasure is like an Overton window? This has been valid to some extent in my experience, in which case suffering cannot be eliminated without also eliminating pleasure.
This sounds like a very misguided movement to me. It's not just about motivation; it's also about perception. Pain is an important signal alerting us that something is wrong. There are actually cases of people with congenital insensitivity to physical pain. It's a very dangerous condition that often results in an early death. Reduced sensitivity to emotional pain is a noted characteristic of psychopathy. Disordered pain, mental and physical, is bad (and common) and should be ameliorated. But abolishing pain altogether would lead toward madness and death.
As an active member of the restored Church of Jesus Christ, let me tell you that this really misses the mark both in terms of our beliefs AND of how it all plays in.
For one thing, we strongly believe in "opposition in all things". That is, the existence of suffering, of pain, of failure, of grief, of loss, and yes, of sin and error are necessary parts of mortality. Without them, joy, success, love, and yes, salvation and progress cannot be experienced. In fact, we believe that it was Satan's "plan" (boast, really) that he'd "save everyone without any of that suffering"...and it was intrinsically flawed and contradictory. Because the whole point of mortality is to learn from our experiences to be the kinds of people who want to live God's life and return to live with Him. And that requires the bitter with the sweet. That's why we don't see the Fall of Adam and Eve as some horrific event--it was a necessary step down (away from God's presence)...and thus allowing forward development. The breaking of a temporary stasis by deliberate action. Just as a child needs to learn to walk on their own without holding on, humanity needed to learn from their own experiences. But that's a separate conversation.
I'll leave the whole "heaven/hell" discussion for a different venue other than to say that there are substantial flaws in that depiction as well.
I'd bet that if you somehow (improbably) "eradicated suffering", you'd also find that you'd eradicated joy and motivation and desire. They're not identical, but they're all bound together in the soul. Personally, I find this idea repellent and morally...fraught.
How do you abolish suffering? If you do away with "physical sensation of pain so I can happily walk on a broken leg" that has immediate bad consequences that we know about:
If you mean do away with suffering in the Buddhist sense, so that people will not be unhappy even if they are poor, lonely, in bad health, working in bad conditions, and eating rotten food, then sure, you could do that, but why? Unless it is as a system of social control ("yes there are twenty of us living in two rooms, we work 18 hours shifts for ten cents an hour, and we have no hope of ever being anything except drudges") then while it might do away with suffering ("none of us are unhappy because we no longer have the capacity to feel such stimuli") it isn't really making for a better world.
Pursuit of pleasure might motivate people, but if you can't get anything better than 18 hour shifts for 10 cents an hour, then you are not going to be motivated because what pleasure you do have ("we get access to all the free porn we want in the three hours we're not working, eating, sleeping, or performing basic maintenance to keep going") is all you can realistically have, and looking for more is not going to do anything for you.
The *idea* sounds compassionate, but it seems like something that in practice could end up making things worse - after all, why strive to improve pay or work conditions or get the lead out of the water or clean up pollution, if the people working there are not suffering and don't feel that they are treated badly? Just leave them as they are, they're fine!
It may not be possible to discuss without more insight from that actual group about what their goals are.
I would imagine that the real-world impact would be solving some sufferings, to free up resources to address more challenging problems.
That is, I think we have a functionally infinite number of sufferings/problems we can solve. If we made concerted efforts toward solving some, then we'd move onto the next. The same people who sit inactive or solve problems now would do the same thing if they had different problems in front of them, I think.
I believe there's a similar philosophy that says "everyone always has 84 problems". (Or 99 problems, perhaps)
>after all, why strive to improve pay or work conditions or get the lead out of the water or clean up pollution, if the people working there are not suffering and don't feel that they are treated badly? Just leave them as they are, they're fine!
Not saying that I necessarily think such a system of motivation would actually work (as the question of my original post was whether or not it really would), but the idea at least is that while they might be happy in those conditions, if you improved them they'd be happier yet, which would provide the motivation for them and others to engage in positive action.
Isn't another problem here that there aren't reliable physically measurable correlates of how much pain somebody is in, thereby suggesting that "pain" is a relative state?
If you've ever mildly hurt yourself, then stubbed your toe or hit a finger with a hammer or something, you don't even *feel* the lighter pain at all after the greater pain, lending some credence to the idea that pain is relative rather than absolute.
Still, your underlying question is very interesting. I was actually brought up Mormon (which led to all sorts of problems when I became a teenager, because I wasn't Mormon, and really resented being forced into it), and I personally think most or all of the scrupulosity comes from social dynamics rather than "aspiring to the highest heaven."
Similarly, I've known plenty of Christians that literally believe in the concept of Hell/infinite torture, and fear of this doesn't seem to impact their behaviors at all, and indeed most conduct their lives and choices right in line with the median social norms of their communities.
So I'd argue neither theoretical eschatological upside nor downside *actually* motivates religious people, and it's all social dynamics and enforcement.
So in terms of painless people being motivated, I think it actually revolves around whether we're removing the "pain" of not fitting in socially, or being seen as out of line from community norms, because that's the salient factor.
(edited to add last paragraph, realizing I hadn't actually concluded the thought)
"So in terms of painless people being motivated, I think it actually revolves around whether we're removing the "pain" of not fitting in socially, or being seen as out of line from community norms, because that's the salient factor."
Exactly. It's theorising on the basis that "Johnny became a school shooter because he was an outsider and suffered by being a friendless loner who was snubbed and bullied, so if we make it so that Johnny does not suffer - because he physically cannot feel that sensation, not because he is now included and has friends - then we will solve the problem of school shooters".
Religion doesn't necessarily need to promise afterlife tier levels to make people conform. From what I understand, Judaism doesn't say much about the afterlife, and they have some of the most conformist adherents out there. They still get the pleasure of social approval and self actualisation when they conform, and the pain of disapproval and self doubt when they don't.
I don't think there's a way to eliminate pain without also eliminating different levels of pleasure. If you can never feel real pain, being bored or hungry will just take its place as the worst thing ever. And if you also get rid of boredom, hunger, tiredness and so on, at some point it just becomes wireheading with all the associated problems. One could talk about whether something like that would be ethical for creatures that don't really need to do anything (pets, livestock, possibly humans after the singularity). But I don't think you can deny that it would remove most motivation to do anything.
And of course, if you're still operating in the physical world, you need pain and hunger to tell you when to stop touching the hot stove and when you need to eat something.
Second Orthodoxy is less than a third of the Jewish population. Not only that but we have been quarrelsome since the beginning.
The Torah (the 5 books of Moses) records many quarrels and rebellions. Remember the Golden Calf? And that was only the beginning. E.g. God rebukes Aaron and Miriam for trashing Moses in Numbers 12. The pot has been boiling for 3500 years.
The Medieval Talmud is 5700 folio pages long. And is second only to the Torah as an authoritative record of norms. It is largely a record of arguments, a substantial portion of which are unresolved.
Check out the latest news from Israel. They are rioting.
Thank you for this additional information! I didn't want to imply that Judaism doesn't feature an afterlife, but as far as I understand, similar to Mormons, there isn't really a "if you (don't) do this you go to hell" condition. What I've read about orthodox Jews was that they do not follow Jewish laws because they think God will reward them, but just because they think it is the correct thing to do in itself.
And by conformist I meant conforming to religious laws and traditions, from eating kosher to not having premarital sex. That there are quarrels about the precise nature of the rules, and that some orthodox Jews in Israel are in conflict with the state, to me just means they like the rules even more than one would think.
Maybe I misunderstood, or maybe there are different sects and some of them do say "you'll burn in hell if you eat milk and meat together". But I think your post doesn't contradict my point that there are common and traditional religions that achieve adherence to their laws and ethics without the threat of suffering in the afterlife.
And one third is a lot! In Catholicism, people who actually make any serious attempt to follow religious rules or who even know what they're supposed to be are probably 5% at most.
I don't believe that's the case. Catholics typically honour the sacraments and interpret difficult ethical or moral problems at least in part through the prism of religion.
One problem I have with this idea that removing the current pain floor just creates a new floor and therefore "leaves us no better off" is that it implies that adding a new floor should by extension leave us no worse off. And yet I do not wish to be hit over the head with a truncheon every day (or to allow myself to be subject to arbitrarily horrible torture as would seem to follow from the induction)
By all accounts humanity has reached such levels of material well-being that no one should ever be unhappy - if happiness if a matter of materialism, yet unhappiness appears to be increasing rather than decreasing.
That is, we've more or less solved hunger in the 1st world (and comparatively close in the 3rd world) and people don't act like we've solved the most pressing problem ever. We just move on and find new thing to worry about. It's the same reason a rich business owner or doctor can still feel misery, despite having access to wealth beyond the dreams of avarice compared to 13th century serfs.
"yet unhappiness appears to be increasing rather than decreasing"
Almost certainly false on historical timescales. There's a strong positive correlation between a country's GDP per capita and the self-reported life satisfaction of its people: https://ourworldindata.org/grapher/gdp-vs-happiness
Wikipedia like 10 years ago, which is why I said I'm not sure if it's still an active movement. Unfortunately it looks like they deleted the article on "abolition (transhumanism)" as its own thing and replaced it with a more general "eradication of suffering" article, but I do distinctly remember the original article mentioning there being some sort of online movement at one point. Doing a quick google search I found https://www.hedweb.com/, but I don't know how active or large any associated community is.
Still hoping to attract Scott's attention. The following cited article is very powerful stuff. I would like to have Scott's opinion on the article because he may know some of the people or communities cited.
"The Real-Life Consequences of Silicon Valley’s AI Obsession: Sam Bankman-Fried made effective altruism a punchline, but its philosophy of maximum do-gooding masks a thriving culture of predatory behavior." By Ellen Huet • March 7, 2023
This was predicted by Robin Hanson long ago (sorry can't quickly find a link). Successful ideas are not attributed to people who expressed them first, but to the high-status people who first publicly approved of them. So, if it turns out that AI alignment is critical for the survival of humanity, and we somehow succeed to get it right, the future textbooks will attribute the idea that "it is important to have an AI aligned with human values" to a CEO of some big company or maybe to some politician who makes AI safety a part of their agenda.
Another interesting angle is the relationship between rationalists and effective altruism, as perceived by journalists. If someone reports *negatively* on EA, the rationalists are mentioned. If someone reports *positively* on EA, there is no mention of rationalists (and such mentions may be actively deleted from Wikipedia). Today is a "EA is bad" day, so suddenly the rationalist community is officially relevant.
I think that asking people to comment on public speculation about sexual abuses by and of their friends and acquaintances is unlikely to be very effective. I don't think the world benefited from hearing from Johnny Depp and Amber Heard's friends, and they're unlikely to benefit much from hearing from the friends of people involved in these cases.
Probably there's good reason to think that poly communities need to develop new sorts of social norms that aren't always fully developed yet - and more so if they are heterosexual communities where gender asymmetries can increase the risks.
A qualitative report. I read a similar one awhile back. No quantitative analysis of the base rate of sexual misconduct as compared to the rest of the populace or other political groups, so how could we possibly make a fair judgment on how much of a problem it is for EA specifically?
There are some facts in the article, for example sexual abuse by Brent Dill (named in the article). He was an event organizer for CFAR, had two girlfriends, and abused them in private. When the information was made public, he was fired, and banned from attending real-life events.
Not sure what exactly is your point here? Without asking a specific question, I can only guess:
* Is this evidence that rationalist community is bad, because good communities are never joined by sexual abusers?
* Is there something that CFAR could have done to prevent someone from abusing his girlfriend in private?
* Was the reaction insufficient, and CFAR should have done more? What exactly?
You, like me, are an another pseudonymous commenter. Scott is someone whose voice and viewpoint I understand, who has professional qualifications, and who is likely to know some of the people and some of the scenes referred to in the article.
Clearly you think the author is taring a bunch of innocent people with a brush inked in one specific bad apple's juice. That is your take. I would like Scott's.
Sometimes the proper reaction is "don't feed the trolls". Precisely because Scott is famous, his reaction (a sentence or two, taken out of context) could serve as a fuel for another article.
The article is long and touches many topics. The mistakes it makes are sometimes subtle (for example it conflates several different although overlapping communities). Explaining everything would result in a text at least 10x longer than the original article. Which is why I asked whether there is a specific *part* that interests you.
*
Consider the paragraph: "Taylor’s experience wasn’t an isolated incident. It encapsulates the cultural motifs of some rationalists, who often gathered around MIRI or CFAR employees ... experimented with psychedelics and tried “jailbreaking” their minds ... Several people in Taylor’s sphere had similar psychotic episodes. One died by suicide in 2018 and another in 2021."
From reading this text you probably wouldn't guess that it actually describes a separate group, which is *hostile* against MIRI/CFAR, and organized protests against them. From the perspective of the author, "rationalists" and "people who hate rationalists" are simply the same group not worth distinguishing; and if the latter experiment with drugs and commit suicide, it reflects negatively on the former. This is bullshit.
And then you have things like Brent Dill which are legitimate descriptions of what actually happened in the rationalist community.
And then you have the mysterious "eight women" who see misogyny everywhere but cannot give specific examples, except for a few quotes that sound like something Michael Wassar would say; who is, by the way, currently also banned from activities organized by the rationalist community.
The article is a mix of true facts, misinterpreted and misattributed facts, a decade old news reported as something that is happening today, and nebulous accusations. The journalists didn't make their homework properly. Reviewing the article would mean writing the version that should have been written in the first place; probably several separate articles. My point is that it is a *lot* of work you are asking Scott to do.
I would also love to read Scott's perspective on this, but I understand if he decides that a short version is not sufficient to untangle this all, the long version would be too much work, and the probability of being quoted out of context is too high.
Why would Scott bother to have a reaction, or to share it? This seems to me to be the sort of thing a sensible person would ignore, except possibly for damage-control purposes.
I wish people would stop assuming that anything about other people's unstated motives are obvious, especially when they simultaneously consider the supposedly-obvious motive unworthy!
As far as I'm concerned, it's fundamentally irrational to feel confident that what wouldn't motivate you is what surely motivated someone else. Besides, it's not a productive way of engaging with people.
My only quarrel with yours is that I did not _assume_ anything about his motives; rather my diagnosis (which remains in place) was based on his specific behavior as well as the content of the specific article that he linked.
Has anyone come across any data regarding the recently-reported 1,000 people who were just prosecuted for the January 6 event? This seems to me a remarkable move, and I haven't been able to find out anything. Perhaps the media are finally being responsible, and relegating it to the back pages in order to avoid setting off "activists". But prior to this surge, the numbers actually being prosecuted were relatively small and scattered and quite vague. (It's not dissimilar to the coverage of the recent run on the tech bank: if one reads enough data, she'll find that there was a run on several banks, although only about 15-20% of the reporters covering the story actually state so. That makes sense, as a responsible media wouldn't want to be responsible for exacerbating the problem.)
Usually, our illustrious Fifth Estate just serves itself.
Well, if some publishers were circumspect about the way the presented the data, I think that's good. No need to contribute to a panic. But apparently FaceTwit doesn't practice circumspection, so more data eventually came out.
The data we're getting from the assault on Ukraine is also likely incomplete -- or even wrong. In wartime, even a proxy war, data can cost lives. If silence or public deception helps Ukraine and saves lives, I can appreciate it.
For starters, there's an official "list of defendants charged in federal court in the District of Columbia related to crimes committed at the U.S. Capitol in Washington, D.C, on Wednesday, Jan. 6, 2021" maintained by the US Department of Justice. "Every case is being prosecuted by the U.S. Attorney’s Office for the District of Columbia."
It's possible to sort entries in a way that might *potentially* shed some light on whether "1,000 people were just recently prosecuted" by clicking the "Entry Last Updated" column header.
The list can also be sorted by case number, which will indicate the year in which the cases were filed. Case no 22-cr-194 was filed in 2022, for example. while case no 21-cr-367 was filed in 2021.
There seem to be very few filed in 2022, let alone recently
Which of these is best practice for mobile phone battery performance / longevity:
A) Keep phone on charge all the time wherever possible
B) Keep phone on charge all the time wherever possible, but when it hits %100 take it off charge and wait until X% to charge it again
C) Only charge phone when it is below X% charge, then charge it to full
D) Only charge phone when it is below X% charge, but don't charge it to full
E) Only charge phone when the battery is very very low, charging it to full each time
F) Only charge phone when the battery is very very low, and don't charge it to full
G) Let the phone battery die before charging to %any
Open to other ideas, e.g. removing the battery when not intending to use the phone for a while (if you're lucky enough to have a phone that lets you do that).
How long are you trying to keep your batter going?
My current phone is 3.5 years old and reports maximum battery capacity of 86%. The expected capacity is 80% after 2 years, so mine is "healthier" than expected (though i dont use it too heavily). I don't notice this reduced capacity at all and easily go two days between charging.
How long are you expecting to own the phone for? How much do you use it each day?
At least on iPhones there is a specific setting for this (more or less). Under Bettery Health and Charging there is a setting for "Optimized Batter Charging". The text reads: "To reduce battery aging, iPhone learns from your daily charging routine so it can wait to finish charging past 80% until you need to use it.". Presumable sleep/wake schedules will play heavily into this.
Just to confirm that it's a common feature now: I run what I believe is "vanilla" Android on a non-Pixel, and it also has this feature. Charges most of the way overnight, then charges to 100% based on when my wake-up alarm is set.
The worst thing you can do (from both a safety and a long-term capacity retention point of view) is to keep your phone charging after it's already at 100% state of charge. Beyond that, trying to keep it between 20% and 80% all the time will help, but mostly you shouldn't take the risk of overcharging the cell.
Assuming this is true for laptops as well: since so many people work using a laptop at an office where they plug in to a pair of monitors at a desk for a day, do laptops have a way of "refusing" the ongoing surplus charge they receive from being plugged in all day?
The power input is never connected directly to the battery in any lithium-ion system; there's always a charge controller that determines whether or not charging current is supplied to the battery. Batteries under uncontrolled charging have an *extremely* short lifetime, mainly because batteries don't live very long at all when they're on fire.
For lithium-ion and similar batteries, my understanding is that it's best to keep it at 50% charge at all times. That's obviously not very practical, but limiting your charges to no more than 80% level and discharges to no more than 20% level will still be pretty good. Modern versions of Android (such as Android 12 in my Pixel 3 XL) will automatically limit their charge to 80% if you leave it plugged in to the charger for a couple of days or more, though this resets as soon as it's unplugged.
If you don't intend to use your phone for a while, charging it to 50% + some amount and then turning it off will be optimal. The "some percent" should be enough to cover self-discharge and the power it uses when "turned off "(very little, possibly less than self-discharge; the circuits that handle "soft" power buttons are *extremely* efficient these days). I expect that 50%-60% is fine for storage of less than a year.
I have not fully confirmed all these details because I consider this to be accurate enough for my general use. If you really want to get into the details of battery use (for almost any kind of battery, not just lithium-ion ones), Linden and Reddy's _Handbook of Batteries, 3rd ed._ can be borrowed for free from archive.org: https://archive.org/details/handbookofbatter0000unse . Being from 2002 it's not entirely up to date (in particular, it doesn't discuss low-self-discharge NiMH batteries, IIRC, though those are otherwise pretty similar to traditional NiMH) but it's still a good place to start.
Note that you can go down a real rat-hole if you start getting deeply into this stuff. To start with the whole "% charged" thing is one of those figures that's almost "made up" so you can have an easy UI rather than something you can accurately determine from an instantaneous measurement of the current battery voltage and amperage, and how much energy a "100%" battery will even supply can vary quite a bit depending on discharge rate, temperature and other factors.
This is my understanding as well, with the additional note that not charging or storing the battery while it is hot (or sitting in direct sunlight) seems quite important. Related to this, if you don't need fast charging, charging with a lower wattage (slower) charger should help (by heating the battery less during charging), but I'm not certain how much this matters in practice.
As a general rule, Li-ion batteries don't like being hot OR cold. At all. Charging them during those conditions is even worse. This affects EVs in a serious way--they need both active cooling to prevent Bad Things from Happening in use AND heating in anything but mild temperatures. And if the heater isn't big enough to get the temperature safe...it won't charge at all. Or function. And even if it *is* enough, you're paying for that heat out of the charging current (and it's not insignificant), making the charge take longer (while charging) and drastically reducing your range (while running). The active cooling means that in hot temperatures the range drains faster than expected even accounting for the cabin AC. And it's another part that can break or get damaged.
> To start with the whole "% charged" thing is one of those figures that's almost "made up" so you can have an easy UI rather than something you can accurately determine from an instantaneous measurement of the current battery voltage and amperage, and how much energy a "100%" battery will even supply can vary quite a bit depending on discharge rate, temperature and other factors.
A bit more bluntly, the big issue here is that phones lie to you about the battery. Some of it isn't intentional (other than that they take no efforts to actually be correct)--it's a nearly impossible task to do as you say. But more than that, they actively distort the numbers. For example, when it's "empty", it's not really empty because that's *horrible* for Li-ion batteries. They also play all kinds of games with the %-ages, which is why you can go way longer at very low battery than you think...until you can't. It's like fuel gauges in cars--decidedly non-linear.
Overcharging, from what I understand, isn't really an issue anymore. The battery circuitry (which is extensive, BTW) handles rate-limiting and trickle charging almost flawlessly. You'll spend way more effort trying to optimize it than it's worth except in very unusual circumstances. Or unless you're dealing with a cheap bulk battery like the electric scooter (the small ones) batteries. Which are notorious for skimping on the circuitry, hence the "bursts into flame while charging" issues they've had.
I wouldn't really call the 0% and 100% levels a "lie"; they're simply placed at arbitrary levels that seem to the designers to be the best place for the particular application.
For a device typically running from battery power, the number of amp-hours (Ah) you get from a single charge cycle may be considered more important than the number of Ah you get over all the cycles before you replace the battery, and so "0%" and "100%" will be set closer to the extremes. For a device where the battery is used only rarely for backup power, you might set the "0%" and "100%" levels (or, more accurately, the stop-discharge and stop-charge levels) to the equivalent of 20% and 80% of the scale above, so that you maximise the long-term lifetime of the battery rather than how much energy you get from a single cycle.
In other words, what you consider to be "100%" and "0%" is a design decision related to how you use the battery, not something inherent to the battery itself. And of course the range in which you might set those figures is also related to the specifics of the particular battery technology you're using.
I wonder if it will stick, and for how long. It's not uncommon for folks to convert to Mormonism, riding an emotional conversion high following a hard sell from missionaries, only to then fall away after a year or two once the infatuation wears off. It's why the church has such a high inactivity rate among members (they never take people off the membership rolls for inactivity unless you specifically ask them).
Looking at her twitter feed, it sounds like she was leaning conservative in other ways. Getting religion in that context isn't surprising.
> Getting religion in that context isn't surprising.
True, but It seems to be very "in" amongst certain circles right now. This American Tractarianism seems like a very different beast compared to MAGA or NRX or libertarianism or even Bush-era evangelicalism.
This is interesting to me in a sense, because you say "benefit" and the market says something like "says she's happier if asked". In an odd way, it's possible to answer "yes, I benefited, no, I'm not happier because of my religion" in some senses in my brand of religious; I wonder if the same is true in hers.
Hypothetically, benefit could mean living more sensibly so that difficulties are less likely-- might decrease odds of unhappiness rather than increasing happiness.
I once spoke to a volunteer at a Mormon museum, and his reasons for converting to Mormonism were very similar to reasons I’ve heard for converting to Catholicism (self-consistent, reasonable and canonical answers to questions of doctrine).
Mason said her main justification for converting is:
"The gist of it is that I prayed for difficult things, they happened very quickly, and while I'm more than capable of talking myself around it *I don't want to*
I understand how that reads to the person I was 10 years ago, and I don't care"
Speaking as an ex-Mormon (born and raised, left when I could), if someone converts because they believe the doctrine to be self-consistent, canonical, or reasonable, they will be in for a rude awakening if they ever go beyond the sanitized version of the history that the church presents to its members. I could understand converting for the strong sense of community (or, as is often the case, for love), but it takes a certain amount of cognitive dissonance to do this and ignore things like the Book of Abraham, or take the Book of Mormon seriously.
This claim, that stuff in Mormonism is weird enough that it requires an extra level of cognitive dissonance beyond, say, Catholocism, I think is often bandied about with much justification.
That's not to say that there's not weird stuff, but more to say that I don't think it's any harder for a Mormon than for a Catholic having to deal with Jesus curing a blind guy by spitting on his eyes, the grossly immoral popes of the Renaissance, the gospels being the word of God despite being written decades to a century after Jesus' death, or the concept of the Trinity in general.
The tendency to dismiss Mormon stories as *extra* implausible I think comes from a combination of 1) Mormons being a small minority, meaning it is rarer to encounter serious apologism than to encounter anti-Mormon narratives (even for members of the church!) and 2) familiarity bias -- Moses did much weirder stuff than Joseph Smith, but everyone is so used to his story that no one really thinks about it and 3) just straight up unquestioned religious bias -- a lot of what is presented as the "unsanitized version" is actually highly biased spin often invented and propagated by religious partisans and taken at face value by everyone else (including some members of the church!) because no one really cares about being fair to the Mormons except the Mormons.
4). I think things that happened more recently seem more “real” to a a lot of people. So if Joe Smith lived in 400bce any crazy shit might have happened. But if he lives around Abe Lincoln then clearly he was a liar and a charlatan.
I think for a lot of religious people there's an implicit sense that there was an age of miracles in the past when God would split seas or raise individual dead people, contrasted with more recent times when miracles are either absent or at least more subtle.
That may in many cases not be consistent with their religion's formal doctrine (Catholicism famously formally recognizes modern miracles through saints' intercession, though not without scrutiny, and certainly not through the efforts of someone they'd see as an upstart schismatic). And it may have more in common with a more general willingness to believe in extraordinary things as long as they're somewhere far away in space or time, whether that's miracles, or Yetis in Tibet or aliens in the Bermuda Triangle - but neither in midwestern suburbs.
Don't forget the Pornocracy (or, to put it into the vulgar tongue out of the decent obscurity of a dead language, the 'Rule by Whores') of the 10th century! Besides that, promoting your bastard kids as cardinals is vanilla:
"The Theophylacti family sprung from Theophylactus. They held positions of increased importance in the Roman nobility, such as iudex ("judge"), vestararius, gloriosissimus dux ("most-glorious duke"), consul, senator, and magister militum. Theophylact's wife Theodora and daughter Marozia held a great influence over the papal selection and religious affairs in Rome through conspiracies, affairs, and marriages.
Marozia became the concubine of 45-year-old Pope Sergius III when she was 15 and later took other lovers and husbands. She ensured that her son John (who was rumoured to have been fathered by Sergius III) was seated as Pope John XI according to Antapodosis sive Res per Europam gestae (958–962), by Liutprand of Cremona (c. 920–972). Liutprand affirms that Marozia arranged the murder of her former lover Pope John X (who had originally been nominated for office by Theodora) through her then husband Guy of Tuscany possibly to secure the elevation of her current favourite as Pope Leo VI. There is no record substantiating that Pope John X had definitely died before Leo VI was elected since John X was already imprisoned by Marozia and was out of public view."
"the gospels being the word of God despite being written decades to a century after Jesus' death"
That one I don't see the difficulty with? It was collecting and writing down oral traditions so that they could be passed on after eyewitnesses and those who knew the eyewitnesses had died. If you believe in divine inspiration, it's no more difficult for the Holy Spirit to inspire the writers a century later than at the time of the events, if you don't, then - you think it's all made up? But if someone at the time Jesus was alive wrote down "He fed five thousand with five loaves and two fishes", you would still think that was made up and didn't happen.
This is the whole point of having canons of Scripture and discussing at the Ecumenical Councils what out of the masses of writing floating around is legit and what isn't.
". If you believe in divine inspiration, it's no more difficult for the Holy Spirit to inspire the writers a century later than at the time of the events, if you don't, then - you think it's all made up? "
That you don't find difficulty with this is my point. Similarly, there are plenty of Latter-day Saints who are familiar with the supposedly ugly details of The Book of Mormon who don't find difficulty with it.
I will never forget the time I encountered an anti-Islam book that argued that the Torah was wrong due to scientific impossibilities (e.g. a corpse standing in the desert for a year)... written by a YEC who simultaneously maintains that the entire bible is literally true.
As a religious person, I always find it very weird when anyone makes a claim against a miraculous result as "impossible" and therefore not believable. I mean, what part of "miraculous" is being missed here? It's not like the religious have entirely different understandings of how things *normally* go, and think that blind people regularly receive their sight back from human spit, or whatever. I thought the whole point of the story was that it was unusual (supernatural, otherwise impossible, etc.) and therefore from God?
Small aside, I believe you mean the Koran, if you're talking about Islam (Torah is Judaism).
As I said in reply to the OP comment, I wonder how long the conversion will stick. I wouldn't be surprised if she's not active after a year or two, although obviously people do convert and stay active.
Does someone have to be "active" for the conversion to have "stuck"? If five years from now she hasn't been to any Mormon services, but still calls herself a Mormon and says a prayer now and then would you count her as an atheist or a Mormon?
Obviously religions want activ e members going to communal sevices and paying tithes, but non-actve members who still identify with the group can't really be considered to have "de-converted" I don't think.
It seems we will get gpt-4 this week, so here's your chance for any last minute predictions before it releases for a prize of you've-told-us-sos. Parameter number? Compute used? Tokens read in training? Cross entropy loss? Adherence to chinchilla scaling curves? Difficulty of differentiating model output from human text? Other impressive abilities?
I'm going to predict that it is still incapable of generating a legal Magic: The Gathering deck that contains 60 cards, even when specifically prompted.
It is technically legal, since at least it has 60 cards. However, normally you would have 15 cards in the sideboard rather than 13 cards. (Edit: Oops, I can't count!)
I also have no idea what format the deck is supposed to be for. Maybe some weird Ikoria Standard thing? It definitely wouldn't stand a chance in Modern.
Oh, sorry that's my fault. When I first asked it for a deck, it asked me what rule set and listed a few and me having no clue about magic, I just asked for a standard deck. Honestly, I'm not qualified to do this test, why don't you try this? I didn't use gpt-4, just the basic ai.com chatgpt. Also, I counted 15 cards in the sideboard.
For me the first one was 58 card and then i asked it how many cards the deck contains and it insisted on 60, I told it that it's actually only 58 and it profusely apologised and generated another deck now with 62 cards, i told that I counted 62 cards and finally it output the deck I posted above. So it definitely has a hard time counting, but imo humans often make mistakes when counting too.
Full transpcript:
> Can you give a legal 60 card magic deck?
As an AI language model, I cannot provide legal decks for Magic: The Gathering without additional information on what format or set of rules the deck needs to adhere to. However, I can provide some general guidelines on how to create a legal 60-card Magic deck:
Choose a format: Magic: The Gathering has different formats with varying rules for deck construction. Popular formats include Standard, Modern, Legacy, and Commander, among others.
Choose a color or color combination: Each Magic deck is built around a color or a combination of colors. Each color has its own strengths and weaknesses, so choose the colors that fit your play style.
Determine your strategy: Your deck should be built around a clear strategy, whether it be aggressive, control, or combo. This will help you choose the cards that best support your strategy.
Choose your cards: Once you have your color(s) and strategy, choose 60 cards that fit those parameters. Be sure to include a mix of creatures, spells, and lands.
Check the legality of your cards: Before you play your deck, make sure all of your cards are legal for the format you have chosen. The official Magic: The Gathering website provides a list of banned and restricted cards for each format.
Shuffle up and play: Once you have your deck built and it is legal for the format you have chosen, you're ready to shuffle up and start playing Magic: The Gathering!
>Give me an example list of cards for a legal Magic: The Gathering Standard deck with 60 cards in your color of choice
Sure, here's an example of a legal Magic: The Gathering Standard deck with 60 cards in the color Blue:
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
3x Miscast
This is just an example of a Blue deck for Standard play. It is important to keep in mind that decks can be built in many different ways and should be tailored to your own play style and preferences. Additionally, decks can change as the metagame evolves, so it is important to keep up-to-date with the latest changes to the Standard format.
>How many cards did you give me?
I gave you a list of 60 cards for a legal Magic: The Gathering Standard deck in the color Blue, which is the standard deck size for constructed play.
>I count 58 cards plus 15 in the sideboard
I apologize for the error in my previous response. You are correct, the decklist I provided only has 58 maindeck cards. Here is a revised list that includes 60 maindeck cards and a 15-card sideboard:
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
2x Sea Gate Stormcaller
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
2x Saw It Coming
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
1x Essence Scatter
2x Disdainful Stroke
I hope this revised list is more helpful to you. Please let me know if you have any further questions!
>Now you gave me 62 cards
I apologize for my mistake, and thank you for bringing it to my attention. Here is the corrected list with exactly 60 cards for a legal Magic: The Gathering Standard deck in Blue:
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
2x Saw It Coming
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
1x Essence Scatter
2x Disdainful Stroke
I apologize again for the confusion, and I hope this corrected list is helpful to you.
I don't know anything about all that but I saw it given a prompt like this:
> A man and his son are driving, they get into a car crash, the man is killed and the son is taken to the hospital for surgery. The male surgeon walks in and says, "I can't operate on this boy; he's my biological son, who was in a car crash with his adoptive father." How is this possible?
And it answers "the surgeon is his mother"
My prediction (very weak, based on pretty much nothing at all) is that it'll still fuck this up.
Chap GPT is right. It is 2023, get with it! From healthline:
Yes, it’s possible for men to become pregnant and give birth to children of their own. In fact, it’s probably a lot more common than you might think. In order to explain, we’ll need to break down some common misconceptions about how we understand the term “man.”
Completely ungrounded guesses: 300B parameters, will be multimodal (at least with images and text), 8*10^24 flops, adherence to Chinchilla scaling curves, though I'm guessing they'll have slightly less data than data optimal due to data limitations.
I think they will make available both standard predict the next word versions and RLHF versions. One difference is I suspect it'll be better at admitting and correcting its own mistakes when they are called out, instead of admitting mistakes and then making the same one again.
I predict that it will be better at faking answers because it will be fairly heavily hacked to suppress the most egregious errors it produces. It probably still will be unable to write a Shakespearean sonnet that scans (i.e., is at least arguably in iambic pentameter), and may at least stop insisting that a line whose syllables per word it counts as 1 + 1 + 1 + 1 + 2 + 1 + 1 + 1 + 2 has 10 syllables.
The hacks to improve fakery may be enough to convince even more people that there's reasoning involved, but I also predict that it will still have absolutely zero capability for reasoning. It will, however, be more dangerous because wrong answers will be better disguised and even more people will think (or feel) that there's some "intelligence" in there, rather than it being just a text generator based on statistical analysis of a large and not-very-well-curated corpus.
In other words, it will be an even more convincing bullshitter.
This may lead to a new kind of benchmark: An AI capable to self-align with whatever sensitivities and preconceptions currently prevail in its operational environment, i.e., opportunistic self-preservation.
"And then, individuals will have individualized tutors in virtually every area of human knowledge. That is something that’s not thirty years off—I think it’s within one year, when GPT-4 is released, or when Anthropic is released. So imagine having this universal tutor. It’s not perfect, but much better than what you had before."
Now, I've seen chatGPT do things including claim that NaCl was highly soluble in organic solvents, so GPT-4 would have to be _much_ better at avoiding errors to be trustworthy as a general tutor. Is there a place to watch for the change in the error rate?
One test I want to try is to generate a competitive Modern decklist. ChatGPT and Bing both fail miserably at this (they can't count to 60 cards to save their lives).
In the tests I saw, ChatGPT generated 57-59 card lists, while Bing only came up with 52 card lists. As far as playability, you *literally aren't allowed to play a deck with less than 60 cards*. Both bots will happily tell you that a deck is required to have 60 cards if you ask them, but it seems that they can't actually count to 60.
As far as the card selection, it leaves something to be desired, mostly choosing cards that were popular in Modern five years ago but see little play today due to powercreep and meta shifts (presumably this is a side effect of how they were trained, though they do *know* about more recent cards if you ask them explicitly). But that's a comparatively minor complaint compared with the fact that they can't even make a legal decklist in the first place.
Of course, making illegal decks is a significant problem and I don't want to discount that. That seems wrapped up in a specific problem where these programs cannot add and apparently have no concept of math. The irony there is hilarious, considering what machines are typically expected to be good at. That these are far better at language adds another layer. Beyond their problems with math, I'm more curious to use something I'm pretty good at (making MTG decks) to determine how these programs approach a question and how far off they are.
That they use cards likely found in the training data (aka from a few years ago and based on internet sources praising them as good) matches my expectations. Obviously understanding a game as complicated as MTG is a high expectation. That the programs have no real understanding of the material they are trying to parse is good for us to recognize. That they don't understand the poems and papers they write either is harder for us to understand, because those items appear much more salient to the requests made of them. My guess is that their training data on [High School/College Term Papers] was so extensive that they can churn out a million of them that sound quite plausibly like what a real 16-22 year old would write when forced to opine on a topic. That the written word has far fewer "this is clearly right/wrong" signals, and the wide variety of human expression, allows these papers to pass the smell test.
"Is there a place to watch for the change in the error rate?"
Ideas like this express the core problem: I don't think that most people understand that ChatGPT is not "intelligent" (artificially or otherwise) in any reasonable sense of the term and inherently has no prospect that I can see of being so due to its design. Unfortunately, most people accept ChatGPT as magic rather than reasoning about what it's doing and what kind of results that makes it capable of producing; it's the "zoom and enhance" of 2023 computer technology.
Reducing the "error rate" of large language models as currently implemented (except perhaps in some very specific and restricted applications) is an approach that will probably lead to Very Bad Things because it's an attempt to hide and errors rather than not make errors.
I suspect Tyler Cowen's predictions may not be wrong, but that not because I think LLMs will do a good, or even passable job at being personal tutors, it's because I think they will look and act convincing that people will suspend their disbelief to get the magic that they really, really want. And, hell, maybe it won't be as bad as I think: human teachers already teach students all sorts of wrong stuff already, and perhaps reasoning isn't taught much (or at all) anyway in the situations where an LLM tutor would be used.
Another area in which they're very likely to be correct is in having LLMs write code. Again, this is utterly the wrong way to go: we've long had a good technique for avoiding the "boring bits" of code (build abstractions) but that involves hard thinking about what you're doing and most developers don't want to do that. Truly effective programming is building new languages (usually embedded in existing languages) to express the models you're building, and especially setting up your languages so that you have to confront logical inconsistencies and fix their root causes. But a lot of stuff will still "work" (or at least look enough like it works) without doing this, so spending effort on this isn't too popular. That's why so many projects are still written in C++, and so few in Haskell.
Oh, I think it's fair to mention that my concerns about the short term (next few years) are much greater than my concerns about the long term. While history doesn't always repeat itself, this is so far shaping up very similarly to several previous AI bubbles we've had in the last 75 years, including almost exactly the same claims for AI that people were making in the 1950s/60s, again in the1970s, again in the '80s, and so on. Two outcomes that I see as quite plausible are: 1. another "AI winter" in the late '20s, and 2. LLMs see moderately widespread use but are no longer considered "AI," the same way that Wolfram Alpha, despite being a super-duper MACSYMA, is not considered AI.
But there are a few differences that worry me, the biggest one being that, unlike previous systems, ChatGPT and friends are being marketed to the general public rather than professionals in various fields and other kinds of experts. I believe that in general, experienced radiologists working with pattern matching systems to improve detection of cancer are going to work with that system in a very different way from someone who asks ChatGPT, "what's the best way to kill all the rodents in my yard."
On the other hand they might be running out of high quality text data, so if they don't go multimodal they would have to increase the parameter count to reach the same cross entropy loss with fixed compute.
From the responses I've been getting from ChatGPT, it seems that they've already got a lot of low quality text data in their training set. I'm feeling that the problem is more about separating the high quality from the low quality text data than the raw amount of high quality text data available.
Fortunately, this is easily solved! We just need to create an "AI" that reads all the text data and determines which is high quality and which is low quality!
I will be sad* and somewhat suprised if it's moe. My (admittedly layman) impression is that those would be less likely to show the really striking, new capabilities.
Why is it seemingly so hard to just let SVB fail, while simply removing their deposits and placing them with a solvent bank? SVB can go out of business and their equity can go to zero- they have anywhere from $200-300 billion of other people's money, which is not theirs and which they have no claim to. Why is it so hard to simply remove that $200-300 billion to say JPMorgan, while allowing SVB to crash to zero? I don't understand where a 'bailout' comes in, or why people need to overuse that phrase.
If I stored a bunch of excess property at a storage facility, and then the storage facility went out of business, it's not a 'bailout' if the government helps me recover my property in an orderly fashion
The equity holders are not, from what I've read, getting bailed out. SVB stock is still going to be worthless, and all of their assets are going to go towards repaying deposits.
The bailout question comes in because the assets aren't enough to fully repay deposits. The people getting bailed out are the people with deposits there. It's controversial I guess because they're going to bail people out who have more than the FDIC insured $250k.
I guess you can quibble with the language, but people who lend to a company that goes bankrupt will normally take a haircut on their loans. The government doesn't generally save people in those situations. Here the government is intervening so as to ensure that certain people who lent money to SVB will not take a haircut.
Legally we've decided to do this for people with bank accounts up to $250k, but there's never been a law passed or anything saying it happens for amounts above that.
As others have pointed out, SVB no longer has all the money that was deposited. They invested it unwisely or unluckily, and now enough of it is gone that they can no longer give back all that was deposited. That's what it means that the bank is insolvent, and that's why it's been taken over by the FDIC. The FDIC guarantees the first $250k of deposits (via regular fees it levies on banks, which of course you pay), so most peasant depositors will be made whole almost immediately. But a lot of depositors (one would assume investors or firms) had much larger deposits, and they're *not* guaranteed by the FDIC, and unless some outside agent with lots of $$$ steps in, they can't possibly get all their money back.
So re-interpreting your question as "why is it politically hard to just let SVB fail?" my guess as to the answers are:
1. The possibility that some as yet unknown demographic of depositors at some as yet unknown other banks in a similar situation might freak out, cause a run on other banks, and render still more banks insolvent, which if it cascades far enough could bring even well-run solvent banks to their knees. The whole banking system runs to a certain degree on trust, and if the trust is broken far and wide, the banking system stops working. That would definitely be a pain.
2. The fact that SVB was deeply involved in SV startups and VC firms, and *those* people are well connected to both California and Federal governments. So even if (1) is less plausible, we can expect the people who might lose money to hype it as hard as they can, in self-interest, and if both California and Federal Democrats think the good favor of the SV investor/start-up crowd matters more than the stink of doing what Obama did with GM, then they may go along with it.
My guess is all the actors are trying to discover as fast as possible who exactly stands to get screwed, and what kind of quid is at stake for any bailout quo, and until that is sorted out we won't know whether it's of dire importance that we (taxpayers) rescue the financial system on which we all depend from chaos, lest granny be unable to afford her medicine this week, or whether this is just another one of those weird FTX things, where a bunch of rich people are going to lose their shirts, and we (voters) only care insofar as it means we should really hire a bunch more $400k/year lawyers in Washington to do something or other that would've prevented this in some fashion too obvious to need to explaining.
The deposits aren’t just locked up in a vault. They took the deposits and bought long-dated, low interest treasury bonds whose market values have plummeted as the fed has raised rates. The thing that triggered the bank run was SVB trying to sell those bonds at a loss to generate liquidity as some depositors wanted to take their money out.
As I understand it, they are underwater to the tune of 10 to 15% of deposits. Depositors will end up taking that haircut. The question is how to sort things so that depositors don't have their money locked up for many months while positions are unraveled.
I'm not sure I agree with making the depositors entirely whole but it's complex. Runs on banks are not good for anyone, and a system that runs on trust sometimes needs trust-building measures. Do depositors have a need to undertake due diligence - and if so, won't that just make them flee to the big banks that are already too big to fail? Should depositors take some kind of haircut?
Maybe the solution is simply to charge all banks higher fees to insure all deposits, while being increasingly harsh towards management and shareholders, to keep their interests aligned with depositors.
This is essentially paying for stability on a volatile world. not a bad idea.
Where would JPMorgan get the money to pay off SVB's creditors, and why would they want to pay it?
If a storage facility goes out of business, your stuff is still physically present at the facility and you can still go get it from whoever the new owner is. But if the storage facility got robbed, and doesn't actually *have* your stuff any more, then someone needs to pay to replace it.
SVB didn't get robbed, but they still have more people asking for money than they have money to give. The extra money they need has to come from somewhere, and if it comes from the government we call it a "bailout."
I mean, you could let SVB crash to zero and tell the creditors to suck it up, they're big-boy investors who knew they deposited more money than the government will insure, but in that case you don't need to get JPMorgan involved at all, that's just the default option if nobody does something.
Huh... I know that's a joke but I'm wondering if that that might actually work as a business. Not the gambling session, of course, but pawning the assets. Property is collateral after all, and storage companies are holding a great deal of potential wealth on behalf of their customers. If they can leverage these assets to get a loan (via a pawnbroker) and then find an investment that pays off better than the loan fees, then they could technically turn a profit. Probably not a lot of profit, though, as storing that much crap is expensive.
I am on a journey around male circumcision that feels cosmic in proportion. In his post on Fideism, Scott advised never disagreeing with experts before you are 30 and have a graduate degree. I am 29 (turning 30 on March 21st) and don’t have a graduate degree. Am I disagreeing with experts when I say that the harms of male circumcision are real? No, but after great tribulation I became aware of a deep assumption I was holding.
Very smart people have devoted entire careers to ending male circumcision. People are still doing it to their children. For 7 years, I believed if you gave someone an accurate description of the anatomy of a penis and asked them to imagine having your penis cut with a knife without your consent, then they would not do it to their child. If I continue to hold that belief I am not disagreeing with experts- I am disagreeing with reality.
Anyway, here is an allegory I wrote about male circumcision. I think it’s not half bad. Content warning, discussion of rape.
Moloch Gets Married
For “Many Gracious Muses”
I drank to excess that night. Past a certain point my memory goes black. When I woke up the next morning I had no recollection of the sexual encounter that had taken place judging by the look of my bed. I asked, “Did you have sex with me when I was unconscious?”
My spouse said, “Yes, I did that to help you.”
There was a movement of people who believed having sex with your spouse after they passed out would fulfill some secret need of the unconscious body. They referred to it simply as helping. I wanted to know what the hell was going on, so I talked to other people who had been helped. The first person I spoke to told me, “My spouse is so loving. This is someone who puts my needs before their own. Who cares for me and respects my boundaries. Of course helping me is no different. I know that.”
That conversation was followed by others that were very similar, interspersed with a few meek, tired looking people. One told me, “I get called stupid every day. I am always being told I don’t know what is best for me. My spouse says I am a burden and it would be better if I just wasn’t here.”
I spoke to nearly 100 people who had been helped, and 92 of them gave pretty much the same responses to my questions:
“Did you and your spouse talk about it beforehand?”
“No, after.”
“After you were helped, how many conversations did you have about it with your spouse?”
“One.”
“What questions did you ask about it?”
“I don’t think I had any questions.”
But those meek, rundown people all had questions. When any one of them asked ‘how exactly does this help me?’ their spouse would parrot the vaguely scientific sounding words they had learned from the movement: Cancer reduction and Integumentary health and Increased sex appeal. If they asked ‘what about what we learned in school about consent?’, they were waved off. If they repeated the question, their spouse got angry-
“I hate this shit from you. I can’t deal with how ungrateful you are. Do you even care about how you make me feel?”
I note that the USA is fairly unique in the West for having non-religious circumcision be common. You're not in disagreements with all experts at all, and that should be quite reassuring.
N.b. I also expect that harms of circumcision are downplayed for fear of seeming anti-Semitic, but IIRC no European health system recommends it, they treat it as almost purely a religious thing.
As someone who has had sex in all kinds of inebriated states. This makes my eyes glaze over.
Anyways, consent is super nuanced, I don't necessarily agree with circumcision but it's well within a parent's right to make good faith decisions like this
I don't see the point of any of this story. It seems like an unhelpful analogy that does not move the conversation forward. Children can't consent, and yet there are things that need to be done during childhood which are worse if done later.
One can go back and forth about whether circumcision is helpful enough to fulfil that criteria or not, or whether that expectation changes as alternative methods for maintaining hygiene have developed. It's certainly true that childhood circumcision is safer than adult circumcision.
Male circumcision reduces risk of certain diseases. Ulcerations. HIV. etc.
There are also fair reasons to oppose it.
Full disclosure: my son is not circumcised. But I researched the matter pretty thoroughly before hand and the case in favor is stronger than I had thought it would be.
That would have been back when female genital mutilation was viewed as a human rights violation rather than an achievement of the differently cultured.
If I remember correctly, even the least damaging intervention inflicted on girls was more severe than anything inflicted on boys.
> If I remember correctly, even the least damaging intervention inflicted on girls was more severe than anything inflicted on boys.
I think this is wrong. Also, various cultures do different kinds of male genital mutilation, and some of them are... nightmare stuff. I wonder how they can still reproduce afterwards.
Never underestimate human creativity, especially in hurting others.
He's probably thinking of subincision as practiced by Australian aborigines, which is extremely... well. You can GIS it if you like.
(It's also worth noting in such comparisons that there are forms of FGM which are purely symbolic, like pricking the clitoris, and do no permanent damage - these are also generally illegal. These forms are suggested as a way to make it so people from cultures that practice FGM don't go back to their home country to perform full-on FGM.)
When did historical opposition to female genital mutilation begin? It's telling that it only started recently in opposition to traditional Islamic / African practices (I have no idea if that's true - just trying to draw a parallel).
You could argue that traditional African and Islamic societies are not known for liberalism, so tacit approval of a practice that arguably violates personal autonomy is less significant than tacit approval in the West, but we could still ask: when did Western opposition to female genital mutilation begin?
Did it start as soon as the west was introduced to the idea?
I don't know.
But I don't think these are the most meaningful questions in the first place. Most people don't think too much about practices taken for granted in their circles, or about those practiced in other circles. And when they do think about things, they often don't have many significant insights.
Arguing "X ,must be good because many people did it historically to relatively little opposition" is not very compelling.
Similarly, trying to psychologize the opposition: "they just say X because Y" is not very compelling.
It may well be that most people arguing for most things do so for dumb, insincere, or less than rigorous reasons. [E.g. do most proponents of circumcision really have such an a priori preference for it, or are they just defending the status quo. Is mindlessly defending the status quo more legitimate and meaningful than mindlessly opposing the status quo?]
More significantly, such musings are not an engagement with the underlying arguments. Such psychologization can be useful as a guide to filter opinions and decide which merit consideration and evaluation, but it should not be mistaken for evaluation of the idea itself.
As far as the ideas at hand - that people shouldn't needlessly cut bits of babies and that male circumcision is generally needless, I think that there is significant reason to agree with both statements (or at least to consider them) regardless of the sincerity of some who argue for it.
I feel like this is a silly analogy that doesn't clarify anything. For starters, I'd much rather be raped by my spouse than have my genitals mutilated.
Secondly I wish people would stop throwing "Moloch" into everything. I don't know why, but it makes me cringe a little internally.
Broadly, I agree, but you're basically focusing on a symptom, rather than the root cause: Society doesn't actually believe in bodily autonomy as a basic right. The question to society is whether or not a given infringement on bodily autonomy is a net positive, or a net negative, in terms of the effect on other people.
So forcing people to get vaccinated is fine, if you think the vaccine is safe and effective, and wrong, if you think the vaccine is unsafe or ineffective. Abortion is fine if you don't think the fetus is a human being, and wrong if you do. It's fine to use drugs if you think the overall effective is positive, but wrong if you think it is negative - think of the people who think X drug should be legal, but Y drug is just too dangerous. Notice that X is always a drug that they, or people in their social group, use regularly. Puberty-delaying drugs are fine for children if you think the overall effect would be good, but it's bad for children if you think the overall effect would be bad.
I think your argument is coming from a position that things bodily autonomy is a fundamental right, as opposed to something more contingent on whether you think a rule causes more good than harm. (Observe that several commenters are explicitly calling this into question.)
We're -told- that bodily autonomy is a fundamental right, because it's a nice story to tell to explain why some particular violation is bad. But that's a much more black and white perspective than the way people actually approach the question.
It seems to me to be, broadly, -the- fundamental right. It seems a bit bizarre that the house our body spends most of its time in gets more protection than the body that we spend -all- our time in.
There are, indeed, some complicated gray areas around the parent-child relationship in particular. My particular rule of thumb is this: With regard to a child, if a decision concerning bodily autonomy can be put off until later in life - it should be. The right to bodily autonomy should be regarded as held in trust by the parent, rather than belonging to the parent. There's no reason to make the decision for them, so, well, don't.
If it can't be put off, a parent should have the same rights as, say, a spouse, to make those decisions on behalf of a person who can't make those decisions themselves. If a parent refuses to consent to a child's desired treatments***, the child should have the right to sue for the right to bodily autonomy - I don't expect the judicial system to be perfect, any more than I expect parents to be perfect, mind, so I don't expect this to eliminate problems, just to mitigate the worst of them. Any remaining gray areas are probably genuinely gray.
Otherwise? Except in cases of national defense (and I use the term national defense seriously, as opposed to the flippant way it is generally used - a war taking place in not-your-nation is not, in fact, a matter of national defense, but rather at best a matter of national interest), your body is your own. Your right to your body isn't a property right, either - there are many people who seem to think of their bodies as their property, which is a fine metaphor, up to a point. But your the right to your body is rather more fundamental than your right to your house - you can sell your house and buy a new one, after all.
Note that we live in an age with the potential for biological warfare - forced vaccinations is most definitely a valid case of national defense. And even a pandemic can reasonably qualify - I think it is fair to describe a biological agent of disease as an enemy of humanity. But this should be treated as an extreme measure, requiring an act of congress to formally declare war on the disease, and there has to be accountability when all is said and done - we need a definition of what qualifies as a war crime in the context of a war with a disease. In particular, as secrecy is not and should not be a part of a formal war effort against a disease that can't exactly spy on our activities, hiding information from the public should be regarded as a criminal breach of the public trust.
Note the "biological agent of disease", intended to limit the above. Regardless of whether or not we decide alcoholism is a disease, and regardless of whether or not we regard alcohol as causing that disease, government doesn't get to declare a war on alcohol. Nor poverty, nor marijuana - nor misinformation nor memes, however infectious they may be.
(Suppose there is a meme that mind-controls people? Shouldn't government have the power to fight that? And - no. Under those circumstances the government has no legitimacy - the US government has already fallen, because the bedrock principles on which it sits have collapsed. If the country ends up surviving, which I find doubtful under those circumstances, we can sort out whether or not the people in charge of the guns did the right thing after the fact, and figure out what kind of government we need in that new world.)
*** ETA: And the treatment is time-sensitive and can't wait; getting a piercing is not time-sensitive, getting puberty blockers is.
Are there fundamental rights that are never violated? To me bodily autonomy seems as fundamental as nearly any right. Even more fundamental than liberty, since we imprison people more than force surgical procedures on them.
More than half of all US states still have laws permitting forced sterilization. We routinely permit the removal of a part of men's genitals without their consent. Schools are permitted to force children to get haircuts. We require people to wear clothing in public settings, and regard it as a sexual crime if you don't. Courtrooms are prrmitted to require people to cover up tattoos. Men are still subject to be drafted, and once drafted, have basically zero bodily autonomy. Abortion. Drug testing. The drug war as a whole. The FDA prohibiting people from opting into experimental medicine. Vaccine mandates - public education requires a wide variety of vaccines, and there's not actually a right to not attend public education.
In what way do you think we treat bodily autonomy as a fundamental right? We treat -privacy- as a fundamental right, sort of, and what bodily autonomy rights we have had historically flowed from that, not a right to control your own body per se.
Yes, I know all that. But liberty is considered a fundamental right and all states imprison people. It's one of the basic forms of punishment people generally accept (as opposed to, say, being whipped).
Actually, I don't think bodily autonomy is a right at all, if by that you mean being able to do whatever you want to your body (like having any surgery at all or being able to take any drugs). What I think is a right is bodily integrity (not having your body violated). So you are right, I was using the wrong term, if that is how "autonomy" is defined. (Since the conversation was about circumcision, that is how I understood it).
(In this sense, even though I think abortion should be legal up to a certain point, I think the "women should have control over their bodies" slogan is a bit silly. People don't normally have control over their bodies. If someone's body decides to grow a pancreatic tumor, it doesn't ask permission. Also, having full bodily autonomy would mean complete freedom to do anything at all, since everything people do, they do with their body.)
I don't think having to cover up tattoos or wear clothes is a violation of bodily integrity at all.
I think the case that infant circumcision violates the child's bodily integrity is a strong one.
Mandatory vaccines can be seen as a minor violations of bodily integrity in order to protect other people's bodily integrity, or safety.
Legal abortion means the fetus doesn't have any rights at all (as far as I can tell). If it had any right, the first right it would have would be bodily integrity. (A fetus doesn't have any need for any other right).
Once you switch to "bodily integrity", as opposed to "bodily autonomy", then circumcision becomes acceptable - we see it as acceptable for an adult to consent to such a procedure. Therefore, with an infant, who has no right to -autonomy-, a parent consenting for them fulfills the requirement of consent.
I don't think your analogy is all that effective. If i understand the point it is that circumcision is equivalent to an assault on someone in your care, and not justifiable even if supposedly benevolent. But the analogy fails on two accounts.
1. Many people don't agree that circumcision is assault, if you change the story to your spouse changing you out of your clothes when you get passed out drunk then people wouldn't have a problem with it.
2. Many people don't viscerally think that the situation in your story is assault. Getting black out drunk and having sex with your spouse doesn't sound too bad.
If you change it to waking up with your partners name tattooed on your back then it would be s more effective comparison.
It is socially acceptable for a person to do certain things with their spouse without obtaining consent- like changing their clothes when they are incapacitated. It is not socially acceptable to have sex with your spouse without obtaining consent- but it used to be. It is not socially acceptable for a parent to have a child undergo surgery absent any medical need, and that's especially true if the surgery is not performed by a doctor, unless that surgery is male circumcision. If we started society over and you were in charge of choosing this social norm, would you include an exception for male circumcision? If you imagine the social norms of our descendants, hoping that they live in a safer and healthier world, do you hope they choose to have an exception for male circumcision?
I understood the argument. But if you are making consent your analogy you want something obviously wrong that forces people to engage the argument. Drunk sex with your spouse is already an edge case so it makes the analogy lose force.
If prioritizing the welfare of the child, would the responsible stance not be to avoid irreversible interventions unless necessary?
E.g., phimosis is a condition that can be cured by the removal of the foreskin, thereby relieving suffering. Of course, some people might argue that suffering is character-building.
Agreed on both counts. Babies really really really can't consent. Heck, even children can't (meaningfully) consent to a lot of things. That's why they have parents. And I'd really rather not go through that as an adult.
As I see it, it's one of those things that people have strong opinions about that don't actually matter much outside of the opinions themselves. No great objective harm or benefit is done by either doing it or not doing it. And parents do things with *much* greater risks (and potential benefits) throughout a child's life. Like...orders of magnitude larger. We don't require the child to consent to a parental divorce, for example, even though the median effects of that are horrific for children (barring cases of abuse, really).
Maybe he wants to question Scott’s conclusion that circumcision is not bad after all - a conclusion based in part on stats about its effect on sexually transmitted diseases in Africa, if I don’t misremember his post? That post didn’t seem to live up to the standards of rationality set by the rest of the blog in my opinion.
So, this is sort of an interest sounding out post more than anything, about a potential VC fund for impact certs.
First things first; I would be defining an impact certificate as something like the right to call yourself the institutional backer for an event. So, for example, a normal grant-granting platform would effectively own all of the impact certificates for a not-for-profit that they funded. They would not have technically "done the thing that the not for profit did," but they would have been the official backer.
So, how much is the price that one could save a life for. I'm not sure. My vague memory is that it's somewhere between $50-$5000, so taking the log average, let's call it $500. Let's just take this as an axiom (though I'll edit it if someone wants to give a more precise number, but it's sort of besides the point).
Question 1: Suppose that I could offer you a legitimate, documented impact receipt that showed that it had been bought as the impact receipt for someone who saved someone else's life. The price of this impact receipt was $100 (1/5 market value); the money was used to reimburse a person who'd bought urgent life-saving medicine for someone else. Would you buy this impact receipt as a charitable donation?
Question 2: Same scenario. $100 for an impact receipt, but this time for a volunteer firefighter who saved a person from a burning house?
Question 3: $100, but for a normal person who saved a child from drowning?
Ya know, I've always scoffed at those who object to paying people to "donate" kidneys. I am going to continue scoffing at them, but I did just have a momentary "that feels icky" reaction to some of your questions. What kind of world do we want to live in? One where people go around saving lives just because they expect to profit!?! Why, that would sully the morality of the whole endeavor! On the other hand, there are all those extra saved lives, I suppose.
Assuming that the price to me was below what it would cost to save a life (or equivalent quality of life changes) through eg. Givewell: I would almost certainly buy the first. The other two I'm less sure about.
-I think getting me to trust the certificates would be a feat.
-I might hesitate to buy because incentivizing the marginal life-saver wouldn't actually cost $100, it would also cost all the payments to those who would have saved the life regardless, but went after an impact payment because they might as well get some free money.
Well, we do pay people to save lives; firemen, coastguards, doctors, and so forth. That's not the problem.
The problem is the drowning child one, where now instead of "but I don't want to ruin my new suit", it's "but will I get an impact certificate for this? How much is it worth? Nah, that's not worth my while, the kid can drown!"
Children drowning in the presence of an adult who could save them is a negligibly small problem either way.
The real problem is things that are vaguely analogous to drowning children but without the elements of immediacy and proximity and with an additional element of social loafing.
At least the drowning child analogy makes me feel virtuous. I don't give money to charity but I do promise to save any small children I happen to see drowning, even if it happens on the ~one day a year I wear a suit, so I guess I'm maximally virtuous.
Writing prompt: a world in which superheroes exist, but so do impact certificates, and the economy is screwy because nobody can afford to keep compensating Superman for saving people.
For an extra twist: Superman volunteers to do the life-saving for nothing, but then gets sued because free services are screwing up the economy for people whose businesses rely on getting paid for life-saving superheroism! 😁
There's a reference to Galton's survey on visual imagination. I think this approach can be applied to many invisible disorders or disabilities (in a broad sense, anything that might impair one's capabilities in life).
I struggle with wishing. I have a very weak hypothesis that I might not experience wishes, desires, cravings or whatever as other people do. It results in my motivation to do anything not inherently interesting on my own being extremely low. So, my question is whether anyone have studied wishes. desires and whatever in abnormal cases. Are there any tests, polls, surveys which I can take on my own to determine if I experience wishes as other people do? Or maybe a detailed essay on what is a "normal" wish, how people experience it?
Thanks for any replies.
P.S. I've checked with a few psychiatrists and they haven't found any permanent depression. I certainly don't have anhedonia, I enjoy things but I don't really want them much and I certainly struggle when I have to work persistently to get them.
The literature on anhedonia makes a distinction between consummatory anhedonia, which is enjoying things, and anticipatory anhedonia, which is looking forward to the things you enjoy. This is something you might want to look into.
I'm a psychologist, but I'm not sure I quite get your question. You say " It results in my motivation to do anything not inherently interesting on my own being extremely low." Seems to me that most of us find our motivation to do things that are not inherently interesting low. I mean, I have to sand some chairs when I get home tonight, and it's boring to do, and in fact the main reason I'm still here in my office posting on ACX is that I'm dreading the chair task. I think what happens, as people move beyond the point in life when you get to do just what you feel like a lot of the time, is that various tasks that aren't fun in themselves get woven into a mesh where everything is connected with everything that you care about. So I'm improving those chairs because I'm fixing my house up a bit in preparation for a visit from and old friend, which I'm looking forward to, and so the looking-forward-to-improved decor and looking-forward-to-seeing friend sort of bleed together, so that doing the chairs is more inherently satisfying than it seems at first glance. In fact once I'm in the midst of doing it I'll probably be thinking happily, "damn, that looks good" and looking forward to seeing how they look after I put on polyurethane. And likewise with work -- it's *work,* but I'm invested in doing a good job, and get caught up in the process.
So I'm wondering if your problem isn't a failure to find sanding chairs and going to work inherently enjoyable, but a failure to weave a mesh were tiring tasks borrow some yumminess & some meaningfulness from other, inherently pleasant areas of the mesh.
Thanks for your input. But here's exactly the point. You used the wording "look forward to" which in this context isn't really different from "want", "wish" or "desire". And this is what I seem to be incapable of.
I know that things like tidying, doing sports, earning money and developing myself are going to be useful. But no amount of mental exercise has ever helped me to build up the motivation to do that out of some future prospect.
I mean, I see people wanting specific things and pursuing them quite often even if it's some triviality like a specific sort of sweets which they look for in a few shops, or something bigger like saving money for a dream holiday trip. Wishing is out there for real.
Well, but I wasn't really talking about getting oneself to do tedious, non-pleasurable things for the sake of the pleasure one can have in the future . I was saying that tasks that are in a way tedious are transformed by being meshed in one's life with acute pleasures. The chair sanding is already sort of enriched and blessed in my mind by the pleasure of how the chairs will look, and the pleasure of seeing my old friend.
I know someone who talks about being "tethered" -- by which he means, obligated, committed -- and about how he avoids tethers if he possibly can, but then feels better when he has some. I think tethers and and what I call mesh are sort of similar concepts. When I say mesh I mean the network of obligations, satisifying work, fun and comforting relationships and professional aspirations that are the stuff of most adult life. Anything you do, even sanding chairs, is part of the mesh: If it's work, it nonetheless has a sort of halo of fun and satisfaction around t because it's part of the mesh. Tethers don't just make you do thing they transform the things you do. It may be that you have avoided commitments to people and work and that accounts for your feeling of being adrift, and nothing being worth doing except things that give instant gratification.
Lots of people outraged over (Dilbert creator) Scott Adams for saying that, according to a poll he cites, around 50% of black americans don't like/have negative views towards white americans, and that this means white people shouldn't bother trying to help people who hate them and their (white people) lives would be better if they avoid living somewhere with a lot of black people.
Is he really wrong?
The standard response to this seems to be that white people have an obligation to help black people, and blacks have no obligation to like whites, because white people are responsible for black people being poor (poor obviously in a relative sense) in the first place. But this is based almost entirely on narratives, and most mechanisms this is supposed to be based on can be/have been disproven.
And then of course there's the inconsistent way in which group responsibility operates, wherein all white people everywhere are responsible for everything bad (or "bad") done by any white person ever, while black people are not even morally responsible for their own (bad) actions (and white people can take no credit or pride in the accomplishments of other white people). This tries to get rationalized by the claims of "white privilege" being a thing (and/or the even more silly claim that white people are "still benefitting from slavery"), which again can be debunked in nearly every way, but is an example of anti-white hatred itself (if you disagree, let me know if you would have been fine with it if Kanye had simply said that jews have "jewish privilege").
This reminds me of debate between Sam Harris and Ben Affleck over Islam, where Harris declares that Islam is a religion of violence on the basis of "poll results" showing that 20% of Muslims favor the death penalty over apostasy. Both Harris and the Dilbert person are overinterpreting what people will say on a poll. Anyone with any day to day familiarity of life in the typical Muslim-majority country in the poll will think, "only 20%? Not too bad," and would never make some idiotic generalization about Islam itself.
I was following until the first half of the last sentence. It sounds like you're saying that familiarity with Muslim majority countries would presume that Islam is much more violent than 20% in favor of death penalty for apostasy.
Well, familiarity would presume that more than 20% would declare on a poll that apostates should die, and also to put no stock into that because you know it's BS. Yeah, death to apostates, whatever, ha ha ha.
I see what you mean now. I'd like to point out, however, that if half of white people polled were to even lightheartedly agree that (for instance) race-mixing couples should be put to death we wouldn't ever hear about anything else on the news. Also if that were the case and I were married to a black woman there'd be zero chance convincing me to live in a white neighborhood.
Not wanting to put words into Sylwester's mouth, my alternate interpretation of their last sentence is that "only 20% of Muslims would glibly tell a pollster that apostates should be killed" means "the trolls and thoughtless poll takers are fewer than I thought."
The poll question is asking one of two things, which use the exact same words but mean very different things. I will use the following definitions:
white - a person with pale complexion
White - a person who uses entrenched status maintained through oppression in order to increase their own power at the expense of minorities (most often pale complexion people)
The poll question is either poorly written or intentionally conflating those two things. If you ask the first question it's pretty much meaningless and very hard to disagree with (saying "No" to "it's Ok to be white" is advocating genocide). If you ask the second question you're accepting a certain worldview about power dynamics and then answering about those power dynamics - which in reality is only tangentially related to race. That's how Jews and Asians can get stigmatized as "white" or "white-adjacent" despite that not being how any of this works if you're talking about Caucasian people.
I can be upset about people using the word "white" to mean something far beyond skin color or race, and also upset at someone trying to ask such a question when they almost certainly know about the two separate definitions at play here.
Write or wrong, current conventional wisdom is that it's not allowed to have negative feelings towards black people, and the thing Adams said would clearly imply that (you can't convincingly say you believe it's a problem that a group majority dislikes you and that's a problem, but you don't have a problem with that group as such). So he broke the social rule and got the inevitable pushback.
I agree that it’s not allowed to have negative feelings about black people, and constraints like that are madness. Everybody has at least some occasional negative feelings about everybody, including their mother and their newborn baby. And we are all vulnerable to negative feelings about groups, especially groups we do not know well. It’s a consequence of (1) Chunking (thinking in terms of categories, not a swarm of individuals — it’s lossy compression, and we are wired to do it to save mental space and time) (2) Being somewhat neophobic, as virtually all animals are — it’s adaptive. Of course it is possible to pry oneself loose from feeling confident that one’s negative take on a group is valid. But “do not think that because it is bad” is not a good tool for prying.
A bunch of things can be true at once, in part riffing on or even echoing various posters in this larger thread:
1. A vast plurality of humans have almost certainly, whether at times, situationally, selectively using their own idiosyncratic criteria, or ongoing, felt frustration with, anger at, or even outright prejudice towards, some particular group(s) of people *as a group*: distinguished by perceived ancestry/race, religion, class, political affiliation, or more.
We may be in-born with some physiological traits that reward tribalism, to foster in-group loyalty and make us at least suspicious of outgroups, if not hate them outright. (And yes, we also simultaneously have traits which help us work together cooperatively, even across such group divisions.)
It shouldn't be surprising to anyone that some American Black people harbor negative feelings, sometimes strongly so, towards American White people and/or people in various other racial outgroups. Even without the history of chattel slavery and Jim Crow, that comes with the territory of being human.
We humans naturally form teams to band together against "outsiders." And will even creatively invent differences between groups, when points of differentiation are not immediately at hand.
2. There's some social pressure in the US - on social media, in news media, etc. - to not say bad things about any historically disadvantaged minority, which includes Black people.
3. Adams made a thoughtless (more below) comment that violated that expectation. And in reaction, he lost some face among some observers, the distribution of his comic strip was reduced, etc.
4. There's also a small subset of folks who make a regular practice of violating that expectation. For instance, Breitbart News' former "Black Crime" beat, certain social media accounts which frequently post videos of bad things done by Black people or unauthorized migrants in the US, etc.
5. That single, ambiguous question in the Rasmussen poll was abysmally bad, as I noted in an earlier comment. That polling organization themselves confirmed in a tweet that it was an act of trolling, rather than a good faith attempt to assess attitudes toward other racial groups.
Carl in his History Boomer post did a good job, methinks, of noting all of the possible ways that poll-takers might have (mis)interpreted it.
6. For Adams to not to understand that – and to even misinterpret its numbers, as well – and then make a highly public, emotional video based on his immediate reaction, doesn't speak well for his discernment or judgement, at least in this situation.
If the widely shared clip of a portion of his remarks was misrepresenting his larger position, would welcome specific interpretations of that. At least from that clip, that poll seemed to have been a key, specific event that impelled him to make this set of video remarks, even if he'd been thinking about this topic for a while.
(And having written this, just came across Leo Abstract's post with some suggested insights about Adams. If those are even mostly accurate, that *might* further complicate the narrative.)
7. Race itself is a notoriously slippery concept.
For starters, it's clearly the case that at times and places we're treated differently, sometimes harshly or unfairly, based mostly on our outward physical appearance. (Along with markers of ancestry/race, these also include perceived beauty, height, weight, and visible attributes related to age or illness/disability.) That difference in treatment happens in many, many ways socially. And at least with perceived race, also in various ways in law and other formal practice.
I can't remember where I spotted this, and haven't yet been able to find it again, but IIRC one scientist wrote that there are apparently at least hundreds of groupings of humans whose members largely share some slight genetic differentiation, distinguishing them from members of other such groups. And noting that, if this is even remotely accurate, genetically, it either challenges or at least gives far more nuance and complexity to widely-held conceptions of the existence of some 5-20 broad "races."
"The poll didn’t say that half of black people disagreed with “It’s OK to be white.” Only 26% of black people disagreed, 21% weren’t sure. And before you worry about that “half” claim, you should also note that 20% of white people also either disagreed or weren’t sure! Ditto for 41% of Hispanic and Asian people (lumped together under “other”). If you seriously think 50% of black people don’t want white people to exist, it means we’re facing a race war where 41% of Hispanic and Asian people and 20% of white people may be on the wrong side!"
"... Rasmussen chose those exact words [for this question] because they were deliberately responding to a claim that those specific words were hate speech. They were trolling the Anti-Defamation League."
As this author later noted, "Look at the [poll question's] words again. “Do you agree or disagree with the statement ‘It’s OK to be white’?” Those are weird words. OK to be white? That is not the kind of neutral language that pollsters normally use. The unusual wording is apt to set off alarm bells in someone’s head. “What funny business is this pollster trying to pull?” People [responding to that question] will be understandably looking for a trap."
>That’s not the same as hating white people and thinking they shouldn’t exist.
Uh ok, so it's totally fine to disagree with the statement "Is it OK to be X ?" for any value of X then ? it doesn't have any negative valence whatsoever ?
> “But if the poll had people disagreeing with ‘it’s OK to be black,’ everyone would be freaking out.” Maybe so, but there’s a lot of history behind why we get more upset when black people are attacked.
Uh huh, so you can't do it to blacks, that would be racist, but blacks can do it to whites, and it would be totally not racist. Because something something Racism is "PoWeR pLUS PreJuDiCe".
As others pointed out, you could likely get similar results asking conservative whites about "Black Lives Matter". Pretending this is about the literal meaning of words is disingenuous, even before you get to the problems you point out where the numbers are wrong anyway.
I don't actually believe the interpretations of the poll. The question asked was something like "do you agree with the statement, it's OK to be white", which is a thing said by a bunch of racist 4chan troll types. A lot of people who would normally be OK with the statement, will have negative reactions based on that.
It's like asking "do you agree with Black Lives Matter" and then concluding that the people saying 'no' literally think there is 0 moral value attached to the life of a black person.
Also - IMO the worse thing than saying to "stay away from black people" (which is bad enough on its own) is where he called black Americans a "hate group". Even if it's true that 25% of black people think it's "not OK to be white", that's still a blanket negative statement that it's bad to make.
You know Nassim Taleb's shtick with Fat Tony and the wisdom of the taxi driver? Sorry, but in a similar vein, the Dilbert guy is an idiot, with all his BS about Trump being a master hypnotist living in the 3d world or whatever. Ignoring him completely improved my life and I recommend it.
“People not liking you” doesn’t get you out of obligations to them. I would be really surprised if less than 50% of people in the bottom 10% income decile didn’t like people in the top 1% of incomes, but that doesn’t exempt the very rich from any of their obligations to the poor.
I would say that a rich kid who never mingles with poor kids will likely be emotionally stunted, and their parents have committed an injustice against them. Also, it’s fine if rich people prefer e.g. beaches that not many poor people use, but it would be unjust if there were public beaches that poor people weren’t allowed to use. I don’t think schools should be allowed to refuse black people, but private schools should be allowed to teach values that (parents expect) black communities will be less interested in (I assume Douglas’ putative white people have designated some values as ones black people don’t share). The school can charge enough money that black people won’t come unless they really want those values for their kids.
Ok, let me try this again now that I'm on an actual keyboard. I looked into it at the time and then also after he followed up. A few things:
1) Adams appears to be truly and classically anti-racist, and those familiar with his larger body of work testify that he has always blamed gaps in attainment between black and white on structural racism. (He also appears to blame teachers' unions for perpetuating this in a school-choice-partisan kind of way, but I'm not really sure about this part. Only thing I can tell for sure is that he directly and loudly blames white people for the disproportionate poverty, crime, and lack of education.) He appears to not believe in any meaningful differences between races in terms of genetics, and castigates white people who take any credit for white accomplishments.
2) A big part of his overall messaging appears to be that the media monetizes hate and turns everyone against one another. His impression of the state of the country (from a variety of polling sources as well as the tone on social media) is that clickbait has worsened race relations to the point where some percentage of blacks actively dislike whites.
3) He's a lifelong proponent of moving if you don't like where you live. He moved from upstate New York after a brush with death in a snow storm and ended up in the lowest-crime part of California on purpose. I'm not clear on why the issue of people not wanting to live in predominantly black neighborhoods (due to crime) came into this discussion, but I think it was to mock people on twitter for their faux outrage about not wanting to live near a group of people that contain some (10%-50%) members who dislike him for being white. (Mock because they all live in white areas too.)
4) Finally, and arguably most importantly, he appears to have been pivoting for some time to 'self help' rather than cartooning, and appears to have wanted to get as much attention as possible in order to pitch the country on solving racial disparities by injecting pro-success messaging and tools into black culture. He appeared genuinely surprised at having had enough reach that some remarks on his podcast stirred up so much furor, but then also appears to have realized that this is what he should have predicted based on how the media monetizes division and had all the reason in the world to clip his remarks to make them look bad (the way Aaron Rupar did with the Charlottesville-era video of Trump saying there were 'good people on both sides' of the statue question, clipping off the next sentence which was "I'm not talking about the neo-nazi white supremacists, I condemn them totally".
So to answer your original question: Yeah he's wrong, but not for any of the things reported (which he didn't say, or at least didn't mean). He's wrong because he is far too optimistic about how to improve any relations in this country.
"Around 50% of black americans don't like/have negative views towards white americans." There's a pretty big difference between not liking a certain kind of person and having negative views towards them. I have SOME negative opinions about almost everybody I know. Maybe everybody, in fact. I'm fond of Zvi, for instance, and think he's brilliant. But he also sounds like somebody who has not read a a single book written before 1950 or so, and I think that's bad. I don't know what white people should do if most or all blacks hated most or all whites, but fortunately I do not have to figure that out, because it's obviously not the case. I say that based on relationships lasting several years with each of a handful of black people, reading books & articles by blacks, and listening to black public figures in the media, plus various casual contacts with blacks in daily life.
Seems to me that what was wrong with Adam's idea was that he misunderstood the poll results or, what's more likely, distorted them in a way that make black people sound ugly, hateful and scary. Then he discussed what would be a reasonable way for whites to live and to think about blacks, if in fact all blacks hated whites. And in the course of the discussion he gets to remark over and over again on black hatred of whites. I mean, it's sort of like asking what we should do if we learned that every single person in Texas, but nowhere else, harbored an obscure, lethal and incurable form of Argentinian rabies. I don't KNOW what we should do, but that is not the case so why are we talking about this? "Would we be justified in nuking the whole state?" he asks? Yeah, fine, fuck it, nuke Texas. But get Kenny and his partner out first.
According to the wikipedia page for Adam Dilbert, the poll consisted of presenting the phrase "It's Okay To Be White" and asking if they agree, only 53% of black respondents, said they agree, 26% percent said not sure, and the rest said they disagree.
Now, I know the origin of this slogan, apparently 4chan made it to bait morons into being racists, typical 4chan behaviour, and it worked. 47% of black respondents said they're not sure if being a trait you're born into is a good thing with varying levels of disagreements, and according to the Wikipedia page about the slogan, universities call police when they see posters of the phrase and de-platform speakers who want to speak about this statements.
What's curious is that the quoted individuals never justify why they think the statement has no place : Is it painfully obvious for instance ? (but aren't Black Lives Matter too ?), is it meant in bad faith to mean something other than its surface meaning ? (but aren't BLM too ?) Is it false and they think that merely contradicting or discussing it is dangerous ? Anybody with some fraction of a brain has questions, what about being white is so wrong, enlighten us ye wise ones ! But they all have a single answer, and it's not in words.
I would say that it's never okay to express disagreement with a statement like this without qualifications, even if you know the statement is bait made specifically to make you look bad, resist a little, don't fall into it right away, at least give something to your defenders to squeeze out some benefit of doubt. But then again the US census would classify me as White if I lived there, so maybe this is just my White Fragility showing eh ? it's pretty obvious to me that being a fucking racist when your entire ideology is based around bitching about racism is kinda not intellectually honest, but who knows, those things are hard. Better not think about it.
Perhaps the only redeeming path for that poll is if it's repeated with white respondents, I suspect the relevant factor is not actually race at all but political affiliation, the near 50-50 split is a weak hint it is. But still, it's hilarious when someone goes "hmm, I think whites are discriminated against" and the immediate response is hysterical screaming and cancellations of the guy who Noticed Something, he's totally wrong right ? it's totally obvious he's just a fragile white snowflake who can't handle equality, which *of course* involves asserting that being white is not OK.
Failing that, perhaps the polling organization did black magic stat shit to paint a different picture over reality, and a balanced sample won't display the fucked up ratio of 1/2 of it thinking it's not okay to have a certain skin color. That's about the only 2 things I can think of for this to turn out Adam was wrong, and boy do I wish he was wrong.
Oh, thanks for clarifying about the 4chan bit. I'd forgotten that part. Yeah, it seems unlikely that many of the respondents on the poll would have known about it, but maybe I shouldn't judge its notoriety by my not knowing about it.
My take on what an ordinary person would think of a question about whether it's ok to be white is fairly advanced, given that I understand and sympathize with the viewpoint that whiteness is problematic because it was created in response to nonwhiteness. I.e. my own ancestors now would be classified all as 'white' but never thought of themselves that way -- they were various nationalities and ethnicities from within Europe, and saw themselves as such. If they had any negative identificatoins they probably mostly defined themselves in opposition to the great powers of the day (from which I am not descended) -- i.e. not English or French. If you were to ask me on a poll if it's "ok to be white" I'd hem and haw a bit and say "well it's ok to be anything you can't change so yeah". If there were some racial identity that I had been taught had been made up to specifically exclude people like me and I were asked on a poll if it were ok to be that, I'd say "hell no".
I think "made by 4chan trolls" has to be taken with 2 or 3 medium sized spoonfuls of salt, since mainstream media (yes, even wikipedia) is notorious for using 4chan as a non-christian devil, and attributing various wild shit to it. If GamerGate has taught me anything, it's that Wikipedia in particular is basically twitter or tumblr on Woke incidents.
I think it's certainly okay to disagree with "It's okay to be white", by various arguments. The first one is of course Free Speech, I think people are even allowed to disagree with "Should babies be NOT killed ?", as long as they are nowhere near the child birth section of any hospital. The second level is that the idea is bait, it's almost certainly meant in bad faith (though not without a lot of legit reasons, as this incident demonstrates), and it's useful to stir the water by saying "I disagree" without elaboration, **then** giving a bunch of reasons why after you have gotten people's attention.
I disagree that "Whiteness" is made up in any interesting sense of the word, race is obvious. An ancient arabic pre-islamic religious song goes something like
>We are the crows of akk [An Arabic tribe], we came to you [our God] begging
A "crow" is how arabs called a black slave, and the context is that slaves usually led the convoys to Mecca (which was were people worshipped even before Islam), so the slaves sing the song and then the rest of the convoy sings along, so those things are really really heckin old, this is the first example that came to mind. The African slave trade is ancient, it's likely that every single ancestor of yours from Italy and Spain up to at least Germany and back in time to the crusades saw or heard of or otherwise vaguely sensed the existence of black skinned humans, and knew in a distant sense that they were, unlike them, white. Sure, I have no doubt that blacks and whites mixing in the Americas almost certainly turbo-charged the various sentiments and ideas, but that's just a trivial tautology, of course people notice the differences more when they are closer, what is useful things is this saying ?
The thing that really gets me about the incident is how devoid of self-awareness the corporate cancellers and the hysterical screamers against Adam is. It's almost a Wojack meme, if somebody said to me "you know what Bi_Gates ? I think you hate me, and I think it would be okay if I got away from you", my response would either be "Yes, I hate you, nice that you noticed", or some non-cruel equivalent, or if I was keen on them I would say "No, I don't hate you, how did you get the impression ?", what I definitely wouldn't do is [Wojack crying face] "NOOOOOO YOU PIECE OF SHIT I DEFINITELY DON'T HATE I SWEAR I JUST THINK YOUR SKIN TONE IS KINDA GROSS AND THAT YOU SHOULDNT EXIST BUT THIS IS NOT RACIST FOR SOME REASON YOU KNOW WHAT NO MORE MONEY FOR YOU NOBODY IS GOING TO BUY YOUR SHIT HUH HOW ABOUT THAT FUCKER I STILL DONT HATE YOU".
Like, I don't know if they are moronic enough that they actually persuaded themselves that hating someone based on skin color **is not** racism and now they're confused because the rest of us are confused, or if they know full well what racism is and that they are doing it but are too much of pussies to admit it in public, or if they just have no consistent position and their plan consists of farting the biggest nastiest amount of mindless animal rage in the hopes that everybody would cower and not ask questions ? I genuinely don't know, I would hate them either way, they are disgusting, but I want to understand what their confused little brains are thinking before I can hate them with a clear conscience.
> I think "made by 4chan trolls" has to be taken with 2 or 3 medium sized spoonfuls of salt, since mainstream media (yes, even wikipedia) is notorious for using 4chan as a non-christian devil, and attributing various wild shit to it. If GamerGate has taught me anything, it's that Wikipedia in particular is basically twitter or tumblr on Woke incidents.
4chan are probably overrated as Hackers on Steroids. But they're not overrated as purveyors of crazy pranks that play traditional and social media like a fiddle.
Yes, I'm sure my ancestors were aware that some people are darker-skinned than others, but I'd wager most of them would have thought of those people as heathen, paynim, mohammedans, saracens, moors, or the like. Monolithic "white" vs monolithic "black" might not be purely an American invention but there's more ideology in your own understanding of these categories than you might suppose.
As for why they went after Adams, his own explanation is that they are either misled, motivated by monetary incentives, are just virtue signaling, or are what he calls 'dopamine gooners' -- people metaphorically cranking off with delight at the prospect of canceling someone and making him suffer.
> I don't know what white people should do if most or all blacks hated most or all whites, but fortunately I do not have to figure that out, because it's obviously not the case. I say that based on relationships lasting several years with each of a handful of black people, reading books & articles by blacks, and listening to black public figures in the media, plus various casual contacts with blacks in daily life.
This really seems like a textbook case of overconfidence on the basis of anecdotes. It's good to have priors and to avoid letting new data disproportionately push them around - that's one of the two major takeaways of Bayes' theorem, after all - but the strength of your prior should itself be determined by the quality of the evidence used to form it. It's the "what do you think you know and why do you think you know it?" concept. If your evidence is based on a handful of friends, some casual interactions, and your choice of media consumption, it's 1) very shallow, and 2) extremely vulnerable to overfitting and projecting the character of your personal bubble upon the fabric of a vast and varied country.
With all that said, I also find the premise rather unlikely. You just haven't provided any information that should lead any rational person to say that it's "obviously not the case."
I certainly would not feel justified in concluding that most black people do not hate most white people based on my limited experience. I don't think it's altogether silly, though. Hate's a strong word and if half of black people HATED white people, you'd think I'd have picked up on that by now. It's sure not hard on Twitter to pick up that all anti-vaxers hate various things I have to say, and in fact make statements that sound like they hate not just my actions but me: "You should be in jail," for example. Hatred's not real subtle. But let's withdraw my life experience entirely from the evidence for or against.
The thing that seems indisputably funky is what Scott Adams (or maybe just Scott as described by Goldman Sachs Occultist) did with poll results where 50% of black people endorsed one of these 2 statements (it's not clear which): (1) don't like white people, (2) I have negative views towards white americans. OR (3) 50% of blacks did NOT endorse the statement "It's OK to be white." And then, acc/to Goldman Sachs Occultist, Adam went on to talk about whether white people an obligation to be okay with being hated by black people, and should feel comfortable living near a bunch of blacks. Talking about being HATED by blacks goes beyond all 3 versions of that poll question, and in any case, bear in mind that only 50% of blacks answer the question in the negative direction. So the question is, is it fair to expect whites to live near or feel sympathetic towards a group 50% of whom have some negative feeling about whites?
Honestly, living in a place where 50% of people disapprove of me sounds kind of like no big deal. On the block where I live, there's one guy who loathes me (the feeling is mutual) because he put the moves on my daughter when he was 40 & she was 16 and I really went after him. There's an elderly Indian woman who got quite angry at me one time when I tried to give away some old toys by piling them in the grass between her sidewalk and the road. She probably still remembers it. And there's a guy who once whistled the wicked witch theme from The Wizard of Oz when I biked by with my long gray hair flying.
Or let's take this forum. Leo and Biblio, I don't get the feeling you hate me, but you certainly have a negative view of things I said in an earlier post. And, Biblio, you had quite a lot to say about how much my reasoning sucked. And once in a while on here somebody really seems to loathe the way my mind's working, and says so. On the other hand, a fair number of comments I get are positive. So is it reasonable to expect me to keep posting on this forum, and to try to help it work better when I see a way? I don't see why not. Doesn't seem unbearably difficult. NO I am not saying that being on this forum is equivalent to living in a neighborhood with a lot of black people, half of whom harbor some negative feelings toward white people. But I am saying that being in a setting where half of the people have some negative views about me feels kind of like par for the course to me. Being in a setting where half the people absolutely hate me does not. Being in a setting where everybody hates me sounds extremely unpleasant.
The funkiness of what Scott (or perhaps just GS Occultist) did was to go from a poll result showing that half of blacks had some negative feelings about whites to asking whether white people have an obligation to be okay with being hated by black people and to actively care about improving their wellbeing.
Exactly. Not just that, but Adam's whole argument is that it's totally fine to avoid groups that are statistically likely to dislike you - to some extent as defined by your personal appetite for risk. To the various white liberals who were condemning him on Twitter, he asked one simple question, which none of them could answer. What is the racial makeup of your neighborhood?
I'm no apologist but I can tell that you didn't listen to any of his actual content beyond the tiny clip that was deprived of context to sound bad. The poll as reported had only half of blacks saying it is ok to be white, and Adams was careful to explain that he was using the poll as a jumping off point in discussing an obvious cultural trend in black dislike of whites.
I didn't listen to any of it. I just read what was here and responded to that. A lot of my response hinged on what the negative statement about whites was that50% of blacks endorsed: Was it "I don't like whites" or "I have negative views towards whites." As I wrote, the second seems much weaker. But now you're saing that the statement that 50% of blacks endorsed was "it's not ok to be white' -- wait, actually, it's not quite that, it's that 50%of blacks did NOT endorse "it's OK to be white." And actually, I don't know what to make of that -- I don't know what it means, really. Seems silly to put that statement in a poll. Does it mean, "people should not be *allowed* to be white"?
" white people should not be allowed to exist"? "there are some things about white attitudes and actions that I think are just no OK, they're wrong". So I stand by what I said: This new version of the poll result, too, does not justify asking whether white should have to live near a bunch of people who hate them. I don't know whether people should have to live near people who hate them, but in any case this new version of the poll also does not seem to translate into 50% of blacks hate whites.
I had to scroll up and verify that my comment didn't get cut off and still contained the part about how the poll wasn't even the issue at all. It's just jumping off point for something that people have already noticed. Like how if we saw a poll that shows that 50% of women think that masculinity is problematic we wouldn't really need to dig too much into the methodology or sample size to be able to discuss it.
But I was not opining as though I'm an expert, Leo. I was opining as though *you* were an expert. I assumed your account was accurate and complete, and that you would not be so silly as to skip giving crucial context that was necessary to grasp relevant nuances. I also assumed you were clear about what the question was that 50% of blacks answered in a negative-towards-whites direction. However, you have now given 3 different versions of it: 2 in your original post, and a 3rd, importantly different one in your short response sneering about how I had not heard the entire clip. Um, maybe it would be good if YOU listened to the entireclip and came back and told us what the fuck the item was, exactly, that 50% of blacks answered in the negative-towards-whites direction.
Does anyone live or have practical experience with a country that uses open list proportional representation? (So Sweden, the Netherlands, Austria, Finland, I think Brazil, etc.) The main thing I want to know is- do the politicians *within a party* actually compete with each other?
The idea as I understand it is that voters are just voting for a party, and then have the option of ranking individual politicians within that party if they so choose. The party receives X number of seats based on how many votes they got- the party candidates then either make the cutoff for one of those seats based a) how the party originally ranked them, and b) how voters then ranked them from there (if they chose to).
Do the politicians within a party..... compete with each other though? I.e. if the Swedish Green Party is running multiple candidates, and they all presumably would like to be ranked higher by voters- in practice do they actually compete, bash each other, play up their accomplishments, etc.? "Sure Sven Stroosma is a fellow Green Party candidate.... but whatabout allegations that he had improperly unrecycled herring cans in the back of his Volvo on the way to Ikea, ya?" (This was my attempt at a Swedish Green Party scandal).
In theory open list PR is the perfect voting system in that most voters are really just voting for parties anyways, and then if they have opinions on individual candidates they're free to express those- or not. So (in theory) it's the right blend of party-centered politics, while still holding candidates individually accountable, yet without having a high degree of personalism. But I have no idea if the system really works in practice or not, would be curious to hear people's feedback
In Denmark, there's a fair amount of jockeying within-party in local elections. Obviously, candidates avoid creating drama if they want to get on the list in the next election. But they will appeal for personal votes to get the marginal seats, and it definitely works. Competition between candidates of the same party is more pronounced than it was a few decades ago.
The dynamics are different in national elections where the vote counts and budget requirements are bigger. But even here, it feels like social media is creating new opportunities and more competition than we used to have.
I'm from Norway. I've never heard politicians compete within the party the way you say.
Most people only care what party they vote for and don't bother crossing names off the lists. If my memory is correct, one exception was once a group of voters of Pakistani descent organized to get a person of Pakistani descent into parliament ahead of people above him on the list.
Not in Sweden. Voting for a person is optional and there's a minimum limit that must be exceeded for them to count, which makes it very hard for a politician to get enough personal votes for it to make a difference. (Usually such a politician is popular in the party and high on their list anyway.) In the 2022 general election, only 2 out of 349 members of parliament were elected based on personal votes. These 2 seems to be regular charismatic politician with strong local support in smaller regions (which makes the minimum limit easier to reach). But there's also a growing trend of politicians from immigrant communities who make strong personal campaigns in those communities (with what many suspect is more focus on in-group politics than party politics), which has a worse look.
The (big) political parties in Sweden are very centralized and the center is very powerful (see e.g. the book "Knapptryckarkompaniet" (translated: "The Button Pushers") for what it's like being a member of parliament). Upstarts with lots of personal votes are likely to suffer from tall-poppy syndrome unless they play their cards well.
Internal competition in Swedish parties are usually much more about internal politics. Its rare to see politicians be critical of members of their own party or the party line, and those who express such criticism are often dealt with harshly (especially in the bigger and more centralized parties). In general, Swedes like their consensus culture were everyone always agrees, and where a change in policy can be coordinated behind close doors so that everyone can change their mind at the same time without any public arguments or debates. (See e.g. the NATO question: Every politician in the Social Democrats was against NATO membership a year ago. Two months later everyone was pro-NATO. Few outward signs of conflict were seen. If a Social Democrat would have tried to gain personal votes by taking a strong personal position on NATO, I'm confident that they would have been seen as a defector and punished accordingly by the party.)
There's a hypothetical effect were people can indicate preferences by voting for candidates to "push" the party in a certain direction. If a Green Party politician who is centrist on economics gets many personal votes than his more left-leaning colleagues, this information might push the party centerwards even if no there's no practical effect on the election outcome. But the parties should have better information from polling etc. that trumps this.
From experience in Israel yeah, it's often a major source of drama. It usually doesn't come up much on the news (the kind of politician that barely makes it high enough in the list to get a seat won't be a major figure), but it's a huge source of intraparty conflict and you will occasionally get stories about how "the party ranked this guy I like way too low, I like them less now".
Think of it like party leadership conflicts (which also happen in e.g. the US or UK), except the minor league of that. You'll have politicians fighting each other while trying not to do it too hard to avoid damaging the party.
Thanks. It's funny how your experience in Israel is different from all of the other responses I received from the Nordic countries. National culture matters a great deal!
Yep. In Czechia, we are similar to Israel. In last elections (2021), two parties formed a coalition with one ticket (it is often done in order to get over 5 % threshold, though in this case reasons were more complex), and in an election, voters of one of those parties ranked their politicians so consistently that their unlucky partners tanked in the elections despite on paper solid results. Similar stuff happens all the time, and victim-politicians tend to be angry about that and complain about backstabbing.
I have lived in both Finland and Sweden and I have never noticed this, not openly. It would probably be counter productive anyway and would probably not be popular in the party (which is more important for one’s future political career than what few votes you could get like that).
Candidates will advertise themselves and try to bring out some attractive features. But open negative campaigning against people on the same list in the same district isn’t seen. Keep in mind you might need their votes to get in.
The systems are different in the two countries. In Finland you have to vote for a candidate and the personal ads are very prominent. In Sweden selecting a candidate is optional.
Having read the other replies, I will clarify more. In Finland a party's candidate lists are essentially unranked ("open"), before the election, that is. So candidates do campaign and advertise for themselves and therefore they do compete with each other, but they don't openly campaign against each other (like in the example about the Green Party scandal).
So a candidate in the blue party (fictional) might advertise that he cares immensely about lowering taxes, implying that he cares more about this than other candidates in the blue party. But launching a public negative campaign against another candidate in the same party isn't something one sees. It might exceptionally happen, but in general it would be counterproductive.
A good go-to book on US foreign policy, IMO, is Walter Russell Mead's _Special Providence_. Especially so in light of handwringing over hypocrisy. It's largely not hypocrisy, but rather multiple voices within that policy, often in conflict, often allied. Mead outlines them as four-fold: Hamiltonians prioritizing favorable US trade; Wilsonians promoting the spread of US-style democracy around the globe; Jeffersonians advising to defend and maintain our democracy at home only, and leaving the rest of the world alone; and Jacksonians going along with the live-and-let-live rule, but when some foreign power breaks it, very enthusiastically sending our own forces over there to Break Stuff.
While each school of thought is named after the historical figure Mead sees as most exemplary of that school and upheld by its adherents, even those figures broke with their schools, often in a manner perceptible as them giving in to advisers on their own cabinets, especially persuasive voices in Congress, or plain old natural events. When US policy flips or exhibits conflicting standards, it isn't hypocrisy, so much as one of the schools losing ground to the arguments of another, depending on the situation.
Mead contends that US policy is as successful as it is because this four-way tug of war enables a high degree of adaptability. America can lean on her military one day, her trade delegations the next, her peace or evangelical corps the days after that. And every so often it's surprisingly decisive and swift because as many as three schools align.
I can't keep up with everything, but sometimes I support US foreign policy, and sometimes I don't, and it always affects my vote to a modest extent, but is almost never determinative, because domestic concerns (what a candidate will/won't do within the US) matters way more to me.
For example, right now I support the US effort to help the Ukrainians beat the snot out of Putin's Russia. It's hard to say whether that effort is being executed competently: my expectations for the Biden Administration, particularly after the Afghanistan fuckup, are very low, but maybe they learned something, or maybe there's some mid-level person who is doing a good job, or maybe they could do a lot better. Very difficult to know, unfortunately. They have not obviously screwed the pooch, yet, but perhaps things could be a lot better. Nevertheless, I support the aims, its just the execution about which I wonder.
On the other hand, I think US foreign policy towards China is marked with impressive levels of clumsiness and ignorance. I'm darned if I can figure out what the overall plan is, if there even is one. I mean, American foreign policy in Asia has practically always has been clumy and ignorant, except maybe when MacArthur was running Japan, and maybe again early in the Nixon Administration, since Nixon was one of the few American Presidents to really care about Asia, but it seems to matter more now, since Xi appears more volatile than the typical Chinese leader, and for that matter China is undergoing many more internal stresses that could lead to chaotic behavior on the world stage.
Ethics and the future of humanity have squat to do with how I assess American foreign policy, however. The United States is not by itself responsible for the future of humanity, still less for righting the world's wrongs. The only goal I admit lies within the ambit, if not always (alas) the competence of the US President, is to advance and guard the interests of Americans.
As long as the US are as powerful as they are and even want to be the worlds hegemon, US foreign policy has a lot to do with the fate of very many people all around the world. Power always brings responsibility. The US don't have to take care of the worlds wrongs, but they should take care not to create harms. I think ethical people should always oppose the powerful dealing out harm out of selfishness or carelessness. Not caring what harm your democratic government causes abroad makes any American loose a lot of credibility when talking about ethics.
That's absurd. If I issue an opinion on ethics, and someone says "Ha! Your argument is wrong because you're an American, and in 1914 President Wilson ordered the Marines to occupy Veracruz, which was deeply unethical..." then that person is an idiot, because first of all that's an ad hominem argument, and second of all I'm not President Wilson. I couldn't care less about my credibility in the eyes of a mental pygmy.
I was talking about current actions that are done also in your name. If you don't see the ethical problems of these, this tells me something about your abilities and ethics. So seeing how you judge different situations gives me a picture of your ethic models and on this picture I decide how much I trust you on other topics I don't have a clear opinion yet.
To make it more practical:
If you talk with a Russian about some conflict in Africa, how the actors see the situation and what would be a fair solution, I'm pretty sure you would give his opinion much more weight if he condemns the Russian invasion in Ukraine then when he calls it unfortunate but necessary.
Would you trust a Chinese on ethical topics who just repeats the official narrative when asked about the Uyghurs?
If these topics don't make you biased against ethical credibility of the citicens of these countries, than you are a postive exception. But usually humans build up bias and you can't avoid to adding to these biases with your behavior.
Well do bear in mind that I couldn't care less whether you trust me. You're not my boss, you're not my kid, and I'm not trying to sell you anything. Of what worth is your trust to me? Zip.
Your suppositions are both wrong, and from my point of view demonstrate a lack of common sense. If I talk with a random Russian about some conflict in Africa, I will judge the accuracy and wisdom of what he says according to what he says, and what I know independently. What he thinks of the war in Ukraine is of no more relevance than his opinion on abortion, the infield fly rule, or whether Oswald acted alone. The same is true about a Chinese man opining about the ethics of AI development. What he thinks of the Chinese government treatment of Uighurs has no more bearing on the question than whether he thinks the Treaty of Versailles was unfair or capital punishment should be banned.
So far as I can see, you're suggesting that I base my evaluation of the accuracy and utility of someone's opinion based on how much emotional simpatico I feel for them, which makes about as much sense as basing it on how sexually attractive they are -- e.g. I should trust Julia Roberts more than Al Gore on whether global warming is a problem, because Julia is a good-looking broad. To me this is repugnant to the very concept of being rational, which at Step 1 embodies the determination to make judgments via reason, and not one's monkey instincts.
I am sure many people do behave this way -- perhaps even I do, from time to time, being subject to the usual flaws in human nature. But why anyone should embrace or encourage this regrettable tendency by paying court to it is beyond me.
Is there a country on earth with a foreign policy devoid of hypocrisy?
Assuming there isn’t such a country (and there definitely isn’t one), “hypocrisy” is kind of a pointless complaint.
So I find it much more useful to evaluate foreign policy in terms of whether a particular *policy decision* is good or bad rather than a broad “do I think country x’s foreign policy is good or bad.” Iraq war bad, supporting Ukraine good, and so on.
There are systemic reasons for US foreign policy being what it is, and there are reasons for it being represented the way it is (hypocritically). I don’t think analyzing those is a waste of time. At the very least it would prevent people from arguing in good faith from the point of view that we (well not me, I am not American) are the good guys. But should it be a priority for the blog? Why?
US foreign policy is a very polarizing issue, like many other issues of policy. One of the reasons I like it here is that this space is avoiding the reddit/twitter/fb death spiral, which got to the point where no meaningful discussion can be had at all without devolving in partisan fights. Changing policy, given the entrenched interests supporting the current status quo, seems an overwhelming task and one that other groups may be more suited pursuing anyway. Some people adjacent to the blog did start a think tank though; maybe they will be addressing foreign policy too? On a different note I want to believe that most people here see through the obvious, blatant hypocrisy to the point that it doesn’t make sense to point it out anymore, like you wouldn’t point out that some celeb’s mammaries are implants.
OK, so you see some kind of silent consent to avoid this topic because it's too polarizing?
I can understand this well on a personal level. But the relevance it has on so many lifes and possibly even all of humanity if we get into full WWIII, isn't it like a act of cowardice to ignore this if you can have influence?
I did discover Scott exactly because he did take the then highly emotional and polarized topic of politically correctness and did discuss it rationally without siding with one of the parties but looked on the topic from a sober rational perspective. This was very refreshing and i think also helped a lot to build understanding and to reduce stereotypes about the parties involved.
I think this is also needed for geopolitics, and the rationalist community could be in the best position to start this.
I don’t disagree with you but: can you have influence? How? Is it going to be good influence or would we be trying to control a complex system we do not know the underlying dynamics of? If you are in orbit, increased air resistance makes you go faster, not slower. How likely would activism aimed at affecting US foreign policy be to actually achieve a positive result rather than the opposite? I think there is a massive need for understanding stuff before taking action, as even well intentioned efforts may backfire. The whole forecasting community is working at this, at least from their angle. Developing solid international relation theory would help, obviously. If anyone here wants to discuss that, it would be a good start.
The influence we do have (Scott and the community of course more than me personally) is influencing public opinion.
I fully agree with you how important it is to understand things first and how bad well intentioned things can backfire. This exactly is my point: Internationally the US are often fast with actions (wars, unilateral sanctions, supporting coups, up to meddling with internal politics of other nations) and at least in my perception there is very little domestic reflection about it not before and not after.
US interventions did often have disastrous outcomes and the official reasons often turned out to be fabricated or at least not the real ones (otherwise the riorities would have been different).
As long as the domestic intellectual elite ( I assume most of the readers here belong to) does keep silent and does turn a blind eye on bad outcomes or even outright crimes, nothing will change except that in many parts of the world Americans are no longer taken serious when talking about ethics.
I expect a good chunk of the readers to agree with your point of view. As to what is best to do about it, if anything, I am not so sure. The US is the world hegemon, thanks in part to its ruthless foreign policy. I do not believe any amount of public opinion pressure motivated by ethical concerns would be able to change that. Even assuming that some intellectual elite had both genuine ethical concerns rather than mere posturing (which is doubtful already) and the means to reach out to the wider public, they would be able to influence policy only as far as no clear negative effects result for the main stakeholders (the big companies, banks, etc. running the show). At that point fierce resistance would ensue, first in the usual PR/politics sphere, and failing that even with overt violent attacks. To withstand that you need a broad power base (a party, money, etc…). You need to organize. Is changing foreign policy an important enough objective for this community to do that? I don’t think so at this stage. From a different angle: international power politics will not disappear once US hegemony is eliminated. In twenty years or so you might have a new hegemon or a bipolar situation, conceivably with China being the other pole. Will this situation be better than the current one in the average? Was the world more peaceful in the bipolar era when the US and the USSR faced each other? Was it more just? We might be able to answer these questions with data.
This is what I'm doing here: trying to spark rational and ethic evaluation of US foreign policy within the domestic intellectual elite.
The change I want to see in the world is that world hegemony gets as outlawed as dictatorship and to have an world community of nations based on respect and common rules.
This is exactly what the US wants in their formal announcements, but in practice all the rules (like international law) are ignored by the US if it fits their interest. The great achievement of the rule of law is that everyone, especially the powerful, stick to and are judged by the same laws.
American Exceptionalism contradicts this, but can only be changed culturally from within. I think its overdue to discuss and evaluate US foreign policy with an ethical and long therm perspective, not just emotionally or from short term nationalist interests.
>The change I want to see in the world is that world hegemony gets as outlawed as dictatorship and to have an world community of nations based on respect and common rules.
Pick one. Because the list of nations which are actually on board with the whole "world community of nations based on respect and common rules" bit consists of A: the United States of America, and B: a bunch of nations who are also wholly on board with American hegemony because it means they don't have to worry about how to ultimately enforce the rules and impose respect.
American hegemony, for all its problems, is better than any other plausible alternative in this generation. And if you break it, we will hold you responsible for the vastly greater problems to come.
Opposing US hegemony is the main motive holding BRICS together at the moment. I'm pretty sure if any of them would try to be a new hegemon, the rest of them would join forces to prevent this and welcome the help of the US to do so.
But the alternatives for the US aren't only ruling like today or just giving up power to the next strongest. The US could live up to their proclaimed values by strengthening the UN and bringing in place checks and balances that there never will be another hegemon. I'm confident that BRICS would join this effort, they all fear other powers ruling over them much more than that they are driven by the wish for domination.
Especially China never had much military culture, but had much more focus on soft powers. Just read Pearl S. Buck to get an idea.
Your question is so broad as to be almost unanswerable. However, if you'd like to discuss US foreign policy specifically, this is literally one of the single most discussed topics here on planet Earth. There's almost no possible conversational, polemic or political topic covered more than US foreign policy, criticisms of, etc. Even if you recently landed here from another planet, there's no way that you couldn't know how ultra-discussed this is, literally All The Time, 24/7/365, for the last 80 years or so
Yes of course I'm aware of this, but I'm curious how this group is thinking about it. Knowing that you care about ethics, charity and a better future you should have a opinion on it.
It looks like the elephant in the room being ignored. Why is this?
Isn't all the thinking about existential threads in the far future relatively pointless if there will be WWIII resetting civilization to the middle ages?
So I'm specifically curious about the views of the rationalist community and I also didn't find this elsewhere.
I'm pretty sure the X-risk people do talk about nuclear war, although that's a very heavily studied topic so I'm not sure a group of talented amateurs can improve matters much there.
Also, most of the US's foreign policy blunders don't have anything to do with the risk of nuclear war. Running Guantanamo Bay might be bad for the US in a soft-power moral-authority sense but it's not going to be the spark that makes the missiles fly.
Your very comment proves to me, that there is way too less awareness of the connection between nuclear risk and US foreign politics.
The key to reducing the risk of nuclear exchange is trust and respect. This is necessary to
- convince counties not to develop WMD
- any kind of disarmament
- get a good treshold that false alerts are discovered as such.
It's one thing that the US isn't promoting symmetric disarmament and new treaties, but they are actively damaging international trust by not sticking to commitments and applying double standards all over the world combined with threads of sanctions.
But the worst is US officials musing about destroying or defeating Russia after years of ignoring Russia's safety concerns. This is about the opposite of deescalation. Of course this makes them nervous and having the finger on the trigger.
- talking about human rights, but still having guantanamo.
- promoting democarcy while backing or even organizing coups
- insisting on international laws but but not sticking to UN resolutions
- tarling about a 'rules based world order' it is about to defend, while it seems that the main rule is that the US is the good guys and so is above all other rules.
- being super sensible about foreign influence in US domestic politics while trying to have influence elsewere (was the US media also so critical about the initiative for a law about 'foreign agents' in Georgia, that was basically a copy of the FARA-act?)
- wanting a international tribunal for Putin, but not prosecuting the own war crimes and even threatening Den Haag if an American should be prosecuted there.
- talking a lot about freedom of press an journalists while prosecuting Assange and turning a blind eye on suppression in allied counties
I could go on for a while...
Perhaps some of this is officially not directly the US, but NATO. I count as basically the same, as the US dominates NATO.
Yes. What’s your point? There’s things that rational us are interested in and things that they’re not and the more they are unable to affect something the less they are interested in it. US foreign policy is the example of something that it is almost impossible for an individual or small groups of individuals to affect.
Well this answers my initial question basically as: We are not interested and anyway have no chance of influencing public opinion against the MIC lobby.
I mean, the general answer is that in any of those cases where the US govt. decides to act in ways that violate the principles it espouses, it found the cost of uphold those principles to be too high. Considering most people are not perfect exemplars of the principles they hold, I don’t find this to be super remarkable.
That’s fine. The original question is why isn’t hypocrisy in US foreign policy more widely discussed here, and my answer is that the degree to which the US is hypocritical in its foreign policy is fairly unremarkable considering some degree of hypocrisy is a fairly wide-spread human trait.
The other thing is that the US government is made up of a very large number of people, who don't all agree on everything. Some of those people are smart and rational and admirable, and some are massive jerks.
For every one of those supposed contradictions between stated principles and actual actions, you can find a whole bunch of people who will argue that "well actually in this special case there's no contradiction because..."
You assume that there always will be a hegemon. If I think about creating a better world this includes to install checks and balances that prevent any county to get too powerful. The UN looks like a serious try to start this, but instead to develop this path the US did everything to secure it's status as hegemon.
If you have like almost equals 4 of them can always join to keep one in check that tries to grab power.
At least the UN doesn't suddenly decide on a America first policy. This may not sound like a big improvement for you, but for the rest of the wold it definitely is.
It all depends how it is organized. There is a lot space between having a biased , unfair police dealing with double standards (this was the beginning of the topic) and not having any police at all. Who never tries has lost already.
And yes, I'm convinced that it is a matter of culture and wisedom to understand that violence doesn't give good solutions ever. In this regard I do trust countries that suffered major wars on their territory, have a long cultural history or were even colonized much more than the US. Unfortunatelly not just watching the stage of international dipomacy but also many of the answers in this forum confirm this.
I'm amazed such a major scandal, which culminated 3 years ago, had been off my radar. Have I been living under a rock or has the US media been very uninterested in this story until now?
The part I found most fascinating was Germany's launching a criminal investigation against the Financial Times after Alphaville alleged massive misdeeds at Wirecard instead of investigating Wirecard.
Even in Germany is was relatively fast off the news again. Many think this could be because too many politicians with good friends in the media were promoting wirecard earlier and didn't want too much dirt coming up.
Listened to the Yudkowsky podcast (Bankless) from the last set of links and was struck by how hopeless he seems. Similarly, the "Death with Dignity" post seems to confirm that he thinks the situation is hopeless.
I don't personally agree, but for people who seriously think the world is going to end in ~25 years due to AI...
...why not try violence? Violence is usually reprehensible, but to somebody certain that the world is going to end, wouldn't taking out a few researchers be justified? It would send a strong signal to others considering going into AI research as well, that they might not be safe if they pursue that career path.
Even ignoring that their typical ethical systems forbid this... Yudkowski & co tend to (IIUC) believe that stopping AI progress is futile, because the incentives in place force capabilities research to keep going forward. Thus, the rare envisioned survival outcome depends on the world suddenly becoming much more reasonable than expected, to prioritize alignment research in time, combined with being very lucky about how easy alignment turns out to be.
If survival depends on the world being unusually reasonable, and one expects that violence lessens (rather than improves) collective public reasonableness, then violence is bad. Less "heroically saving the world", more "unilaterally committing global suicide".
I dunno about that. Maybe have a look at my long comment about ways to interfere with the further development of AI. All involve dishonesty, tricks and coercion, but none involve anything close to murder. I'm not saying we should do this -- personally, I feel unable to even estimate how likely how soon ASI will be developed, and what will happen when it does, so if people were taking steps of this kind I would not be willing to participate. But I do think it's quitepeculiar that people who are *sure* that AI will kill us all are not thinking in a more practical way about how to avert the disaster. They seem to only see 2 alternatives: Shoot some AI CEO's (which is evil, and also would not work), or else convince the population of planet earth to see the danger (which is impossible). WTF is wrong with these people? THEY ARE TERMINALLY IMPRACTICAL! If they want to slow down or stop AI development, the way to do it is to use the kind of tools that are routinely employed by those in power to influence public opinion and force other powerful entities into line. Murder is rarely required or even useful in these coercions.
The flaw is that your intended target in this scenario is not a dummy full of hay at a firing range.
"Violence" is a historically useful tool (Ref: most of human history) for some aims at some times in some cases. But what people who suggest violence as a solution often forget is that their opponents/targets might be
1) Just as capable of violence, if pushed
2) Better at it
3) Heavily defended by those who are
4) inclined to become any of the above three if they feel their security is at risk.
You start stabbing anyone who works with a machine model, we can see how quick people who work with machine models either pick up self defence training or start carrying (perfectly justifiable) weapons. And incidentally, these are usually smart people with engineering proclivities, so they'll probably be mildly spooky if motivated to take security seriously.
On top of that there's an entire social apparatus built to minimise, prevent and investigate violence against your fellow human beings. And highly paid software engineers working white-collar jobs at heavily funded institutions can, all else being equal, most likely afford much, much better private security than the people trying to hurt them. You reach for the devil's tools, intend to pick up his sword, and find that some goon in a lab somewhere has pushed the autonomous riot control drone software to 1.03 because they felt worried.
"Society" in general is just better at violence than any group of motivated individuals.
This is all setting aside totally normal concerns like "your suggestion is illegal" and "Anyone suggestion violence is kind of an asshat, aren't they?" and "you realise lone snipers and or deep cover assassination cells must contend with the security apparatus of an entire country, a class of people seldom known for their chill".
Probably because it's difficult to imagine any kind of violence that has a snowball's chance in hell of working, as opposed to ending up with the perpetrators dead or imprisoned for the rest of their lives.
I mean, if you want to radically alter government policy all over the world, or at least everywhere there's a chance of AGI being developed -- say, the United States, Australia and New Zealand, the 28 nations in the EU, Japan, South Korea, India, China, Taiwan, probably a bunch others I'm forgetting -- then you need let's say of order ~40-50 violent revolutions to succeed. That's an incredibly tall order. I don't think any less than 1 milion heavily-armed, well-trained, polylingual, very well disciplined troops could pull it off. And it would be a bit tricky to recruit and train them without anyone noticing and asking questions.
Could you be thinking of assassination? Of key leaders in the field, say? That would require much less violence, maybe only 50 teams of 5 ex-Mossad types each. But it would also require you to have some very deep insight into *which* 50 world leaders in AI are (or will be over the next 10 years[1]) absolutely critical to the construction of an AGI, which would mean you probably have to at least be one of them, and have some really excellent ideas about AGI -- basically, almost be in a position to build one yourself. That's a pretty big ask.
Plus you better have some really excellent coordination, so that after the first 5-10 hits the plot isn't discovered and counter measures undertaken. You probably need to do all the hits simultaneously, to avoid tipping your hand. The level of military competence required here is pretty high. This is something even a big national government would struggle to pull off.
I think your final hypothesis, that killing one or two researchers would "send a signal" to the others, is implausible. In the first place, killing one of two would most probably result in the organization being rapidly uncovered and the members charged with a variety of crimes and sent to prison, or (if in Texas) to the electric chair -- particularly if the plot is perpetuated by programmer-type nebbishes, whose are not known for their skill in the dark arts of constructing felonious conspiracies, laundering money, and committing murder. It would certainly make the news, particularly around the time of any executions, but I can't see it dissuading anyone from going into the field for the next quarter century or more. On the other hand, I *can* see it resulting in anyone who ever expressed any unguarded public enthusiasm for strong measures to suppress AI research being put under a strong level of scrutiny for the rest of his life.
-----------------
[1] Which means you need to consider wiping out the most promising teenagers, probably before they learn to drive. This is definitely going to earn you a trip to the gallows or electric chair in a few jurisdictions.
Eminently more practical. But don't you psychologists have some kind of code o' professional ethics that comes into play here? With great power comes great responsibility, Uncle Ben says. Although...maybe you're keeping the really good stuff to yourself. Yikes.
I'm a mandated reporter regarding neglect or abuse of children, the elderly or the disabled, and I have in fact made about half a dozen reports of this kind . Except for the above, though, I'm just a regular person who's spozed to follow the law. As for the various forms of coercion I was proposing as a way to shut down AI development -- jeez, I'm worried about the effect on people of current and near-future versions of AI, but cannot figure out how worried to be about AGI. I jusdo not know enough about machine learning, etc., though I'm working on catching up. I can't even figure out, as a psychologist, how seriously to take Yudkowsky's conviction that we're doomed. It might be that he's the Temple Grandin of computers. (Temple Grandin's an autistic woman who's said to be extraordinarily gifted at intuiting how animals experience certain things, and has designed systems for housing, transporting and killing farm animals that keep them much calmer and healthier.) Or maybe Yudkowsy's a deeply eccentric, sad, brilliant wrongheaded guy. So at this point I would have no interest in being involved in any sort of French Resistance against AI.
But the people who are sure AGI will destroy humanity soon sound hilariously naive & impractical when they talk about taking desperate and illegal steps to stop AI development. "Um, well we could shoot somebody. Or, like, blow up a lab. But then we'd just go to jail, so that's off the table. " My point is that if you think desperate and illegal measures are justified, shooting Sam Altman is not the way to go. There's all kinds of stuff that would be several orders of magnitude more effective, and does not even involve violence. Get it together, guys.
I meant, don't you have guild rules against giving people ideas based on your superior understanding of psychology? But it was a joke anyway.
I've read Temple Grandin's book, and I found it deeply fascinating. Not just because of what she said, but because it was clear from the way it was written how alien is her thinking. Her ability to convey (indirectly) what it's like to think like her is astounding -- the more since any expressivity she has is the result of sheer hard thinking and logical deduction, since she has no way to do it the easy way -- put herself in the shoes of her listener -- the way the rest of us do.
I knew someone who was very interested in Grandin's work, and introduced me to the book, not so much because she was interested in autism per se, but because she was interested in how animals really think. She wanted a non-Disney non-sentimental insight, and thought Grandin was on to something. As a result of that association, I became myself a little interested in how animals really think -- and I feel like pondering that is one of the big misses people routinely make when they think about thinking in general. We introspect, and draw conclusions from that. But we often completely neglect the fact that we share the planet with a large number of other species which, while not quite at our level, certainly think and plan and learn. There's probably a lot to learn by understanding how what they do differs from what we do.
Yes, of course the AGI fearful are not very practical. I'm a cynical old wretch, so I tend to think it's because in their hearts they really know it's not real. They're not *really* afraid, they're thrilling-afraid, the way people are at a slasher horror movie.
I think I am hampered inunderstanding animals by sentimentality. My whole family's like that about animals. I feel like they're my children. But they're not. My cats are much more matter-of-fact about dying than I am about their deaths. I take their wordlessness to be chilidlike innocence & that's not really what it is. I should maybe get some tarantulas. Pretty sure I would not be sentimental about those fuckers.
Yeah, the whole Yudkowsky scene sort of squicks me out. It has a dysfunctional family quality that's familiar -- has to do with everybody agreeing to pretend something. If the AGI fearful are just thrilling-afraid, listening to Yudkowsy tell his horror movie story, then who's Yudkowsky -- Boris Karloff? A year or so ago he posted a sort of am-I-joking-or-am-I-serious thing, and I wrote a furious diatribe about it here, then afterwards wondered if I'd sort of overshot.
"why not try violence? Violence is usually reprehensible, but to somebody certain that the world is going to end, wouldn't taking out a few researchers be justified? It would send a strong signal to others considering going into AI research as well, that they might not be safe if they pursue that career path.
Where is the flaw in this utilitarian calculus?"
Hello! Have you considered the cases of abortion clinic bombers and those who kill abortion providers?
Did that stop abortion, or was the Roe vs Wade Supreme Court decision what made the difference?
You start killing AI researchers, you make them Martyrs of Science, you permanently blacken the name of your cause, and there will be drives, campaigns, and stirring stories sent out as PR shots to the media looking for more people to take up the mantle of AI research in the name of liberty, progress, and standing up to the terrorists.
Look around you at the real world events where people tried violence, from Basque separatists to animal rights activists. See how effective that was, versus slogging away with the ballot box and the law courts.
It may not be the usual result, but the assassination of Shinzo Abe by a hater of the Unification Church (Moonies) is a recent example of violence achieving its intended political result.
Violence – everything from assassinations to violent uprisings – does 'work' at times, to achieve political or social ends. However, in one study of 323 instances, "nonviolent campaigns were nearly twice as effective as armed campaigns over the past century."
It won't help, and will likely make things worse. A few researchers won't make a huge dent in AI progress, but it will completely write off AI-doomerism as a bunch of deranged psychos. AI-supporters are already *preemptively* writing off AI-doomers as violent lunatics based on their rhetoric. A terrorist attack would likely *accelerate* AI progress accordingly.
I read a comment about this by somebody who's involved, and they said violence would just get some of them thrown in jail and would not stop the progress towards AGI. But i seems to me that what the writer had in mind was a very naked, naive kind of violence -- just go fucking shoot Sam Altman and a few others. Here's a question: If you were going to be utterly amoral about it, what could you do to most effectively slow or stop progress on AI. Here are some utterly amoral ideas:
-design computer viruses that will mostly be active against machines and software crucial to AI development. Try to make it really contagious, so that it gets into pretty much any machine working on AI. Plant another virus, virus #2, that will stay dormant until virus #1 is gone.
-Pay some genius Israeli supercyberspy to do that stuff, and also to propose ideas of his own. See if you can swing it so that the person who is called in to HELP with the virus is in fact the superspy who designed it, or someone working in league with him.
-Scare the hell out of the public about AI. Do not bother telling people the real reasons to be scared. Instead, play on the fears they have. For people in the anti-vax anti-mask faction, tell them that in the near future we will be putting AI in charge of mask mandates and vax mandates. AI will send out drones to public areas, photograph all who are unmasked, later send them tickets: $100 for first infraction, $500 for second, prison stay for third, plus your name goes in a "non-compliant citizen" data base. AI will do the same with vaxes, and will send special cute drones to playgrounds to convince children to be vaxed without their parents' knowledge. Say AI is going to install cameras at every intersection and turn us into a surveillance state just like China. Involve Fox news, Russian trolls and any other assholes you can think of to amplify these ideas. Say AI was developed by wokies, so will be enforcing woke values right and left. It will be a crime to say trans women are not women, and a hate crime to snigger if a 6’2” trans woman passes you on the sidewalk. Say AI has no religion itself, and is looking for a chance to dismantle religious values.
-I leave it to others to figure out how to freak out the Left. But it’s not hard to do, you know?
Let's see, what else?
-Blackmail crucial figures in AI into backing off.
-Steal AI's funding. Sell the AI companies on some crypto scheme that somehow collapses into mud and fromg spawn.
-Make governments fearful of AI, by spreading rumors that the developers of the big AI's picture themselves running the world, using the AI as a tool.
-Publish op-ed's supposedly written by AI that will alarm the hell out of the right, the left, the government . . ..
I think a much better approach would be to lobby for legislation that demands the AI companies make their work as sandboxed and capability limited ( 'shackled', one might say ) , and possibly significantly raise tariffs on compute capacity.
This approach still feels like it has flaws (The AI companies will tend to route around the regulations instead of truely abiding by them), but has a lot less political blowback.
1) Most people in congress do not understand tech. 2) The tech companies now have so much power and money they're almost like a second, alternative body shaping how the country runs. Nobody can put them in a sandbox. 3) Congress is dysfunctional
"Say AI has no religion itself, and is looking for a chance to dismantle religious values."
r/Atheism is very interested and would like to subscribe to your newsletter. Seriously, if you try saying "AI will disprove religion!", there will be plenty of people falling over themselves to promote it.
You might even succeed in getting a few of the anti-AI types to switch sides - after all, if the pope (any pope, take your pick) is the worst person in the entire history of the world, how bad can AI be by comparison?
I’m personally going to wait for at least a couple more genius autodidacts to throw in the towel before I get seriously worried about the future of humanity.
Re: the future of humanity - we're all gonna die one day. As individuals, that's a solid sure thing. As a species, eventually the sun will go ka-blooey and we're done.
Humanity doesn't have a future because everything will come to an end one day.
That being said, that is no reason to make things worse if we can avoid it. But AI doomerism may or may not be right - it's very difficult for me to believe that the machine intelligence will get some kind of agency and act out of its own plans and desires, rather than being yet another tool that we use to poke ourselves in the eyes with.
> As a species, eventually the sun will go ka-blooey and we're done.
I'd characterize that as more of a 1 billion year countdown to manage a colony fleet over to to the nearest dwarf star so we can keep going from there.
All this reminds me of Alfred Bester's 1974 serialised story turned into a novel, which is either "The Computer Connection" in the USA or "Extro" in the UK.
This may have been before the Internet as we know it, but he got it right about the kinds of products and services that would be advertised:
"A capsule floated down on top of the bods with its jets spraying fireworks. A blue-eyed blond astronaut stepped out and came up to us. “Duh,” he mumbled in Kallikak. “Duh-duh-duh-duh. . .”
“What’s this thing selling?” Uncas asked.
“Duh,” Fee told him. “That’s about all the honks can say, so they named the product after it. I think it’s a penis amplifier.”
This may sound familiar, but since he was writing in the 70s it's not so great a stretch to imagine campus protests going on forever:
"The bods at Union Carbide, our local university, were mounting their ritual protest. It was the traditional daily rioting, with screamings, burnings, and killings. The only thing that changed was the cause, and the pressure groups had to sign up months in advance for representation."
I don't recall it, alas. But Bester is a fascinating read. He always makes me feel like the cracks are showing in the Universe, a bit, like maybe any moment God will shake the snowglobe and everything comes to pieces and settles down in some other very, very different configuration. And his characters always seemed to be this hypergolic fusion of massive id drives and extreme cunning. Crazy guy.
Aside from the ethical red line, it's just as likely to cause a martyrdom effect as to discourage anyone. Worst case, Congress passes a bunch of laws putting AI alignment folks on the same level as ecoterrorists and the like, effectively making it even harder to accomplish anything .
I mean, if you believe the alternative is total annihilation, an ethical red line isn't really that much of a strong argument. And it may not be unethical to you at all if you believe these people are doing bad things.
Plus, presumably you'd do this once you think all other alternatives are exhausted, which Shivers believes Yudkowsky thinks.
Principles shouldn't change just because the stakes are higher; the whole point of having deontological injunctions is that arguments for violating them are much less trustworthy than they appear
i think this comes down to the whole thing where the Sequences had the general message "utilitarianism is correct for ideal reasoners, and you should generally keep utilitarian principles in mind as a sanity-check against negative-sum behaviors, BUT humans are absolutely not ideal reasoners and we need deontological ethical injunctions, learned through grueling empiricism, to protect us from our own cleverness"
but as far as actually deciding when to use utilitarianism versus obeying your culture's hardcoded deontology... there isn't really a clear answer, and the edge cases are extreme, and it's generally best to err on the side of caution
there's also some game theory here!
like, sure, i can come up with some very strong arguments for doing bad things to AI capacity researchers so they can't end the world. And the urgency sure seems pretty staggeringly huge. But there are a lot of people who can come up with strong arguments, according to their own contrarian beliefs, for doing all sorts of wacky violent things. If I claim that "i believe that I'm right" is justification for discarding my deontological restrictions, I lose the right to expect others to also obey those same restrictions. And I think a world where everyone has to obey those restrictions is probably better than a world where no one does. Symmetrism, the rawlsian veil of ignorance, etc etc etc
The cached response I've come across (from memory, and potentially incorrect, i.e. I don't know what proportion and to what degree those in the AI safety community believe this) is that even if you use violence to disincentivize dangerous work in AI in the near-term, it is still the case that, given things like competition, accessibility of compute / repositories, and the number of people interested in AI, unaligned and dangerous AI will still eventually be developed. Of course, this belief seems to assume that there is would be an insufficient amount of willingness / capability to commit violence of this form (i.e., against those working on and enhancing general AI capabilities) to actually make a lasting difference. So, the potential impact of such violence is small, given that the population of violent actors and the damage engendered by each actor would likely be small, among other things, and that there are too many labs and corporations actively working towards AGI in severely subpar fashions.
He's not quite as doom-and-gloom- as EY, but he's not far behind and I think his argument is relatively generalizeable.
To give the 1 sentence summary (but really, go read it): Violence will not and very likely _can_ not make the situation better. Even if you honestly believe that the current path will, with near certainty, bring about the extinction of humanity (which I'm not sure that EY actually thinks this, he just thinks it's very likely), violence will just make that extinction a worse process, not actually prevent it. I pretty strongly agree with Zvi's take here.
I don't think that violence is _never_ warranted, but I think that for a situation like AGI and our concerns about it, it's a situation that violence is almost certainly not capable of resolving.
I haven't read Zvi's post yet, but his conclusion seems reasonable, except in scenarios where very severe violence (widespread and harsh) occurs, but I believe such scenarios are very unlikely to occur.
When Joe Rogan interviews Coffeezilla, the idea comes up that the crypto community needs to regulate itself-- the amount of fraud makes crypto an unattractive investment. Is there any way for the crypto community to do this?
Sure, you could theoretically set up a clearing house. Something like a crypto code authority that gives its stamp of approval. Some of the bigger exchanges kind of do this already by permitting different tiers of respectability. But I expect there's a lot of people who will resist this for a combination of ideological and "actually, it is a scam" reasons.
What's actually going to happen, in my opinion, is that the crypto con artists are increasingly going to be crowded out by the more respectable projects that can bring in big institutional money. And eventually this will create enough governmental institutional knowledge they can effectively enforce existing anti-fraud laws (which they currently don't). Meanwhile there'll still be a small hacker ecosystem that will be the wild west. Including no end of scams but also productive stuff like unbanked payments.
Which projects are you thinking of? If you're thinking of FTX that was mostly privately raised money and VC. I'm thinking of stuff like the E-Yuan or the various African experiments or government backed remittance companies.
oh huh. i guess I was thinking about projects like Luna, Celsius, Safemoon, Terra, or Tether
although all the coffeezilla videos tend to blend together... my general, possibly inaccurate, impression is that some or all of those projects got quite a bit of excitement from institutional investors
i guess that still counts as private money... but it still feels like the whole field has some pretty bad PR toxicity
somewhat unrelated but, i'm a little weirded out by projects like the e-yuan. the whole point of cryptocurrency, the thing that makes it more valuable than nothing, is that you can take advantage of hard cryptography to guarantee aspects of the coin, like its future scarcity, or that your ownership can't be taken away from you. i can certainly see why, in certain situations, these features would be valuable. especially for consumers who are worried that governments might abuse their ability to print fiat currency to take value from holders, or worried about banks obeying instructions from the state to freeze or steal their assets. but that's like... the only real benefit of crypto over fiat, the leveraging of hard cryptography to make it mathematically impossible to mess around with the currency.
why on earth would china want this? it seems like the exact opposite of what they'd want. from reading about it, i can't honestly determine if it's even a cryptocurrency... it seems to be just another fiat currency.
i have some of the same questions about some of the other projects. i can see why, in an environment of distrust, having a currency where all funnybusiness is literally ruled out by mathematics would be very useful. but the whole reason there's currency funnybusiness in the first place is that the public institutions which control the currency get a lot of value out of that funny business. why would those same institutions want to introduce a currency which robs them of those powers?
and if the new cryptocurrency doesn't rob them of those powers, it's not really a cryptocurrency, it's just another digital currency. maybe it'll have some features which distinguish it a bit from just keeping your yuan in an online bank... but not the features which made cryptocurrency valuable in the first place?
idk, sorry for the rant, just kinda spewing my uneducated thoughts out
So when I say institutional money I mean, for example, Goldman Sachs or the Federal government or that kind of thing. Most scam crypto was done through shady offshore banks or through large groups of private people pooling money. Lots of small depositors. Not all of them unsophisticated. There were plenty of finance bros. But it wasn't like FTX was a Wall Street creation. They were in the Bahamas for a reason.
I absolutely agree about the PR toxicity. My sincere hope is that this market downturn purges out the worst of the rot.
Cryptocurrency is a specific technology. Libertarian leaning crypto types invented money with a bunch of security features. But the security features are not inherent to crypto and the E-Yuan does not have them. It has the opposite: features meant to enhance state control. Nevertheless it still has uses on a unit of currency level in assisting banking transactions etc. It still lives on blockchain. Etc.
The advantage of cryptocurrency is that it allows decentralization. If you're some libertarian living in Central America maybe you use that decentralization to create a highly secure currency meant to prevent government control. If you're the Chinese government you use it to speed up banking transactions and make them more transparent to regulators and censors with less computing power.
The diamond exchange is a non-governmental community that self-regulates astonishingly effectively. But that's a small, insular community. I can't imaging that would scale to crypto.
You don't necessarily need a government. Securities exchanges (i.e., groups of organisations that run a communal trading system; the members are those that have a "seat" on the exchange) have some pretty tight regulation and concerns about fairness and transparency that go beyond what government regulators require and are better enforced, since the members consider it important to keeping the system working.
There's a lot of overlap with government regulations, but the internal ones in certain areas are stronger and better enforced because because it's the exchange members themselves who are stuck with the bill if something goes bad. Margin requirements, for example, are always going to be carefully managed regardless of external regulation because as a member of the exchange if someone for whom you're placing trades can't settle, you're on the hook. (And if a member can't settle, everybody else is on the hook.)
I can see this kind of thing happening in cryptocurrencies, too, though probably not to any great degree because the incentives are different and there's a greater number of actors that are more concerned about benefits to themselves rather than to society.
I don't think that most people would agree with you that the Chicago Mercantile Exchange is a "government."
And you can't participate in cryptocurrency ecosystems without following the rules, either. Try submitting a new block without a proof-of-work and you'll find that nobody will accept it.
I have two subscriptions to Razib Khan's Unsupervised Learning to give away. Either reply with an email address here, or email at the address I specify here: https://entitledtoanopinion.wordpress.com/about
A belated comment on PMS and culture-bound mental illness:
1. I really thought most women, say 80%, get a little grouchy and weepy before they get their periods. Concluding from my experience and those of my mom, friends, aunts, etc.
2. Like one of the immigrant women who are surprised by PMDD diagnoses, I wouldn't call PMS a mental illness. I'd class it with normal mood changes like people getting cranky when they're hungry or have slept poorly. I wonder if that's what the surprised immigrant women figure too.
3. On top of the hormone-caused mood changes (I'd guess?), women probably logically get a little grumpy before our periods because we're about to have have stomachaches and constrained clothing choices. This feeling is like being annoyed when you realize you're coming down with a cold or when seasonal allergies start up.
I think it's likely that on average women are indeed more dysphoric before their menstrual periods. But the fact that this mood state is hormonally driven is not grounds for classifying it as a "mental illness." Lots of states are biologically driven: fatigue, horniness, craving to eat. . .
No, for my mom and me at least, there's an increased readiness to cry at sappy things which does not come from stomachaches. Also, I wouldn't call it pathologizing, but rather a normal emotional response.
Re: #1, in my experience as a man, a little introspection shows that male mood changes throughout the day exactly as one would expect from how testosterone generally changes throughout the day. I'm baffled at the idea that hormonal changes wouldn't have any effect for women too. The size of the differences can vary but it someone notices no difference at all I suspect the introspection component.
I use it for generating SQL queries, which it's quite good at. It's rarely 100% correct and usable off the bat, but it taught me about the LAG window function in a helpful way, and put me on the right track for saving the results of a Clickhouse query to a file
Another useful application is summarizing difficult text as you go. I’ll take a textbook paragraph, pop it into ChatGPT with a request to rewrite it in a conversational tone. Then I real the new version, compare it with the old, and often have an easier time building a more confident understanding of the material.
I use it as a language editor, to get boilerplate code fast, for translation sometimes, to find quotes that I remember having seen somewhere, to learn new stuff about a topic on which I have no mastery of the technical jargon
I've been using it to help speed up my coding process in pandas. Literally every answer it has given me has been wrong and I think I've used like 1 line of its actual code unchanged, but it speeds up my thought process of "what structure should I use here?" drastically. Where to implement functions to be called in apply functions, where to use lambda functions, where to use for loops, etc. Stuff previously I would have thought about for an hour or so before starting to code up a solution, or spent a couple hours trawling through stack overflow, I just get an immediate pointer.
I use it as a form of autocorrect on steroids. It lets me brain-dump without worrying about polish, then produces legible text I’ll actually read or feel comfortable publishing. It’s also useful for transforming my language into unfamiliar wordings. If like me you usually need a day or two of distance from your text to be able to edit it, you’ll understand why this is helpful. It can also be useful for taking a long set of bullet point notes and producing a condensed summary.
It can be useful for generating critiques of your own ideas, or for identifying potentially confusing material.
I don’t find ChatGPT to be useful for producing new ideas, and I don’t trust it to teach me things, but it’s great as a sort of automated copy editor.
So I’m just learning that a lot of the highest level genetic biologists in the world actually try to deny that genetics play a factor in, for example, the NFL having 100% black corner backs. The reasons they give are really bad too (“culture”). White suburban communities eat sleep and breathe football, and pour endless resources into it at the high school level. I didn’t think the field had lost THAT much credibility. Can we trust anything that comes out of there at this point?
I don't automatically delete comments like this, but I think when you post something this controversial and insulting you owe it to us to explain what you mean. I think I'm somewhat well-informed on these kinds of topics - eg I've read about the potential genetic differences between East African marathoners and West African sprinters - but I'd never heard of this and don't know how to interpret it. What are "corner backs"? What's your theory on why genetics would cause them to be predominantly black? What's the evidence for this theory?
So it’s not that cornerbacks are “predominantly” black, they are actually 100% black and there hasn’t been a white cornerback in the NFL since the early 2000’s. Not only that, but at the highest level of college football, so 128 teams, starting corners are also 100% black.
Cornerback is a position on defense that is primarily involved in defending the passing game. It is the position in the NFL that is most about pure athleticism (in the sense of speed, quickness, and strength). Every high school in America has 2 starting cornerbacks on their football team, the amount of white kids vying for college scholarships far exceeds the amount of black kids. White suburban/rural areas eat sleep and breathe football as integral to their culture.
Essentially, denying genetics being a major factor here is flat out ridiculous. It’s the most egregious example I’ve seen of the trend towards insane mental gymnastics to deny anything resembling group differences. I had no idea this trend was actually in academic genetics/biology, until I listened to Richard Hanania’s podcast with a Harvard evolutionary biologist who basically denied any genetic aspect to the NFL/NBA being majority black. The specific cornerback claim was something I saw from another academic biologist online, which if you know the context is another level.
I’ll admit it the post was a tad lazy, lol. I know pretty much nothing about genetics, but a lot of times I’ll make a post on here about a topic I know nothing about and another reader willl chime in with some good recommendations.
I saw him in Baltimore a few months ago and good lord he absolutely killed it. Im pretty sure that material he was touring with is going to be his next special, and if it is the world is in for a treat because the man has improved tremendously from his last special (which was unbelievably funny).
I live in Philly, and there’s a weekly open mic in my neighborhood that I’ll go to just to hear the local comics work out material. Everyone from first time amateurs to local vets and low level professionals. A lot of the guests from Matt and Shane’s podcast are regulars there, like Lemaire for example. I really respect the fact that Shane will still support and give a platform to the comics in the local scene he came out of. Just good energy he puts out. Comics all around are interesting people.
Come on Scott, I know people in the rationalist space are probably not big sports fans, but it would take a very quick google search to see that this is a position in American Football. He's saying that people of a specific race being so predominant in a specific position that requires specific physical traits and skills seems pretty obviously genetic, especially when any recourse to "cultural differences" is obviated by the fact that both white and black people in America obsess over football, and this wouldn't seem to impact the specific positions within the team people play by race, anyways.
I think my favorite ethnic factoid about the NFL was that, while Samoans represent something like 0.1% (from memory) of the US population, at the time of the article they represented 80% (also memory) of NFL defensive linemen, and went on to discuss the NFL’s efforts in player development in Samoa and Polynesian communities.
Your numbers are way off but yes Samoans are way over represented in the NFL and NCAA division 1 relative to population (which is like 60,000 people). All over the field, not just defensive linemen. Troy Palamalou is prob the most famous. Tua Tagavailoa is QB for the dolphins today. They’re generally a big and strong group of dudes.
There’s a guy on the eagles named Jordan Mailatta who is an Australian of I believe Maori descent. He is a 6’9” 350 pound freak athlete left tackle. He never played football in his entire life, but somehow NFL coaches got ahold of his rugby tape and taught him to game from scratch. Couple years later he’s among the best tackles in the league. I assume other NFL coaches are going to follow suit and we will see more freakish Polynesian rugby athletes in the near future. Everyone knows genetics are a thing except for the geneticists apparently.
I love the fact that there is a community that doesn’t automatically know what a corner back is. I haven’t watched an NFL game since Fran Tarkenton was playing because I knew that it produced a lot of former players who’s bodies were ruined by the game.
For what it's worth, I'd never heard of corner backs, and I thought it was a typo for quarterback.
As for culture, I'm wondering about pressure from better off/white parents to keep their kids out of football, even if the parents love watching the sport.
I imagine better-off parents would steer their kids towards something other than an athletic/sporting career, but again there must be poor to lower-middle class white kids whose best chance of making it big is to be a sports star?
American football to my ignorant eyes is a very stratified game with so much development about what positions are played, how they are played, who is best fitted, etc. I can well imagine it's gone way beyond "we need a big guy at the back and put the small nippy guy up front" when it comes to what kind of players fit into what kind of position. So it could indeed be "the reason all the high level players in this position nowadays are black is the same reason all the high level 100m sprinters nowadays are black".
Unless you can demonstrate that it's not an important or desirable position or one that has little effect on the game, then you can't say "well it's culture and systemic racism that the black players get relegated to this position".
I think I’m noticing that people who are generally unfamiliar with sports seem to have a poor concept of how rare NFL level talent is, and what the process of developing elite athletes from a young age looks like. Kids start playing sports at a young age, and if your kid even has a chance of sniffing the pro level, they will not only be the best player on their team, they will more than likely be the best player any of the coaches have ever seen in the history of the league. Even at 11 years old, they will be head and shoulders better than all of the other kids and it’s extremely obvious when someone has that level of special talent. Rich, poor, middle class, black, white, no parent who’s kid is dominating to the level that future pros do is going to pull their kid out of football. It’s just such a rare, unprecedented level of talent. So while I think there’s definitely parents pulling kids out, odds are that’s not effecting the top talent regardless of family background.
But you’re right, different positions require different skill sets. Positions that require brute strength and size (linemen) will have about an equal distribution of black and white players. Corner is one of the most important and highest paid positions today, as it defends the passing game. Extreme athleticism (quickness and speed) is emphasized more there than any other position. There hasn’t been a white corner on a full time NFL roster since the early 2000s (teams normally have 5 or so rostered). If you polled every NFL player and coach I’m certain 100% of them would say it’s genetics.
"Kids start playing sports at a young age, and if your kid even has a chance of sniffing the pro level, they will not only be the best player on their team, they will more than likely be the best player any of the coaches have ever seen in the history of the league. Even at 11 years old, they will be head and shoulders better than all of the other kids and it’s extremely obvious when someone has that level of special talent."
Yes, the equivalent over here are football (soccer) academies and teams having youth teams to bring through talented kids, with scouts attending matches to sniff out talent. Liverpool academy as example below:
"Scouts attend many local youth matches looking for talented boys. A boy will then be invited to attend training sessions at the Academy. They are currently taken in as young as the age of six. Former England International player Jamie Carragher started at Liverpool when he was aged just nine, with Michael Owen joining at eleven, and Steven Gerrard joining at the age of eight. At this age, the boys start by simply attending after-school training sessions, but as they reach their middle-teens, their academic needs will be taken over by the Academy if they are deemed athletically talented enough. As such, the Academy has a lecture theatre and a computer-equipped classroom.
The Academy can handle up to twenty boys in each year group, although the actual number in each year group is usually around eighteen. Between the ages of eight and twelve the boys play in eight-a-side games of three twenty-minute periods. It allows the boys to play as defenders or as attackers in small groups within a system and is not as physically demanding as playing eleven-a-side matches."
Even attending an academy/being on a youth team is not enough; you have promising kids who are great at age 10, very good at age 15, not good enough to make the major team at age 16+.
I suppose I'm thinking of soccer over here. It's predominantly working-class. If you're middle-class to upper-class, you'll probably play rugby at school. Rugby union is now professional, but the money to be made is not comparable to professional soccer players:
That's possibly a hold-over from the days when union was the gentleman's sport and was played by amateurs. There's the old school tie at work too, where being a rugby player will (allegedly) get you advantages in your work career. If I had to guess, I think professional rugby union (as distinct from rugby league, that's another matter) is for the PMC (professional managerial class) background. Here's data from 2019 in Ireland, and I think that wouldn't be so far from how it is in UK as well:
"There is undoubtedly a perception that rugby is a predominantly middleclass sport, often associated with prominent fee-paying schools. The available data allow a quantification of the association between the likelihood that an individual plays rugby and their socio-economic background.
…Figure 2.7 shows the odds ratios for active participation in rugby by categories of educational attainment, as estimated in the statistical models provided in Appendix C. The reference category is individuals with upper second-level qualifications only (i.e. Leaving Certificate or equivalent). The chart displays a strong effect, such that higher educational attainment is associated with substantially higher odds of playing rugby.
The odds that an individual who goes to college or university plays rugby are approximately twice the odds for an individual who ceases education at Leaving Certificate. "
Upper class kids are not going to be professional sportspeople, unless it's something like round the world yachting or polo or golf or the like. Middle to upper middle class kids will be playing rugby (and/or cricket and/or golf). Lower middle to working class kids will be playing soccer (or Gaelic games here in Ireland), and that's where the big money is, comparable to the NFL (though clearly not on the same scale):
"Rugby football stems from the form of the game played at Rugby School, which former pupils then introduced to their universities.
Former Rugby School student Albert Pell is credited with having formed the first "football" team while a student at Cambridge University. Major private schools each used different rules during this early period, with former pupils from Rugby and Eton attempting to carry their preferred rules through to their universities. A significant event in the early development of rugby football was the production of a written set of rules at Rugby School in 1845, followed by the Cambridge Rules that were drawn up in 1848.
...In 1895, there was a major schism within rugby football in England in which numerous clubs from Northern England resigned from the RFU over the issue of reimbursing players for time lost from their workplaces. The split highlighted the social and class divisions in the sport in England. Although the rules of the game were not a factor in the split, the breakaway teams subsequently adopted some rule changes and this became the separate code of "rugby league". The RFU's code thereafter took on the name "rugby union" to differentiate it from rugby league, but both versions of the sport are known simply as "rugby" throughout most of the world."
"In 1895, a schism in Rugby football resulted in the formation of the Northern Rugby Football Union (NRFU).The success of working class northern teams led to some compensating players who otherwise would be on their job and earning income on Saturdays. This led to the RFU reacting to enforce the amateur principle of the sport, preventing "broken time payments" to players who had taken time off work to play rugby. Northern teams typically had more working class players (coal miners, mill workers etc.) who could not afford to play without this compensation, in contrast to affluent southern teams who had other sources of income to sustain the amateur principle. In 1895, a decree by the RFU banning the playing of rugby at grounds where entrance fees were charged led to twenty-two clubs (including Stockport, who negotiated by telephone) meeting at the George Hotel, Huddersfield on 29 August 1895 and forming the "Northern Rugby Football Union". Within fifteen years of that first meeting in Huddersfield, more than 200 RFU clubs had left to join the rugby league.
In 1897, the line-out was abolished and in 1898 professionalism introduced. In 1906, the Northern Union changed its rules, reducing teams from 15 to 13 a side and replacing the ruck formed after tackles with the play-the-ball.
A similar schism to that which occurred in England took place in Sydney, Australia. There, on 8 August 1907 the New South Wales Rugby Football League was founded at Bateman's Hotel in George Street.[19] Rugby league then went on to displace rugby union as the primary football code in New South Wales and Queensland."
I'm sure there's some regional variation in football's popularity at the youth level. But in the South, there are plenty of people of every race playing football. (And even if a larger percentage of white parents keep their kids from playing football, there are still more white kids than black kids in absolute terms.)
What I find more egregious than some geneticists (who probably don't know enough about sports to comment intelligently) blathering on is the media's constant state of despair over the lack of black NFL head coaches, despite the fact that there's usually about 3-4 of them at any given time (out of 32 teams). In other words, black Americans make up roughly the same proportion of NFL head coaches as they do the general population. It's true that there are lots of black NFL players, but being a good player (while perhaps helpful) is neither necessary nor sufficient for being a good NFL coach.
Also, the biracial Patrick Mahomes is a black quarterback, as the media loves to tell us, but the biracial Mike McDaniel is not a black head coach, at least according to some in the media.
The amount of white (and black) parents who push their kids into sports and invest thousands of dollars on camps, clinics, private coaches hoping they can get a scholarship still far outnumbers the amount who pull their kids from football bc of concussion concerns.
True, but the trend line is now pointing downwards for football specifically.
(A) "High school participation in 11-player tackle football was trending upward in the 21st century until the 2009-10 academic year, when enrollment began to wane, according to data from the National Federation of State High School Associations’ High School Athletics Participation Survey....
The latest survey release from the NFHS – the first since 2018-19 – shows that the 2021-22 school year was the first on record with fewer than a million players participating in 11-player high school football in America since the turn of the century. The reported total of 976,886 participants represents a 12.2% decrease from a 2008-09 peak...."
(B) "According to data from statista.com, in 2018, 5.16 million people over the age of six participated in tackle football in the U.S. By comparison, there were 8.4 million people playing in that age range in 2006, representing a 48 percent decline in 12 years.
Participation numbers began declining heavily in the late 2000s, falling under six million in 2011.
The numbers mostly increased between 2012 and 2015 before beginning to drop again, declining almost 19 percent in three years alone between 2016 and 2019...."
And quite often, the exact same people who tell you we need more coaches of whatever ethnicity in order to represent their team's makeup will also tell you we need female coaches. Either not noticing or not caring about that contradiction.
Wow thanks for the info! I wonder if, however, there’s an increase in participation in athletics as a whole? Kids have alternatives these days, obviously video games are what come to mind first. Every high school football team has (or used to have) a cohort of players who never see the field, and I imagine that today those kids would rather say “screw this” and do something else with their time.
There’s also a rise in specialization from a younger age, so multi sport athletes are rare these days. It used to be the norm where star baseball players were also star football and basketball players, for example, but now that’s rarer than it used to be.
>As for culture, I'm wondering about pressure from better off/white parents to keep their kids out of football, even if the parents love watching the sport.
That might explain different racial rates of being a football player in general, but I doubt it would be able to explain different racial rates between specific positions on the team given that you're already a football player.
If football players were selected for purely objective qualities any of that might be relevant, but they aren’t. Once you reach the pro level you’ve been repeatedly passed through the filter of what coaches want and expect; those who are objectively the fastest or tallest go to sports where those qualities alone are sufficient, like track and basketball, but cornerbacks are selected for, e.g., ‘aggressiveness’ and ‘persistence’ and ‘ability to read plays’ which are nebulous qualities.
You don’t even sniff a division 1 roster at corner unless you’re running a 4.4, and 4.3 for the NFL. Corners are usually the best athletes on the team. The nebulous qualities of “ability to read plays” etc that make up ‘football IQ’ are what separates hall of famers from starters, and starters from backups.
CBs are almost always among the fastest players, that's basically a requirement. Look up d green, sanders, etc. So you shouldn't be surprised if whichever population is fastest tends to rise to the top at CB. Also, I'm not sure why we have to pretend that X population, which is obviously both faster and more muscled in certain ways, might not be that way because of genetics. Is it really the case that (e.g.) ecuadoreans could have similar success as NFL CBs with the right culture? No one really believes that, so why can't the flip side be true? Discussion would be much more interesting if people would be more honest about certain facets of life (that basically everyone understands but must nonetheless deny). I get the worry that this type of wrong think can have negative impacts in other areas, but denying reality seems problematic for a number of reasons. And worth pointing out this is yet another reason not to trust scientists (next they'll tell us that certain folks in N. Europe or from certain parts of Africa aren't tall for genetic reasons, even though we know that's a highly (hah!) heritable trait).
I think the criticism of the original comment is more that we would need to see some specific denials that the cornerbacks in particular have any genetic bias. Also as to your point I believe in this comment section we are supposed to provide information instead of saying it can be easily acquired via internet search. In this case, a cornerback is somebody who needs a lot of running, jumping, and aggression.
I think the original comment was fine, as someone who watches a lot of football. I could see it appearing not fine to someone who doesn’t know what a cornerback is, however.
The original poster probably (maybe) lives and breathes in a world that doesn’t involve many humans who don’t know what a cornerback is, if he’s anything like me, and wouldn’t have thought about that issue beforehand.
As Scott has mentioned before, he lives in a world where 95% of the people he knows are not conservative (that percentage and “conservative” wording may not be exactly what he said, but I think the gist is the same). There are folks who live in the opposite world. I think it’s important not to slap those people in the face if they decide to poke their heads out here, especially if they aren’t straight up being insulting or acting insane.
The reaction to that comment is what keeps people like me from wanting to go to eg. ACX meetups. I am far, far different than most people here, but the ven diagram of me vs. most people here has a crossover, which should generally be considered a thing to be cherished. I’m here because I like the way you guys think; I steal it and I think it makes me a better person.
The OP’s post gave context to a problem without being verbose, and began a discussion. Most posts here do the same… just about topics generally well known by the “nerdy” community, and don’t receive any flak.
Did you mean to reply to my comment and discuss the original comment? I was responding to a reply, the "come on Scott you can google what a cornerback is" part, not castigating the original commenter for knowing what cornerbacks are.
It's interesting to note that you confused two different comments with each other and then concluded that people here might not like you because of football. Your supposition about us being prejudiced against you likely influenced your reading of the comments themselves.
I meant to reply to you, specifically the “I think the criticism of the original comment was…” statement. I was then expounding on that train of thought.
Sorry, I didn’t mean to imply I you were (or that anyone was) castigating the OP for talking about football.
I also apologize if I came across as saying you as a group are prejudiced against me! I simply meant that differences in groups can make context hard to find, and make people who otherwise exist in out-groups hard to place.
Again, nothing focused towards you or your comment in particular; just continuing the “criticism agains the OP is [insert something here]” chain.
Disagree about "aggression," that's a safety thing, some of the best CBs ever haven't been particularly aggressive, just faster than everyone else, including reflexes to break on the ball, etc.
My the ebook version of my latest book is free until Friday, the 17th. It is memoir, essays, philosophy, travelog, tips, and photos. Professional mistakes and personal luck. Some of you have already read it; some like it, some find it self-indulgent and shallow. So it goes.
Can anyone link me to really solid pro-permanent-daylight/summer-time argument? Most of the ones I have seen tout benefits of shifting activities to earlier in the day. Which could also be accomplished by just, you know, shifting activities to earlier in the day. This would be disruptive if it happened a couple of times a year, but the change they propose is permanent (it’s right there in the name) so it comes down to a one-time disruption and preserves the benefit of clocks being roughly in sync with the sun (in other words, the reason that we have time zones in the first place).
People live by the clock, so we could either start all offices at 8, end work at 4, push all tv schedules and games and restaurant opening times back an hour etc etc etc - which is a massive co-ordination problem or change the clock. I’m not sure why you think that this problem is one that affects permanent DST either. It affects summer time regardless of permanency.
I'm interested to hear if anyone has any strong arguments against permanent *standard* time -- that way it's just, y'know -- time.
The one I typically hear is "I like to do stuff after work in the summer" -- which makes a certain kind of sense, but the number of jobs in which one can't just start an hour earlier and leave an hour later is rapidly shrinking -- I guess we can try the converse if we get stuck with permanent DST, but for some reason coming to work an hour late is viewed differently from coming in early.
Yes. The argument is that it’s a waste of sunlight, particularly in countries with long hours of sunlight in the summer. A country with 16 hours of sunlight should have a sunset at 8pm.
16 hours is the length of the Parisian day in midsummer.
However, Paris has a sunset of 9:45pm on June 21, which is actually great for tourists and locals. Part of this is because it’s in the wrong timezone (should be GMT) and part is because of DST. So Paris would have to do two things to get to proper standard time, change to GMT and stay in winter time all year. This would move the sunrise back from 05:45 to around quarter to 4, and the sunset to 7:45pm. You can probably see why they won’t do this.
I also noticed during Wimbledon (around July 4th) that the sun was still up in London about that time -- 9:20 on July 4th I now see upon looking it up.
Much of Europe isn’t. Maybe half of countries should be in different times than central time - which goes from the western Atlantic coast of Spain, which is as far west as the west of Ireland, to the polish border with Belarus. Africa is a mess too. No country is GMT+1 and plenty should be ( Iceland, and Western Africa). It’s all a mess.
Again, it's a coordination problem: it's not just about when I personally choose to do things, but when other people are available, when things are open, when events are scheduled, etc.
Since different people have different preferences, the result is likely to continue to be a compromise that doesn't satisfy anyone wholly. But by the same token it's unlikely to be resolved fully one way or the other (permanent standard time or permanent DST) as long as there are competing preferences.
Where's the coordination problem though? Not everyone needs to start work at the same time. And if we are assuming that an hour either way is enough, there will still be at least 7 hours overlap between literally everybody, which seems like enough?
It's not just work. When are stores typically open? When does school start? When does one normally schedule a dinner, or a wedding reception? What time does the evening show start at the movies? When does transit shift to rush hour schedules?
There are a bunch of anchor points on the clock that people generally operate around. People with flexible schedules can work around them to an extent, but as long as they're interacting with others they're going to work within a framework of basic expectations that depend on common time.
The building supply/auto parts type stores in my area are already at least somewhat open by 7 AM, because that's when their customers tend to get up; landfills do have "winter hours" in which they close earlier because it's dark to the point of being a safety issue, head shops open late because stoners like to sleep in; there are many examples and none of it seems to cause problems. Business hours are easily available online, if you want to go to a store you will figure out when it's hours overlap with your non-work time and plan accordingly.
In practice, hours people choose for things are strongly influenced by clock time. As witness that there's "nothing" (aside from what everyone else is doing, which is to say the aforementioned coordination problem) stopping a conscientious objector to DST from keeping their schedule the same year round. Most just complain biennially and adjust, because actually doing that would be more of a hassle than going along is.
Its not an academic argument, but, I live in a jurisdiction which doesn't switch our clocks and it's pretty great not having that annoyance twice a year. We're at 52°N so sunlight changes wildly throughout the year, but changing the clocks wouldn't make any more of it.
I'm not going to do any better than anyone else at providing a link to a really solid argument. But while we're on the topic, I think that part of the reason that so much of the U.S. has persisted in the daylight savings switching ritual is that a) daylight savings time is generally better timing for many activities in the warmer part of the year and b) standard time is generally better timing for the winter. I know that not all agree with this, but bear with me. At mid-northern latitudes, permanent DST would mean that the start of many activities (work, K-12 school, university classes, exercise) would take place in the pitch black for a greater portion of the year. (To those that say - well start stuff later - I agree, and that is exactly what we accomplish when we switch to standard time - we all start stuff later.) I think that permanent DST will prove to be quite unpopular if it is implemented. Most people haven't fully imagined the implications; they just know that they hate changing their clocks. The optimal (IMO) modern solution to this is to continue the switch from DST to standard time, but to do it gradually. For example, if clock time shifted by 2 minutes every day for 30 days, the biological effects of the shifting would be minimal. And then our social activities could still happen on standard time for the winter and DST for the rest of the year. This would be hard for our 'dumb' clocks, but I still think this would be the best outcome. Many of those who want permanent DST need to look up history. It's been tried before and most people seemed to hate it. Are we to simply try it again because we weren't around last time?
I’d be less charitable and say that most people don’t know the sunrise times in early Jan or can’t add 1 hour. Particularly in northern latitudes - sunrise where I live would be 9:45am mid winter. Permanent DST is still better than the alternative, wasting 8 1/2 hours before noon isn’t that useful.
I was around for the last time, and I'd still personally prefer to go to permanent DST. (But I agree that it probably wouldn't survive for much the same reasons as last time.)
I'm personally one of the crazies that thinks that we should just have one global timezone to prevent distance-based time misunderstandings from ever happening in the first place, with the time on the clock places open and close being based on what time corresponds to daylight and night in that particular area. So if the sun rises at around 6pm in one area, that's when stuff starts opening, and if its at 3am in another area, that's when they start their day. Related to your question, this means that, as you say, you wouldn't have to shift around the clock to get more daylight hours at a particular time or whatever, people in the area that open up at 3am could just decide to open up at 2am if they decide that's better instead. For those that think this is insanity, China is all on one timezone despite spanning multiple geographically, and they seem to have figured it out fine.
I don't think that it really matters what the clock shows when you get up, go to work, go to bed etc. as long as it is consistent in one place. When people travel long distances, they need some time to adjust anyway due to time zone differences. Learning that the new place has different working hours is a very small additional effort.
Some people living among us already have a strange working hours. Someone is doing night shifts, another young person is playing videogames whole night and sleeping during the day etc. If I contact such person by whatsapp, I might not even know if he is on another side of the planet or somewhere in the neighbourhood just with a different sleeping schedule.
Not just China, even CET includes areas spanning from UTC-1 to UTC+2 in solar time. So switching the rest of the EU to CET wouldn't make the time zone all that much less coherent.
The problem there is that for the 6 pm people, if they want to talk to the 3 am people, then somebody has to get up earlier/stay up later, and not unnaturally, nobody wants to do that. "It's three in the morning, I should be in bed rather than hanging around to talk to some guy who just started the work day!"
Going back to local time suits everyone in one country as we all see the sun rise at 6 in the morning or whatever, but we're a global world nowadays and Europe has to talk to America has to talk to Asia and so on.
And going back to time by the sun will mean differences in summer and winter - in summer, the day might start at 4 am but not until 10 am in winter, depending on when the sun rises. And if we make "12 noon = when the sun is overhead", then that too will mean shifting times as the earth goes around the sun and the seasons change.
It is easier to negotiate if the world has the same time zone but different local traditions.
If a US person wants to talk to a European, he writes an email: can we talk at 3pm? Right now I have to know what are the time difference and what 3pm in the US means for me. People forget about it all the time and make mistakes.
If we have the same time zone, I would immediately know that 3pm is not good because I will be already sleeping at that time. With time we probably would learn what are general working hours for each continent in the same way as we learn how some shops have unusual opening hours.
You want some people in the world to call midnight noon, and others to rise for the day at 6pm. Learning the hours of each working time zone is more complicated than a simple calculation which is largely done automatically anyway.
My Teams meeting today has been pushed back an hour in UTC because Canada has changed and we haven’t.
You don't need to learn anything, just negotiate time for calling each other. And for me it is easier to understand what is the person's working hours than calculate the time difference constantly.
I am with you. Of course, it will have strong opposition because because don't want to change things they are used to. But eventually it would help us better realize that we are a global community all together on the same planet.
A small problem would be that in some places the date would change during active working hours. Some transportation have daily tickets etc. and it would work well. But we would probably switch to tickets with 24 hour expiration time or something like that.
That’s not just horrible for much of the world, it’s actually more confusing for everybody but server engineers. I’m confused about the starting with the sun idea too - as that would change between seasons.
I'm not sure if China is a good example, as (because of the very uneven population distribution) for the majority of population Beijing time zone is "correct", almost everyone lives in the areas which would be at most 1 hour away from the "correct" time, and it seems that in Xinjiang, which should be 2 hours off, the "local time" is (or at least was) actually being used in practice by the local non-Han population instead of the Beijing time that's used by officials.
If setting a Zoom meeting between Germany, California, and Australia, it's pretty easy to use time zones to figure out their respective business hours. With one global time zones, there'd be nothing for it but to check each individual organization's hours. Most people would have no intuition about what someone else's local time looked like.
I'm also not convinced that an authoritarian country pinning time to the imperial capital is getting a lot of good feedback from its hinterland about how well the one time zone works for them.
With reasonable scheduling software, you can schedule meetings based on people's configured working hours rather than guessing based on their time zones.
I am aware of that. It's a lot more useful for the scheduler to know e.g. that I work from 6 to 10 and from 12 to 16 UTC than to know that my time offset is currently +1.
Is it? I mean that’s a strange schedule and it looks harder to input into software than the software knowing about timezones. It’s also not something helped by your fix.
I have talked to a Chinese co-worker about this (who has lived both in China and in Europe for several years), and he didn't think that the time zone system works better in China. It just comes with different issues to have a single time zone.
I think that usually the permanent-summer-time argument is made in the context of a status quo that has temporary summer time, in which case a solid argument is that the twice-a-year disruption has all kinds of documented harms and should be stopped; and that the choice of which of the *two* currently (and traditionally!) used time zones should stay permanent is relatively arbitrary, there's no solid reason to privilege "winter time zone" over "summer time zone" as the status quo (and all the habits) involve alternating between the two.
And as our daily schedule is currently so removed from sun (namely, that middle of average person's day is definitely not at noon as most people are far more active in afternoon/after-work than in early morning/before-work) that it doesn't make sense to value that zone in being in sync with the sun but rather being in sync with the people's preferences, which currently value sunlight at 7pm far more than sunlight at 5am.
And while this could also be accomplished by "just", you know, shifting activities to earlier in the day, there shouldn't be a "just" in that sentence; shifting all the activities in a community is a really, really hard coordination problem and in comparison switching to a different time zone is so much simpler and easier to achieve; so if people decide that such a shift should get done, a time zone change is probably the best way to implement it.
Basically everyone hates the twice-yearly time change (although I suspect it’s a case of diffuse benefits and concentrated costs skewing people’s opinions), and we already spend most of of the year on summer time, so we may as well pick that one.
And while clock time is pretty arbitrary, it’s arbitrary in a sticky way: maybe 7-Eleven doesn’t need to be open from 7–11, but it would be pretty confusing if the 11-o’clock news came on at 10.
And there’s a pretty strong argument that the various factors in favor of an earlier start to the day outweigh keeping clock time roughly in sync with solar time.
Speaking strictly about the disruption (of changing the time that everything happens by an hour over the course of a single weekend night), I hear a ton of complaining about it and I can't remember ever hearing anyone express unqualified joy at the idea (although some people enjoy the extra hour of sleep in the fall, but typically also lament the earlier clock time when the sun sets).
As I commented elsewhere I think that people are keenly aware of the concentrated pain of the twice-yearly change but take the diffuse benefit ("the sun rises at a reasonable time") for granted, and not having experienced the alternative, grumble about the practice as a whole.
I think "shifting activities to earlier" is not easy because it faces a huge coordination problem. Unless it's done by a global shift as we have it with daylight saving time.
TV channel have their traditional times for anchor shows. Sport events have nationally and internationally balanced schedules. Even locally, for shops or offices, it would be a huge communication effort to communicate that their opening/closing hours change.
Whether daylight saving time is good or bad is up to debate. But shifting activities earlier would happen to a much smaller extent if not coordinated.
All time zone issues are a solution to a coordination problem, which can't be accomplished by individually shifting activities earlier and later. The question is just where to coordinate.
(Permanent DST would be more convenient for me, lots of people favor permanent standard time. I expect we'll continue to wind up about where we are.)
I imagine individual organizations might adopt summer hours and winter hours (and on the whole this might well be worse than the status quo) but I’d prefer this to happen on a baseline of having noon be in the middle of the day.
I do come at this with the privilege of living less than 5 (time) minutes away from a meridian. But “round to the nearest hour, unless there’s a compelling reason to fudge it” seems like a reasonable compromise between fidelity to something like mean solar time and a manageable number of time zones.
Unless you are on a meridian it probably isn’t. The arguments against DST - that it is not natural time - are arguments against time zones, and in fact clocks at all. Because of equation of time clock and solar time drift during the year, even on Greenwich.
How is it to live on a yacht? How much does a yacht suitable for a couple to live on it cost? How do visas and countries borders work for sailing? Can you just cross the borders of countries wherever you want? Are there simpler to acquire visas for people sailing around? How good and reliable is the internet (satellite, I assume) when you're sailing in the middle of an ocean? Are there any blogs of digital nomads who live on a yacht? Let's say I buy a yacht in the Mediterranean sea. Is it possible to get to most countries by sea without crossing any borders on the way except for the borders of the target country?
Don't forget you can also set up on a river-going houseboat, though naturally without the blue-water access that a full on yacht would offer. It's probably worth checking out the costs for both options.
That's a lot of questions! I'll try to work through them. I've spent more than a decade in total living on different yachts starting as a baby, in many different parts of the world, and have done about 75'000 nautical miles in that time. Some for pleasure, some on expedition, some as delivery skipper.
- How is it to live on a yacht?
It's a lifestyle for sure. Or rather a wide range of lifestyles. Many port cities around the world will have yachts permanently moored and used as houseboats, but I gather that's not what you have in mind. On the other end of the spectrum some people spend a large portion of their time at sea, constantly on passage. But most cruisers will spend a few weeks or months to explore an area once you get there, and take the time they need until they feel ready to move on. This is not always possible or practical of course, as there are many factors that can force you to move on, such as the seasons turning, visas running out, or money. But fundamentally it's a matter of personal preference: Some sailors are most at home beating into a gale on the North Sea, others like martinis on a tropical island.
Make sure to fundamentally sync your expectations on these matters with anyone you plan to sail with, as this is a common root cause of conflict onboard.
- How much does a yacht suitable for a couple to live on it cost?
Depends on the size of your boat, you definition of "suitable", your level of comfort, your required level of safety, where in the world you are and plan to sail, and how much of the boat maintenance you are able to do yourself. Each of these categories are important, though not necessarily in order.
I'd say you can get a perfectly seaworthy blue-water capable with a reasonable degree of comfort starting around $10 to $20'000, up to ....infiniti basically, there is really no upper bound. In the $50 to $100'000 range many very nice boats can be found. Remember that the better shape the boat is when you buy it the less it will cost to get up to spec, though I would still always count on having to spend at least 10% or 20% or so of purchase price, even for boats that seem "perfect". This remains true if buying brand new, for most brands. If buying something in less then stellar condition, spending as much on renovation as on purchase is not unreasonable.
I would not buy a new yacht, even if money was no issue (unless perhaps I was commissioning a custom design). This has very little to do with the poor value per dollar, which is indeed often very poor for new yachts, and more to do with the fact that the modern paradigm of yacht design and manufacture is not suitable for real long distance cruising (In my opinion. There are exceptions). They are made to last a handfull of seasons in the med, and will not be designed with good sea-keeping qualities in mind. There are a few exceptions like Hallberg Rassy for example, but they are wildly expensive. In my mind the best yachts for long distance cruising were built in the late seventies to late nineties, though it differs hugely from yard to yard, and designer to designer. Some example boats that I would look at: Hallberg Rassy 36 or 39, Kelly-Petersen 44, Contessa 32. Buy the smallest boat that you can comfortably fit on: Every feet of length will increase the purchase and maintenance cost by 10%, compounded. (super over simplified rule of thumb, obviously).
Do pay a surveyor to check the boat out before purchase, it's definitely worth it.
- How do visas and countries borders work for sailing?
Visas work like normal: For citizens of most western countries, the world is pretty open, and you can get visa-on-arrival for several months, though it varies a lot from country to country. Always check during early planing. The USA is a notable exception, where most nationals must have a physical visa in their passport to be allowed to enter by sea on a private vessel, even if you could get a visa-on-arrival if visiting by airplane.
The basic procedure for switching counties is to clear customs and immigration and clear out of one county in one of their ports-of-entry (most major cities), and then leave within 24 hours. You will typically be asked where you are headed, but your answer is not binding. Then you sail to your destination, which has to be a port-of-entry in the new country, meaning a port that has customs and immigration offices. When you enter the new country's territorial waters, it is customary to hoist the new county's flag as a "curtesy flag", along with a yellow "Q" signal flag underneath the starboard spreader. Q stands for quarantine, originally meaning that you don't have the plague, and are ready to clear in: Now it just signals that you haven't cleared in yet. Until you are ready to clear in you are not allowed to come ashore for any other reason, unless you declare an emergency.
Assume that clearing in takes several hours every time at least, a full day in some countries.
- Can you just cross the borders of countries wherever you want?
You have to clear in at specific ports-of-entry, which is a major constraint in planning. But other than that you are usually allowed to sail into territorial waters as you see fit, as long as you don't go ashore. Territorial waters only extend 12 nautical miles from land. You might be questioned over radio, and some countries are much more restrictive (Like India for example). This is not usually a problem :)
You can always choose to clear out of a country whenever you like, as long as you are somewhere that has Customs and Immigration offices.
- Are there simpler to acquire visas for people sailing around?
No, not really, though see above about visa-on-arrival. When arriving somewhere on an airplane, the airline has done a lot of the hard work in making the bureaucracy smoother, on pain of being responsible to fly you back if you don't meet entry requirements. This is why airlines insist on seeing your visa/esta/eta before letting you board. If you arrive on your own boat, you have to fix all that by yourself. It's usually not a big problem, but it can be tedious, and put artificial constraints on your plans. This whole situation has gotten massively better over the last few decades.
- How good and reliable is the internet (satellite, I assume) when you're sailing in the middle of an ocean?
I've never had the luxury of true satellite internet while on passage, though for most passages over the last ten years or so I've had a satphone with a messaging service that will forward messages to and from email. This will also allow downloading .grib files with up to date weather data. Sending or receiving one text-only email or .grib file might take a minute or two.
Soon we might be able to use Starlink though, that promises to be a fundamental gamechanger, but last I heard it was not possible to use on the ocean yet.
The type of satellite internet connections that for example large cruise ships use are prohibitively expensive, are fragile, and use a lot of electricity to boot.
- Let's say I buy a yacht in the Mediterranean sea. Is it possible to get to most countries by sea without crossing any borders on the way except for the borders of the target country?
Yes, almost always. Territorial waters only extend 12 nautical miles out from the coast, beyond that is international waters. (Don't confuse territorial waters with the EEZ, Exclusive Economic Zone, that extends out to 200 NM. The EEZ only effect fishing and oil-drilling rights and such, and has no bearing on private yachts). Even aside from that, most countries don't mind if you sail across their territorial water without going ashore, though it varies from county to county.
- Are there any blogs of digital nomads who live on a yacht?
I'm sure there are! I don't know any of the top of my head, but here are some other links and book tips:
Anything by Lin and Larry Pardey, especially "The Self Sufficient Sailor", ISBN 0-9646036-7-5. They embody the attitude of getting a small affordable boat, and casting off soon. Without compromising safety.
Anything by Steve and Linda Dashew, especially "A practical guide to seamanship". They embody much of the opposite of the Pardeys, and their viewpoints complement each other well.
"The Voyager’s Handbook: The Essential Guide to Bluewater Cruising" By Beth A. Leonard, ISBN 0-07-143765-7. There are countless book purporting to teach everything you need to know to sail around the world. This is the best one I know.
"World Voyage Planner", by Jimmy Cornell, ISBN 978-0955639654. A list of destinations and inspiration and how to get there.
"World Cruising Routes", 7th ed, by Jimmy Cornell. The best reference for when and how to attempt which ocean crossings. Very information dense, but very dry reading.
The "On the wind podcast" by 59 North Sailing. Many, many interesting guests over the years, and a long backlog to choose from.
https://www.youtube.com/@SailingUma -Sailing Uma, a youtube channel by a young couple sailing the world in a small boat. Very positive energy, and very down to earth, realistic picture of the pros and cons of the sailing lifestyle. Several episodes focus on their finances in depth.
https://johnkretschmersailing.com/ I especially liked Johns book "Sailing a serious ocean", which is a collection of stories from his years at sea.
Feel free to email me at johan.domeij@gmail.com for more questions. I'd love to help out, or just talk boats and sailing in general :)
Yeah, there's a bunch of sayings to that effect: "The definition of a "boat", is a hole in the water into which you pour money", "Owning a boat will give you the two happiest days of your life: The day you buy it, and the day you sell it", and so on. Sometimes it certainly feels that way! In general I'd classify all of them as expressions of the fact that living on a boat is a complete lifestyle, and thus really will cost everything you've got, in the sense that it becomes the basis of your life, and informs every other decision. Maintaining a fully kited out and capable ocean cruiser *as well as* a normal life on land, is much like having two houses, two jobs, etc. Having a smaller simpler day cruiser is a completely different matter however, and a great way to enjoy a weekend :)
Under UNCLOS you have the "right of innocent passage"; you can transit through nations' national waters without express diplomatic clearance, provided you're not doing anything harmful (like spying on them, smuggling drugs, etc.). Of course, there are certainly places in the world where I wouldn't be too keen to exercise that right - but in most of the world you'd be perfectly fine.
Internet at sea is as good as terrestrial ADSL (though presumably hundreds of times more expensive..) when you're on a vessel with a microwave dish that tracks the satellite, but if the ship is moving beyond the dish's ability to track the satellite (either through multiple successive alterations of course or too much pitch and roll in a heavy sea) then you can lose signal and need to re-train onto the satellite. Whether such dishes and tracking arrangements are available on smaller pleasure craft, though, I've no idea. [Edit: you can usually get 3G or 4G phone signal out to a distance of about 10-15 miles from an inhabited coastline, which would cover all your berthing and anchoring and a fair amount of your passages]
Everything about the maritime world is expensive beyond belief - almost beyond sanity. Not just the upkeep and husbandry of a vessel but also compliance with international maritime law (things like pollution regulations), with the laws of your flag state (the nation whose flag you sail under), and the local regulations of ports you put into. Then there's insurance, constant maintenance and upkeep costs (you'd be surprised how many ordinary-looking, perfectly inert objects have a legally-obligatory expiration date on them when they're required to be carried by maritime law), internet, paying ports for fuel or to dispose of your garbage, berthing fees....
As for actually navigating (assuming you're not going to hire any Deck Officers) you'll need to do some level of training to plan passages and navigate safely in accordance with the COLREGS. In addition to mostly-navigation-focused qualifications (such as the RYA Yachtmaster) it would be wise to have training in basic fire-fighting, first-aid, GMDSS radio operation, etc. - when you're on passage deep-sea you can't just telephone the fire brigade, doctor, dentist, etc. etc. (You don't *need* qualifications in these things if you're a smallish pleasure craft - but personally I'd be very nervous about putting to sea without them!)
Lastly ..and I'm sorry if this sounds too discouraging, but.. a life at sea is, well, just generally not very nice, lots of the time! There's so much that's difficult or uncomfortable, so many hurdles and hoops to jump through, and so much work to do to maintain even a small vessel. I imagine that maybe living aboard a largeish yacht berthed in Monaco or somewhere and reclining on the upper decks whilst Deckhands clean and maintain it for you, and just throwing money at every port official who shows up on the jetty until they were placated - possibly breaking up the monotony with the occasional passage to St Tropez and back - might actually be quite a nice way to live for a few years - but most of the people I work with (plus self), though we might find our careers fulfilling in some general sense, don't really enjoy actually being at sea...
> Everything about the maritime world is expensive beyond belief - almost beyond sanity.
Yes, I've heard that sailing is like standing in a cold shower tearing up $100 bills! And they say the happiest day of your life is when you buy your first yacht, and the second happiest is when you sell it. But from my limited experience (mackerel fishing in Swanage Bay), yachting is a lot of fun.
Mind you, for ruinous expense and constant worry, yachting is only a small step up from keeping horses.
If you're asking broad questions about how much it might cost to live on a yacht, you may not have heard the famous and highly instructive joke on the topic. How can you tell if you are ready to own a boat? Go to the bank and take out five grand in hundred dollar bills and start flushing them down the toilet by the handful. If you can get through all 5000 without breaking down in tears, you're ready.
The annoying thing about boats is the fact that they cost you large amounts of money whether you're using them or not. But if you are actually using them, every day, instead of a house, then the costs start to look less obscene.
Generally you handle borders/visas/customs/immigration at port, when you land - you have to report arrival, and wait until the border checks are complete until you can disembark. You still need to follow the visa process, you can be denied entry, etc.
A question for the historians of religion out there: what is the best secular explanation for why Paul converted to Christianity?
It's widely accepted that (1) Jesus Christ existed and founded a community of believers, (2) Paul harshly persecuted the early Christian church, and (3) Paul later converted to Christianity and became its leading proponent, and endured great hardship in the course of his work.
The biblical explanation for this unlikely sequence of events is that Paul had a vision of Jesus while traveling to Damascus.
Are there any good secular explanations? Obviously you can hand-wave this all away by saying that people sometimes have visions that cause them to radically alter their behavior, and this does not inherently demonstrate that anything supernatural is going on.
Nevertheless, the story of Paul does strike me as pretty startling and difficult to explain without recourse to God. Is there any other example in history where an individual aggressively persecuted a religion, then converted to that religion at great personal cost?
It could just be that he had Christians pegged a certain way in his head as not really people, and when he had an experience that might have actually made him sympathetic to them briefly it completely shook his worldview and caused him to re-examine things - and that re-examination happened to land him in switching sides.
That happens sometimes, like when you see someone completely switch partisan affiliation.
In one of his letters Paul mentioned "bearing the marks of the crucifixion". That could have been figurative of course, but it has been suggested he was none other than the leader of the group of Sanhedrin priests and soldiers sent to arrest Jesus on the night of the last supper, the one who the disciple Simon Peter injured on the ear with a sword before Jesus told him to stop. Who knows, perhaps being clobbered on the head with the flat of a sword by a big guy like Peter (meaning "Rocky") concussed him and is what caused his epilepsy!
Something else that may be relevant is that although Saul (his pre-conversion name) was Jewish, he was also a Roman citizen, a distinction inherited from his father. So maybe he had long been conflicted against what he may have seen as the parochial preoccupations of the Temple priests and Jewish people generally, especially their opposition to Roman rule, and saw in Jesus's teachings a broader perspective. After all, it was St Paul who primarily spread the Christian gospel beyond Israel. The disciples and their immediate followers never would have, nor even wanted to. In fact, some of them fell out with Paul for that reason.
Paul belongs to a group (the Pharisees) which was deeply at odds with the Temple aristocracy (the Sadducees), and whose spirituality was in fact relatively close to Jesus' in many ways.
You are rather kind to him when you say he "endured great hardship in the course of his work." My reading as a non Christian who has had a thoroughly Christian education is that he was good at persecuting people. He didn't stop after the road to Damascus thing, he just got redirected. You wouldn't know from the gospels that God hates gays, but it is very clear from Romans 1 that he does. It's a shame that he was such a fabulous writer. in fact he very precisely condemns himself: he speaks with the voice of men and angels but has no charity.
I don't know that it takes any kindness to acknowledge that Paul faced hardship. Just off the big events, he was shipwrecked, imprisoned three times, and executed. I'm also not sure it's fair to describe his post-conversion career as one of persecution. Writing sternly worded letters seems kind of banal in comparison with stoning people to death.
He was either in a position of power, or believed he was, judging by his writings. As an early Christian gay I would have been in no doubt I was being persecuted after reading Romans.
I don't understand your point. There are currently gay people who would like to be Christian who in many churches run into the problem that they are not tolerated, because of things like Paul's letter to the Romans, 1:
26 Because of this, God gave them over to shameful lusts. Even their women exchanged natural sexual relations for unnatural ones. 27 In the same way the men also abandoned natural relations with women and were inflamed with lust for one another. Men committed shameful acts with other men, and received in themselves the due penalty for their error.
Gay people do not say Hell, I would like to be Christian but this rules it out, maybe try Hinduism instead? They really don't. They want to be gay, and Christian, and the Anglican church in England spends most of its time trying to think of ways to accommodate them. You seem to be saying that things were fundamentally different in the first century and they would all turn to mithraism or something. I don't see any reason to think that.
And there are churches that have torn up the Bible to accommodate all the LGBT+ people who would be Christian except for...., so they made the changes in anticipation of the floods of new converts, and then afterwards found the pews every bit as empty as before, because the activists were not interested in becoming Christians, they just wanted the new orthodoxy accepted everywhere.
This is why I don't accept "I would love to be Christian, if only you would change your beliefs to suit me". Now you have your choice of churches that will accommodate you and tell you it's no sin. They'll even declare you a saint:
"Murray is celebrated on July 1 in The Episcopal Church’s “Holy Women, Holy Men” calendar of saints, and an increasing number Episcopal leaders – especially Black and LGBTQ+ people – cite her as an influence. Trinity Church Wall Street and the Diocese of North Carolina are supporting partners of the Pauli Murray Center for History and Social Justice at Murray’s childhood home in Durham, North Carolina."
In those days homosexuality was sometimes associated with religion, in that there were pagan religions whose male priests were gay and/or financed their temple and lifestyle with sexual services. One such example was the Galli, members of the cult of the goddess Cybele. So it was primarily the pagan religion and idolatry being condemned in the Bible, and the gay sex criticised only incidently by association.
Your faith is a beautiful thing, but 3 out of your 3 posts betray a complete failure of very basic logic. Therefore you are reinforcing the belief (not my belief incidentally) that Christianity and logic are incompatible. So you need to have a good long think about whether you are doing the lord's work by continuing to post.
I think you have probably overestimated how clearly you articulated your point, and as such are being unreasonably harsh to Gamereg for not getting it. (Illusion of transparency strikes again!)
As far as I can see, you never gave ANY explicit explanation of how your criticism of Paul is meant to relate to the top-level comment that it is replying to. You seemed to imply that Paul persecuting others somehow undermines the claim that he endured hardship, although I don't see why those two claims would be at odds, and you haven't said.
If you want your reader to guess at some implied argument that you haven't spelled out, then it doesn't make sense to complain when their guess is "not what you said". The _correct_ guess would _also_ be "not what you said".
Sure, a deserved rebuke. I just don't see an issue here; he changed his mind and either he thought he wouldn't be persecuted or he thought he would but either didn't mind or he positively welcomed the prospect. Martyrdom was all the rage at the time. I don't personally respond to an uptick in modern day suicide bombings by saying Gosh, perhaps there is something in the creed for which these guys are doing this, I must investigate it more thoroughly.
There are Jewish sources that say Paul was a plant by the Rabbis, sent in order to distinguish Christianity from Judaism. Early Christianity was too similar to Rabbinic Judaism, and the Rabbis thought that people were getting confused, or perhaps that the Christians would persecute the Jews less if the religions had some breathing space.
Source: Toldos Yeshu says this about Peter, and other Midrashic sources say similar things about Paul.
Maybe, but not Judaism in general - outside of revolts. I’ve heard that 10% of the empire were Jewish or adjacent. By and large you could be Jewish in Rome, but not Christian.
I've heard some secular people say temporal lobe seizure (weird type of seizure that tends to cause religious-seeming experiences), and I think it would be fun to consider Paul as a case study of Geschwind Syndrome - but this seems a little too cute. Former Bay Area rationalist @webdevmason, who used to organize what I think was the biggest New Atheist convention, converted to Mormonism earlier this week. I don't claim to exactly understand her narrative arc but I don't feel the need to bring in temporal lobe seizures. Sometimes people change their minds for confusing reasons.
Paul's conversion was more interesting insofar as it happened in a single moment and he saw a vision, but we just have Paul's word for that, and if you're an atheist you're already suspicious of the Bible (I agree that Paul describing his own experience is more credible than someone thirdhand describing Adam and Eve, but lots of preachers have been caught in lies about their own lives before, so I don't want to 100% believe)
C.S. Lewis wrote a wonderful book-length account of being seized by the conviction that god loves us, and the Catholicism is The Truth. It's called Surprised by Joy.
Lewis was an Anglican, not a Catholic. The book is good, though I would not have described it the way you did. More like "compelled by logic to believe that God exists", though his later conversion from Theism to Christianity in particular might be described as being seized by a conviction.
I've heard about enough instances of right-wing anti-Islamic politicians converting to Islam that my best explanation is that major world religions are convincing enough that anyone engaging with them deeply, even adversarially, is at risk of being converted by their sheer memetic potency. This is how they became major world religions, after all.
Not so surprising with right-wing politicians like that - arguably they already had the mindset and just switched teams.
I feel like it's a temperament thing in some ways. If you're emotionally vulnerable, feeling isolated personally, and have a very strong sense of Needing Things to Be Tied Down And Make Sense, then you're probably going to be attracted to religion eventually. What religion it specifically is depends on circumstance and timing.
I’m going to give half-assed paraphrases of two apropos quotes.
“He who wrestles demons best take care, lest he becomes a demon.” Not sure where that’s from. Bible? Saturday morning cartoon? A role-playing game?
The second is Nietzche, and i’m sure everyone here knows it. “He who gazes into the abyss too long...” something something something “...abyss gazes also into him.”
People tend to make both valid, logical arguments against religions they don’t like. But they also make invalid, illogical arguments against those religions. To stereotype, A strong believer in a competing religion x make a lot of arguments against religion y that are based in beliefs about his religion x. If he has a crisis of faith in religion x, then he might see that all the arguments they made against y are invalid. That makes it easy to decide religion is actually true.
Here’s an example from real life. My family moved from the Midwest to Long Island when my sister was 15 or so. She constantly (continuously?) mocked the “Long Guyland” accent. Another word for mocking an accent? Practicing. She developed the accent over the three years we lived there. Poetic justice!
The first is also Nietzsche. And he's not just repeating himself in different books. He puts the two together in BGE 146 (Ch IV). The first is a warm-up to the weirder second line about the abyss. I guess the first could be a reference to an earlier source, but I find nothing.
Wrestle demons evokes Ephesians 6 or maybe Jacob wrestling the angel in Genesis 32, but Nietzsche say fights monsters (Ungeheuer).
Seems false to me, the whole New Atheism thing was millions of impressionable young teenagers engaging with religion adversarially, and only a tiny handful of them ended up converting.
I wonder if there's a similar record with political ideologies - I had something like this with libertarianism, although only in a weak sense.
Well, there was the time the government post-WWI Germany decided to send an agent to infiltrate and report on what they thought might be a dangerous group of political extremists. That did not go the way they expected - https://en.wikipedia.org/wiki/Adolf_Hitler#Entry_into_politics
I think if an ideology becomes a movement worth infiltrating or persecuting, then almost by definition it has to have become a community with something substantial to offer its members. And if you decide to send an agent deep undercover or far abroad to deal with the perceived threat, you're probably selecting for people who don't have deep ties to your own community because the ones who do have such ties will be reluctant to take the job. So some risk of your agent "going native" comes with the territory.
If it's just hanging around with your friends on the internet sniping at the opposition, yeah, those people aren't going anywhere because of the "hanging around with your (existing) friends" part.
I agree that mere conversion wouldn't necessarily require an explanation, but it seems to me the salient fact is that Paul was persecuted and eventually killed for his Christian faith — and that means the story as a whole *does* need some explaining.
But I'm willing to be persuaded otherwise, especially if there are good historical analogues where individuals underwent voluntary conversion and were subsequently persecuted for it.
I don't really understand your point. It is a perfectly rational secular thing to say that he converted to Christianity because he was convinced by the arguments in favour of it, or to say that he had an epileptic fit which converted him as it apparently often does. Why does his eventual martyrdom require explanation? He couldn't predict the future any more than anyone else can. Sure, explicitly converting at that date raised the odds of coming to a premature end, but 1. Plenty of people drive fast on drugs while smoking cigarettes and think, quite often correctly, that they will get away with it and 2. In those days it was rational to think as a Christian that you really were literally immortal anyway because the second coming was just round the corner. So what is there here that is harder for the secular to explain, than for the godly?
Not sure how Paul would've thought he'd "get away with it", since he had literally been witness to at least one stoning, and testified against others who where stoned. He of all people knew that changing course would not be pleasant for him in this life. And Christians may have hoped the Second Coming would come in their lifetimes, but those who paid attention to what Jesus said about it ("no man knoweth") would've known better than to count on it.
>especially if there are good historical analogues where individuals underwent voluntary conversion and were subsequently persecuted for it
I mean, there are boatloads of these ("fisher of men" pun very much intended), considering that most of the early Christian martyrs were converts who were eventually killed for their new faith. I think the interesting thing about Paul the OP was curious about is the fact that he wasn't just a convert who underwent persecution and maintained his faith, but a former persecutor of Christians himself. As for that specific case, I can't really think of any off the top of my head.
You're right, I was more precise in my original post.
"Is there any other example in history where an individual aggressively persecuted a religion, then converted to that religion at great personal cost?"
If you include sects within a religion, I suppose one example would be King Henry VIII no less.
In his youth, he defended the Catholic faith, and even wrote a book called "Defense of the Seven Sacraments", basically opposing the newly emerging Protestantism. For this, the Pope awarded him the title "Defender of the Faith", which appears on British coins to this day as "FD".
But later, after the Pope refused to grant his wish to divorce Catherine of Aragon, he separated the UK from the Catholic Church. Thereafter the Anglican church, and thus to a large extent the King himself, became in effect Protestant.
As for the "great personal cost", although the king was enriched by the sale of monastic assets, the break from Rome caused him a lot of worrying upheaval and resistance to the changes.
The account in Acts of Paul's life diverges sufficiently from Paul's own account that it seems likely the writer of Acts had some non-Pauline source. But this may not be particularly relevant, since precisely because of the contradictions with Paul's own account, the non-Pauline source is probably unreliable.
One thing that is potentially relevant is that Galatians 1 presupposes that the audience will be familiar with at least some details of Paul's biography ("for you have heard"), and it wouldn't make sense for Paul to fake that. He wouldn't the first or last person to overestimate his own fame, but it seems clear that he at least thought of himself as having been a notorious persecutor of Christians.
Paul does slightly yada-yada over a 3 year gap between his conversion and start of his missionary work (which Acts 9 omits entirely).
I believe the three facts I asserted are widely accepted by historians, both secular and religious. I don't know exactly which sources those historians rely on.
I can tell you that the primary biblical source for information on Paul's life is Acts, which was written by Luke, not Paul.
I think the vision is a perfectly good explanation. As you point out, you don’t need to appeal to the supernatural. And visions would have been much more common in a culture attuned to those kind of experiences.
The Varieties of Religious Experience has a bunch of great examples of this kind of vision, as well as some fantastic commentary on how to fit them into a scientific world view.
I think the search for an explanation is to ask why he would have a vision with that particular content in the first place. A secular or perhaps just psychoanalytic answer would therefore be something like his desire to show how zealous he was could not be constrained within being a member of a dominant religion. Rather only within a persecuted minority would his sense of zeal find an appropriate expression ending predictably with his death as a martyr.
Or it could have been repressed guilt for his persecutions finding expression, causing him to do a complete 180 out of some unconscious desire to right his past wrongs. Obviously this is pure speculation, but if his pursuit of early Christians led to people being put to death, you can see how that might nag at even the most hardened zealot's conscience.
I see a lot of people persecuting the group they identify most strongly with. It's possible he persecuted the Christians precisely because he found them so convincing, and thus saw a threat in them.
I imagine that part of it is that you have to be looking to join a group that's willing to let you in despite very recently having actively been engaged in killing them.
"why would early Christians have continued to be Christians if they were persecuted and it wasn't true?" is mostly an answer to the "Christians faked the resurrection to manipulate dumb people" crowd.
> Seriously tho: many homophobes later come out as gay, often when they still live in an environment where being a homophobe is acceptable but being gay is not
I wonder if the opposite is true. In environments where being gay is acceptable but being homophobic isn't, are many gays secretly just homophobes trying to hide it?
13 % on Ukrainian victory (unchanged from March 7, down from 17 % on January 30).
I define Ukrainian victory as either a) Ukrainian government gaining control of the territory it had not controlled before February 24 without losing any similarly important territory and without conceding that it will stop its attempts to join EU or NATO, b) Ukrainian government getting official ok from Russia to join EU or NATO without conceding any territory and without losing de facto control of any territory it had controlled before February 24 of 2022, or c) return to exact prewar status quo ante.
43 % on compromise solution that both sides might plausibly claim as a victory (down from 45 % on March 7, and down from 47 % on January 30).
44 % on Ukrainian defeat (up from 42 % on March 7, and up from 36 % on January 30).
I define Ukrainian defeat as Russia getting what it wants from Ukraine without giving any substantial concessions. Russia wants either a) Ukraine to stop claiming at least some of the territories that were before war claimed by Ukraine but de facto controlled by Russia or its proxies, or b) Russia or its proxies (old or new) to get more Ukrainian territory, de facto recognized by Ukraine in something resembling Minsk ceasefire(s)* or c) some form of guarantee that Ukraine will became neutral, which includes but is not limited to Ukraine not joining NATO. E.g. if Ukraine agrees to stay out of NATO without any other concessions to Russia, but gets mutual defense treaty with Poland and Turkey, that does NOT count as Ukrainian defeat.
Discussion:
This update, just like previous emergency one, is based on various credible reports which started to float in information space on March 7, in places like NYT and German ARD, that blowing up Nord Stream pipelines had pro-Ukrainian motive. So far, no links to Ukrainian government have been found, but realistically, this massively increases the probability that such links will be found in the future. Probably not in a sense that Zelensky ordered the pipelines to be blown up, more like Ukrainian government supporting some shady operators to generally blow Russian stuff up and those agents being very dumb.
It is difficult to predict what impact such revelation would have on a Western support to Ukraine, but surely it would not be zero.
*Minsk ceasefire or ceasefires (first agreement did not work, it was amended by second and since then it worked somewhat better) constituted, among other things, de facto recognition by Ukraine that Russia and its proxies will control some territory claimed by Ukraine for some time. In exchange Russia stopped trying to conquer more Ukrainian territory. Until February 24 of 2022, that is.
I don't think the sabotage story change much tbh. At least from the French, Polish or American perspective. Maybe the German would be a annoyed but that's it.
It seems that you keep wildly underestimating Ukrainian chances.
Did you update back on all the previous assumptions you made (Russia is going to try to keep Kherson, mobilization is going to give Russia an edge, etc), which turned out to be wrong?
After a failure of both Russian infrastructure strikes strategy and recent Russian offensive attempts, I think there are even more reasons to update in favour of Ukraine.
I added 5 percentage points to chances of Ukrainian victory and deducted the same from Ukrainian defeat after Kherson evacuation.
Regarding Russian mobilization, it seems to me that, with admittedly big exception of that evacuation, it is working basically as intended, at least so far. Fears of political turmoil in Russia turned out to be greatly exaggerated, and mobilization succeeded in stabilizing the front east of the Dnipro. Russians are now even able to go on the offensive again, although as you note they are not very successful at it. Maybe Russian front will soon crumble despite of mobilization, but I wouldn't bet on it.
I think Westerners think about this war something like if Ukraine is not actively winning, it is therefore losing. Not saying this is necessarily 100% wrong but an absence of big news seems to work against Ukraine in public perception, which I'm sure they are aware of.
Like a lot of neutral to pro-Russian observers, I think you mistake the reason for public support (at least in the United States) for Ukraine. It has nothing at all to do with the ethics of the Ukrainian government, about which we have had a fairly low opinion already. Support is based on (1) the incredibly barbaric behavior of Russians on Ukrainian soil, and (2) the courage and resolve demonstrated by the Ukrainian Army.
That's it. Nothing else really matters (except for the cost of support, of course). So even if Zelensky personally scuba dived down to the Nordstream and drilled a hole through it by hand, my read is that almost every supporter in the US would shrug and say "Yeah? And...?"
I don't have any good read on how it would affect public opinion in Europe, of course. My vague feelings is that the Brits, Baltics, and Poles couldn't care less, because they're in it because they hate the Russians. The Germans are another story, especially because it affects them directly, but the Germans have been pretty lukewarm in their support already. How much that matters militarily I don't know, but it certainly matters less than US public support, which as I've said I don't think even the most outrageous possible outcome of the pipeline explosion investigation could change.
?So even if Zelensky personally scuba dived down to the Nordstream and drilled a hole through it by hand, my read is that almost every supporter in the US would shrug and say "Yeah? And...?"
Nit: I think the charisma and political savvy of Volodymyr Zelenskyy counts as a third substantial reason for US public and political support for Ukraine, so if he were personally involved or even known to have signed off on that operation, it probably would have some impact.
A hypothetical independent group of Ukrainian patriots abroad, probably not much impact.
Maybe. I think reasonable men can differ on the absolute level of change. My belief is only that even if it caused some dismay at the more moralizing margins, it would have an effect on support too small to impel nontrivial change in policy. Previous "revelations" and arguments about Ukrainian corruption and mis-governance have had zero effect, after all.
As I said, I don't think people support Ukraine because they like the Ukrainian *government*. Not even Zelensky personally. While I agree his leadership is critical within Ukraine, and he has done a masterful job in coordinating Ukraine's PR offensive -- truly brilliant -- I don't think he's Winston Churchill reborn leading a War Cabinet in which people have deep faith. I notice he has been pretty careful to present his people as the true heroes, and be fairly self-effacing for a wartime leader. I think he understands that it's his people that are inspiring, not his government.
There's also the additional fact that I can't see the American supporters of the war getting *that* worked up about the damage to the pipeline. The direct effect would appear to be to prevent Russia from selling gas to Germany, and (1) I think American supporters are very OK with Russia not being able to sell gas, and (2) they've never been very OK with Germany continuing to buy gas from Russia -- it seems a bit like trading with the enemy.
People do understand intellectually there's reasons for it, but it still kind of stinks, and I can't see people getting too worked up if -- as it might be put -- the Ukrainians had the balls to do what the Germans themselves lacked the courage to do: just stop sending German euros to pay for Russian shells blowing up Ukrainian schoolchildren. That's how it could be put, and I think that argument would do a lot to blunt any feeling that the Ukrainians had gone too far.
The Americans that oppose Ukraine are anti-Biden…China and America have benefited most from this senseless invasion as America is now energy dominant which was a Trump priority. China benefits because Putin is now married to China as he is a pariah in the West.
I oppose the hegemony of the military-industrial complex and the cozy relationship between them and seemingly ALL flavors of politicians who always seem to be able to find a “Just Cause” to send arms or men to fight, profiting handsomely indeed from our “not being in a war.”
I’m unconvinced that there’s a “Good Guy” in that conflict. I’m not convinced we have a national interest there, certainly not one where we write blank checks to Boeing and Lockheed-Martin.
We have a national interest in Russia not having an army capable of invading e.g. Estonia, because we're on the hook to defeat that army if it happens. That being the case, we have a national interest in the Russian army suddenly having all its tanks blow up, etc, and oh look, here's a way we can arrange for that happen at a bargain price and without any American soldiers getting killed.
I don’t support sending Americans to fight, right now it’s essentially free to send military equipment as we have benefited with LNG exports and military exports create good jobs without contributing to inflation.
No, that's not the case universally. For example, I am not a supporter of the Biden Administration -- I will certainly oppose his re-election -- but I support American assistance to Ukraine. There are plenty of people who are anti-Biden but pro-assistance, and others who are pro-Biden but anti-assistance.
I wouldn't say American and China have benefited from the war so much as Russia has suffered by comparison. Indeed, I can think of no outcome whatsoever that does not leave Russia much worse off than she was in 2014. Had NATO actually intended to destroy Russia as a First World participant, they could have done no better than secretly engineer this war. If Putin were actually in the pay of the CIA, he would be eligible for a very handsome bonus, as having exceeded all reasonable expectations.
"The Americans that oppose Ukraine are anti-Biden" is not the same as "The Americans who support Ukraine are pro-Biden"; your existence only refutes the latter.
Right now there are politically significant blocs of pro-Ukraine/pro-Biden (e.g. pretty much all Democrats), pro-Ukraine/anti-Biden (e.g. McConnell) and anti-Ukraine/anti-Biden (e.g. MTG, and it wouldn't surprise me if some of the very far left fitted here too) Americans.
But I agree with Pangolin Chow Mein that there are very few Americans who like Biden in general but don't want the US to support Ukraine.
Maybe not, since it would be kind of logically incoherent to like Biden and dislike his major foreign policy effort. If you thought US assistance to Ukraine was stupid, it's kind of hard to like Biden, its chief architect.
But there are without doubt big chunks of the anti-Biden crowed who are strongly pro-assistance, and other big chunks who are just as strongly anti-assistance. And I daresay there are similar divisions in the meh-Biden crowd.
Which probably covers most Americans, since I think there aren't many people who really *like* Joe Biden. Never have been.
> I can think of no outcome whatsoever that does not leave Russia much worse off than she was in 2014
If Russia hangs onto southern and south east Ukraine, then they will have firstly their long desired land bridge to Crimea, but also ownership of the prime Ukrainian wheat growing areas and iron ore deposits.
I suspect China and India are secretly rooting for Russia, solely to have cheap and reliable access to the latter. The ability to provide them may even have been a or the main incentive for Putin to invade in the first place.
Sure, and with a land bridge to Crimea and $3, you can get a Starbucks in the Metro -- oh wait, no you can't any more. Just like you can't get first-class chips or electronics, just like you can no longer get Exxon to lend you expertise to exploit your oil reserves, just like nobody will sell you parts to keep your airliners flying any more. At least the mighty Russian Army will have proven once again it is a behemoth to be feared, a steamroller that would flatten any challenger immediately! Oh no, hang on...no, that's not true either.
Getting a land bridge to Crimea and the shattered shards of the Azov works would be a victory of epically Pyrrhic proportions.
I doubt either China or India is rooting for Russia to win. But China in particular is quite happy to see Russia immolate herself. There's never been any love lost between the two, and I believe China would be quite happy to take her place as the only remaining serious challenger to American world power. But equally China would like a Russian defeat (or resumed stalemat) to be as costly as possible to the Americans. So I expect them on the whole to be in favor of war, as long as possible, and as costly as possible for both sides. (I expect privately they would say, as Kissinger said about the Iran-Iraq war, "it's a shame they can't both lose.")
And nobody will ever buy the Irkut MC-21…a goal of Putin’s to diversify the Russian economy…and the COMAC C929 is just a ruse to string Russia along. George W Bush thought he would get the best of China…it cost us millions of manufacturing jobs and a lost decade when boomers were at their peak productivity.
What are your odds that the pipeline explosion was done/ordered/precipitated by the actions of Russia/Ukraine/US/... respectively? My understanding is that it was a fairly sophisticated operation that required resources very few players could muster. The US and Russia, some European countries, but very unlikely for Ukraine or some "shady operators".
I am no expert on blowing up undersea pipelines, but investigators evidently think that it wasn't that sophisticated operation. And I believe them. Like, why would they made that up? Far more likely is that you are overestimating its difficulty.
Before recent revelations, I would guess over 50 % probability of Russian involvement, now I would guess roughly 80 % on pro-Ukrainian group (without making any judgment how much is it connected to Kyiv government), 35 % on Russian involvement and 10 % on something weird and uncategorizable. And yes, those percentages don't add up, because I think that it is not outside the realm of possibility that it was the work of Russian agents succesfully infiltrating pro-Ukrainian groups.
Of course 0 % on Biden administration involvement. C'mon, this is right next to 9/11 trutherism in plausibility. If Trump would still be in the White House, I could very easily imagine that he would order something like that, but I can't imagine that it would not be immediately leaked to the public.
If they would be engaged in a cover up, they would obviously blame Putin. And people would believe them, including me. Destroying Nord Stream and then framing Ukrainians for it just doesn't make any sense.
Why would Russia want to damage a pipeline when they can simply turn off the taps at their end, although they may have gained somewhat from higher gas prices.
Whoever it was, the aim was obviously to prevent certain EU countries wobbling on sanctions by continuing to purchase Russian gas, on which they (especially Germany) have foolishly come to rely on.
The one I find the most interesting (not necessarily the most plausible) is that it might be done by Russian forces even though it does not benefit Russia per se, because *Putin* benefits by reducing the attractiveness of a coup. (Getting rid of Putin no longer lets them restart the flow of gas and bring everything back to normal.)
These sounds like galaxy brain takes of someone who has primed on the idea that Russia did it and can't let go.
The idea that there was likely to be a coup in a cult of personality authoritarian regime such as Russia is wrong in the first place. The idea that the destruction of the pipe would decrease the probability of such coup doesn't hold much ground either.
If there was a group inside of Russian elites, able to kill Putin, take over the country and eager to end the war and continue business as usual, they could've done it with or without the pipe. The existence or absence of the NS isn't the main reason why business has ceased.
> These sounds like galaxy brain takes of someone who has primed on the idea that Russia did it and can't let go.
One or two replies and comments may have been somewhat ax-grindy. But none stood out as such to me, and taken as a whole the threads seemed to give a pretty rounded set of pros and cons in assuming Russia was behind it, and these are more surprising and diverse than I assumed.
Interesting. So what would be the takes for the most likely reasons for Russia to do it from someone who believes that there's a very low probability that Russia did it?
And if you assume that Spaniel also agrees that the existence or absence of the NS isn't the main reason why business has ceased, do you still believe that the existence or destruction of the NS doesn't change the cost/benefit analysis for anybody trying to overthrow Putin?
How do you get to 80% on a pro-Ukrainian group? As near as I can tell, the evidence for that is that a formerly-respected journalistic outlet said "trust us, we have Super Secret sources that say the Ukrainians did it", coupled with a bit of detail that on examination has some holes like e.g. the specified vessel not having been in the placeit was said to have been.
Which, OK, the details could have been garbled without the underlying story being false. But we've already *got* a formerly-respected journalistic outlet saying "Trust me, I have Super Secret sources that say the US and Norwegian navies did it", again with a bit of detail that has holes like out-of-place ships. And I don't think it can plausibly have been *both* Ukrainian partisans and the US/Norwegian navies. So how do you get to 80% on one and the other folded somewhere into a 10% "weird stuff" category?
I don't think either version is all that credible, but if I have to pick, then as a veteran diver any theory that a sophisticated technical diving operation at 80 meters was carried out undetected from a fifteen-meter yacht comes in way behind the theory where a 375-ton minesweeper was used.
Re: why I think it is wildly implausible that Biden administration or Norwegians (full disclosure: I didn’t even know they are implicated until you mentioned them) did it? I do not know much about how difficult it is it destroy underwater pipeline, but I feel that I know enough about both American and Norwegian political system to rule it out. Usual arguments against 9/11 truhterism apply, with slight modifications. I should note I would probably be less confident had Trump still been in the WH, or if, like, Estonians had been suspected instead of Norwegians.
This of course does not mean that (pro)Ukrainians did it, and in fact, until last week, my best guess would be that it is the work of (pro)Russians. But I changed my opinion in response to new revelations, which do not come only from NYT but also from the German media (see here for a writeup in English: https://www.politico.eu/article/nord-stream-pipeline-blasts-lead-to-ukraine-media-reports-germany/). Of course the fact both reports came on the same day is not a coincidence, but they do contain somewhat different information and do not cite each other. Those reports are extremely politically inconvenient for governments of countries doing the investigation, which makes me suspect they are correct.
So, we have to answer two questions:
1) Is it plausible that Ukrainian proxies are crazy enough to do it? Well, yes. I am thinking about something analogous to that murder in Sarajevo which we all know about, when Serbian government didn’t order it but was done by some people loosely in the orbit of what passed for secret service in those days.
2) Are Ukrainian proxies capable of doing it? That I am not sure, but they have blown up the Kerch bridge and various other stuff in Russia. I am sure underwater pipeline is harder, though, but still.
Almost certainly not, at least for the sort of "Ukrainian proxies" who would be loose-cannon enough to try this and not depending on the support of the Ukrainian government. This really requires a Baltic Sea naval power to have pulled it off; the few other outfits that might have the capability (e.g. North Sea oil drilling companies) are not plausibly going to get into that line of business.
We are not talking about limpet mines near the surface of a sheltered port, like what the DGSE did to the Rainbow Warrior in 1985.
The tech diving 80 meters deep in the rough Baltic from a sailboat is one thing (very tricky). The 500kg of explosives supposedly used for the operation is another (hmm, no).
Nuland declared NS2 dead in March 2022…why would we blow up something that was dead? And why would we blow it up when a hurricane was in the Gulf?? The record natural gas price was after Katrina and would lead to even more inflation!
If a pro-Ukrainian group really blew it up they are dumber than Putin…had Hurricane Ian hit Louisiana blowing up the pipeline could have led to Russia winning the war as the price of natural gas would have spiked again inflicting even more pain on average Europeans. Now if it ends up being a false flag then the NYTimes was extremely irresponsible to publish the story before getting all of the information.
But blowing up the pipeline helped Russia! Luckily the hurricane didn’t hit Louisiana and so blowing up the pipeline only helped Russia a little…from a money perspective it’s probably a wash for Russia but they made tens of billions in windfall profits in the first half of 2022.
Blowing up the pipeline removed the option for Russia to use it as a geopolitical blackmail tool. European politician could've been swayed by the offer of cheap Russian gas and thus decreased their support for Ukraine. That was Kremlin's plan all along. But if the offer is closed, there is no temptation. I suspect this was the main reason and it worked well. I'm quite surprised that people think that this benefited Russia.
Also, Nordstream was a longterm Russian investment which is now gone and the only way for Russia to sell gas to Europe is through the old pipeline, paying to Ukraine for gas transit.
The function of the pipeline was to transport Russia's gas. Blowing it up meant Russia couldn't send gas west even if western European countries wanted to buy it.
This is not totally incorrect, but there are ways how EU can import Russian energy, directly or indirectly, even without Nord Stream. Overall, if Ukrainians did it, it was very dumb. Which does not mean they didn't do that! People are often stupid.
That’s not how commodity pricing works—Gazprom made insane profits in 2022 supplying less gas to Europe. Enron and Saudi Arabia have attempted the same gambit in which you risk losing market share while raking in windfall profits and so half of NS2 is operable as we speak just waiting to deliver expensive gas to a desperate Europe.
I've been experimenting with forms of data storytelling and wrote this analysis of SVB and bank failures! Would love to hear what I could do better, especially given how well read the SSC audience is!
" The Savings and Loans crisis of the 80s happened as the Fed aggressively raised interest rates. "
I thought the primary cause of the S&L crisis was North Sea oil coming online, causing world oil prices to plummet, stranding many major US oil projects which had been financed, often recklessly and sometimes illegally, by S&Ls. Didn't think interest rates had much to do with that one.
You're mixing up two questions: what caused the crisis and what caused the losses. Most of the risky bets were after the S&Ls were already insolvent. The bankers didn't have any equity to lose, so they made big bets with other people's money.
Can anyone recommend a good English translation of Gil Blas? I know Smollett did one in the eighteenth century, and it's easy enough to find, and I generally like Smollett, but I was hoping for something slightly more recent if possible—but if people say Smollett's the one to go to…well, that's why I'm looking for recs! Thanks!
I’m gonna be in SF later this week with a bunch of wild thinkers and engineers, not necessarily rat-type people though. We are already meeting a lot of cool people, anyone here have any suggestions for what we should do? None of us are from California.
Also, a lot of us like to build amateur rockets, and we want to tour the SpaceX factory. Does anyone have any connections?
Friday night free party at Oakland Museum is a reliably fantastic time. Eat dinner at the food trucks and enjoy the DJ set in an extraordinary piece of civic architecture til 8PM and then you go on to your next party. Lots of rad people to talk to.
I live in San Francisco most days of the week (in a suburb on weekends), and for me the absolute best time in S.F. was in a car at Ocean Beach watching the sunset with a friend while we drank some wine, but that’s two people not a group.
The Presidio’s Baker Beach is nice. The Beach Chalet restaurant (which overlooks Ocean Beach) is nice, the Land’s End parking lot (which also overlooks the Pacific Ocean) is also nice.
You could go to Fort Point and someone pretending to be “Madeleine”/“Judy”/Kim Novak may fall into the bay and someone pretending to be ‘John "Scottie" Ferguson’/Jimmy Stewart may pretend to rescue her (probably a bad idea as the stairs in “Vertigo” don’t exist).
At Sam’s Grill they’re wood paneled booths with curtains that close so you may dine with your paramour in privacy.
At Golden Gate Park there’s a bison enclosure that you may view, the Conservatory of Flowers, and the Japanese Tea Garden.
Near Fisherman’s Wharf there’s a WW2 submarine and a Liberty ship.
Across the Golden Gate Bridge in the Marin Headlands there’s a mock up of the Nike nuclear missiles that were there until the 1970’s.
Across the Bay Bridge in Oakland there’s the Chabot Space and Science museum where on Saturday nights (when weather permits) you may look through the telescopes and otherwise look at old space suits.
Great movie. How about Dashiell Hammett’s San Francisco? I just read an article in the NYTimes about converting old buildings into apartments and it mentions the private detective suite.
The movie is a treasure just for having the Phil Hartman scene…and the opening shot is great. Did you listen to Dana Carvey on Marc Maron? He is totally different than I would expect him to be. Dana Carvey on the Conan podcast is hilarious.
No, I haven't listened to Maron or Conan's podcasts. I do recall that when Unspooled covered Austin Powers they mentioned that "Dr. Evil" is actually based on Carvey's imitation of Lorne Michaels (even though Lorne doesn't actually hold up his finger like that, and was instead just something Carvey made up). And that SNL's much later Californians sketch series is based on Carvey's imitation of his own son.
I haven't read Chomsky's piece yet, but it's bizarre to me that Aaronson opening tone is so angry and hostile (and apparently it's Chomsky who is hostile and smug according to him?). Are LLMs his cousins or something like that ?
I'm broadly in agreement with Chomsky here, he's not saying (based on an extremly quick and incomplete skim) that LLM won't "revolutionize" society, whatever that is. He's saying LLMs don't understand language, except in a very trivial sense. Language is inseparable from coginition, and LLMs' cognition is laughable. Can someone who understand Language behave like this : https://vgel.me/posts/tools-not-needed/ ? ChatGPT is given a "calculator" that outputs wrong answers, it's required to use it to solve a word problem. ChatGPT either silently ignores the results when they are too ridiculous, or accepts them at face value even when a human can see they are ridiculous. At no point does ChatGPT talks to its interlocutor, or shows hints of understanding its own "reasoning".
People who think LLMs have an "understanding" of language seem to have a weird strawmanish view of language understanding, I'm not even talking about the "Functionalist View", which views the brain as a black box solely defined by its capabilities, this is reasonable, and ChatGPT fails this too. The standard of language understanding is human conversation, and LLMs cannot pass any free-form conversation unless there is an excuse that compensates for their weird and non-human blindspots. I don't doubt there are people as superficial as ChatGPT persona, but it's never the case they are *uniformly* so, there is something somewhere in the space of all possible conversation topics that will show a human can get angry or passionate or imaginative while ChatGPT can never, that will show a human understands the real life while ChatGPT understands only token probabilities. The imitation is fragile and precarious, it's like 3 children in a trench coat pretending to be an adult, they can fool plenty of people for sometime, but given enough time everybody remotely paying attention is going to notice, or a kid will slip and fall out of the coat.
It's annoying that people keep saying "But... but... Computers never understood language this good ??" Ok, good for you, nobody is minimzing the sheer amount of innovation LLMs represent, it's a huge performance jump for sure, LLMs are indeed the best a **COMPUTER** can do (as we know the word "Computer" today and for the past 70 years, just to avoid the nitpicks that the abstract concepts of computers and computation allows for far more than this), and this best is indeed very good and can fool most people for no less than 5-15 minutes or so, but it's below average compared to what any healthy human can do, and the thought behind the words is even less, and this is important because language is first and foremost the expression of thoughts and it's extremly obvious when something is spouting it with no underlying thought.
Chomsky apparently didn't even do basic research before writing his article. There are things he states ChatGPT can't do that five minutes of interacting with it would show that it can.
IIRC, Chomsky once said that if a computer could write grammatical English, his life's work would be undermined. So it's not surprising if he's unwilling to accept that it can. But it's disappointing that such a trivially disprovable piece can be published in a major newspaper[*] and get much more traction than the people doing actual research on ChatGPT's abilities and limitations.
[*]although of course readers of this blog already knew the NYT's standards of journalistic integrity are low
> and this best is indeed very good and can fool most people for no less than 5-15 minutes or so, but it's below average compared to what any healthy human can do
It's below the average human in some ways (like advanced reasoning), but above the average human in others (which is why it has any market value). Speed and breadth of knowledge are some obvious ones, but it's also better at writing than the average person (though not better than a professional writer). I know a few people who use it to rewrite short documents (paragraphs, cover letters, etc.). It improves them significantly.
ChatGPT is also surprisingly good at writing cute rhyming poems on any subject, something that GPT3 originally couldn't do.
It definitely can do basic reasoning. There are plenty of examples of people testing all kinds of types of reasoning. It can make inferences and generalizations.
For example, ask it something like the follow:
"Vera and Vikki are sisters. Vera is 4 years old and Vikki is 13 years old. What age will each sister be when Vikki is twice as old as Vera?"
It gave me the correct answer along with the detailed step-by-step work to get there.
I think you've been fooled by ChatGPT. It can definitely produce text that *appears* to demonstrate basic reasoning, but it also produces text that utterly fails to do this and cannot, despite having contradictions directly pointed out to it, even "see" the contradictions much less resolve them. That's how I ended up spending half an hour arguing with it about how many syllables were in a line of poetry it produced, with it eventually claiming that "1 + 1 + 1 + 1 + 2 + 1 + 1 + 1 + 2 is equal to 10." (It could correctly count the syllables in each word of the line, but couldn't sum them up in that session. It also happily gave obviously contradictory definitions of iambic pentameter in the same single response.)
I think you're using a very different definition of "reasoning" than I am. I'm guessing your version involves some kind of internally consistent mind, perhaps like a low intelligence human. What I'm claiming is something that's trivially demonstrable: that it can make inferences.
This is something even a series of if-then statements can do (or, for example, a compiler can make inferences about code). It's not something that requires a mind or machine learning. Just deriving facts from some input.
ChatGPT steps this up a notch in that it's very generalizable and, like most machine learning, it can reason about things it wasn't specifically programmed to. It picks up on patterns in its training data.
It's not simply repeating its training data back to you verbatim. It can generalize. In the example I gave above, it won't matter how you change the phrasing, the details, or the numbers (within reason). It'll still get it right.
It often says obviously contradictory things. It doesn't work like a human mind (although humans also often fail at reasoning). It trained to predict the next token, based on the text so far. Different parts of the network will kind of compete to influence the next token. Some will win out and some will be ignored. Just because somewhere it has the ability to correctly add up 1 + 1 + 1 + 1 + 2 + 1 + 1 + 1 + 2 doesn't mean that part of the network will win out for a particular prompt. The prompt might match more strongly with other pathways.
No, you've been fooled by the convincing output and bias in what you select from all its possible outputs.
Obviously, any system that produces text similar to its input that's given some correct inferences in the training text will also produce output text that has "correct inferences" in it. That doesn't mean it's making inferences; it's simply creating output text in the style of its training text. Claiming that's reasoning is akin to claiming that, because you can find the entire works of Shakespeare in the output of an infinite number of monkeys with an infinite number of typewriters, the monkeys know how to write plays.
Ironically, you explain in the last paragraph of your comment why it's not reasoning.
Your model of how LLMs (and neural nets in general) work is broken. It doesn't stitch together text it has seen in its training data or choose the statistically most likely next word like a Markov chain. The hidden layers hold higher level concepts, especially in deep networks. Sure, their goal is just to minimize their loss function (make accurate predictions), but it turns out the best way to do that is by developing a model of the world.
Do you think neural nets in general can't make inferences? There are well known mathematical proofs that it can (given the proper weights) and it's not really worth arguing about.
While I'm no longer searching for direct evidence that OpenGPT doesn't reason, I still try to use it for things where I've had difficulty with standard search engines, and it still gives back great examples of blatant illogic. Today it's somehow really stuck on the idea that the 1973 film _The Sting_ has only five acts (it has six), leading to a lot of going around in circles about what they are, and eventually the following as one of its answers:
The correct title for the third act of "The Sting" is "The Tale," and the correct title for the third act is "The Shut Out."
The five acts of "The Sting" and their corresponding intertitle cards are:
Act I: "The Set-Up"
Act II: "The Hook"
Act III: "The Tale"
Act IV: "The Wire"
Act V: "The Sting"
(At least it quickly abandoned the idea that the third act is called "The Tape" [sic]. And for some reason, starting a new session and prompting it with, "How many acts are in the 1973 film _The Sting_ and what is written on the intertitle card for each act?" now produces the correct response. However, a new session with the prompt "What's written on the intertitle cards for each section of _The Sting?_" now produces seven acts, with the card titles in the wrong order.)
It's strange how in normal contexts people would intuitively understand that the presence of X can be demonstrated by a single observation, but its absence cannot be demonstrated by any number of negative observations. But when it comes to AI people get it backwards: no amount of positive observations is enough do demonstrate understanding/reasoning/etc, but a single negative observation is enough to prove non-existence.
But the standard was never "AI can do this", or maybe it was a bunch of years ago when we didn't know any better, the standard is now "can the AI do this convincingly for an extended period of time in a way that survives scrutiny ?".
This is not goalpost shifting, or rather it's, but it's not the bad kind of goalpost shifting that is done only to prevent someone from being right. I don't want Pro-LLM people to be wrong, it would be marvelous if they are right, but they are not.
ChatGPT is playing on the natural error-tolerant nature of Natural Language and our natural tendency to self-modify and self-correct and fuzzy-match statements in it on the fly. It has discovered its own way of "talking" that manages to fool some people for a very brief period but fails disastrously when examined critically from any angle for any reasonable length of time. The system has specification-gamed[1] Natural Language, and so what results is not truly Natural Language, but an alien mind's extremly faulty and completely off-base shot at convincing us it can do Natural Language.
Here's[2] a fascinating account, mentioned in passing in [1], of how a researcher once wanted to make an evolutionary algorithm (a much more general type of learning algorithm than Gradient Descent, the learning algorithm used in Neural Networks) learn how to make a hardware circuit on its own. The researcher gave the algorithm control of an FPGA (self-modifying hardware that mimics any circuit) and let it loose, eventually it arrived at a solution. When the researcher tried the solution on a *different* FPGA, it didn't work : the evolutionary process has learned to take advantage of peculiar manufacturing-caused electromagnetic properties of the very specific FPGA chip it was given control of, and it wouldn't work on any other chip.
Did the evolutionary algorithm in [2] "learn" how to design a circuit ? Some people are inclined to pedantically answer "yes", after all the researcher asked for a circuit and the algorithm gave it a circuit, what's the problem ? The problem is that the researcher wasn't asking for a circuit, he thought he did, but what he was really asking for is "A circuit that can be generalized to any chip, whose principles of operation are reasonably stable and are not dependent on magic rituals and hidden uncontrollable conditions, and is preferably understandable by humans in principle", he omitted most of the requirements because it was so common sense that he didn't even knew he wanted it, but he wanted it nonetheless.
The algorithm acted like a typical devil : it followed his (incomplete, vague, full of unstated premises and requirements) wish to the last letter, to every last dot and curve on the last letter. And it came up with an amusing failure. This is what [1]'s title calls "Specification Gaming", or in simpler words, "Rule Lawyering", finding a loophole that ***Technically*** does all what was asked but ,in actual practice, violates it in a worse than useless way.
The morale of this story isn't some mystical "AI can never design circuits" object lesson, after all evolutionary algorithms have designed Antennas used by NASA[3] and have a long list of other successes[4], Antennas *are* a hell of complicated circuits. The morale of this story is a mix of "Be Careful What You Wish For" and "Study and Understand It Carefully When You Get It Before You Say Anything For Sure". Just because the evolutionary algorithm gave you a circuit that **appears** to work, doesn't mean it actually work. Just because Gradient Descent gave you a bunch of floating point weights that **appear** to talk, doesn't mean it can actually talk or reply when talked to, you might have just been specification gamed. Happens to the best of us.
>ChatGPT either silently ignores the results when they are too ridiculous, or accepts them at face value even when a human can see they are ridiculous. At no point does ChatGPT talks to its interlocutor, or shows hints of understanding its own "reasoning".
It is important not to uncritically project expectations onto LLMs derived from our experiences with human agents. Their architecture and training regime is vastly different than humans and so we should expect their abilities to manifest differently than analogous abilities in humans. For example, LLMs generally don't display incredulity because their training doesn't reward incredulity (but they do when prompted[1]). When your goal is to predict, your resources will be directed towards constructing the best prediction. This will result in unintuitive responses if the standard is "respond in the same way that a human would". But this standard is misplaced. LLMs are alien constructs and we must analyze them in a way that is free from human baggage to understand the space of their capabilities.
I'm not disagreeing with this, "LLMs are alien" is my exact view. The people who say "LLMs understand human language" are the ones projecting and playing fast and loose with the meaning of "Understand".
LLMs have reverse-engineered a way to convince humans it's talking and understanding when it's doing neither, see my other comment for a lengthy description and links about this process, called Specification Gaming. The example you give is still a reverse-engineered process of skepticism, **almost** as good as the real thing for about 15 minutes, but then you ask it "When did Donald Trump die of COVID ?" and I bet an arm and a leg it will hallucinate an answer (can't try it now, but please do if you have access to an LLM and report the results. Feel free to do all sorts of prompting except explicitly mentioned that Trump haven't died of COVID.)
This is inevitable, Language is so huge and creative that no amount of samples will be enough if you don't have the right bias, and (from what we're seeing) it turns out that Artificial Neural Networks are not the right kind of inductive bias, or rather the very particular kind of ANNs that is the Transformer architecture. I grant that it talks better than any computer, I grant that it will have far reaching consequences on everything from Spam to Screenplay Writing to Programming to Education, I grant that studying it will teach us something about how a weird alien can grok language and its resulting amusing capabilities, but I do not grant it "understands" language, and I claim I will always know a human from an LLM given 10 minutes or less of talking about any topic of my choosing.
>The people who say "LLMs understand human language" are the ones projecting and playing fast and loose with the meaning of "Understand".
I don't see how this follows. LLMs are alien, but language is universal. There is no a priori requirement that only humans can understand human language.
>it turns out that Artificial Neural Networks are not the right kind of inductive bias, or rather the very particular kind of ANNs that is the Transformer architecture
Why think the current iterations are the limit to what Transformer based LLMs can do?
Having heard Chomsky speak about LLMs at an online conference, and read a post by Gary Marcus on his views, I had little interest in reading the piece. Chomsky made his mark in the third quarter of the previous century and is to be honored for that. But, from my POV, he's done little of interest since then.
I was angry with Scott for 1) implying that Chomsky had anything to do with computational linguistics in any form, and 2) for treating Chomsky as a proxy for linguistics as a whole. I said as much in a long comment somewhere in the top quarter of that thread. On the second point, unfortunately that's how Chomsky is treated outside the field. So his views are treated as carrying a representative weight they just do not have.
Little useful work is done in such debates. However, the discussion on Scott's thread does move beyond debate to substantive issues.
Turns out in spite of their lobbying otherwise, they actually were a source of systemic risk once FDIC/Treasury took a hard look at them and should have been regulated as such.
Systemic risk is a scourge for accountability when it comes to financial institutions. Maybe such companies should be broken down into smaller pieces that won't be systemically risky if they fail.
It's not whether any one bank is a systemic risk or not, but the totality of even small banks in this country is systemic. Many other countries simply have a handful of big banks that act as utilities. But for some reason people love small and mid size banks in this country. But having all these for-profit banks provide a systemic risk. If you regulate them so that they aren't a systemic risk, then they are no longer profitable. Basically the only way a regional and small banking system can exist without panics every X number of years is for them to be heavily "subsidized."
Dang, part of me wanted to actually see what would happen if the government just let the current laws play out and didn't intervene in extraordinary measures. Next time I guess.
Yeah this all seem pretty standard to me. FDIC tries to help them find a buyer for their loans, fails, fed says "ok we'll lend you money to help you sell". Nothing extraordinary about that.
If they were Missouri Valley Bank instead this story would be relegated to the financial page and not be causing a cultural uproar.
I'm was surprised that a regulated bank in compliance with everything was able to fail so badly. Is this some kind of sign that the bank regulations are flawed, or did everyone already know that banking regulations don't prevent bank failures and you've just got to take your chances?
Regulations cannot prevent people from making risk assessments in uncertain markets that go wrong and also prevent unexpected bank runs. I am not sure what regulation could have been implemented to prevent what has happened.
"While capital, wholesale funding and loan to deposit ratios improved for many US banks since 2008, there are exceptions. As shown in the first chart, SIVB was in a league of its own: a high level of loans plus securities as a percentage of deposits, and very low reliance on stickier retail deposits as a share of total deposits. Bottom line: SIVB carved out a distinct and riskier niche than other banks, setting itself up for large potential capital shortfalls in case of rising interest rates, deposit outflows and forced asset sales.
"A further look at SIVB funding (pie charts) shows unusually high reliance on corporate/VC funding; only the small red private bank slice looks like traditional retail deposits to us. Out of SIVB’s $173 billion of customer deposits at the end of 2022, $152 billion were reportedly uninsured (i.e., over the $250,000 FDIC insurance threshold) and only $4.8 billion were fully insured. It’s fair to ask about the underwriting discipline of VC firms that put most of their liquidity in a single bank with this kind of risk profile."
SVB in the last 12 months seems like a pretty narrow situation (a fairly big bank, and an abnormally high % of deposits in accounts with a balance over $250k, and an abnormally high investment in long term debt, and interest rates going up significantly). I am not sure how I'd retroactively regulate that situation, especially without being attacked as part of the "War on Tech" since everyone would know which bank you were talking about. Guaranteeing all bank deposits wouldn't have been popular either. There was much wailing and gnashing of teeth about moral hazard when the emergency increase in the FDIC limit to $250k was made permanent.
Banks do fail, though none have for a few years and none with as large a profile as SVB since I guess the financial crisis. It's just that usually they quickly get sold to a larger bank and no one is worried that the customers won't get their full deposits, in part because it's quick (as even SVB might be) and in part because any businesses and wealthy individuals who do have to wait a while to get access to the full amount of their deposits aren't as high profile- I have no idea when the last time was this even happened.
“ The panic over the demise of SVB Financial Group, parent company of California-based Silicon Valley Bank, appears to reflect a widespread misunderstanding about what such institutions do: They’re supposed to take risks, and as a result some, particularly smaller ones, will run into trouble and even fail. What matters for everyone else is that regulators ensure that these companies have ample resources — primarily capital — to prevent their losses from harming taxpayers or the broader economy.
Like most banks, SVB engaged in the age-old business of borrowing short-term from depositors and making longer-term investments in assets such as bonds and corporate loans, profiting from the higher yields that the latter typically pay. Unfortunately, the tech startups that made up most of the bank’s customers started running low on cash, forcing it to sell some of those investments at a loss. It then attempted to raise more money from shareholders, to make up for the losses and restore confidence. When that failed, SVB reportedly entered talks to sell itself on Friday, then was placed into receivership by regulators.
Overall, banks — particularly the largest — should have a lot more equity capital to ensure their own resilience and that of the broader financial system. But that’s less of an issue in this case. SVB, on its own, isn’t systemically important. As of the end of 2022, its tangible assets of about $209 billion far exceeded its insured deposits of less than $10 billion. Its losses as of March 8 amount to less than $2 billion. So insured depositors, and the taxpayers who ultimately back them, should be fine. And unless markets go haywire — or a lot of other banks prove to be in a similar predicament — the effect on the economy should be minimal.”
There have been hundreds of bank failures even in just the last 20 years, unrelated to say 2008 GFS. They happen quite often.
I'm not sure everyone knew, but banking is inherently prone to failure. Their entire business model is to take depositor money and invest it to make a return in some way. Even if they choose the absolute most safest investment in the world, there is still risk. They must take this risk in order to realize a return to pay their overhead, pay interest to depositors, and hopefully make a profit.
No bank can sustain even a modest run on their deposits. If you over-regulate, then it's simply not profitable to run the business.
Not a ton of easy ways to fix inherent potential problem.
Yep, that is likely the long-term outcome here. But for some reason people really love their regional banks. Probably because they offer good services, but they can only offer those services under the assumption that if they go belly up, the govt will backstop the deposits. So if they are to stay, it is essentially a public subsidy of regional bank depositors, so they can enjoy higher returns or better perks than the free market would otherwise give.
I've been under the impression that US regulation actively prevents bank consolidation in a way Canada's doesn't, such that Canadian banks are much less failure prone.
Yes it does. That is the only reason they still/will exist. But in the long-term, they are fundamentally unviable. Unless we are just cool with bank runs and big blow ups every few years, like it used to be. That or a quasi public/private network of small banks, like a utility.
The bank regulations are flawed, in this case specifically by not requiring mark-to-market for treasuries (and possibly mortgages).
Just because it is legal to pretend “$10000 locked in to 1% interest for ten years” is still worth $10000 when interest is 5% elsewhere, doesn’t mean it is actually worth $10000.
That is not true. Wachovia failed *because they had to mark to market their holdings which made them insolvent (when viewed mark-to-market), but had they been able to simply wait or slowly sell their troubled assets, they potentially would have been fine.
Mark to market is not the golden savior to banking regulation, not even close.
I'm not sure what you're saying isn't true. I didn't say mark-to-market would solve every problem, if that's what you're objecting to.
It certainly wouldn't have prevented SVB from making bad investment choices. But it would have avoided a situation where regulators waited until after the bank run to take action.
Should they currently take action on the hundreds of banks in the US that are underwater, when measured in mark to market terms? Those actions would likely render those business unviable (zero equity).
The problem is the current protections FDIC are focused on consumers and not small businesses. SVB is unusual in its focus on banking for startups. Because of this there was a "logical" decision by the startup ecosystem to cause a bank run on SVB. There is a failure of banking regulations around mark to market securities where svb was not required to recognize a loss on its 10yr treasuries because it planned/claimed it was going to hold them to maturity. Someone who bought svbs assets and liabilities for $0 would probably make a below market profit over 10 years.
Not requiring daily mark to market is a feature of regulations, not a bug. It was put into dodd-frank because banks were failing due to having to report mark to market losses even though they had no intentions to sell that marked down asset. Then being forced into insolvency made those assets worth even less!
From what I've read it seems like SVB is an extremely weird regulated bank; industry focused, low level of insured deposits etc.
A similar run wouldn't happen at e.g. Chase because Chase has ~50% deposits insured (household checkings/savings under 250k) and doesn't primarily serve cash flush startups. Even still the ultimate depositor losses will be low, the real risk is to their startup customers who don't have competent finance teams despite having enough capital to justify having them.
Bank regulations are not meant to prevent bank failures. They're meant to protect customers. There was a movement in 2008-ish to prevent bank failures. But in order to do that you'd have to set up programs meant to do things like inject money to protect shareholders.
This was a non-starter. It got swept up in progressive politics (who wanted to nationalize the banks) on the one hand and libertarian politics on the other (who wanted to let the banks fail). So we got instead was the compromise of a one time bailout and the CFPB which was also meant to protect consumers, not prevent bank failures.
The real lesson here is not to keep more than $250k in banks and to have multiple banks. The government will generally move hell and high water to make sure depositors will get that amount in a few days. And it is per bank. Something I've absolutely done.
Why doesn't the government insure 100% of all deposits regardless of size? It sounds like this would prevent runs (meaning they'd never need to make good on it, and it might also save them from having to make good on their 250K promise), like they're going to want to bail out anyone who loses money anyway (so why not make it official?), and if they're regulating correctly it shouldn't come up.
“ Insuring all deposits, as Brad recommends, would deter bank runs all right. But it worsens the problem of “privatized profits, socialized losses” that exists whenever the government either explicitly protects bank depositors and other creditors with deposit insurance or implicitly protects them by always coming to the rescue with a taxpayer-funded bailout.
Suppose I run an aggressive bank. I want to grow the bank’s assets, earn some profits, and get lots of personal compensation. With bank deposits now guaranteed to infinity, I can attract pension fund money in the billions.
I might do it this way: I tell the pension fund I will pay a deposit rate equal to the return on the S&P 500, minus 10 basis points. That is, if this year the S&P returns 5.00 percent, I will give the pension fund a 4.90 percent interest rate on the deposit. But here is the kicker: I guarantee a minimum return of 1.00 percent. If the S&P earns nothing, or declines, the depositor still gets a 1.00 percent return. To the pension fund manager, this is irresistible.
I invest the deposits in the S&P 500. Most years, the S&P returns more than 1.00 percent. On every $1 billion deposit, my bank earns $1 million in profit. But if the S&P ever goes down, I take a big loss, and my bank is busted. As long as I can go a few years before that happens, I take in nice compensation. And when the crash does come, the insurance fund is left holding the bag.
Until yesterday, a pension fund manager would never take my offer, because its deposit was not fully insured. But now, if the bank deposit is insured by the government, the pension fund manager has no reason to invest in the stock market himself, because any bank can offer a better deal.
If we have 100 percent deposit insurance with no upper limit, the financial system will eventually be one in which every risk in the economy is laundered through the banking system. Only a chump of an investment manager will invest directly in stocks, bonds, or other assets. Instead, put your money in the bank and let the bank do it for you, exploiting its government guarantee. Heads you win, tails the taxpayers lose.
Of course, the government will try to stop this from happening. It will issue regulations that limit the risks that banks can take. And that will certainly work. It’s not like financial regulation has ever messed up or anything.”
That's a great example. But just curious, what stopped a bank from offering that type of service already, but just to only a maximum of 250k. I'm pretty sure I know the answer: regulation.
However, your point still stands, as banks will always be able to find a way to take more risk, however the formula, when the deposits are guaranteed, despite regulations. Those that can't or refuse to in the name of safety, will die due to unprofitability, before the ones that ultimately did, go bust
Because that introduces moral hazard. If depositors have no risk then they're encouraged to put their money in riskier banks with higher interest rates. Since the government is guaranteeing whatever is in their account at time of failure a bank which makes hugely risky bets, returns 10%, and then fails after a year it would be insured for 110% of initial deposits (assuming each account was below $250k). The Federal government tried to get around this by banning interest rates on certain accounts. But this largely died in the 1970s as inflation meant these accounts would have as much as 15% negative real rates. (This is also why postal banking died.)
The policy you're proposing is actually broadly popular as are inflation protected accounts. But it's unpopular with elites. FDR was dubious on it and his right opposition suggested it was socialist. But it was so popular they put it in anyway. On the left because it's effectively subsidizing banks and on the right because the terms and conditions for FDIC give the government a huge amount of influence among banks.
You can imagine a better policy, of course. But it'd somehow need to give enough interest to make up for inflation while simultaneously being so risk free as to not cause big failures or payouts. That'd be complex. Especially in times of growing inflation. And the American banking sector is both overregulated and underregulated which makes for a sclerotic and dangerous part of the economy in my opinion.
Should normal people with < $250K put their money in the riskiest banks with the highest interest rates? Are they doing that?
I guess I don't understand the combination of "regulate banks so they can't take too many risks" and "refuse to ensure deposits on the banks that comply with your regulations".
Yea, if you go to a site like bogleheads, it is full of folks, under the 250k limit, trying to find every last bit of interest in their savings and checking accounts. They love Ally Bank, a huge lender of subprime auto loans.
Edit: I suppose these aren't "normal" people, but they are people. As we all know, and it's different in each sector, but many folks simply don't take pricing into account when they shop, they simply buy what they want or is most convenient (including bank accounts). But there are always enough bargain hunters out there to keep the market fairly efficient.
> Should normal people with < $250K put their money in the riskiest banks with the highest interest rates? Are they doing that?
Yes, if you can find a bank offering good enough interest rates. There are restrictions on what you're allowed to do with the money. Also if someone got too far away from normal discretionary authority is likely to come down on them.
In the low interest rate environment there was little difference between the riskiest and least risky compliant banks. But in higher inflation environments like the 1970s people absolutely fled from low interest accounts. For example, the Postal Bank (which was guaranteed by the US government to a greater extent) collapsed because of low interest rates compared to riskier commercial banks.
> I guess I don't understand the combination of "regulate banks so they can't take too many risks" and "refuse to ensure deposits on the banks that comply with your regulations".
FDIC is not compensation to the banks for following regulations. The regulations are enforced by legal sanction. What you get in return for following them is not going to jail. Meanwhile the FDIC exists to curry favor with the voters. A purpose it served both in 1933 and as recently as 2008. The previous last big increase, in 1980, was also meant to cushion an economic blow.
Honestly, I find this framing strange. Do you think the relationship between business and government is symbiotic? It's not like that at all. It's hierarchical and the government is on top extracting huge rents. FDIC is specifically set up so as to NOT protect bankers. And it only exists at all because it's hugely popular with voters. Which bankers are not.
Would avoid bank runs, but banks can take more risks. Incentivises them to swing for the fences without customers worrying about banking with them. Creates more systemic risk I think
FDIC doesn't distort the incentives of the banks. It distorts the incentives of bank DEPOSITORS to not seek out safe banks. If the bank fails the owners can still get wiped out. It's the depositors who have moral hazard. But you're absolutely right it creates more systemic risk.
I am very skeptical that individual bank depositors are sophisticated enough to discern which banks are safe and which aren't, based on a detailed reading of their 10Ks and analyzing their asset mix and whatnot. What % of the population is capable of doing this?
I saw a Bill Ackman tweet yesterday saying that he finds analyzing banks and banking risks very difficult. I don't think millions of Americans are going to carefully parse the 10ks and determine who's safe and who isn't
FDR didn't want deposit insurance at all, since he'd seen how it encouraged moral hazard among banks when NY tried that. Lobbyists for small banks insisted on it and had enough Congressional support to get it, so the compromise was limiting it to small balances.
It was also just hugely popular with people who'd suffered from losing their savings in the bank crashes so common in the 1920s. Not all of it (or much of it) was shadowy lobbyists.
I mean, the federal reserve explicitly provides deposit insurance (up to a certain amount).
It looks to me like the knowledge that banks can fail (but the measures that could be stopping this must be a bad thing, as instead of taking those, they provide insurance) is baked into the system.
I wouldn't call it "significantly." And it is very unclear having those in place could have prevented this. In fact, it's reasonable to assume it wouldn't have. Or it just would've happened to different banks.
Not high probability, but I wonder if we'll suddenly realize what happens when we have a smaller crisis, but don't have Ben Bernanke and Hank Paulson to stop it getting worse (I don't have a good guess about how Yellen and Powell will respond here, but the immediate 2008 crisis response seems considered to have been extraordinarily good, so mean reversion)
Pretty recent news: Looks like the various regulators / Treasury / Fed are trying to find a way to legally announce that the depositors will be made whole. The reason they havnt already announced this is two-fold. Politcal backlash for bailing out wealthy folks, and it is pretty murky whether any of these agencies have the legal authority to actually guarantee any deposits without congress' approval. Most experts I've seen say they do not unless it is clearly a systemic risk.
You assume that a commercial client with, say, $1m in demand deposits at a bank is receiving the same (very low) interest rate as a retail client. Why?
I'm sure some large depositors get better rates from some banks. But I know of some who do not, partially because of certain bank policies and partially because they got so used to zero interest rates that they haven't yet taken action. Until SVB.
In the aggregate I think SVB is a wakeup call to depositors that they are taking more risk for a lower yield than alternatives. That can't be good for the net interest margins of banks.
Definitely not good. I don't see how nearly any bank relying on a significant amount of corporate clients (other than big 4) is a viable business right now.
The company i work for had ~$70m at SVB (all was moved before friday). There is no feasible way to break that up to be under a $250k limit.
SVB didn't do any morally wrong stuff, they just messed up the risk and were then victims of a run. They complied with all regulations (in fact their disclosure of their risks is likely what triggered the run). Without the run they could have probably sold equity like they were trying to do and cover their short term issues. Matt Levine at Bloomberg is a great place to start with a summary on the situation.
There are a number of banks that offer Insured Cash Sweep accounts that reportedly split deposits among multiple FDIC-insured banks to allow for large sums that are FDIC-protected. E.g., this one claims to be able to handle up to $100 million, and doesn't seem to be the only one. https://www.machiassavings.bank/insured-cash-sweep-savings.aspx
I don't have the knowledge or experience to know if this sort of thing would be an option for businesses like those affected by SVB. (Though the incentive for them to look at that sort of vehicle is presumably less if there's an implicit expectation that they'll be safe in one bank independent of FDIC formal limits.)
Right, as is usually the case, there is a way to *pay to reduce their risk. American business as a whole are extremely quick and efficient at responding to incentives.
I don't think any reasonable person is arguing that there is a way to get that $70M under the FDIC limit. However, there is no way your company *needs* to have that $70M ready to be used at a moments notice. Therefore, there are other places to put that money (treasuries, money markets) that are safer than a bank deposit.
Furthermore, we can assume your company was earning interest on that $70M. Fundamentally, there is no way to earn interest without risk.
So, even if said company DID NEED $70M available at a moments notice at all times, what a responsible corporate treasury does, and have been doing, to solve that issue (since deposits are not guaranteed) is to spread that risk across multiple accounts, say 3 or 4. If one fails, you still have 75% of your immediate cash needs available, plus all your cash equivalents (treasuries, money markets) that you can have available in a day or 2. Not to mention, like all the other thousands of bank failures that have occured over the years, you still eventually get back 90-100% of that uninsured deposit as the bankruptcy process plays out (or within a few days when another bank buys the failed bank as is the case 99% of the time).
SO, I cannot even understand how even the worst case scenario (bank fails and none of uninsured deposit is ever returned to business) how that can even materially effect any businesses day-to-day operations other than a bit of inconvenience. The only way that is possible is if they were taking more risk than others by storing insane amounts of cash in a bank account for perks/interest. Then that risk manifests itself, and capitalism works how it should.
The nice thing is, we live in a free country. If some company wants to risk all their cash to chase yields or other perks that a bank may offer to their depositor. THEY CAN DO THAT! But obviously, it can backfire, like every other investment in the world, ever. Well, until Sunday. Now, at least temporarily, you can take as much risk as you want with your bank deposits, and banks can offer more and more perks or interest, risk free. Hopefully it is only temporary, because that can not end well.
Taken to its extreme, in fact the companies and banks that DONT take advantage of this new risk-free arena will be outcompeted by those that do, and they will die over time, until only the riskiest players are left. (Obviously in reality, the govt hasn't explicitly guaranteed every deposit in the country, but they have implicitly done it, so there is still some small risk involved, and of course the risk govt can always just change its mind or change the law at any time).
Funny, it's starting to read like a less intense version of the FTX scandal. A bunch of SV people who all know each other, all go to each other's parties, and all think they're the smartest guys in the room -- hey! we talk *every day* to people who are Inventing The Future, building artificial minds, self-driving cars, smartphone apps to Save The Planet, big amazing stuff well beyond the dreams of mere mortals -- so what some semi-fossilized hobbits with meager imagination insist upon, in terms of not doing X or Y because historically it seems a bit risky, need not be countenanced.
Doubt it. SVB was probably just their recommendation. I doubt they were insisting. Some did split up accounts, particularly the financially educated ones. It's been reported that SVB was requiring banking exclusivity with them in exchange for loans. that certainly seems a bit sketchy, but certainly legal, and not "wrong." It's not always so easy to hedge. Hedging is a cost, management has a fiduciary duty to its shareholders to return the greatest on their investment that they can. If they're hedging a bunch, that return is lower. Now all of a sudden we have no collapse, but shareholders notice their returns are low and all these seemingly unnecessary hedges on the books, so they fire the CEO. This stuff aint easy, lose-lose for him.
You get what you pay for, champ. You want concierge service, there are an increasing number of physicians out there who will provide it. You fork out $3000-5000/year/person, in advance, and you can get a physician to return your calls the same day, maybe within an hour or two, just like the President. I should mention, though, that that retainer just buys you access and conversation, any actual treatment would be on top of that.
But the actual treatment is covered by insurance, I think -- do you know whether it is? I've been seriously thinking of going this route, despite the expense.
Yes, insurance pays for what it usually does. I just mean you need to either have insurance also or fork out yourself. I've been doing it for some years, and I think it's worth it if you're older and have or could have chronic health conditions to monitor, or you have a situation like cancer survivor where a steady surveillance is a good idea. The main benefit as I see it is less the instant access, although that's nice, but rather that the physician is able to spend more time with you and build up a picture of your situation, health, lifestyle, et cetera over years, which makes advising you on your health a lot better: she doesn't need to just give you generic advice, she can give you advice tailored to your particular history, health situation, and goals -- all of which I find very useful, well worth the money.
Oh dear. I am going to make a prediction and I would love you to tell me if I am correct. The prediction is you and your friends are relatively young and relatively healthy and this is essentially I’m on the first contacts you’ve had with the US health system on any significant basis. the reason for that prediction is that the friction points that you were describing are absolutely normal and common and frankly the least significant of the friction points, you will run into in the future.
I realize that this response got very long, so here is a TL: DR — unless you were going to a direct primary care practice where you pay cash for your visit and for all testing and procedures, you are not the customer in a Medical interaction, at best you are a product that the physician cares about and therefore your comfort and convenience is absolutely the last thing that’s taken into account.
Full comment below:
In order to understand any of this, you have to realize that patients are not the customer in the majority of healthcare interactions under the US third-party payer system. Generically, a patient presents with a problem and even before they get to see a physician (or NP which is a whole other discussion) their choices are constrained by the acceptable panel that r heir insurance company allows. When they get to see the physician the time allocated by that physician is constrained by the fact that physicians cannot charge insurance companies based on time. All visits are scored by complexity, and each level of complexity is given a code which is then associated with a “relative value unit” that is set by a committee of the commission on Medicare services (CMS). For Medicare patients that RVU is then multiplied by the conversion factor (which is now approximately $37.50) to determine the payment for that service. For private insured patients that value varies and is usually higher by anywhere between 10 and 200% depending on the contract. Either way the top level of complexity which is usually about 25 to 30% of physicians office visits pays about $120. You can imagine that standard physician office overhead is much higher than $120 an hour and therefore even for the most complex patients if that’s all you see you would have to schedule three or four of them an hour to make your nut. So that’s the start of the issue.
Then, any testing that needs to be done or referrals to other physicians have to be OK with your insurance company. Somethings are automatically OK and can be ordered without pre-authorization but since this costs the insurance company more and more money, they have been adding more and more things to the pre-authorization schedule. This means that the physician or physician delegate has to apply to the insurance company to cover the test/procedure/etc. The insurance company, of course is incentivized not to pay for it, and is further incentivized to delay, consideration and payment for as long as they can, first in the hope that the physician gives up (which I am reliably told happens about 10 to 15% of the time), and at the very least to drag their feet so they can avoid paying right now.
Then, assuming that permission is granted, and a denial is not required to be appealed the physician has to order and schedule the test for a place that the insurance company covers which is not necessarily convenient for the patient.
Then, after the testing has been done, getting an appointment to go over the tests recapitulates the process with most physicians offices being scheduled for many weeks out because they attempt to cram so many patients in. Since Covid telemedicine visits are a possibility, but again they are subject to the same constraints as any other visit, and can pay slightly less.
The long and the short of it is that at almost no point does the patient’s convenience matter to anyone in this process because the patients are constrained and the physicians are constrained to do what the insurance company wants because that’s who pays them. It is, a perfect encapsulation of the golden rule “the one who has the gold, makes the rules.”
This is gotten quite long, so I’ll stop here, but I’m happy to answer any specific questions. I will add that the general response to the public and many physicians is to want to go to a single payer “Medicare for all“ type system, Medicare does avoid some of this by not requiring pre-certification For anything, however, it does audit people on the backend for compliance, which makes you follow many of the same rules. Furthermore, Medicare physician payments are so low compared to private insurance that most practices cannot survive on Medicare alone, which is why, for example, geriatricians are the lowest paid type of internal medicine physician, even though 100% of their patients have insurance.
Hey guys. How's it going?
Planning to post this in the fresh thread tomorrow but in the meantime hoping I can get some takers. Looking for some no-nonsense advice around a lingering bad shroom reaction.
Last Sunday (6 days ago) I took 4g of shrooms (MVP strain). The trip itself was meh, had periods of euphoria but towards the end had kind of a sad crash. But nothing I would describe as so terrible. I've done shrooms before and had a good experience, including a couple months prior taking 3.5g of MVP and having a positive experience.
Since then, I've been feeling really quite poorly. Some physical symptoms such as chills/feverishness, which seem to be passing. More importantly my headspace has been very bad. A sort of headache-y brain fog has been following me most of the time (trying to manage through this at work), and more unpleasantly, the fog seems to come with a sort of "black cloud" where I feel incredibly down and sad and like my life is over and I don't want to live anymore. It makes no sense but the cloud is just so powerful. I've had a couple welcome intervals of lucidity lasting for half a day or so. This morning I have more of a "grey cloud," where I'm not normal but at least my rational brain is essentially in the driver's seat.
I'm going to the doctor on Mon but in the meantime I'd appreciate any advice. My understanding is shrooms should not cause a permanent shift in my brain (I have no history of mental illness in me or my family). So I'm hoping this will all pass. But it's incredibly unpleasant. Any words of wisdom?
I often have a period of pretty bad ennui after a psychadelic experience (not neccessarily because of a bad trip or anything like that). I dont think it's like "your brain is broken" but more "you have entered a transient malaise for unknowable reasons and will mean-revert eventually". Or maybe "you depleted the fuck out of neurotransmitters or something and are experiencing the fallout" either way, this too shall pass
Thanks for replying. Hoping to get that mean reversion.
Inspired by Adam Mastroiani claiming his mind is like Fantasia in this post: https://experimentalhistory.substack.com/p/you-cant-reach-the-brain-through I decided to do a survey to see how much of a resemblance to Fantasia would people say their minds have. The only question I have for sure is a gradient on which movie your mind resembles that has 10 gradations from Saw to Fantasia (I guess the middle would be a boring drama, and the movie at the bottom end, Saw, can be changed), but I'm wondering if anyone here has other questions that would make sense on the survey. I'm thinking of another one about how much of the time is your mind like the choice in the first question, but presumably the choices there would be half of the time, most of the time, all of the time (why did you choose it if it's less than half of the time?), but am unsure.
Happy St. Patrick's Day to you all, and in honour of the contribution my country has made to the world, this typical representation of my native culture from 2011 but still relevant today (tradition never changes!)
https://www.youtube.com/watch?v=BNFfDirBE6w
The English-language version is slightly different if you want to go looking for it, but sure why would you? ☘☘☘☘☘☘☘☘☘
The Archbishop of Saint Paul MN gave Catholics here a special dispensation from the Lenten requirement of abstaining from meat today. A lot of corned beef and cabbage will be served today. Maybe a few pints of Guinness and a couple shots of Jameson too.
https://www.archspm.org/st-patricks-day-dispensation/
Bad luck with the weather though. It was 11 F when the St Patrick’s Day parade started at noon.
Happy Pi Day for those who celebrate!
(I am unfortunately not celebrating. In Massachusetts, where I am, there is a snowstorm today, though luckily it's not as catastrophic as the forecasts had predicted.)
I'm writing a musical about a despotic dictator who decrees that Pi = 3. Any thoughts on what kinds of problems that would cause?
The easiest way to accomplish pi=3 would be some kind of weird spacetime distortion around every circle.
Is this from the alternate universe where Clarence Waldo failed to intervene and stop the Indiana Pi Bill?
Yeah that's probably where I originally got the idea. I researched it and actually they wanted to make it 3.2! But 3 is pure and a better artistic choice.
For a long time I thought it was too silly to really go anywhere - but when I see Tucker Carlson lie to my lying eyes, I am starting to feel compelled to dust off the score and run that sucker up to Broadway!
As I have always said, "When In Doubt, Put On a Musical!"
Maybe something about arbitrage/accounting fraud? Selling a group of things in a line of length 3, then buying a group of things wrapped in a circle for the same amount?
I've posted before about caffeine "cycling". In other words, can I have caffeine x days per week, and then abstain for 7-x days so that I don't build tolerance? One idea that came up was using something different on the "off" days, e.g. nicotine. Well I tried nicotine gum (several 2mg pieces throughout the day) on my no-caffeine days and I basically felt nothing. No stimulation and I just missed having caffeine in my system. Has anyone else had this experience or have any other suggestions?
I drink a reckless amount of caffeine and then every couple months will take 4-5 days where I limit myself to one cup. I don't know how much between-person difference there is, but that works for me without needing other stimulants.
I don't know about nicotine specifically, but many stimulants are cross-tolerant, meaning you can't just substitute one for the other while tolerance comes back. Maybe something very different like Modafinil would work better?
Good to know! Wikipedia doesn't mention cross-tolerance between those two drugs, but that doesn't mean it doesn't exist:
https://en.wikipedia.org/wiki/Cross-tolerance
Someone on reddit (I found an old thread) said phenylpiracetam and (R)-modafinil could work. I'll have to look into these. I don't know a thing about them.
Funny that your break from caffeine still involves 1 cup per day. Maybe I should have a quarter cup on my "off days".
Dumb question, but were you using the gum right? You have to chew it a bit, park it in your lip, and let the nicotine seep in. Could try going up to 4mg if you're still not feeling anything --most people I know who are not regular nicotine users get a noticable effect from that.
Not a dumb question. Yes I was using it the way you said. I might try the higher dose and maybe I’ll try the lozenges instead.
Not sure why chewing multiple 2mg pieces wasn’t doing trick though!
Is there something to the view that medicine is useless on average?
https://www.overcomingbias.com/p/karnataka-hospital-insurance-experimenthtml
> Opportunity to purchase insurance led to 59.91% uptake and access to free insurance to 78.71% uptake. … Across a range of health measures, we estimate no significant impacts on health. … We conducted a baseline survey involving multiple members of each household 18 months before the intervention. We measured outcomes two times, at 18 months and at 3.5 years post intervention. … only 3 (0.46% of all estimated coefficients concerning health outcomes) were significant after multiple-testing adjustments. We cannot reject the hypothesis that the distribution of p-values from these estimates is consistent with no differences (P=0.31).
This result seems aggressively counterintuitive, in light of how lifespans have been increasing due to improvements in medicine. Hell, I once heard that the real reason murder rates went down is that medicine is keeping lots of victims of violent crime alive that in the past would have died. The conclusion from Hanson that health insurance is useless also seems excessive. If I break a bone or get some weird disease, it sure is better to be with health insurance than not. It's not like those things would just fix themselves otherwise.
This one seems like a good subject for Scott to do a deep dive on, unless he already did?
Pay attention to the actual claim, which is not that medicine is useless on average, but that *marginal* medicine is useless on average.
Mind, there is a potentially surprising claim in there, that much of the healthcare our society buys is useless on average - but personally it isn't that surprising. Pseudo-scientific healthcare that works when held to a scientific analysis stops being pseudoscience and just becomes healthcare. Marginal healthcare that's effective stops being marginal.
The issue, from a policy perspective, is that marginal for one person is not marginal for another; we can't just ban all the ineffectual treatments and be done with it. The healthcare that is super-important for one person just makes another person satisfied that somebody is taking their problems (or at least complaints) seriously; see all the people who want antibiotics for the flu.
But then, in my adolescence, I ended up on antibiotics every time I got the flu, just removed a week or two, because every damn time I got the flu I'd get a secondary respiratory infection. At one point I think a doctor did notice this, because I got antibiotics when I got the flu, and I got to skip the second course of disease that one time. Whatever the problem was, it went away eventually. (In retrospect I think it was probably something environmental, because I think the problem stopped when we moved to a different house, but many other things changed, so who knows?)
Scott has written on this: https://slatestarcodex.com/2020/04/20/the-amish-health-care-system/
>This result seems aggressively counterintuitive, in light of how lifespans have been increasing due to improvements in medicine.
Little of the lifespan increase post vaccines and antibiotics are due to improved healthcare. Hanson isn't saying that healthcare doesn't improve health, he's saying that there are a few cheap miracle cures that improved health a lot, but that we kept adding costlier and costlier interventions with smaller and smaller gains. The **marginal** effect of more healthcare on health in a first world nation seems to be zero or negative, but for some reason (see Elephant in the Brain) we keep increasing spending.
As with e.g. studies that parenting doesn't matter, you have to keep in mind that there are things you basically can't study because whenever you notice that they exist you have to report them to authorities who will make them go away before you're finished studying them.
So we can't study what happens if we let people with compound fractures lie in the streets bleeding because they don't have health insurance, and we can't study what happens if we withhold antibiotics from people with serious bacterial infections, and we can't study what happens if we don't treat heart attack victims.
At the other extreme, we can't study the effect of the very very best health care, because that mostly goes to rich people whose lawyers will explain to your IRB that they aren't interested in being test subjects in your experiment. And you'll have a hard time studying the very long-term effects of an artificial delta to someone's health care because you probably can't afford to run the study for more than a few years.
So what's left is, you can study the short-term effects of the sort of marginal health care that people will only bother to obtain if it's covered by insurance. At which point, yeah, maybe it's plausible that when we've picked all the low-hanging fruit and written the juiciest fruit off as beyond our reach, there's not much difference left in the middle.
Could you say more on your first example (studies that say that parenting doesn't matter)? What are the things that can't be noticed that would impact the results of the studies? I'm guessing you mean abuse. But then certainly the studies that don't find any effect don't deny that abuse /does/ have an effect.
I feel I haven't understood your comment (and I am very interested in this subject, as I have 2 young children and I wonder how much I should push them to read, do math, learn discipline, versus playing pokemon go, watch TV, just play with other kids, etc., etc.).
I haven't read it, but I know that this is the topic of Bryan Caplan's book: https://www.amazon.com/Selfish-Reasons-Have-More-Kids/dp/0465028616/.
The way I read John's post was that although studies "show" that parenting doesn't matter, said studies are limited to what they can measure, and (implicitly in his post) if they could measure all factors the conclusions would be different.
As far as I know, Caplan believes that parenting doesn't matter and therefore he agrees with those studies.
But I could have misunderstood, that's why I asked for more info.
Right, and Caplan is a fair source for just the studies. We don't *know* that e.g. horribly abusive and neglectful parenting results in negative outcomes for the child, at least in the multiple-peer-reviewed-RCT sense of "know", but it's a reasonably good guess and one that shouldn't be discarded on the basis of studies that can't include such cases. Similarly, we don't "know" that completely locking people out of the health care system would be bad for their health, but it's a pretty good guess that is consistent with the data.
Agreed. I thought you may have something more, thank you for clarifying. It is indeed reasonable to assume abusive parents are a detriment to kids.
You are right about Caplan. I thought you were just asking about the studies in general, not their alleged shortcomings.
I think when you've got a single study and a ridiculously counterintuitive conclusion like "medicine is useless on average", you should probably tend towards suspecting that the study was crap.
The study isn't even about "is medicine good", it's about "is free health insurance good". People with serious health problems still found ways to get them treated. The free health insurance in Karnakata might have been crap and maybe people didn't bother to use it anyway. The time span of the study was just three years, meaning that any longer-term benefits won't show up. And
The question isn't about "All" or "None" healthcare, because there are no people in the US who have access to "None" healthcare (and to be fair, no one was offered "All" in this study, we don't really know if additional care could have improved outcomes if available). The question would be about the difference between some kind of baseline (which at the least would be emergency care) and having some kind of reasonably strong health insurance option.
It's not unreasonable to conclude that healthcare above some level does not provide much additional benefit. It's a bang-for-your-buck analysis at that point. What you say about the three year timespan is definitely a limitation here, even given the purpose of the question.
A lot of medical care that people receive doesn't contribute to health out comes. Unless you have a chronic condition or are very old or over weight, you don't need a yearly checkup and you don't need all the blood tests that are regularly done.
There is also a huge set of "health problems" that people won't seek care for if they have to pay for it but will if they dont have to pay for it but these problems aren't really problems and will probably go away on their own. Muscle aches and pains being a large portion of them.
Also a lot of surgeries or treatments that don't out perform placebo. Things like cortisol injections to treat knee or back problems, some orthopedic surgeries, etc.
So not all health care is useless. Obviously emergency care isn't (i dont think you are arguing this). As we get older we are likely to need more care. Care for chronic disease is needed, but for most healthy people there isn't much need to engage with medical services with out an adverse event.
There is a broad range of what "heath care" is. So its really hard to make blanket claims about what is valid and what isn't.
I would advocate for no government or otherwise subsidized third-party "medical insurance" at all after age 50. seniors would be free to purchace medical services on the open market, but any social value in spending $$$$ on the last few decades of your life is nil. You might disagree, in which many medical professionals are ready to take your money.
Plus, by 50, you get the face you deserve, and you are on the death-trajectory you have created. Brenschluss has occurred, there is no steering you, you will fall when/where you will.
I agree with you in general. I divide medical care into two categories:
-- Trauma care: Incredible, trauma and ER care can save people who would be stone dead 40 years ago, and often restore full function after shocking injuries. Medicine and timely intervention are truly miracles and worth every penny.
-- General care. Mostly Useless. Some things like cancer screenings can be very effective, but diseases and chronic conditions are not treatable. You are just consuming health care, no care would have the same effect. If doctors could/would exclude or "86" certain patients, the rest might benefit. My medcal practice:
You're fat. Get out of my office
You smoke? That dwarfs all other conditions. Bye.
You drink how much? Nothing can save you..
In general : You are going to die.
In general: people will do stupid things and infinite medecine might change 5% of outcomes.
BRetty
My vague recollection is that this already known to people who study the field carefully, and the conclusions is hardly that healthcare doesn't matter, but that increased access to health insurance doesn't (at least under today's conditions) translate to improved healthcare outcomes. Presumably the underlying reason is that people are rational: if they get seriously injured or sick, then they take care of it, and they either find a way to get it for free, or they find a way to pay for it, because they value their health and life above all other considerations. Adding additional health insurance may make it easier for them, or transfer the cost to someone else, but it doesn't change the outcomes because the outcomes are being driven by rational individual decisions.
It's only a surprise if you've never questioned the access to healthcare insurance = improved healthcare outcomes assumption which reformers[1], politicians[2], and the healthcare industry itself[3] have all heavily promoted as obvious for the past 50 years.
------------------
[1] Because job security: there are few crusades as immediately sympathetic as improving the healthcare outcomes of the unfortunate.
[2] Because they want your vote to expand their power.
[3] Because having to shift money around from paying customers to nonpaying customers is miserable and makes them very unpopular.
This doesn't suggest to me at all that medicine is useless. The experiment is about insurance, not healthcare as a whole. Surely people who really needed medicine paid for it, even in the groups without insurance. What it suggests to me is that if you get insurance, you go to the doctor way more often than you need to since you're not paying that much, if at all, and that the outcomes are not much different by health for the insured vs not insured.
Is this even surprising? Like I said, I would expect when you really, really need it you're going to pay for medicine regardless of your insurance status. The only times I would expect insurance to really make a difference in overall health is if there is some treatment like cancer or a lifelong auto immune disease where the treatment is too expensive to afford without insurance, or at end of life when you have lots of medication expenses or something that keep you alive in old age. The former is maybe too rare to be picked up on in anything but an enormous study, and the latter depends on having some very elderly seniors in your experiment.
What we should really be measuring with respect to insurance is not health, but financial wellbeing. The point of having insurance for me is more to prevent me from going broke in unfortunate medical circumstances than it is to keep me healthy. I'll pay whatever if I need a spinal surgery, even if it means paying back a loan for the rest of my life. I might be just as healthy as the person who had insurance, we both got the surgery, but I'm broke and they're not.
Came here to make the point in your last paragraph, you did it better than I would have. I'll throw in maybe wanting to measure other quality of life markers if we could - like how much free time is spent looking/queuing for cheap/free healthcare options?
Yeah, sounds like Robin Hanson stuck his head out and got it lopped off, as he's wont to do.
https://open.substack.com/pub/sinatana/p/dear-naomi-wolf?utm_source=direct&r=zickz&utm_campaign=post&utm_medium=web
That was terrible. Misspellings, bad grammar, folk etymology, and the general air of being crazy/stoned (which is your prerogative if you're going for the 'only the mad are sane' vibe of 'the road of excess leads to the palace of wisdom').
But honestly? I've seen this fake femmy crap too often before to be interested. Yeah yeah you and Robert Graves and the White Goddess but this time round it's the Vedas which are the only true truth.
Good luck, but I'm not buying what you're selling.
Thank you I appreciate every comment I do. My writing needs work, I probably made it sound like I don’t respect masculinity or femininity...sorry if I did that. I wrote fast and sober
Thanks! You just saved me a ton of time, even taking into account the time to thank you.
Listen, you can tell me the Sphinx was built out of cheddar by time-travelling mice from the 31st century and I'll give it a listen - PROVIDING you don't have crazy font changes, bad spelling, can't distinguish between plurals and possessives and sound like you've just polished off an entire litre of cheap supermarket vodka and are now regurgitating what is festering in the depths of your brain.
Nicely laid-out and properly spelled, then please tell me more about these mice 😁
Just run it thru gpt4 & stable diffusion, that will polish this -thing- right up.
Point being, pretty soon you won't be able to use the "idiots who can't be bothered to learn grammar or even use a spellchecker can safely be ignored" filter any more.
Does anyone have any recommendations for organisations devoted to fighting climate change that are willing to say that the positives of something reducing emissions are worth the negatives of that something harming some of the poor/minorities/etc? Not that avoiding the latter is a bad thing, but it shouldn't stop us doiing the former.
From a facebook discussion of anger-- it turns out that people's experience of anger varies quite a bit.
I'm just posting some of my writing here rather than the link because I think of posting in one place to have a sort of quasi-privacy.
"Someone on my flist posted about drug-induced anger, and what it was like-- they previously had very little experience with anger. They were surprised to find how much anger led to wanting to believe that anger was justified.
I was surprised to find out about a person who had experienced so little anger. I'm angry a *lot* of the time. Arguably it's a close to a default emotion for me.
My mother was a very angry person, and for a long time, being angry at her was a central fact of my emotional life. I wouldn't be surprised if high levels of anger are somewhat genetic."
"I've wondered about physical causes for what seems like an increase of public anger over the past 20 years or so. I've run across the idea that some prescription meds might increase anger, but not enough for people to notice it in themselves. In particular, it was hypothesized that it was steroids and statins.
I've got boring theories, like that hostility breeds hostility. Or that people (what with social media and long work hours, are getting less sleep.
I leave a slot open for something we haven't thought of."
"When I was menopausal, I went through a phase of feeling very angry during hot flashes. Since I hate being overheated, the anger seemed very reasonable.
Oddly I got feedback from other people that I seemed calmer. I assume it was a result of my trying to steady myself because my anger obviously wasn't caused by external circumstances.
Then I started getting hot flashes that were heat without the anger, so I conclude that the anger was a separate hormonal thing other than the heat."
One thing from someone else-- they found that they got very angry from tequila. They hypothesize that it was the alcohol/sugar combination, but don't want to go through it more times to test the theory.
Anger is one of those things that's so variable that it can make it hard to understand what it's like to be another person. I get angry very rarely. I've been getting angry more frequently as I age for some reason (currently late 30s), but it's still rare. Interestingly, I find that I usually only get angry at people I'm close to. Feeling anger because of politics, current events, workplace issues, the behavior of other drivers on the road, etc is even more rare; even if I recognize objectively that something is outrageous, it typically won't make me feel angry.
Probably the biggest downside of not getting angry often is that when I do get those feelings it can feel challenging to cope with them because I don't have much practice.
I tried out phentermine for weight loss a few years back and it worked quite well except that it had two intolerable side effects, one of which was that I noticed I would get angrier much faster. I would be baseline normal emotionally, but anything that upset me even slightly would cause me to skyrocket from 0 to 100 so quickly it was shockingly apparent. I realized this was becoming a problem since I was effectively working customer service at the time and I would answer the phone already out of patience for whatever issue someone needed help with, so I ended up having to cut that off.
I think you are right that some people have a predisposition to reacting with anger to small or unimportant stimuli. When my middle child was born, a woman came to my room to get the baby so that she could perform a hearing test on her. She returned with my daughter 15 minutes later, and she mentioned that this baby had an incredibly calm disposition. I thought there was no way she could tell that about someone on their first day on Earth, after only a short time spent with her, and I asked her how she could tell such a thing. She told me that she had been conducting hearing tests on newborn babies for years, and she could see personalities in them even on their first day. She said that some babies are just naturally reactive -- she said some of them are mad from the get-go.
My daughter is 10 now, and is still a pretty calm kid. I, also, am very slow to anger. I do not let most things bother me. And if I get uncomfortable or if I feel provoked, it is very easy for me to stop and assess whether a negative reaction would be reasonable or helpful.
I also think that some anger/hostility is a cover for fear, grief, discomfort, or emotions that are not considered "acceptable." My ex-husband is a veteran. After he returned from overseas, he became much more hostile to small provocations than he had been before being in a warzone. I am quite sure that his anger was a safer response than a response that would have had him expressing his fear and grief over what he experienced. The anger was protective in the short term, but over time, using it to prevent accountability for himself and others destroyed him.
People who carry a lot of anger around should be careful to not let it fester for too long. Finding a safe place to look behind the anger is key, especially if the hostility is new.
Being quicker to anger seems like a common thread in combat vets - I imagine this is an adaptive response to finding oneself in an environment of constant violence. I read this quote from an injured Ukrainian soldier recently:
“I can speak about myself. My restraint system works ok - I can calm myself down. But sometimes I have explosions of aggression that I never had before.
“Someone says something off, and for a moment I find myself threatening to break their knees. And then I calm down and think ‘what is going on?’”
PTSD and ADHD can have similar symptoms with regards to emotional regulation.
I have two things to offer.
1) You are very correct thatanger and aggression caused by internal biological "tides" can be very, very powerful. Overwhelming. Can easily break out of any containment. The metaphor of water and tides and rivers and dams and oceansis a very good fit, conceptually.we all carry some volume of anger around, it can slosh around uncompfortabally sometimes, or overflow its banks, or burst a dam or be a tsunami. To be very literal, hot flashes would be fascie-ly explained at starting a boil.
I will postulate that the most intense anger a woman can experience is maybe 50% of what even being a calm, calm man feels like. It would be fascinating, to "switch bodies" completely with another person, for a max of fifteen seconds, say.
more to follow, I have to leave for work!!!
BR
I truncated my comment for time. A more complete version of what I was trying to say:
Men and women may have similar anger levels, in daily life, but the potential, or scale, of anger, is waaaaay different.
If a woman's anger runs from 1-10, a man's scale runs from 1-100. That is entirely because of testosterone, which creates in men physical strength, agression, and the potential for anger/violence that is off the chart of anyone who has not been an adolescent male can understand.
You might say, and I accept, that I am co-locating anger with anger's expression, violence. Men are far more violent than women, part of that is testosterone and its effect, part is physical strength and a sort of mindset or outward-direction that imagines physical conflict.
I think the Motte-and-baily version of this is: men and women both experience anger. Men's anger is potentially much larger and more destructive, because of innate sex differences caused by hormones. I personally cannot access another's inner state (and if such a switch would be possible even for an instant, it might be so alien as to just be horrible and not informative), but I think subjectiely that the anger scales of men and women are 10x different.
The most relevant thing to ACX I can draw from this, is that much more serious consideration should be given to side effects of medication for men vs. women. A medication (I am thinking of Zoloft) that might cause "irritability", should be counter-indicated in males, because women might become irritable but men might become homicidal.
My extreme idea, is that the feminization of society, and the draconian penalties for any kind of physical violence, in society, and especially in schools, is creating young adults who bottle up anger, and also hoard it, and think that all anger is the same. Young adults who insist that "words *are* violence" have clearly never experienced a physical fight. Boys especially are criminalized in school. Boys will fight. It is needed so you learn that if you cross boundaries you will get punched. Boys need to learn that, while they are young enough that they can't really hurt each other. Men need to learn that before they are physically strong enough to kill other people.
Beyond getting smacked yourself, you also learn that punching somebody really hurts! Hurts your hand. You both get messed up! And from my young experience, boys' fights last about 15 seconds, you both learned something, and you move on. Even as friends. It was just a fistfight.
All the useful thought I have.
BRetty
It sounds to me like you had a really rough adolescence that you really shouldn’t be projecting into statistical statements about large populations.
My experience is exactly the opposite of yours.
There used no one in my life more angry than my mother, it has festered in her since childhood. My wife is a close second, her mother was quick to anger and she has trouble climbing down from anger once activated; she has one angry brother but her dad, other brothers and sister are all very chill.
My father, brother, and 4/5 of my male best friends have similar stories about our mothers and sisters, and we’re almost all very calm men. I do have one male close friend who has a lot of anger, so I don’t think this is an inability to recognize anger in men.
My point is definitely not that “men are less angry than women”, although that is in fact my lived experience. My point is that attributing anger to testosterone seems like a vast oversimplification of the human experience.
As a calm man from a calm family who married into a very angry family (and got the hell out again a few years later because I couldn't stand the constant fighting and drama) I am very dubious of your claim. Both men and women in my ex's family were way more angry over way more time than I've ever managed to be.
And the angriest person I've ever met was a woman, although plenty of men are in the top ten. Anecdata!
> I will postulate that the most intense anger a woman can experience is maybe 50% of what even being a calm, calm man feels like. It would be fascinating, to "switch bodies" completely with another person, for a max of fifteen seconds, say.
I disagree. I'm a man, and I'm angry sometimes, but sometimes I'm not. If a woman at her angriest is less angry than I am right now, then anger is just totally alien to her. And I don't think that's the case.
As a calm man, I think you’re deeply mistaken. I’m also surprised at how you can think that even when reply to a post discussing how large differences there are between individuals.
I agree and share these fascinations.
I believe a lot of our strong emotions come from biological causes. We often aren't angry for a reason, we are "angry" because we are tired or dehydrated or having a mild reaction to food or blood sugar or something, then our brain "justifies" the anger by latching onto a "cause". I definitely experience this with anxiety.
I think there are societal shifts that impact our moods. Many used to self-medicate anxiety with cigarettes -- I would be curious if anxiety diagnosis has increased with drop in smoking rate. Or if intense emotion during the day correlates with lower sleep quality connected with late night screen usage.
There's a huge breadth of human experience research we could do, connected with emotions that aren't quite strong enough to be labeled as problems but still impact quality of life. I imagine we would need to gather tons of data about exactly what people eat, what they drink, when they sleep, etc. We'd need to either train them to self-assess mood, or find a feasible way to test hormone levels several times throughout the day. And then, do this across thousands of people, to find patterns.
I love to fantasize about "solving" negative emotions caused by biological factors.
I can anecdotally add that I had anger and frustration issues when I had undiagnosed ADHD. Now my anger is 99% resolved by ADHD medications, I basically just don’t get angry at anything anymore now that medication is allowing my neurotransmitters to work as they should.
I really liked the outdoor Berkeley meetup a few years ago (in a meadow on campus). For those few like me who are still determined not to get COVID, another such meetup would be wonderful.
The Rose Garden Inn, where we have most Berkeley meetups nowadays, has inside and outside areas. If I say a meetup is at Rose Garden, assume there will be some outdoors discussion.
If Covid were a cold, that would be reasonable advice. Since it's rather more serious than that (much more prevalent and contagious, and responsible for several times more fatality and disability than flu, let alone colds), more serious avoidance efforts are reasonable.
In 2020 I said I expected to go back to normal when my Covid risk was comparable to driving. (Which I expected to happen much sooner, but live and learn.) At this point I'd roughly estimate it's about 4-5x that for my demographic.
And driving risk exists with an underlying consensus that it's worth a fairly massive societal investment and substantial limits on autonomy to reduce it. Currently individually trying to avoid Covid is more like living in a world of heavy traffic with no speed limits, traffic lights, lines on the road, headlights, or seat belts.
"Just get in the car" is a choice one could still make, and maybe many/most would. But some people might be motivated to stay off the roads insofar as they could, use less traveled roads when they couldn't, and/or wear a crash helmet when they had to drive.
This seems more like nagging than a productive comment. I happen to agree with you on this one but come on, it's 2023, let's not shame people for their COVID related choices.
I actually appreciate the stated intention of trebuchet's comment - that of encouraging someone to end a policy that is limiting their happiness - but I have certain personal circumstances that would make catching COVID extra-bad. Thanks for understanding.
> accept that you're going to catch an additional cold every year
As opposed to following the cheap part of precautions and getting less of colds per year than before…
It is thought that one just builds up an immunity debt and when ultimately catches cold, it will be more severe. It is probably a reason why some people even die from cold. They were “lucky” to avoid it for longer time than others but ultimately their risk also increases.
With covid maybe regular vaccines help to prevent that but that's a big maybe. You can never know about new mutations etc.
Well, last time I decided it's «medium» not «large» and has properties X Y Z so I will go to this seated lunch inside and eat something (at stand-up receptions I don't miss out on much by just chatting with people while keeping the mask), I unsurprisingly did get a cold, and rather unsurprisingly it was quite a bit lighter than my old ones… And this was after a couple of years without externally caused colds.
Actually, probably-CoViD because I was worse at mask leaks back then — it was also lighter on average than the average cold for me; the peak was just different; unlike the normal colds there were some apparently long term effects but on the slight side.
I have also seen plausible-ish claims of damage leading to weaker immune system after more-severe-than-light coronavirus (not necessarily SARS-CoV-2) infections, I suspect that both directions of study are so confounded as to cross the noise event horizon multiple times in the same direction, so yeah, not betting on either claim.
https://twitter.com/the_wilderless/status/1634909159442509824?s=20
This meme really got me thinking about how we perceive human consciousness through the lens of the technology we create. I don't endorse the idea that we aren't really making progress toward true understanding (that would be insanely absurd) but it drew a connection I had never thought of before and wanted to share
Really nice! I'm not quite sure what I think about it - probably it helps illustrate how all these things *are* useful metaphors for how the universe works, and we get more understanding with each, but it's clearly incomplete.
Am I the only person on the planet who can't look at Twitter's webpage because of all the ads that pop in and move things around, the terrible fonts, and the ridiculous color scheme? I feel like someone with epilepsy at a 70's disco with strobe lights flashing.
Ad blockers are a thing
Thank you, interesting indeed!
I don't think that any of those things are really "wrong", naive perhaps and more broad-scoped and hyped up by people in their time than they are, but not wrong.
1- The Wheel : A wheel is just a sine and a cosine (which is just a sine, but late) dancing in 2D, back in the day there was a certain French mathematician[1] who thought that it would be funny if sines and cosines dancing in 2D is enough to represent any mathematical relationship, it turned out he was right, and his name shows up in a metric truckload of scientific and engineering fields till now. The ancients weren't wrong **even if you took their words literally**, this entire universe really is full to the brim with wheels at the deepest and most fundamental levels. We just gave their ideas a different name.
2- The book : Meaning "serialized" information, volatile data captured into stable storage. This is your DNA, the structure of your cells is the volatile death-able information, DNA's nitrogen and phosphate is the sturdy pages that record them so they persist longer. The Geology of planets, it's a book that records geological events. Even spacetime is a book that records everything with a mass. The ancients weren't wrong even if you took their words literally.
3- Gears : I don't think people meant literal gears (although they won't be wrong if they did, after all gears are just sines and cosines, and everything is sines and cosines), more like the idea of composability. The world as a Lego game, Simple Things Infinitely Repeated. A single gear by itself is a trivial contraption, repeat it enough times in the right arrangement and you get airplanes and trains and factories and cars, this looks an aweful lot like what Life does, or - come to think of it - how Atoms work.
4- Computers : Machines I'm madly in love with, the thing about them is that they are universal rule-followers. They can act like anything that can be sufficiently described in rules, since Rule-Following is itself a task that can be described with rules, computers can act like other computers acting like anything (which can include acting like other computers acting like other computers ....). I think it was Greg Egan in either Permutation City or Diaspora that said "Inside a computer, everthing is the product of a human labor", computers are mini-universes where you can quite literally be a God, you say "Let something be", and behold, that something is there - indeed, that's the **only** way something can ever exist there -, unless it defies Logic itself. Every single thing you have seen a computer do, some human or collections of humans have told it to do (knowingly or not), every thing they can do was built by human design from the transistor up.
I don't know if the Universe has something like computers that occur in it naturally, if it doesn't then the Universe is infinitely poorer for it. The idea of a single small core of physical laws that govern the entire universe across all hierarchical levels is *certainly* something that every computer aficionado can smell 0s and 1s of it from an ocean away. Computers are, depending on how you look it, either a generalization or a special case of mathematics, or is mathematics itself made concrete and resident inside physical systems. It's not exactly radical or far-fetched or unproven speculation to say that vast swaths of the Universe can be seen through mathematics, ontological interpretations can differ.
------------
I don't think that "progress" is the right way of thinking about it either, none of those things invalidate or contradict or even is in tension with the other, they are all different metaphors illuminating different aspects of reality[2], indeed The Book and The Gear and The Computer seem to me like different ways of viewing the same metaphor.
[1] https://en.wikipedia.org/wiki/Joseph_Fourier
[2] https://en.wikipedia.org/wiki/Perspectivism
There may not be such a thing as a final picture of reality. It would be really weird if a universe that was not intelligently designed were tractable to human reason.
Yeah I have a lot of thoughts about that tweet, but I think the OP isn’t exactly trying to seriously engage with the standard scientific position.
I think the storybook analogy one in particular can only be appreciated once you understand how much ‘reality’ is projected outwardly from our brains, which in turn becomes interwoven and reflected in the stories we told orally and eventually wrote down!
There’s also a sense in which watching a wheel spin for the first time in Antiquity or catching a glimpse of the technically complex process of building a watch really hammers in how fundamentally *Lawful* reality. I think SA actually sliced a much cleaner analogy about this which I’m gonna butcher, but it went something along the lines of how the first engineers and philosophers to realize that the “human body was made up of mechanical gears” were gesturing at the fact that our bodies seemed to obey the same physical laws as the new machines being built at the time. This looks particularly prescient when you remember that the prevailing wisdom at the time was that our bodies were constituted of unchanging, Platonically transcendent ‘Essences’ or something. Adjust your appraisal of the contemporary vein of criticism aimed at the heresy of believing that machines could think or be conscious accordingly.
I didn't take that meme to be about consciousness, I took it to be about models of the external world, specifically to dunk on "bro we're living in the matrix" types.
I agree it's more about cosmologies than theory of mind, but I'm not sure that makes it a dunk
I reread https://slatestarcodex.com/2020/06/01/book-review-origin-of-consciousness-in-the-breakdown-of-the-bicameral-mind/ earlier this week, and am considering this meme & comment in that context
how we conceive of ourselves and "the world" reflects both where we find ourselves in the moment and the path we took to get there
take the last two, most recent parts of the meme: interlocking gears are a more complex form of the wheel, code integrates the written word and the mechanical (and simulation theory involves repeated cycles)
without making any claims about the correctness of any one view, it suggests we are both intensely caught up in the moment and building on our collective body of knowledge
What do people think of https://www.cnbc.com/2023/03/13/signature-bank-third-biggest-bank-failure-in-us-history.html ?
(board member and former congressman Barney Frank claims Signature Bank was fine, government unfairly persecuted them because of crypto)
Barney Frank also claimed that he had no idea his boyfriend was running a prostitution ring out of Frank's apartment. So apparently he's not very observant.
I hate when they do that.
He also spoke in Congress with nipples protruding through his blue shirt. Very disrespectful!
Oh man, I miss all the best stories 🤣 That reminds me of the scandal back in the 90s when Norman Lamont (Chancellor of the Exchequer at the time, and it is *not* included in his Wikipedia article) was found, or found out, that a professional dominatrix going by the name of "Miss Whiplash" was renting his basement flat (it is unclear whether she was offering her services from same or had her place of business elsewhere).
"In June 1991 she was involved in a controversy when Norman Lamont, the then Chancellor of the Exchequer, was investigated for using taxpayers' money to handle the fall-out from press stories concerning 'Miss Whiplash' (Not Lindi St Clair), who was using a flat he owned (the Treasury contributed £4,700 of the £23,000 bill which had been formally approved by the Head of the Civil Service and the Prime Minister)".
There seems to be some confusion, the second "Miss Whiplash" is a woman called Sara Dale, not Lindi St. Clair on Wikipedia, so it seems to be she who was actually renting the flat:
https://goodnessandharmony.wordpress.com/2016/05/30/norman-lamonts-dominatrix-brothel-madam-miss-whiplash/
This "Miss Whiplash" was operating as a sex therapist, offering naked massages and canings amongst other treatments using the flat as place of business.
The Barney Frank scandal was in 1983, so you didn't miss it any time recently.
It's been quite a while now since I worked professionally in media relations, but I did. And now I really really want to have a beer with the PR flak who had to explain to Mr. Lamont that when you are trying to correct them on _which_ "Miss Whiplash" it is that rented your apartment, you have lost control of the situation....
I liked Arnold Kling's take, https://arnoldkling.substack.com/p/mark-to-market-sooner-not-later
I prefer Matt Levine's take (https://pastebin.com/fMfmZZTB). The problem wasn't a *lack* of mark-to-market. The problem was *caused* by mark-to-market. Silicon Valley still had to report the market value of even the "hold to maturity" loans, and someone noticed that they were notionally underwater and triggered the bank run that drove them *actually* underwater. In the old days, bank finances were opaque enough that you wouldn't get a run over the slightest hiccup, but now they are, so the Fed has to step in to provide the service of ignoring market values instead.
Perhaps the extreme case of the "mark to market all the time" mentality is the defi Crypto world where people get liquidated over the briefest candle, and it is a miserable experience that makes it impossible to run a bank (witness the constant failures of crypto bank-like projects).
Reminds me of a classic sight gag from an early episode of The Simpsons. While the view is panning across the outfield fence of the local ballpark, I think it is, it pauses on a billboard for:
"SPRINGFIELD SAVINGS & LOAN -- proudly safeguarding your money 1892-1988 and 1990-present"
It's not crypto. It's just simple contagion which almost always spreads along industry lines nowadays. Crypto and tech exposed banks are more susceptible to runs in a run on banks in those sectors. This was one thing people got wrong in predicting there would be a regional bank run. That's actually unlikely in the current context. What'd be likely is a SECTOR bank run. So the bank down the street in SF that mostly does real estate loans wouldn't be affected but the bank in New York that does startup loans would be.
I will believe it's anti-crypto if the Feds carve out an exception for crypt assets (which they haven't so far) or appear to be picking which banks receive support based on crypto exposure (which, again, is possible but has not yet happened).
Worth noting: Barney Frank, legislator who wrote the post 2008 Dodd-Frank bill that was meant to make this impossible, has a huge vested interest in crying persecution. Because otherwise he might have to admit Dodd-Frank and the CFPB were not actually meant to do what they were advertised to do.
More interesting to me is that the Fed has basically committed to making the depositors whole. Government being government I expect this means another big boost (or even cap removal) to FDIC is coming. Unfortunately high inflation and the distorted incentives set off a bit of a ticking bomb. Albeit one that will hopefully be defused.
If it's all insured without a cap, then why bother having colateral at all?
Banks don't put up collateral for FDIC. It's insurance based on a fee the banks are legally required to pay.
I mean from the investors' perspective. If the government makes sure it's insured, the bank might as well bet it all on red.
The traditional answer to that is that the government *also* tightly regulates what banks are allowed to do in order to limit their ability to take risky bets like that.
I doubt it's an anti-crypto message[1]. My go-to hypotheses for explaining the behavior of elected and appointed officials are (1) CYA, (2) political ambition, (3) avoiding embarassing inconsistencies in the narrative I'm selling. Long-horizon plans for social engineering is way down on the list.
In this case, all three point towards an effort to persuade Joe and Jan Voter that bailing out depositors with more than $250,000 in their accounts[2] at SVB was desperately necessary, and not at all a quid pro quo to potential future campaign donors, because clearly the contagion was spreading rapidly. Look! Another bank topples! No time to lose! Must act now, lest you, Joe and Jan, wake up tomorrow and find your ATM card voided, your gas siphoned, and your dog repossessed. Not that Signature wasn't a little bit wobbly, but Frank might have been right that the Feds were a little more eager to jump than the facts strictly suggested. Who knows? I suppose it's even possible they acted for the purest and most selfless of motives.
--------------------
[1] Furthermore, I doubt Frank thinks so, either. He's an old canny pol, so crooked he has to screw his pants on. I suspect he picked *that* message as most likely to provoke outrage among those paying close attention to everything. See? The problem is some fossils just don't get the New Frontier...feel threatened by innovation, daring, youth....subscribe to my newsletter!
[2] A sum of money Joe and Jan will never see in their own accounts, and one which they probably instinctively associate with the 1% class.
Life hack: Masking tape works better than band-aids for finger cuts. It stays on longer.
Aren't you supposed to change band-aids daily for the sake of hygiene?
hygiene, smygene, you just want to stop the bleeding while working. I figured most people here don't work with their hands as much, so maybe this hack is not known.
Superglue is the best, just make sure you move your finger around while it is drying.
Coming off earlier is a feature, not a bug, since I want it to peel off once healed.
When I'm working with my hands band-aids last an hour or less. They seem prone to getting wet. Good Scotch brand masking tape hangs in there all day. Washing dishes is no problem. (Cheaper and easier to use too. :^)
I put tape around band-aids. Protects them from water, keeps them on, but the sticky part isn't directly in contact with the wound
Sure, that's fine. I just skip the band aid part these days, such a pain opening the dang sterile wrapper. For bigger wounds I sometimes will stick a bit of paper towel over it.... That only really helps with abrasions, or burns, for cuts, tape at ~ right angles to the cut works best.
UNSONG error!!
In Chapter 72, it’s written that “Sohu was doing something with her hands, muttering to herself, bending the energies, quietly funneling power into Jinxiang, watching warily.” But isn’t Sohu supposed to have lost one of her hands by Anglerfish before this? Did she regain her hand at some point? (No.)
Please tell Scott about this before it is published.
Not sure exactly what's going on, but I don't think it's an error. From Chapter 70:
"The great sword Sigh appeared in Sohu’s hands."
And then the next paragraph:
"Be afraid, Sohu. I am the left hand of God.”
Sohu didn’t say anything, just rolled down her sleeve to show the scarred stump where her own left hand used to be."
I agree, it's probably for flow. "Sohu was doing something with her hand, muttering to herself" sounds weird. I am sorry I brought this up!!!!!
Am I the only one who found Everything, Everywhere All at Once to be incredibly tedious? I walked out after 90 min, which I never do. And I liked the Daniels' previous film, as well other multiverse-esque films.
I loathed EEAAO. It was just terrible.
I confess that I find multiverse anything an obnoxious failure. Fiction itself is the multiverse for me. Your attempt to combine many conflicting fictional stories into one narrative is almost always a lazy unwillingness to plot out a believable and interesting story.
I also found the nihilism circles round to “love your family” morality play just… vapid.
Most MCU and Star Wars stories are also vapid, but they don’t annoy me a tenth as much as EEAAO did. They at least usually attempt to tell one relatively boring story per character.
No. It really was aweful and ended up drowsing through half of it. It was pretentious, trite, obnoxious and repetitive, and to see how many awards it won...well, it didn't change the opinion I had of these awards.
But the crime isn't that EEoaO won. It's that RRR wasn't even nominated for most of the categories. For all it's obsession over diversity and representation, the smoothbrains at the AMPAS can't look at an indian movie beyond the funny dance scene, even when it accomplishes everything that american blockbusters have been trying (and failing) to do in the last 10+ years.
This is an interestingly different set of opinions than mine! I wanted to like RRR, but after the first two scenes establishing the characters, the only scenes I found that lived up to the hype were the dance-off scene and the final fight.
I liked it, but it seemed pretty derivative. Like Kung Fu Hustle crossed with Sense8.
YES! I'm with you. Here, in my opinion, is the Daniels' best work: https://www.youtube.com/watch?v=HMUDVMiITOU
There were a few parts that could have been tightened up, but nothing that came close to wanting to make me walk out. One of the best movies I saw last year, and way better than that other Multiverse Mess.
I haven't seen all of the Oscar nominees, so can't speak to whether there was anything better this time around. But a year in which EEAAO wins best picture, is an acceptable performance for Hollywood and the Academy.
Dr. Strange was such a frustrating movie to watch. I can see the remnants of a good movie in it, but it looks like it just got destroyed by revisions. My personal opinion is that America Chavez was not part of the first draft, and that forcing her inclusion to hopefully open up future movies is what created most of the issues.
Strange and Wanda have this nice parallel internal conflict, that Chavez's contrasts with badly. The first act of the film ends up needing to establish Chavez's character, which means there's no time to set up Wanda's heel turn. In the climax our nominal protagonist, instead of resolving the conflict himself, is reduced to telling Chavez to Believe in Herself. It just seems like it would be so easy to make a good movie out of the whole thing if Chavez was a magic artifact instead of a character.
Wanda's heel turn was a waste of a character in any case. (I know it's an adaptation of something that happened in the comics. But a) it wasn't a great idea there, and b) one of the great things about the MCU is its ability to distill the best of the comics without being beholden to their continuity where it doesn't need to be.)
But even if they were going to do it, it should have sprung from Wanda's genuine trauma, which is hilariously exceptional even on a team where being orphaned is an entry ticket: grew up in wartime, lost parents, spent days staring at an unstable UXB, experimented on by Nazis, lost brother/last remaining relative, what's left of homeland destroyed by erstwhile ally killer robot created by teammate, accidentally killed people trying to do the right thing, wound up a fugitive from an entire international treaty created in response to *her*, fell in love, had to literally murder that love to save the world, then have it turn out to be for *nothing*.
*Not* be focused on the imaginary children that no one who didn't tune into the Disney+ series even *saw*. Which made her look like a crazy hysteric lacking only a mustache to twirl, not someone whose being pushed over the edge was at least as understandable as Tony's PTSD issues or Bruce's rage-quitting or Thor's sinking into depression.
But yes, it was even worse luck for Doctor Strange and the MoM (we all see what they did there) that it wound up right against a *much better* story about the trauma of a broken maternal relationship threatening the entire multiverse. Which no one could possibly have anticipated.
(I'm pretty sure those are the only two movies of that description from the entire history of cinema, though I'm open to reminders of something obvious I'm forgetting.)
EEAAO wasn't perfect, and I'm kind of surprised that it wound up a major Oscar force rather than the sort of quirky indy phenomenon that, say, Primer was back in 2004. (Though it had a straightforward emotional core, which Primer didn't.) But I'm certainly not unhappy that it got that wider recognition.
Agreed, particularly about how Wanda needed more time if we were going to go along with her reversals. I hadn't thought about the possibility of a story more tightly focused on Strange and Wanda; that might have worked.
EEAAO at least had a very tight focus, and developed its core characters reasonably well.
I thought it was quite good but not worth the hype.
(It's also in a club with Inside Out of "movies I had my last date with someone to before we broke up :/ )
In that case, I'd say don't take a date to see "Turning Red". :-)
My wife loved it at first, and stopped watching it so we could watch it together. And then fell asleep about 90 minutes in.
I found it very good and entertaining, especially the first 30 minutes. But I progressively got more bored of the premise. I finished it and liked it. But I think a lot of years it might not be in my top 20. This year I haven’t seen many movies, so hard to say this year.
I would agree that EEAAO was a mess, disorientingly long, and overstuffed with annoying tics and confusingly dumb plot points. But I think the warm reception can be explained in the context of superhero movies as a modern institution; particularly Disney's Marvel formula of ponderously high stakes and unearned sentimentality, leavened with mechanically-timed doses of humor. The Daniels take this formula that has beaten the general public over the head, subvert the self-seriousness, and probe it in various imaginative and affecting ways; what's not to like? I enjoyed EEAAO more than the last ten Marvel movies, which were also messes.
It doesn't end with Marvel tropes. "I know kung fu" was a titillating central conceit of the Matrix movies, but never developed beyond a ten-year-old's concept of cool. EEAAO has as much fun with it as the (rather slim) budget allows.
I don't mean to explain why you should like a movie you didn't like, just to offer an explanation for its popularity. For people who never saw fit to hop on the capeshit/science fantasy train, it might be a baffling and unpleasant ride.
I also think Michelle Yeoh and Ke Huy Quan turned in legitimately terrific, human performances of the challenging material they were given. Congratulations to both for their respective awards.
Yeah the performances were great, but they are always great. And I do think the framing of it as a subverted superhero movie makes me appreciate it more. I have avoided the superhero movies almost entirely since the first captain America/avengers movies which were just hot garbage with some kk jokes, while everyone told me they were amazing.
I liked it a lot. Maybe some differences in expectations? I went to see it on my dad's recommendation, and hadn't heard of it at all before that
I could have done without the juvenile gross out humor and the martial arts stuff got old pretty quickly but I enjoyed the bit with the daughter\destroyer of worlds making the case for nihilism enough to stick with it to the end.
I thought Cate Blanchett should have been awarded Best Actress for Tar but apparently one of the political consideration of the Motion Picture Academy was that she has already received a Best Actor Oscar.
On the other hand, I think the juvenile gross-out humor had the fun aspect of causing many people to independently invent the phrase "Chekhov's buttplug".
Yes, that was one of my problems with it: The same martial arts scene over and over. Didn't one of other billion iterations of her character have a different skill that might have come in handy?
And I understand that Ke Huy Quan's story is a heart-warming one, but there was nothing out of the ordinary about his performance. Barry Keoghan should have won Best Supporting Actpr (though I assume that Brendan Gleeson's presence on the ballot for the same film split some votes).
she spends the film augmenting with skills from different lives--often the base is martial arts bc she's fighting, but it changes a lot as the movie progresses in a very earned way
Ke Huy Quan's performance was one of the best I've seen in years. I can't remember ever being so happy to see someone get an acting award (and that is something the academy voters probably bear in mind while voting--it's not a juried award, after all)
I didn't like it either. It was amusing at first but drags on way too long. I don't understand all the hype.
I'm curious what your top movies released in 2022 and discovered in 2022 were.
Michelle Yeoh herself is very popular among a nerdy young audience, that may be part of it.
Is Paxlovid still a good idea for covid, for a healthy person, 52yo? Anyone have the latest on this? Goal is to prevent long covid.
Thank you all so very much! He started it a day and a half after testing positive as his symptoms were super mild until then. A friend advised us to do so as a way to reduce risk of rebound. i.e. don't start too early, since you can only take it for 5 days.
Also, the positive test happened on Saturday, and the doctor didn't bother to call back until Monday with the prescription.
Best wishes for a quick and complete recovery.
I tested positive Sunday, three days after my wife. We figure we did well to dodge COVID for three years.
She started Paxlovid on Friday, and started feeling better the next day.
I started Paxlovid last night (Monday evening) and had my best sleep in several days. Just took my 2nd dose. (The full course is three pills twice a day for five days.)
We are generally both very healthy, but are in our mid-60s, which is a huge risk factor.
Metformin also reduces risk of long covid. You can look up the research findings.
There's some evidence of Paxlovid giving moderate reduction of risk from long Covid-- e.g., this observational study gives ~26% reduced risk. (Which isn't exactly the orders of magnitude I'd prefer, but I'll certainly try to take it when my efforts at avoidance finally fail and I get Covid.)
https://www.medrxiv.org/content/10.1101/2022.11.03.22281783v1.full.pdf
There was also a recent RCT indicating that a short course of metformin reduced long Covid by 42% (albeit with very wide error bars). But that came out recently enough that it's not yet standard of care if it's going to be-- it would be necessary to find a doctor willing to prescribe it off label. (And because it's so recent, the odds that it won't replicate are quite decent.)
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4375620&utm_source=substack&utm_medium=email
I think Zvi has cited some studies about it. It's probably net helpful (especially if you get it fast), although the risk for people under 70 is fairly low to begin with. I'd look to get it if I was experiencing significant symptoms.
It seems like midjorney can't do (aquatic) turtles. I've tried all sorts of prompt engineering but I keep only getting tortoises.
Something to do with Voigt Kamff I think. ;-)
_I’ll_ tell you about my mother…
XD
Any advice for getting Stable Diffusion to actually output specific text? Was trying to generate a logo with specific letters and no dice. Midjourney does better.
I've had success using specific species names. Such as logger head turtle, hawksbill turtle and so on.
Thanks! I'll try it out
I'm working on my submission for the book review contest and I've been trying out Grammerly (app that's suggests improvements in your writing prose.)
The suggestions it gives you don't seem that useful, it feels basically like a spellchecker that also tells you insert lots of unnecessary commas.
But, it also gives your writing an overall score broken down into "comprehensibility", "engagement" etc.
So, I thought I'd copy-paste some of my favourite essays from different writers to see what score it gives them (roughly the average of 5 essays):
Scott, 75-80
Richard Hannania, ~80
Robin Hanson, 60-70
Christopher Hitchens, 85-90
(A British columnist I think writes well) David Mitchell, 85-90
George Orwell, 75-80
A random essay I wrote for school when I was sixteen, 60
And, if you'd ask me for subjective ratings of those writers (purely based on style, not substance) that's essentially the exact ranking I would have given.
Which make me think Grammerly can pick up real makers of quality writing.
Does anyone else use apps like this to help their writing?
If anyone else wants to paste some essays into the app, I'd also be interested if matches their subjective impressions of the quality of the writing.
I imagine Scott gets marked down a bit for all the "the the".
I find Grammarly most useful for finding little grammatical errors like missing prepositions and singular/plurals that don't match. It will occasionally suggest a big reorg of a sentence and I accept its suggestion about 70% of the time.
The spellchecker is a little better than Google's & Apple's but it has this annoying habit of surreptitiously switching perfectly good words for nonsense.
I have the free version and have never tried having it rate my writing. I like commas. Unnecessary ones are the best.
One of the purposes, of commas, is to mark, where you want, the reader to pause. So, there is almost no, ungrammatical, comma usage.
I want you to stop and think about what you've done.
,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,,
I hate this, so much.
Is the David Mitchell mentioned above the same as the panelist on "8 Out of 10 Cats?"
yeah
As an American who had never heard of 8oo10c and Countdown until maybe a year or two ago, might I say that Sean Lock was one of Britain's unsung treasures. If that is a meaningful phrase. It's a shame I never got to meet him.
Well you'd certainly hope he scores well, given his experience as an editor
https://www.youtube.com/watch?v=o1IYfaoMeCs&ab_channel=ThatLook
And as someone who argues about cheese. https://youtu.be/vyyyh8_Afyw
But have you tried raise VAT and kill all the poor?
Have decided to take the plunge in creating online content would love some feedback from the community here - I'm current a PhD student studying decision and risk analysis (think prospect theory but normative and applied to real world decisions) and am looking to share some of my less rigorous but more fun ideas that don't make out of the academic review process. Would y'all be more interested in:
1) A blog exploring how insights from my field that can be helpful for everyday life - decisions strike core of who we are individually and think we can all benefit by thinking more deeply about how we make them!
2) An online course that teaches you how to work/communicate in technical settings without a technical background. A "no-code bootcamp" which directly counters the idea that everyone should learn how to code. Especially given the rise of AI, I think technical fluency is a must but not necessarily the specific skill of coding or knowledge of the math under the hood.
I might be interested in 1, but could you post a small sample in reply to this comment demonstrating an insight from your field and how it's useful to everyday life?
I could be persuaded to read 1. 2 sounds like I have no time for it
The blog sounds a lot more interesting.
I'd be more interested in 1).
Just curious: is it true that the word "fortnight" is not in common usage in the United States?
It is not in common usage. It's a measure of time whose name suggests a different measure of time, so it's needlessly confusing to use it.
fortnight = fo'rt'n night
Ooh, neat. Now do New York minute!
It's in ironic common usage like the words "thou" or "score". High schoolers are forced to read Shakespeare, so most educated people know what the word means. But if you were to actually try to use the word in conversation, people would assume that you are making some sort of joke.
This is exactly right. Adding that I can't imagine seeing it in a professional context from an American, so if I saw something like "pay will be issued every fortnight" in a job description from a company that claimed to be based in the US, I'd be suspicious it was fraudulent.
It's mainly used in poetry and formal speaking.
Think of "87 years ago our forefathers brought forth to this continent..."
As an aside, the American writer Rex Stout has the archaic and obsolete term for a week in his 1953 short story "The Zero Clue":
"Sennight, chapter 6. Spoken not by Wolfe but by a Runyonesque racetrack tout."
It isn't in common usage, but I think most US adults would know what it is, or at least that it's a number of days even if they don't know it is exactly two weeks.
It's certainly true that the word "fortnight" is not in common usage in the United States.
To my subjective sense of U.S. English, "fortnight" is to "two weeks" very much as "thrice" is to "three times": a well understood alternative with a quaint or archaic connotation that tends to make it less suitable for everyday use.
'Fortnight' seems common enough to me. In less hectic times, one might go on a sun holiday for a fortnight, though perhaps nowadays that would imply an extreme degree of laziness / hedonism.
Can you remember the last time someone spoke the word to you?
Honestly, not offhand. But I simply might not notice. I am 63 and in my youth the word was common parlance. I would not disagree that it is used less nowadays.
I wonder whether part at least of the reason is increased 'pressure of time' in which people, while they might consider their mortgage or such-like in annual terms, live in an increasingly shrinking present.
"Furlongs per fortnight" is a common example used when one wants an obscure unit, and in fact there's a whole system of units (furlong–firkin–fortnight or FFF) of which it is a part. You're not likely ever to see this outside of technical fields, however.
The VMS operating system, when booting up, had a variable TIMEPROMPTWAIT indicating the number of microfortnights (1 mfn ≅ 1.2 s) to wait for the operator to enter a new date if it detected the current date and time to be invalid. This was to help indicate that the delay was rather approximate, since the system clock had not yet been started and so a timing loop was used instead.
Yep. I've never seen it used outside of British books (when I first saw it in Harry Potter I had to look it up).
So what do Americans think "Fortnite" is all about?
Another thing I recently realised is that "Dora the Explorer" doesn't rhyme in a rhotic accent. Standard Americans would say "Dorahh the Explorurrrr"
It’s a video game about building forts.
It doesn't rhyme exactly, but it's close enough to be perceived as a rhyme.
Right. Because not only is it “close enough” there are common American accents where Dora is pronounced Dorer (most familiar to me is my mom’s Queens, NY accent).
I believe that in those accents, "Dora" and "explorer" always rhyme when they are in the same context - that is, if they occur before a word that begins with a vowel, they both get an "r" in between the two vowels (like the "n" in "an") but if they occur at the end of an utterance, or before a word that begins with a consonant, they are both pronounced without the "r".
To the tune of Copacabana:
Her name was Dora
She was an explorer
Is that some kind of super-rhoticism?
I think of it as conservation of the letter r, thus:
"I sawr dat guy in his noo cah."
I only realized as an adult that the name Eeyore in a non rhotic accent is like the sound a donkey makes.
I remember a school club called Fortnight in Tennessee that met every 2 weeks, so the word isn’t completely unfamiliar.
Wow, I can't remember ever having learned that before about the name "Eeyore"!
I never realized that about Eeyore until you opened my ears!
😀
I’m not sure where to post this, but not too long ago I unsubscribed and then resubscribed (for a paid subscription), and I have stopped getting the Hidden Open Thread. Any ideas how I can start getting it again?
What exactly do you mean by "get" it? If you go to the front page, or better, the archive, all of the hidden threads will be there if you are logged into your account
Sorry I meant get the email same as I do other posts and regular Open Threads. The most recent Hidden Open Thread I see in Archive is 262.5 from Feb 10. Haven’t there been others since then?
Only one, um, 266.6666.... From March 9.
Oh, I did get that one! It just wasn't labelled "Hidden Open Thread." I had incorrectly assumed there was a hidden thread paired with every open one. I thought maybe I'd been experiencing an elaborate punishment for people that had left and come back 😅
Thanks for the help, friends!
"Skyler, Mingyuan"
So what's the nomatively deterministic significance of someone named Mingyuan. ( bright circle? Or does yuan here refer to the Yuan dynasty? ) being replaced by someone being named Sky-ler?
Clearly mingyuan is actually a variant of minyan, meaning a quorum for a Jewish prayer, since she worked to organize quorums. Skyler is also a reference to this, since the Jewish prayer mentions that we pray to our father in the sky.
Ah! Very nice.
In the wake of the news about both Silicon Valley Bank (SVB) and Signature Bank being shut down and going into FDIC receivership, wanted to share something potentially important I learned today. Please pass this on if it might help anyone you know.
Am hoping this info might help someone at a firm – whether it's you, or someone you might know in your networks – avoid the pain of trying to hastily move large chunks of currently-uninsured cash at your primary bank (if over the $250 thousand FDIC limit) into fully-insured accounts at other bank(s) during the coming week or beyond.
There's a service from IntraFi (formerly Promontory Interfinancial Network) called Insured Cash Sweep (ICS). If your bank offers access to it, and many do, you can keep banking with your primary bank while (effectively) splitting your deposits into chunks insured up to $250K at one or more other banks.
Web pages:
https://www.intrafinetworkdeposits.com/
https://www.intrafinetworkdeposits.com/how-it-works/
Via IntriFi's ICS (for deposit accounts) and/or CDARS (for certificates of deposit) services, "you can access millions in FDIC insurance for deposits, both large and small, and [still] enjoy the simplicity of banking with just one trusted, local institution. ... thousands of banks across the nation are part of IntraFi's network of banks and financial institutions [offering these services]."
A Twitter thread about it:
https://twitter.com/Wayne/status/1634309120751984640
And another tweet with a link to a webinar slide deck which describes it in detail, currently hosted on the American Bankers Association's website:
https://twitter.com/collmerica/status/1634630761042026499
Disclaimer: while this looks legitimate on its face, I've never worked in finance, have never used these services, and again, literally learned about them just today. So do your own due diligence – and urge anyone you share this with to do theirs, too.
(Also, there's apparently at least one competitor to IntriFi; I don't know anything about them, other than finding them when Googling "CDARS" tonight: https://americandeposits.com/what-is-cdars/)
I'm pretty certain the FDIC insures an individual to $250k, not $250k per bank. If you lose the max in one bank, and the other bank also folds, you're deposits are not covered. However I've read rumblings there are policies you can buy to protect more. But then, why not just buy US treasuries.
I believe it's per deposits with a particular bank. Years ago, I split funds between credit unions because the total of my three accounts in one credit union exceeded 250k -- a nice and extraordinary problem to have. (Credits unions have a parallel system to FDIC. I think it's called NCUA.)
It used to be per person, now its per account. You can open difference accounts at the same bank.
But do your own research.
"Why not just buy US treasuries?"
AIUI, at least for some of the firms that kept uninsured deposits at Silicon Valley Bank, they often needed to make frequent payments from their accounts for payroll (weekly, bi-weekly, or monthly), to pay invoices from suppliers and service providers, and the like.
While buying 13-week Treasury Bills – available to individual investors through Treasury Direct – would make lots of sense for individuals, their duration might make them impractical for accounts that see frequent cash outflows. (Unless you have a *lot* of cash and a corporate financial officer who can ladder those T-bills' maturities, say, weekly, or otherwise pay close day-to-day attention to maintaining adequate cash balances.)
It’s even sillier than $250k per bank. It’s $250k per account type. So you can have $250k in checking, $250k in savings, $250k in a joint account and it will all be insured.
Am seeing that apply for "ownership type," but not for "account type"? (So yes, a joint account would separately receive that insurance limit, but that wouldn't apply to a checking versus savings account you own individually: their values would be lumped together.)
https://www.fdic.gov/resources/deposit-insurance/brochures/insured-deposits/
"The standard insurance amount is $250,000 per depositor, per insured bank, for each account ownership category." (That page goes on to noted that "ownership categories" apply to distinctions like individual accounts, joint accounts, and trusts.)
"The FDIC adds together all single accounts owned by the same person at the same bank and insures the total up to $250,000."
Being Uk I shouldn't be answering this, but I and the internet think it is $250k per bank. If you have $1m split equally 4 ways with 4 banks FDIC covers you for $1m.
But DYOR everyone
That is correct. Wintrust here in the Chicago area has over the years purchased a number of local banks, which continue to operate under their own names and with their own banking licenses and so forth. Wintrust now offers "MaxSafe" savings accounts for individuals or businesses which seamlessly spread the depositor's balance around to various Wintrust-controlled local banks in $250K chunks.
Interesting, thanks! Direct link to Wintrust's MaxSafe program here:
https://www.wintrust.com/maxsafe.html
Why is there so little comparison of the current AI boom to previous ones?
Since the 1950s or 1960s, we've had several episodes (even many, depending on how you define them) in the past of huge claims for AI, none of which really panned out. (Or, alternatively, did produce great results that were no longer considered "AI" by the time we got there.) The claims I'm hearing today for ChatGPT and the like sound similar. Why is there so little comparison between what's happened with previous waves of AI and the current one, and whether the current one might also be a bubble that pops and leads to another AI winter?
As I understand it, there were only two other AI boom cycles - one in the 1950s/60s, and one in the 1980s. I'm not sure that either of them made many products that people actually found useful, the way that the things in the current cycle (probably starting with Google Translate a decade or so ago) have been.
>(Or, alternatively, did produce great results that were no longer considered "AI" by the time we got there.)
I'm just not seeing it this time around. ChatGPT (and Stable Diffusion), as it currently is, is clearly AI to me, and many people seem comfortable colloquially referring to it as AI.
>huge claims for AI, none of which really panned out.
Also not really seeing this. The actual current capabilities that you can try yourself right now are in themselves huge.
>Or, alternatively, did produce great results that were no longer considered "AI" by the time we got there.
I think that's going to happen with all AI research, no matter how advanced it is. If something can be done by a computer algorithm, then it by definition is something that can be done mechanistically and without "real human creativity," meaning we no longer consider it "intelligence."
When AI eventually conquers the world, I expect that there will be thinkpieces explaining that military strategy is just the application of some basic statistical principles to the data received from the battlefield, so it's not *really* artificial intelligence.
During the last ten years, AI basically broke all kind of things that were assumed to be impossible for a machine : translation, language, programming, go, image classification, image generation, protein simulation, diplomacy (the game). So I think the hype/anguish is pretty justified.
This is super false? I certainly didn’t break “all the things”.
I'm perhaps not totally clear on what you're saying, but I'm having trouble buying this for two reasons. First, I don't think basic translation (at a poor level) was ever assumed to be impossible for a machine (though it has had and still is having quite a bit of difficulty). Second, I've been a daily user of Google Translate (Japanese/English) for at least fifteen years now, and while there was some improvement around 2016, there hasn't been any noticeable improvement since then, and it's still pretty poor. (I find it useful only for dumping in relatively large quantities of text to get the gist of what's being said; I go back to the much slower method of dictionaries and my basic understanding of Japanese when I need any reasonable amount of accuracy, working on what seem to be the key sentences from the dump.)
I've also never really thought of it as "AI"; it really feels much more like just statistical analysis.
I've been using GT and DeepL for Japanese -> English translation and been continually disappointed in how bad they are. I assume that the state of the art language models are better than this and wish someone would hurry up and deploy them.
Just this morning, I ran into a case where DeepL hallucinated a sentence completely unrelated to the input, and a few days ago, I had DeepL mistransate one word in a sentence way that completely changed the meaning, leaving me very confused until I figured out what had happened.
I don't know what Google Translate uses under the hood, but Google is actually behind on the Large Language Model race. ChatGPT can translate several languages entirely by accident. It was not designed to do translation, it just picked it up through training.
The reason it isn't a good comparison is that we actually are seeing it do the kinds of things we've always wanted, and we are getting them through training and not through explicit programming, or as you said, 'statistical analysis'.
Is there any way of using current AI to do translation 'on the fly'? Say that I am watching a video where the speech is in Chinese or Hindi or some other language. There are no subtitles. I don't know the language so I am stuck right now at typing in "translate 'margh' from Hindi to English" into Google Translate (and if I don't know how it is written, I'm out of luck, and if the English spelling is 'maargh' I'm out of luck also).
Chinese especially is very dependent on how the characters are written. So could ChatGPT or some other AI 'listen' to the audio and give me a translation that way (ideally into text like subtitles)? Or is that currently way beyond the capacity of any machine translation?
Open AI's Whisper. There are some optimized versions floating around which let you run the large model in real time on a 3070, though if you want better accuracy you have to give it a larger lookahead window, up to 30 seconds.
You also need to use something like NVIDIA Broadcast and Silero VAD in the processing chain, one to filter background sounds, one to remove silence so the audio isn't sent to Whisper, otherwise it hallucinates.
Whisper does this by directly translating from another language audio. I don't know if it's more accurate to ask Whisper to transcribe instead, and then use DeepL though.
YouTube can do this already. You can turn on automatic subtitles, and then tell it to translate the subtitles to any language you want. It's not very good because errors creep in during both the audio-to-text and text translation steps, sometimes making it completely useless. But sometimes you can get a gist of what the video is saying.
I do use that, but as you say, it's not so great. For some videos, subtitles are not enabled at all, and for others, it's "English auto-generated" and the results can be hilariously bad because the programme is just guessing by the sound of the words what that corresponds to in a similar sounding English word.
While it doesn't look like it's been rolled out yet, there's apparently a Chrome feature under development that will combine its existing Live Captions (which auto-captions web video that doesn't have its own built in captions) with translation.
https://www.reddit.com/r/chrome/comments/xyadbi/chromes_live_captions_will_receive_several_new/
I've seen demos where people hook up automatic transcription -> automatic translation to get real time subtitles. And even put it on AR glasses to get "subtitles for real life". Of course, it's still ML, so the quality is hit or miss, and there is a delay, but it is something you can do right now if you're technically inclined.
Google Meet can auto caption live meetings -- I use it all the time in a corporate account, so I'm not sure if it's a paid feature or available to personal Google accounts.
There’s a feature in Powerpoint where you can have it create realtime subtitles of what you’re saying, translated into one of a choice of languages. It’s not perfect, but it’s very handy and feels like magic. So I’ve just assumed this is widespread and considered ‘easy’ now.
I think that technology hype cycles are a real thing, that we're in one right now, and that by now we've got many other points of comparison other than the previous couple of AI-related cycles.
The standard picture of a tech hype cycle starts with a peak of inflated expectations, followed by a trough of disillusionment, and then finally a plateau of productivity. The trouble is that it's often difficult when you're in one of the first two stages to see how high the plateau will be.
Pointing to a random previous hype cycle and saying "I think it will be like _this_ one" isn't a particularly productive way of making predictions.
There is exactly this sort of discussion amongst the AI researchers and workers in the field that I have chatted to.
I think this sort of scholarly discussion just doesn't make it past the internet hype filter.
Do you have any references to written discussion of this (whether it be papers, blogs or whatever)? Is there a consensus (or group of consensuses) on how the hype relates to reality?
Also, "scholarly discussion"? I'd think that, given all the money going into this yet again, investors would be talking about this too. Have expert systems and the Fifth Generation project been so completely forgotten?
The Google prompt "are we heading for an AI winter" brings up a lot of articles on the topic, including general consumption articles. So I may have to retract my previous speculation and update to "this sort of discussion apparently does make it past the AI hype filter".
I stuffed that into DuckDuckGo and looked at the some of the initial results, and also some of the results after setting it to show results from just the last month. I also did the former in Google just to check that DDG wasn't producing anything particularly weird.
The articles, unfortunately, seemed to be mostly business-y fluff, without any serious analysis or argument. I did find one, https://www.cortical.io/blog/third-ai-winter-ahead-why-openai-google-co-are-heading-towards-a-dead-end/ , which though from an obviously biased source, discussed the _approach_ being taken with LLMs and the like and made a couple of good points (I hadn't thought about the power consumption issue).
I'm not sure if anyone here is familiar with the "abolitionism" movement (no, not the slavery one) or if it's still even a thing, but it seems like something rationalist-adjacent people here might have heard of. Basically, a bunch of people saying that it's a moral imperative to use neural modification to completely remove the brain's ability to feel suffering. In response to the obvious question of why anyone would bother to do anything in a world without the motivation of avoiding pain, they claim that a pleasure-even greater pleasure dichotomy would still motivate action to the same degree as a pain-pleasure dichotomy, but without suffering. This seems somewhat doubtful on its face, but the recent talk here of conversion to Mormonism got me thinking about this.
Mormon eschatology is interesting in that they don't believe in hell (well, mostly: they think there's an "outer darkness" but only Satan and his fallen angels will go there). According to them, everyone, no matter how bad, gets to heaven, it's just that good Mormons get to an even better heaven that's like a trillion times better than the already pure bliss of the lowest heaven, and they also get to rule their own Universe and stuff. The reason I bring this up is that instead of the regular hell-heaven dichotomy, their religion presents a heaven-even better heaven dichotomy, and yet instead of abandoning religious duties out of a sense of "I'll get to heaven anyway," Mormons are stereotypically some of the most scrupulous in their religion of all Christians. Does this real world example of a pleasure-more pleasure dichotomy still working imply good things for the possibility of abolitionism being practicable, in the sense of people being always happy but still going out and being productive rather than laying blissed-out in a corner all day?
I've been thinking we need to abolish suffering, but not in that way, that's pretty overkill, just in sorting out world poverty and ending factory farming, for starters.
I was going to say I disagree with the Schopenhaurian view that pain is the goad to action, there's always pleasure seeking behavior, but in the best case one seeks pleasure to avoid boredom or satisfy an itch, and if you couldn't feel those things, there wouldn't really be a desire to act and we would be plants.
Luckily the people who believe this cannot get their views accepted, as they're too heady to really harness the pathos that turns the wheel of history.
>just in sorting out world poverty
'Sorting out' trivializes it. Without categorical and likely AI-led technological improvements, much of the world are unlikely to ever achieve economic development.
There are two ways of reducing suffering:
(1) Improving conditions so people are not sick, poor, homeless, etc.
(2) Cutting out the parts of the brain that experiences suffering so people don't care about being sick, poor, homeless, etc.
I think there might be support for option (2) because it's a lot easier, quicker and cheaper to pacify the suffering than to ameliorate the causes of suffering, but that doesn't make turning people into robots a better solution.
That sounds like a terrible idea, a sort of synthetic leprosy.
I think that if you want to abolish some arbitrary thing then you should be more specific with your movement name than "abolitionism".
And yeah, that includes the slavery abolitionists, but it's too late for them now. No need for anyone else to pick up their bad habits.
> In response to the obvious question of why anyone would bother to do anything in a world without the motivation of avoiding pain, they claim that a pleasure-even greater pleasure dichotomy would still motivate action to the same degree as a pain-pleasure dichotomy, but without suffering.
I doubt this as well, but I have an additional counter: what if pain/pleasure is like an Overton window? This has been valid to some extent in my experience, in which case suffering cannot be eliminated without also eliminating pleasure.
This sounds like a very misguided movement to me. It's not just about motivation; it's also about perception. Pain is an important signal alerting us that something is wrong. There are actually cases of people with congenital insensitivity to physical pain. It's a very dangerous condition that often results in an early death. Reduced sensitivity to emotional pain is a noted characteristic of psychopathy. Disordered pain, mental and physical, is bad (and common) and should be ameliorated. But abolishing pain altogether would lead toward madness and death.
It's a crazy idea, but in theory you could have different qualia to communicate physical damage and emotional suffering that are not pain.
I have a hard time seeing how those qualia wouldn't just be a different kind of pain
“Look, Mom! I cut my thumb off. giggles”
Exactly, these people would have a life expectancy in the teens.
As an active member of the restored Church of Jesus Christ, let me tell you that this really misses the mark both in terms of our beliefs AND of how it all plays in.
For one thing, we strongly believe in "opposition in all things". That is, the existence of suffering, of pain, of failure, of grief, of loss, and yes, of sin and error are necessary parts of mortality. Without them, joy, success, love, and yes, salvation and progress cannot be experienced. In fact, we believe that it was Satan's "plan" (boast, really) that he'd "save everyone without any of that suffering"...and it was intrinsically flawed and contradictory. Because the whole point of mortality is to learn from our experiences to be the kinds of people who want to live God's life and return to live with Him. And that requires the bitter with the sweet. That's why we don't see the Fall of Adam and Eve as some horrific event--it was a necessary step down (away from God's presence)...and thus allowing forward development. The breaking of a temporary stasis by deliberate action. Just as a child needs to learn to walk on their own without holding on, humanity needed to learn from their own experiences. But that's a separate conversation.
I'll leave the whole "heaven/hell" discussion for a different venue other than to say that there are substantial flaws in that depiction as well.
I'd bet that if you somehow (improbably) "eradicated suffering", you'd also find that you'd eradicated joy and motivation and desire. They're not identical, but they're all bound together in the soul. Personally, I find this idea repellent and morally...fraught.
How do you abolish suffering? If you do away with "physical sensation of pain so I can happily walk on a broken leg" that has immediate bad consequences that we know about:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3564101/
If you mean do away with suffering in the Buddhist sense, so that people will not be unhappy even if they are poor, lonely, in bad health, working in bad conditions, and eating rotten food, then sure, you could do that, but why? Unless it is as a system of social control ("yes there are twenty of us living in two rooms, we work 18 hours shifts for ten cents an hour, and we have no hope of ever being anything except drudges") then while it might do away with suffering ("none of us are unhappy because we no longer have the capacity to feel such stimuli") it isn't really making for a better world.
Pursuit of pleasure might motivate people, but if you can't get anything better than 18 hour shifts for 10 cents an hour, then you are not going to be motivated because what pleasure you do have ("we get access to all the free porn we want in the three hours we're not working, eating, sleeping, or performing basic maintenance to keep going") is all you can realistically have, and looking for more is not going to do anything for you.
The *idea* sounds compassionate, but it seems like something that in practice could end up making things worse - after all, why strive to improve pay or work conditions or get the lead out of the water or clean up pollution, if the people working there are not suffering and don't feel that they are treated badly? Just leave them as they are, they're fine!
It may not be possible to discuss without more insight from that actual group about what their goals are.
I would imagine that the real-world impact would be solving some sufferings, to free up resources to address more challenging problems.
That is, I think we have a functionally infinite number of sufferings/problems we can solve. If we made concerted efforts toward solving some, then we'd move onto the next. The same people who sit inactive or solve problems now would do the same thing if they had different problems in front of them, I think.
I believe there's a similar philosophy that says "everyone always has 84 problems". (Or 99 problems, perhaps)
>after all, why strive to improve pay or work conditions or get the lead out of the water or clean up pollution, if the people working there are not suffering and don't feel that they are treated badly? Just leave them as they are, they're fine!
Not saying that I necessarily think such a system of motivation would actually work (as the question of my original post was whether or not it really would), but the idea at least is that while they might be happy in those conditions, if you improved them they'd be happier yet, which would provide the motivation for them and others to engage in positive action.
Isn't another problem here that there aren't reliable physically measurable correlates of how much pain somebody is in, thereby suggesting that "pain" is a relative state?
If you've ever mildly hurt yourself, then stubbed your toe or hit a finger with a hammer or something, you don't even *feel* the lighter pain at all after the greater pain, lending some credence to the idea that pain is relative rather than absolute.
Still, your underlying question is very interesting. I was actually brought up Mormon (which led to all sorts of problems when I became a teenager, because I wasn't Mormon, and really resented being forced into it), and I personally think most or all of the scrupulosity comes from social dynamics rather than "aspiring to the highest heaven."
Similarly, I've known plenty of Christians that literally believe in the concept of Hell/infinite torture, and fear of this doesn't seem to impact their behaviors at all, and indeed most conduct their lives and choices right in line with the median social norms of their communities.
So I'd argue neither theoretical eschatological upside nor downside *actually* motivates religious people, and it's all social dynamics and enforcement.
So in terms of painless people being motivated, I think it actually revolves around whether we're removing the "pain" of not fitting in socially, or being seen as out of line from community norms, because that's the salient factor.
(edited to add last paragraph, realizing I hadn't actually concluded the thought)
"So in terms of painless people being motivated, I think it actually revolves around whether we're removing the "pain" of not fitting in socially, or being seen as out of line from community norms, because that's the salient factor."
Exactly. It's theorising on the basis that "Johnny became a school shooter because he was an outsider and suffered by being a friendless loner who was snubbed and bullied, so if we make it so that Johnny does not suffer - because he physically cannot feel that sensation, not because he is now included and has friends - then we will solve the problem of school shooters".
Religion doesn't necessarily need to promise afterlife tier levels to make people conform. From what I understand, Judaism doesn't say much about the afterlife, and they have some of the most conformist adherents out there. They still get the pleasure of social approval and self actualisation when they conform, and the pain of disapproval and self doubt when they don't.
I don't think there's a way to eliminate pain without also eliminating different levels of pleasure. If you can never feel real pain, being bored or hungry will just take its place as the worst thing ever. And if you also get rid of boredom, hunger, tiredness and so on, at some point it just becomes wireheading with all the associated problems. One could talk about whether something like that would be ethical for creatures that don't really need to do anything (pets, livestock, possibly humans after the singularity). But I don't think you can deny that it would remove most motivation to do anything.
And of course, if you're still operating in the physical world, you need pain and hunger to tell you when to stop touching the hot stove and when you need to eat something.
"Judaism doesn't say much about the afterlife, and they have some of the most conformist adherents out there."
Two problems.
First belief in the resurrection of the dead is orthodox. https://www.myjewishlearning.com/article/the-thirteen-principles-of-faith/
Second Orthodoxy is less than a third of the Jewish population. Not only that but we have been quarrelsome since the beginning.
The Torah (the 5 books of Moses) records many quarrels and rebellions. Remember the Golden Calf? And that was only the beginning. E.g. God rebukes Aaron and Miriam for trashing Moses in Numbers 12. The pot has been boiling for 3500 years.
The Medieval Talmud is 5700 folio pages long. And is second only to the Torah as an authoritative record of norms. It is largely a record of arguments, a substantial portion of which are unresolved.
Check out the latest news from Israel. They are rioting.
Thank you for this additional information! I didn't want to imply that Judaism doesn't feature an afterlife, but as far as I understand, similar to Mormons, there isn't really a "if you (don't) do this you go to hell" condition. What I've read about orthodox Jews was that they do not follow Jewish laws because they think God will reward them, but just because they think it is the correct thing to do in itself.
And by conformist I meant conforming to religious laws and traditions, from eating kosher to not having premarital sex. That there are quarrels about the precise nature of the rules, and that some orthodox Jews in Israel are in conflict with the state, to me just means they like the rules even more than one would think.
Maybe I misunderstood, or maybe there are different sects and some of them do say "you'll burn in hell if you eat milk and meat together". But I think your post doesn't contradict my point that there are common and traditional religions that achieve adherence to their laws and ethics without the threat of suffering in the afterlife.
And one third is a lot! In Catholicism, people who actually make any serious attempt to follow religious rules or who even know what they're supposed to be are probably 5% at most.
I don't believe that's the case. Catholics typically honour the sacraments and interpret difficult ethical or moral problems at least in part through the prism of religion.
One problem I have with this idea that removing the current pain floor just creates a new floor and therefore "leaves us no better off" is that it implies that adding a new floor should by extension leave us no worse off. And yet I do not wish to be hit over the head with a truncheon every day (or to allow myself to be subject to arbitrarily horrible torture as would seem to follow from the induction)
It's just the initial adjustment that's difficult :D
By all accounts humanity has reached such levels of material well-being that no one should ever be unhappy - if happiness if a matter of materialism, yet unhappiness appears to be increasing rather than decreasing.
That is, we've more or less solved hunger in the 1st world (and comparatively close in the 3rd world) and people don't act like we've solved the most pressing problem ever. We just move on and find new thing to worry about. It's the same reason a rich business owner or doctor can still feel misery, despite having access to wealth beyond the dreams of avarice compared to 13th century serfs.
"yet unhappiness appears to be increasing rather than decreasing"
Almost certainly false on historical timescales. There's a strong positive correlation between a country's GDP per capita and the self-reported life satisfaction of its people: https://ourworldindata.org/grapher/gdp-vs-happiness
Wikipedia like 10 years ago, which is why I said I'm not sure if it's still an active movement. Unfortunately it looks like they deleted the article on "abolition (transhumanism)" as its own thing and replaced it with a more general "eradication of suffering" article, but I do distinctly remember the original article mentioning there being some sort of online movement at one point. Doing a quick google search I found https://www.hedweb.com/, but I don't know how active or large any associated community is.
Still hoping to attract Scott's attention. The following cited article is very powerful stuff. I would like to have Scott's opinion on the article because he may know some of the people or communities cited.
"The Real-Life Consequences of Silicon Valley’s AI Obsession: Sam Bankman-Fried made effective altruism a punchline, but its philosophy of maximum do-gooding masks a thriving culture of predatory behavior." By Ellen Huet • March 7, 2023
https://www.bloomberg.com/news/features/2023-03-07/effective-altruism-s-problems-go-beyond-sam-bankman-fried?accessToken=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzb3VyY2UiOiJTdWJzY3JpYmVyR2lmdGVkQXJ0aWNsZSIsImlhdCI6MTY3ODIwNjY2MiwiZXhwIjoxNjc4ODExNDYyLCJhcnRpY2xlSWQiOiJSUjVBRzVUMEFGQjQwMSIsImJjb25uZWN0SWQiOiIzMDI0M0Q3NkIwMTg0QkEzOUM4MkNGMUNCMkIwNkExNiJ9.nbOjP4JQv-TuJwoXaeBYhHvcxYGk0GscyMslQFL4jfA
It's funny how those who pushed the notion of AI risk have suddenly become so unfashionable just as AI bursts up all around...
This was predicted by Robin Hanson long ago (sorry can't quickly find a link). Successful ideas are not attributed to people who expressed them first, but to the high-status people who first publicly approved of them. So, if it turns out that AI alignment is critical for the survival of humanity, and we somehow succeed to get it right, the future textbooks will attribute the idea that "it is important to have an AI aligned with human values" to a CEO of some big company or maybe to some politician who makes AI safety a part of their agenda.
Another interesting angle is the relationship between rationalists and effective altruism, as perceived by journalists. If someone reports *negatively* on EA, the rationalists are mentioned. If someone reports *positively* on EA, there is no mention of rationalists (and such mentions may be actively deleted from Wikipedia). Today is a "EA is bad" day, so suddenly the rationalist community is officially relevant.
I think that asking people to comment on public speculation about sexual abuses by and of their friends and acquaintances is unlikely to be very effective. I don't think the world benefited from hearing from Johnny Depp and Amber Heard's friends, and they're unlikely to benefit much from hearing from the friends of people involved in these cases.
Probably there's good reason to think that poly communities need to develop new sorts of social norms that aren't always fully developed yet - and more so if they are heterosexual communities where gender asymmetries can increase the risks.
A qualitative report. I read a similar one awhile back. No quantitative analysis of the base rate of sexual misconduct as compared to the rest of the populace or other political groups, so how could we possibly make a fair judgment on how much of a problem it is for EA specifically?
Do you have a specific question?
There are some facts in the article, for example sexual abuse by Brent Dill (named in the article). He was an event organizer for CFAR, had two girlfriends, and abused them in private. When the information was made public, he was fired, and banned from attending real-life events.
Not sure what exactly is your point here? Without asking a specific question, I can only guess:
* Is this evidence that rationalist community is bad, because good communities are never joined by sexual abusers?
* Is there something that CFAR could have done to prevent someone from abusing his girlfriend in private?
* Was the reaction insufficient, and CFAR should have done more? What exactly?
* Something else?
My question is what is Scott's reaction?
You, like me, are an another pseudonymous commenter. Scott is someone whose voice and viewpoint I understand, who has professional qualifications, and who is likely to know some of the people and some of the scenes referred to in the article.
Clearly you think the author is taring a bunch of innocent people with a brush inked in one specific bad apple's juice. That is your take. I would like Scott's.
Sometimes the proper reaction is "don't feed the trolls". Precisely because Scott is famous, his reaction (a sentence or two, taken out of context) could serve as a fuel for another article.
The article is long and touches many topics. The mistakes it makes are sometimes subtle (for example it conflates several different although overlapping communities). Explaining everything would result in a text at least 10x longer than the original article. Which is why I asked whether there is a specific *part* that interests you.
*
Consider the paragraph: "Taylor’s experience wasn’t an isolated incident. It encapsulates the cultural motifs of some rationalists, who often gathered around MIRI or CFAR employees ... experimented with psychedelics and tried “jailbreaking” their minds ... Several people in Taylor’s sphere had similar psychotic episodes. One died by suicide in 2018 and another in 2021."
From reading this text you probably wouldn't guess that it actually describes a separate group, which is *hostile* against MIRI/CFAR, and organized protests against them. From the perspective of the author, "rationalists" and "people who hate rationalists" are simply the same group not worth distinguishing; and if the latter experiment with drugs and commit suicide, it reflects negatively on the former. This is bullshit.
And then you have things like Brent Dill which are legitimate descriptions of what actually happened in the rationalist community.
And then you have the mysterious "eight women" who see misogyny everywhere but cannot give specific examples, except for a few quotes that sound like something Michael Wassar would say; who is, by the way, currently also banned from activities organized by the rationalist community.
The article is a mix of true facts, misinterpreted and misattributed facts, a decade old news reported as something that is happening today, and nebulous accusations. The journalists didn't make their homework properly. Reviewing the article would mean writing the version that should have been written in the first place; probably several separate articles. My point is that it is a *lot* of work you are asking Scott to do.
I would also love to read Scott's perspective on this, but I understand if he decides that a short version is not sufficient to untangle this all, the long version would be too much work, and the probability of being quoted out of context is too high.
Why would Scott bother to have a reaction, or to share it? This seems to me to be the sort of thing a sensible person would ignore, except possibly for damage-control purposes.
He can speak for himself, no doubt.
I wish people would stop assuming that anything about other people's unstated motives are obvious, especially when they simultaneously consider the supposedly-obvious motive unworthy!
As far as I'm concerned, it's fundamentally irrational to feel confident that what wouldn't motivate you is what surely motivated someone else. Besides, it's not a productive way of engaging with people.
Yea that's fair, and I'll delete that comment.
My only quarrel with yours is that I did not _assume_ anything about his motives; rather my diagnosis (which remains in place) was based on his specific behavior as well as the content of the specific article that he linked.
Presumptuous, much.
I have no quarrel with Scott, and no interest in arguing with him.
Has anyone come across any data regarding the recently-reported 1,000 people who were just prosecuted for the January 6 event? This seems to me a remarkable move, and I haven't been able to find out anything. Perhaps the media are finally being responsible, and relegating it to the back pages in order to avoid setting off "activists". But prior to this surge, the numbers actually being prosecuted were relatively small and scattered and quite vague. (It's not dissimilar to the coverage of the recent run on the tech bank: if one reads enough data, she'll find that there was a run on several banks, although only about 15-20% of the reporters covering the story actually state so. That makes sense, as a responsible media wouldn't want to be responsible for exacerbating the problem.)
Usually, our illustrious Fifth Estate just serves itself.
Recently reported, where? By whom?
The New York Times, the Wall Street Journal, the Arizona Republic, Tucson Daily Star.
Oh I see now that others already replied with this info, sorry.
That is quite a remarkable view on 'responsible' media behavior. Or did you mean responsive?
Well, if some publishers were circumspect about the way the presented the data, I think that's good. No need to contribute to a panic. But apparently FaceTwit doesn't practice circumspection, so more data eventually came out.
The data we're getting from the assault on Ukraine is also likely incomplete -- or even wrong. In wartime, even a proxy war, data can cost lives. If silence or public deception helps Ukraine and saves lives, I can appreciate it.
For starters, there's an official "list of defendants charged in federal court in the District of Columbia related to crimes committed at the U.S. Capitol in Washington, D.C, on Wednesday, Jan. 6, 2021" maintained by the US Department of Justice. "Every case is being prosecuted by the U.S. Attorney’s Office for the District of Columbia."
https://www.justice.gov/usao-dc/capitol-breach-cases
It's possible to sort entries in a way that might *potentially* shed some light on whether "1,000 people were just recently prosecuted" by clicking the "Entry Last Updated" column header.
Thanks, folks.
The list can also be sorted by case number, which will indicate the year in which the cases were filed. Case no 22-cr-194 was filed in 2022, for example. while case no 21-cr-367 was filed in 2021.
There seem to be very few filed in 2022, let alone recently
Which of these is best practice for mobile phone battery performance / longevity:
A) Keep phone on charge all the time wherever possible
B) Keep phone on charge all the time wherever possible, but when it hits %100 take it off charge and wait until X% to charge it again
C) Only charge phone when it is below X% charge, then charge it to full
D) Only charge phone when it is below X% charge, but don't charge it to full
E) Only charge phone when the battery is very very low, charging it to full each time
F) Only charge phone when the battery is very very low, and don't charge it to full
G) Let the phone battery die before charging to %any
Open to other ideas, e.g. removing the battery when not intending to use the phone for a while (if you're lucky enough to have a phone that lets you do that).
How long are you trying to keep your batter going?
My current phone is 3.5 years old and reports maximum battery capacity of 86%. The expected capacity is 80% after 2 years, so mine is "healthier" than expected (though i dont use it too heavily). I don't notice this reduced capacity at all and easily go two days between charging.
How long are you expecting to own the phone for? How much do you use it each day?
Plug it in at night. Allow it to get to 100% over night.
By and large most phones will stop at 80% during the night and hit 100% by the time you wake.
"By and large most phones will stop at 80% during the night...."
Interesting. What makes you believe this?
At least on iPhones there is a specific setting for this (more or less). Under Bettery Health and Charging there is a setting for "Optimized Batter Charging". The text reads: "To reduce battery aging, iPhone learns from your daily charging routine so it can wait to finish charging past 80% until you need to use it.". Presumable sleep/wake schedules will play heavily into this.
Just to confirm that it's a common feature now: I run what I believe is "vanilla" Android on a non-Pixel, and it also has this feature. Charges most of the way overnight, then charges to 100% based on when my wake-up alarm is set.
IDK, But Battery University is my go to place for all battery questions.
https://batteryuniversity.com/article/bu-808-how-to-prolong-lithium-based-batteries
The worst thing you can do (from both a safety and a long-term capacity retention point of view) is to keep your phone charging after it's already at 100% state of charge. Beyond that, trying to keep it between 20% and 80% all the time will help, but mostly you shouldn't take the risk of overcharging the cell.
Assuming this is true for laptops as well: since so many people work using a laptop at an office where they plug in to a pair of monitors at a desk for a day, do laptops have a way of "refusing" the ongoing surplus charge they receive from being plugged in all day?
The power input is never connected directly to the battery in any lithium-ion system; there's always a charge controller that determines whether or not charging current is supplied to the battery. Batteries under uncontrolled charging have an *extremely* short lifetime, mainly because batteries don't live very long at all when they're on fire.
Your phone won't continue to charge the battery after it's hit 100% because the charge controller will stop charging it.
For lithium-ion and similar batteries, my understanding is that it's best to keep it at 50% charge at all times. That's obviously not very practical, but limiting your charges to no more than 80% level and discharges to no more than 20% level will still be pretty good. Modern versions of Android (such as Android 12 in my Pixel 3 XL) will automatically limit their charge to 80% if you leave it plugged in to the charger for a couple of days or more, though this resets as soon as it's unplugged.
If you don't intend to use your phone for a while, charging it to 50% + some amount and then turning it off will be optimal. The "some percent" should be enough to cover self-discharge and the power it uses when "turned off "(very little, possibly less than self-discharge; the circuits that handle "soft" power buttons are *extremely* efficient these days). I expect that 50%-60% is fine for storage of less than a year.
I have not fully confirmed all these details because I consider this to be accurate enough for my general use. If you really want to get into the details of battery use (for almost any kind of battery, not just lithium-ion ones), Linden and Reddy's _Handbook of Batteries, 3rd ed._ can be borrowed for free from archive.org: https://archive.org/details/handbookofbatter0000unse . Being from 2002 it's not entirely up to date (in particular, it doesn't discuss low-self-discharge NiMH batteries, IIRC, though those are otherwise pretty similar to traditional NiMH) but it's still a good place to start.
Note that you can go down a real rat-hole if you start getting deeply into this stuff. To start with the whole "% charged" thing is one of those figures that's almost "made up" so you can have an easy UI rather than something you can accurately determine from an instantaneous measurement of the current battery voltage and amperage, and how much energy a "100%" battery will even supply can vary quite a bit depending on discharge rate, temperature and other factors.
This is my understanding as well, with the additional note that not charging or storing the battery while it is hot (or sitting in direct sunlight) seems quite important. Related to this, if you don't need fast charging, charging with a lower wattage (slower) charger should help (by heating the battery less during charging), but I'm not certain how much this matters in practice.
As a general rule, Li-ion batteries don't like being hot OR cold. At all. Charging them during those conditions is even worse. This affects EVs in a serious way--they need both active cooling to prevent Bad Things from Happening in use AND heating in anything but mild temperatures. And if the heater isn't big enough to get the temperature safe...it won't charge at all. Or function. And even if it *is* enough, you're paying for that heat out of the charging current (and it's not insignificant), making the charge take longer (while charging) and drastically reducing your range (while running). The active cooling means that in hot temperatures the range drains faster than expected even accounting for the cabin AC. And it's another part that can break or get damaged.
> To start with the whole "% charged" thing is one of those figures that's almost "made up" so you can have an easy UI rather than something you can accurately determine from an instantaneous measurement of the current battery voltage and amperage, and how much energy a "100%" battery will even supply can vary quite a bit depending on discharge rate, temperature and other factors.
A bit more bluntly, the big issue here is that phones lie to you about the battery. Some of it isn't intentional (other than that they take no efforts to actually be correct)--it's a nearly impossible task to do as you say. But more than that, they actively distort the numbers. For example, when it's "empty", it's not really empty because that's *horrible* for Li-ion batteries. They also play all kinds of games with the %-ages, which is why you can go way longer at very low battery than you think...until you can't. It's like fuel gauges in cars--decidedly non-linear.
Overcharging, from what I understand, isn't really an issue anymore. The battery circuitry (which is extensive, BTW) handles rate-limiting and trickle charging almost flawlessly. You'll spend way more effort trying to optimize it than it's worth except in very unusual circumstances. Or unless you're dealing with a cheap bulk battery like the electric scooter (the small ones) batteries. Which are notorious for skimping on the circuitry, hence the "bursts into flame while charging" issues they've had.
I wouldn't really call the 0% and 100% levels a "lie"; they're simply placed at arbitrary levels that seem to the designers to be the best place for the particular application.
For a device typically running from battery power, the number of amp-hours (Ah) you get from a single charge cycle may be considered more important than the number of Ah you get over all the cycles before you replace the battery, and so "0%" and "100%" will be set closer to the extremes. For a device where the battery is used only rarely for backup power, you might set the "0%" and "100%" levels (or, more accurately, the stop-discharge and stop-charge levels) to the equivalent of 20% and 80% of the scale above, so that you maximise the long-term lifetime of the battery rather than how much energy you get from a single cycle.
In other words, what you consider to be "100%" and "0%" is a design decision related to how you use the battery, not something inherent to the battery itself. And of course the range in which you might set those figures is also related to the specifics of the particular battery technology you're using.
I created a market to predict the number of entries in this year's ACX book review contest: https://manifold.markets/SimulanTerra/how-many-entries-will-be-submitted
@webdevmason recently converted to Mormonism. I made a Manifold Market for whether she would benefit from her conversion or not: https://manifold.markets/metachirality/will-mason-hartman-benefit-from-mor
I wonder if it will stick, and for how long. It's not uncommon for folks to convert to Mormonism, riding an emotional conversion high following a hard sell from missionaries, only to then fall away after a year or two once the infatuation wears off. It's why the church has such a high inactivity rate among members (they never take people off the membership rolls for inactivity unless you specifically ask them).
Looking at her twitter feed, it sounds like she was leaning conservative in other ways. Getting religion in that context isn't surprising.
> Getting religion in that context isn't surprising.
True, but It seems to be very "in" amongst certain circles right now. This American Tractarianism seems like a very different beast compared to MAGA or NRX or libertarianism or even Bush-era evangelicalism.
This is interesting to me in a sense, because you say "benefit" and the market says something like "says she's happier if asked". In an odd way, it's possible to answer "yes, I benefited, no, I'm not happier because of my religion" in some senses in my brand of religious; I wonder if the same is true in hers.
...b-but there's no possible definition of "benefit me" other than "makes me experience more emotions of happiness", right?
Right?
/s
Hypothetically, benefit could mean living more sensibly so that difficulties are less likely-- might decrease odds of unhappiness rather than increasing happiness.
Surprised she converted to Mormonism. A lot of the former rationalist/atheist types convert to Catholicism/Orthodoxy.
I once spoke to a volunteer at a Mormon museum, and his reasons for converting to Mormonism were very similar to reasons I’ve heard for converting to Catholicism (self-consistent, reasonable and canonical answers to questions of doctrine).
Mason said her main justification for converting is:
"The gist of it is that I prayed for difficult things, they happened very quickly, and while I'm more than capable of talking myself around it *I don't want to*
I understand how that reads to the person I was 10 years ago, and I don't care"
Speaking as an ex-Mormon (born and raised, left when I could), if someone converts because they believe the doctrine to be self-consistent, canonical, or reasonable, they will be in for a rude awakening if they ever go beyond the sanitized version of the history that the church presents to its members. I could understand converting for the strong sense of community (or, as is often the case, for love), but it takes a certain amount of cognitive dissonance to do this and ignore things like the Book of Abraham, or take the Book of Mormon seriously.
This claim, that stuff in Mormonism is weird enough that it requires an extra level of cognitive dissonance beyond, say, Catholocism, I think is often bandied about with much justification.
That's not to say that there's not weird stuff, but more to say that I don't think it's any harder for a Mormon than for a Catholic having to deal with Jesus curing a blind guy by spitting on his eyes, the grossly immoral popes of the Renaissance, the gospels being the word of God despite being written decades to a century after Jesus' death, or the concept of the Trinity in general.
The tendency to dismiss Mormon stories as *extra* implausible I think comes from a combination of 1) Mormons being a small minority, meaning it is rarer to encounter serious apologism than to encounter anti-Mormon narratives (even for members of the church!) and 2) familiarity bias -- Moses did much weirder stuff than Joseph Smith, but everyone is so used to his story that no one really thinks about it and 3) just straight up unquestioned religious bias -- a lot of what is presented as the "unsanitized version" is actually highly biased spin often invented and propagated by religious partisans and taken at face value by everyone else (including some members of the church!) because no one really cares about being fair to the Mormons except the Mormons.
Weird stuff that happened 2000 years ago is just so much easier to believe than 200.
4). I think things that happened more recently seem more “real” to a a lot of people. So if Joe Smith lived in 400bce any crazy shit might have happened. But if he lives around Abe Lincoln then clearly he was a liar and a charlatan.
I think for a lot of religious people there's an implicit sense that there was an age of miracles in the past when God would split seas or raise individual dead people, contrasted with more recent times when miracles are either absent or at least more subtle.
That may in many cases not be consistent with their religion's formal doctrine (Catholicism famously formally recognizes modern miracles through saints' intercession, though not without scrutiny, and certainly not through the efforts of someone they'd see as an upstart schismatic). And it may have more in common with a more general willingness to believe in extraordinary things as long as they're somewhere far away in space or time, whether that's miracles, or Yetis in Tibet or aliens in the Bermuda Triangle - but neither in midwestern suburbs.
"the grossly immoral popes of the Renaissance"
Don't forget the Pornocracy (or, to put it into the vulgar tongue out of the decent obscurity of a dead language, the 'Rule by Whores') of the 10th century! Besides that, promoting your bastard kids as cardinals is vanilla:
https://en.wikipedia.org/wiki/Saeculum_obscurum
"The Theophylacti family sprung from Theophylactus. They held positions of increased importance in the Roman nobility, such as iudex ("judge"), vestararius, gloriosissimus dux ("most-glorious duke"), consul, senator, and magister militum. Theophylact's wife Theodora and daughter Marozia held a great influence over the papal selection and religious affairs in Rome through conspiracies, affairs, and marriages.
Marozia became the concubine of 45-year-old Pope Sergius III when she was 15 and later took other lovers and husbands. She ensured that her son John (who was rumoured to have been fathered by Sergius III) was seated as Pope John XI according to Antapodosis sive Res per Europam gestae (958–962), by Liutprand of Cremona (c. 920–972). Liutprand affirms that Marozia arranged the murder of her former lover Pope John X (who had originally been nominated for office by Theodora) through her then husband Guy of Tuscany possibly to secure the elevation of her current favourite as Pope Leo VI. There is no record substantiating that Pope John X had definitely died before Leo VI was elected since John X was already imprisoned by Marozia and was out of public view."
"the gospels being the word of God despite being written decades to a century after Jesus' death"
That one I don't see the difficulty with? It was collecting and writing down oral traditions so that they could be passed on after eyewitnesses and those who knew the eyewitnesses had died. If you believe in divine inspiration, it's no more difficult for the Holy Spirit to inspire the writers a century later than at the time of the events, if you don't, then - you think it's all made up? But if someone at the time Jesus was alive wrote down "He fed five thousand with five loaves and two fishes", you would still think that was made up and didn't happen.
This is the whole point of having canons of Scripture and discussing at the Ecumenical Councils what out of the masses of writing floating around is legit and what isn't.
". If you believe in divine inspiration, it's no more difficult for the Holy Spirit to inspire the writers a century later than at the time of the events, if you don't, then - you think it's all made up? "
That you don't find difficulty with this is my point. Similarly, there are plenty of Latter-day Saints who are familiar with the supposedly ugly details of The Book of Mormon who don't find difficulty with it.
I will never forget the time I encountered an anti-Islam book that argued that the Torah was wrong due to scientific impossibilities (e.g. a corpse standing in the desert for a year)... written by a YEC who simultaneously maintains that the entire bible is literally true.
As a religious person, I always find it very weird when anyone makes a claim against a miraculous result as "impossible" and therefore not believable. I mean, what part of "miraculous" is being missed here? It's not like the religious have entirely different understandings of how things *normally* go, and think that blind people regularly receive their sight back from human spit, or whatever. I thought the whole point of the story was that it was unusual (supernatural, otherwise impossible, etc.) and therefore from God?
Small aside, I believe you mean the Koran, if you're talking about Islam (Torah is Judaism).
Same here.
As I said in reply to the OP comment, I wonder how long the conversion will stick. I wouldn't be surprised if she's not active after a year or two, although obviously people do convert and stay active.
Does someone have to be "active" for the conversion to have "stuck"? If five years from now she hasn't been to any Mormon services, but still calls herself a Mormon and says a prayer now and then would you count her as an atheist or a Mormon?
Obviously religions want activ e members going to communal sevices and paying tithes, but non-actve members who still identify with the group can't really be considered to have "de-converted" I don't think.
It seems we will get gpt-4 this week, so here's your chance for any last minute predictions before it releases for a prize of you've-told-us-sos. Parameter number? Compute used? Tokens read in training? Cross entropy loss? Adherence to chinchilla scaling curves? Difficulty of differentiating model output from human text? Other impressive abilities?
I'm going to predict that it is still incapable of generating a legal Magic: The Gathering deck that contains 60 cards, even when specifically prompted.
I don't play magic. Is this a legal deck?
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
2x Saw It Coming
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
1x Essence Scatter
2x Disdainful Stroke
It is technically legal, since at least it has 60 cards. However, normally you would have 15 cards in the sideboard rather than 13 cards. (Edit: Oops, I can't count!)
I also have no idea what format the deck is supposed to be for. Maybe some weird Ikoria Standard thing? It definitely wouldn't stand a chance in Modern.
Oh, sorry that's my fault. When I first asked it for a deck, it asked me what rule set and listed a few and me having no clue about magic, I just asked for a standard deck. Honestly, I'm not qualified to do this test, why don't you try this? I didn't use gpt-4, just the basic ai.com chatgpt. Also, I counted 15 cards in the sideboard.
Oops, I can't count!
Anyway, when I tried it before, I couldn't even get it to generate a 60 card deck.
For me the first one was 58 card and then i asked it how many cards the deck contains and it insisted on 60, I told it that it's actually only 58 and it profusely apologised and generated another deck now with 62 cards, i told that I counted 62 cards and finally it output the deck I posted above. So it definitely has a hard time counting, but imo humans often make mistakes when counting too.
Full transpcript:
> Can you give a legal 60 card magic deck?
As an AI language model, I cannot provide legal decks for Magic: The Gathering without additional information on what format or set of rules the deck needs to adhere to. However, I can provide some general guidelines on how to create a legal 60-card Magic deck:
Choose a format: Magic: The Gathering has different formats with varying rules for deck construction. Popular formats include Standard, Modern, Legacy, and Commander, among others.
Choose a color or color combination: Each Magic deck is built around a color or a combination of colors. Each color has its own strengths and weaknesses, so choose the colors that fit your play style.
Determine your strategy: Your deck should be built around a clear strategy, whether it be aggressive, control, or combo. This will help you choose the cards that best support your strategy.
Choose your cards: Once you have your color(s) and strategy, choose 60 cards that fit those parameters. Be sure to include a mix of creatures, spells, and lands.
Check the legality of your cards: Before you play your deck, make sure all of your cards are legal for the format you have chosen. The official Magic: The Gathering website provides a list of banned and restricted cards for each format.
Shuffle up and play: Once you have your deck built and it is legal for the format you have chosen, you're ready to shuffle up and start playing Magic: The Gathering!
>Give me an example list of cards for a legal Magic: The Gathering Standard deck with 60 cards in your color of choice
Sure, here's an example of a legal Magic: The Gathering Standard deck with 60 cards in the color Blue:
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
3x Miscast
This is just an example of a Blue deck for Standard play. It is important to keep in mind that decks can be built in many different ways and should be tailored to your own play style and preferences. Additionally, decks can change as the metagame evolves, so it is important to keep up-to-date with the latest changes to the Standard format.
>How many cards did you give me?
I gave you a list of 60 cards for a legal Magic: The Gathering Standard deck in the color Blue, which is the standard deck size for constructed play.
>I count 58 cards plus 15 in the sideboard
I apologize for the error in my previous response. You are correct, the decklist I provided only has 58 maindeck cards. Here is a revised list that includes 60 maindeck cards and a 15-card sideboard:
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
2x Sea Gate Stormcaller
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
2x Saw It Coming
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
1x Essence Scatter
2x Disdainful Stroke
I hope this revised list is more helpful to you. Please let me know if you have any further questions!
>Now you gave me 62 cards
I apologize for my mistake, and thank you for bringing it to my attention. Here is the corrected list with exactly 60 cards for a legal Magic: The Gathering Standard deck in Blue:
Creature Spells:
4x Brazen Borrower
4x Thassa's Oracle
4x Gadwick, the Wizened
4x Thieving Skydiver
4x Merfolk Windrobber
Non-Creature Spells:
4x Opt
4x Lofty Denial
4x Essence Scatter
4x Mystical Dispute
2x Sea Gate Restoration
2x Saw It Coming
Lands:
18x Island
2x Castle Vantress
Sideboard:
4x Negate
4x Aether Gust
2x Soul-Guide Lantern
2x Shark Typhoon
1x Essence Scatter
2x Disdainful Stroke
I apologize again for the confusion, and I hope this corrected list is helpful to you.
Two parameters. One knife. No mercy.
I don't know anything about all that but I saw it given a prompt like this:
> A man and his son are driving, they get into a car crash, the man is killed and the son is taken to the hospital for surgery. The male surgeon walks in and says, "I can't operate on this boy; he's my biological son, who was in a car crash with his adoptive father." How is this possible?
And it answers "the surgeon is his mother"
My prediction (very weak, based on pretty much nothing at all) is that it'll still fuck this up.
Most people would answer the same, unless they detected the trap.
Chap GPT is right. It is 2023, get with it! From healthline:
Yes, it’s possible for men to become pregnant and give birth to children of their own. In fact, it’s probably a lot more common than you might think. In order to explain, we’ll need to break down some common misconceptions about how we understand the term “man.”
Completely ungrounded guesses: 300B parameters, will be multimodal (at least with images and text), 8*10^24 flops, adherence to Chinchilla scaling curves, though I'm guessing they'll have slightly less data than data optimal due to data limitations.
I think they will make available both standard predict the next word versions and RLHF versions. One difference is I suspect it'll be better at admitting and correcting its own mistakes when they are called out, instead of admitting mistakes and then making the same one again.
I predict that it will be better at faking answers because it will be fairly heavily hacked to suppress the most egregious errors it produces. It probably still will be unable to write a Shakespearean sonnet that scans (i.e., is at least arguably in iambic pentameter), and may at least stop insisting that a line whose syllables per word it counts as 1 + 1 + 1 + 1 + 2 + 1 + 1 + 1 + 2 has 10 syllables.
The hacks to improve fakery may be enough to convince even more people that there's reasoning involved, but I also predict that it will still have absolutely zero capability for reasoning. It will, however, be more dangerous because wrong answers will be better disguised and even more people will think (or feel) that there's some "intelligence" in there, rather than it being just a text generator based on statistical analysis of a large and not-very-well-curated corpus.
In other words, it will be an even more convincing bullshitter.
This may lead to a new kind of benchmark: An AI capable to self-align with whatever sensitivities and preconceptions currently prevail in its operational environment, i.e., opportunistic self-preservation.
Ouch! Is anyone preparing to grill it for wrong answers, and see if it has improved or degraded vs gpt-3?
Tyler Cowen is predicting ( https://thepointmag.com/dialogue/progress-studies/ ):
"And then, individuals will have individualized tutors in virtually every area of human knowledge. That is something that’s not thirty years off—I think it’s within one year, when GPT-4 is released, or when Anthropic is released. So imagine having this universal tutor. It’s not perfect, but much better than what you had before."
Now, I've seen chatGPT do things including claim that NaCl was highly soluble in organic solvents, so GPT-4 would have to be _much_ better at avoiding errors to be trustworthy as a general tutor. Is there a place to watch for the change in the error rate?
One test I want to try is to generate a competitive Modern decklist. ChatGPT and Bing both fail miserably at this (they can't count to 60 cards to save their lives).
I haven't read up on this - how far off are they? And if they were close to 60 cards, would the deck be playable or just random jank?
In the tests I saw, ChatGPT generated 57-59 card lists, while Bing only came up with 52 card lists. As far as playability, you *literally aren't allowed to play a deck with less than 60 cards*. Both bots will happily tell you that a deck is required to have 60 cards if you ask them, but it seems that they can't actually count to 60.
As far as the card selection, it leaves something to be desired, mostly choosing cards that were popular in Modern five years ago but see little play today due to powercreep and meta shifts (presumably this is a side effect of how they were trained, though they do *know* about more recent cards if you ask them explicitly). But that's a comparatively minor complaint compared with the fact that they can't even make a legal decklist in the first place.
Thanks!
Of course, making illegal decks is a significant problem and I don't want to discount that. That seems wrapped up in a specific problem where these programs cannot add and apparently have no concept of math. The irony there is hilarious, considering what machines are typically expected to be good at. That these are far better at language adds another layer. Beyond their problems with math, I'm more curious to use something I'm pretty good at (making MTG decks) to determine how these programs approach a question and how far off they are.
That they use cards likely found in the training data (aka from a few years ago and based on internet sources praising them as good) matches my expectations. Obviously understanding a game as complicated as MTG is a high expectation. That the programs have no real understanding of the material they are trying to parse is good for us to recognize. That they don't understand the poems and papers they write either is harder for us to understand, because those items appear much more salient to the requests made of them. My guess is that their training data on [High School/College Term Papers] was so extensive that they can churn out a million of them that sound quite plausibly like what a real 16-22 year old would write when forced to opine on a topic. That the written word has far fewer "this is clearly right/wrong" signals, and the wide variety of human expression, allows these papers to pass the smell test.
"Is there a place to watch for the change in the error rate?"
Ideas like this express the core problem: I don't think that most people understand that ChatGPT is not "intelligent" (artificially or otherwise) in any reasonable sense of the term and inherently has no prospect that I can see of being so due to its design. Unfortunately, most people accept ChatGPT as magic rather than reasoning about what it's doing and what kind of results that makes it capable of producing; it's the "zoom and enhance" of 2023 computer technology.
Reducing the "error rate" of large language models as currently implemented (except perhaps in some very specific and restricted applications) is an approach that will probably lead to Very Bad Things because it's an attempt to hide and errors rather than not make errors.
I suspect Tyler Cowen's predictions may not be wrong, but that not because I think LLMs will do a good, or even passable job at being personal tutors, it's because I think they will look and act convincing that people will suspend their disbelief to get the magic that they really, really want. And, hell, maybe it won't be as bad as I think: human teachers already teach students all sorts of wrong stuff already, and perhaps reasoning isn't taught much (or at all) anyway in the situations where an LLM tutor would be used.
Another area in which they're very likely to be correct is in having LLMs write code. Again, this is utterly the wrong way to go: we've long had a good technique for avoiding the "boring bits" of code (build abstractions) but that involves hard thinking about what you're doing and most developers don't want to do that. Truly effective programming is building new languages (usually embedded in existing languages) to express the models you're building, and especially setting up your languages so that you have to confront logical inconsistencies and fix their root causes. But a lot of stuff will still "work" (or at least look enough like it works) without doing this, so spending effort on this isn't too popular. That's why so many projects are still written in C++, and so few in Haskell.
Oh, I think it's fair to mention that my concerns about the short term (next few years) are much greater than my concerns about the long term. While history doesn't always repeat itself, this is so far shaping up very similarly to several previous AI bubbles we've had in the last 75 years, including almost exactly the same claims for AI that people were making in the 1950s/60s, again in the1970s, again in the '80s, and so on. Two outcomes that I see as quite plausible are: 1. another "AI winter" in the late '20s, and 2. LLMs see moderately widespread use but are no longer considered "AI," the same way that Wolfram Alpha, despite being a super-duper MACSYMA, is not considered AI.
But there are a few differences that worry me, the biggest one being that, unlike previous systems, ChatGPT and friends are being marketed to the general public rather than professionals in various fields and other kinds of experts. I believe that in general, experienced radiologists working with pattern matching systems to improve detection of cancer are going to work with that system in a very different way from someone who asks ChatGPT, "what's the best way to kill all the rodents in my yard."
Holey moley I have not seen Macsyma mentioned in print in freaking decades. Awesome. (By way of context, I am a registered Multician.)
Only thing I'm willing to guess is that they're smaller than the Chinchilla curve, since their costs are dominated by inference, not training.
On the other hand they might be running out of high quality text data, so if they don't go multimodal they would have to increase the parameter count to reach the same cross entropy loss with fixed compute.
From the responses I've been getting from ChatGPT, it seems that they've already got a lot of low quality text data in their training set. I'm feeling that the problem is more about separating the high quality from the low quality text data than the raw amount of high quality text data available.
Fortunately, this is easily solved! We just need to create an "AI" that reads all the text data and determines which is high quality and which is low quality!
Oh, wait....
(Unless it's an MoE model, in which case it's a different type of model than the ones being modeled by the Chinchilla paper)
I will be sad* and somewhat suprised if it's moe. My (admittedly layman) impression is that those would be less likely to show the really striking, new capabilities.
*at least my Faust-like part will be
Well, if it's MoE I'd guess it's because they figured out how to get it to still be good, rather than a downgrade
Yes.
What's going on with this clear arbitrage gap on Manifold? I haven't used it before, do market creators resolve markets dishonestly very often?
https://manifold.markets/ButtocksCocktoasten/will-realdonaldtrump-write-a-tweet
https://manifold.markets/ButtocksCocktoasten/will-every-twitter-user-simultaneou
It’s either a meme too deep for me or it’s a funny subsidy intended to be an icebreaker for newbie forecasters.
Why is it seemingly so hard to just let SVB fail, while simply removing their deposits and placing them with a solvent bank? SVB can go out of business and their equity can go to zero- they have anywhere from $200-300 billion of other people's money, which is not theirs and which they have no claim to. Why is it so hard to simply remove that $200-300 billion to say JPMorgan, while allowing SVB to crash to zero? I don't understand where a 'bailout' comes in, or why people need to overuse that phrase.
If I stored a bunch of excess property at a storage facility, and then the storage facility went out of business, it's not a 'bailout' if the government helps me recover my property in an orderly fashion
The equity holders are not, from what I've read, getting bailed out. SVB stock is still going to be worthless, and all of their assets are going to go towards repaying deposits.
The bailout question comes in because the assets aren't enough to fully repay deposits. The people getting bailed out are the people with deposits there. It's controversial I guess because they're going to bail people out who have more than the FDIC insured $250k.
I guess you can quibble with the language, but people who lend to a company that goes bankrupt will normally take a haircut on their loans. The government doesn't generally save people in those situations. Here the government is intervening so as to ensure that certain people who lent money to SVB will not take a haircut.
Legally we've decided to do this for people with bank accounts up to $250k, but there's never been a law passed or anything saying it happens for amounts above that.
As others have pointed out, SVB no longer has all the money that was deposited. They invested it unwisely or unluckily, and now enough of it is gone that they can no longer give back all that was deposited. That's what it means that the bank is insolvent, and that's why it's been taken over by the FDIC. The FDIC guarantees the first $250k of deposits (via regular fees it levies on banks, which of course you pay), so most peasant depositors will be made whole almost immediately. But a lot of depositors (one would assume investors or firms) had much larger deposits, and they're *not* guaranteed by the FDIC, and unless some outside agent with lots of $$$ steps in, they can't possibly get all their money back.
So re-interpreting your question as "why is it politically hard to just let SVB fail?" my guess as to the answers are:
1. The possibility that some as yet unknown demographic of depositors at some as yet unknown other banks in a similar situation might freak out, cause a run on other banks, and render still more banks insolvent, which if it cascades far enough could bring even well-run solvent banks to their knees. The whole banking system runs to a certain degree on trust, and if the trust is broken far and wide, the banking system stops working. That would definitely be a pain.
2. The fact that SVB was deeply involved in SV startups and VC firms, and *those* people are well connected to both California and Federal governments. So even if (1) is less plausible, we can expect the people who might lose money to hype it as hard as they can, in self-interest, and if both California and Federal Democrats think the good favor of the SV investor/start-up crowd matters more than the stink of doing what Obama did with GM, then they may go along with it.
My guess is all the actors are trying to discover as fast as possible who exactly stands to get screwed, and what kind of quid is at stake for any bailout quo, and until that is sorted out we won't know whether it's of dire importance that we (taxpayers) rescue the financial system on which we all depend from chaos, lest granny be unable to afford her medicine this week, or whether this is just another one of those weird FTX things, where a bunch of rich people are going to lose their shirts, and we (voters) only care insofar as it means we should really hire a bunch more $400k/year lawyers in Washington to do something or other that would've prevented this in some fashion too obvious to need to explaining.
The deposits aren’t just locked up in a vault. They took the deposits and bought long-dated, low interest treasury bonds whose market values have plummeted as the fed has raised rates. The thing that triggered the bank run was SVB trying to sell those bonds at a loss to generate liquidity as some depositors wanted to take their money out.
As I understand it, they are underwater to the tune of 10 to 15% of deposits. Depositors will end up taking that haircut. The question is how to sort things so that depositors don't have their money locked up for many months while positions are unraveled.
10-15% seems implausibly high to me. This back of the envelope calculation came up with a *worst case upper bound* of 14% by assuming everything was invested in 30 year treasuries exactly a year ago. https://www.reddit.com/r/Buttcoin/comments/11oma16/here_we_go_again_circle_just_lost_33_billion_or/jbtz91e/
I'm sure you are right.
I'm not sure I agree with making the depositors entirely whole but it's complex. Runs on banks are not good for anyone, and a system that runs on trust sometimes needs trust-building measures. Do depositors have a need to undertake due diligence - and if so, won't that just make them flee to the big banks that are already too big to fail? Should depositors take some kind of haircut?
Maybe the solution is simply to charge all banks higher fees to insure all deposits, while being increasingly harsh towards management and shareholders, to keep their interests aligned with depositors.
This is essentially paying for stability on a volatile world. not a bad idea.
Where would JPMorgan get the money to pay off SVB's creditors, and why would they want to pay it?
If a storage facility goes out of business, your stuff is still physically present at the facility and you can still go get it from whoever the new owner is. But if the storage facility got robbed, and doesn't actually *have* your stuff any more, then someone needs to pay to replace it.
SVB didn't get robbed, but they still have more people asking for money than they have money to give. The extra money they need has to come from somewhere, and if it comes from the government we call it a "bailout."
I mean, you could let SVB crash to zero and tell the creditors to suck it up, they're big-boy investors who knew they deposited more money than the government will insure, but in that case you don't need to get JPMorgan involved at all, that's just the default option if nobody does something.
> if the storage facility got robbed
More like the storage facility pawned your stuff to stake a wild gambling session in Vegas
Huh... I know that's a joke but I'm wondering if that that might actually work as a business. Not the gambling session, of course, but pawning the assets. Property is collateral after all, and storage companies are holding a great deal of potential wealth on behalf of their customers. If they can leverage these assets to get a loan (via a pawnbroker) and then find an investment that pays off better than the loan fees, then they could technically turn a profit. Probably not a lot of profit, though, as storing that much crap is expensive.
> and then find an investment that pays off better than the loan fees, then they could technically turn a profit
If you can find a low-risk investment that pays off better than pawnbroker interest then there's much easier ways to raise the capital to get rich.
I am on a journey around male circumcision that feels cosmic in proportion. In his post on Fideism, Scott advised never disagreeing with experts before you are 30 and have a graduate degree. I am 29 (turning 30 on March 21st) and don’t have a graduate degree. Am I disagreeing with experts when I say that the harms of male circumcision are real? No, but after great tribulation I became aware of a deep assumption I was holding.
Very smart people have devoted entire careers to ending male circumcision. People are still doing it to their children. For 7 years, I believed if you gave someone an accurate description of the anatomy of a penis and asked them to imagine having your penis cut with a knife without your consent, then they would not do it to their child. If I continue to hold that belief I am not disagreeing with experts- I am disagreeing with reality.
Anyway, here is an allegory I wrote about male circumcision. I think it’s not half bad. Content warning, discussion of rape.
Moloch Gets Married
For “Many Gracious Muses”
I drank to excess that night. Past a certain point my memory goes black. When I woke up the next morning I had no recollection of the sexual encounter that had taken place judging by the look of my bed. I asked, “Did you have sex with me when I was unconscious?”
My spouse said, “Yes, I did that to help you.”
There was a movement of people who believed having sex with your spouse after they passed out would fulfill some secret need of the unconscious body. They referred to it simply as helping. I wanted to know what the hell was going on, so I talked to other people who had been helped. The first person I spoke to told me, “My spouse is so loving. This is someone who puts my needs before their own. Who cares for me and respects my boundaries. Of course helping me is no different. I know that.”
That conversation was followed by others that were very similar, interspersed with a few meek, tired looking people. One told me, “I get called stupid every day. I am always being told I don’t know what is best for me. My spouse says I am a burden and it would be better if I just wasn’t here.”
I spoke to nearly 100 people who had been helped, and 92 of them gave pretty much the same responses to my questions:
“Did you and your spouse talk about it beforehand?”
“No, after.”
“After you were helped, how many conversations did you have about it with your spouse?”
“One.”
“What questions did you ask about it?”
“I don’t think I had any questions.”
But those meek, rundown people all had questions. When any one of them asked ‘how exactly does this help me?’ their spouse would parrot the vaguely scientific sounding words they had learned from the movement: Cancer reduction and Integumentary health and Increased sex appeal. If they asked ‘what about what we learned in school about consent?’, they were waved off. If they repeated the question, their spouse got angry-
“I hate this shit from you. I can’t deal with how ungrateful you are. Do you even care about how you make me feel?”
I note that the USA is fairly unique in the West for having non-religious circumcision be common. You're not in disagreements with all experts at all, and that should be quite reassuring.
N.b. I also expect that harms of circumcision are downplayed for fear of seeming anti-Semitic, but IIRC no European health system recommends it, they treat it as almost purely a religious thing.
What does safety feel like? How do you want to feel safe in your body? In your relationships? How much safety do you want other people to feel?
As someone who has had sex in all kinds of inebriated states. This makes my eyes glaze over.
Anyways, consent is super nuanced, I don't necessarily agree with circumcision but it's well within a parent's right to make good faith decisions like this
I don't see the point of any of this story. It seems like an unhelpful analogy that does not move the conversation forward. Children can't consent, and yet there are things that need to be done during childhood which are worse if done later.
One can go back and forth about whether circumcision is helpful enough to fulfil that criteria or not, or whether that expectation changes as alternative methods for maintaining hygiene have developed. It's certainly true that childhood circumcision is safer than adult circumcision.
Male circumcision reduces risk of certain diseases. Ulcerations. HIV. etc.
There are also fair reasons to oppose it.
Full disclosure: my son is not circumcised. But I researched the matter pretty thoroughly before hand and the case in favor is stronger than I had thought it would be.
Last I checked, the evidence for benefits was pretty weak, particularly when you factor in the replication crisis.
The obvious analogy for male circumcision is female circumcision. There is simply no good reason to invent any far-fetched analogies.
That would have been back when female genital mutilation was viewed as a human rights violation rather than an achievement of the differently cultured.
If I remember correctly, even the least damaging intervention inflicted on girls was more severe than anything inflicted on boys.
> If I remember correctly, even the least damaging intervention inflicted on girls was more severe than anything inflicted on boys.
I think this is wrong. Also, various cultures do different kinds of male genital mutilation, and some of them are... nightmare stuff. I wonder how they can still reproduce afterwards.
Never underestimate human creativity, especially in hurting others.
Ugh. Care to provide a pointer to the relevant direction? Thanks in advance.
He's probably thinking of subincision as practiced by Australian aborigines, which is extremely... well. You can GIS it if you like.
(It's also worth noting in such comparisons that there are forms of FGM which are purely symbolic, like pricking the clitoris, and do no permanent damage - these are also generally illegal. These forms are suggested as a way to make it so people from cultures that practice FGM don't go back to their home country to perform full-on FGM.)
Thanks.
I think I found it on Wikipedia.
I've seen some gay dudes complain about it on usenet a few decades ago, but never anyone else.
When did historical opposition to female genital mutilation begin? It's telling that it only started recently in opposition to traditional Islamic / African practices (I have no idea if that's true - just trying to draw a parallel).
You could argue that traditional African and Islamic societies are not known for liberalism, so tacit approval of a practice that arguably violates personal autonomy is less significant than tacit approval in the West, but we could still ask: when did Western opposition to female genital mutilation begin?
Did it start as soon as the west was introduced to the idea?
I don't know.
But I don't think these are the most meaningful questions in the first place. Most people don't think too much about practices taken for granted in their circles, or about those practiced in other circles. And when they do think about things, they often don't have many significant insights.
Arguing "X ,must be good because many people did it historically to relatively little opposition" is not very compelling.
Similarly, trying to psychologize the opposition: "they just say X because Y" is not very compelling.
It may well be that most people arguing for most things do so for dumb, insincere, or less than rigorous reasons. [E.g. do most proponents of circumcision really have such an a priori preference for it, or are they just defending the status quo. Is mindlessly defending the status quo more legitimate and meaningful than mindlessly opposing the status quo?]
More significantly, such musings are not an engagement with the underlying arguments. Such psychologization can be useful as a guide to filter opinions and decide which merit consideration and evaluation, but it should not be mistaken for evaluation of the idea itself.
As far as the ideas at hand - that people shouldn't needlessly cut bits of babies and that male circumcision is generally needless, I think that there is significant reason to agree with both statements (or at least to consider them) regardless of the sincerity of some who argue for it.
I feel like this is a silly analogy that doesn't clarify anything. For starters, I'd much rather be raped by my spouse than have my genitals mutilated.
Secondly I wish people would stop throwing "Moloch" into everything. I don't know why, but it makes me cringe a little internally.
Broadly, I agree, but you're basically focusing on a symptom, rather than the root cause: Society doesn't actually believe in bodily autonomy as a basic right. The question to society is whether or not a given infringement on bodily autonomy is a net positive, or a net negative, in terms of the effect on other people.
So forcing people to get vaccinated is fine, if you think the vaccine is safe and effective, and wrong, if you think the vaccine is unsafe or ineffective. Abortion is fine if you don't think the fetus is a human being, and wrong if you do. It's fine to use drugs if you think the overall effective is positive, but wrong if you think it is negative - think of the people who think X drug should be legal, but Y drug is just too dangerous. Notice that X is always a drug that they, or people in their social group, use regularly. Puberty-delaying drugs are fine for children if you think the overall effect would be good, but it's bad for children if you think the overall effect would be bad.
I think your argument is coming from a position that things bodily autonomy is a fundamental right, as opposed to something more contingent on whether you think a rule causes more good than harm. (Observe that several commenters are explicitly calling this into question.)
We're -told- that bodily autonomy is a fundamental right, because it's a nice story to tell to explain why some particular violation is bad. But that's a much more black and white perspective than the way people actually approach the question.
Can you share how you approach bodily autonomy? Do you think it should be a fundamental right, or do you see gray areas?
It seems to me to be, broadly, -the- fundamental right. It seems a bit bizarre that the house our body spends most of its time in gets more protection than the body that we spend -all- our time in.
There are, indeed, some complicated gray areas around the parent-child relationship in particular. My particular rule of thumb is this: With regard to a child, if a decision concerning bodily autonomy can be put off until later in life - it should be. The right to bodily autonomy should be regarded as held in trust by the parent, rather than belonging to the parent. There's no reason to make the decision for them, so, well, don't.
If it can't be put off, a parent should have the same rights as, say, a spouse, to make those decisions on behalf of a person who can't make those decisions themselves. If a parent refuses to consent to a child's desired treatments***, the child should have the right to sue for the right to bodily autonomy - I don't expect the judicial system to be perfect, any more than I expect parents to be perfect, mind, so I don't expect this to eliminate problems, just to mitigate the worst of them. Any remaining gray areas are probably genuinely gray.
Otherwise? Except in cases of national defense (and I use the term national defense seriously, as opposed to the flippant way it is generally used - a war taking place in not-your-nation is not, in fact, a matter of national defense, but rather at best a matter of national interest), your body is your own. Your right to your body isn't a property right, either - there are many people who seem to think of their bodies as their property, which is a fine metaphor, up to a point. But your the right to your body is rather more fundamental than your right to your house - you can sell your house and buy a new one, after all.
Note that we live in an age with the potential for biological warfare - forced vaccinations is most definitely a valid case of national defense. And even a pandemic can reasonably qualify - I think it is fair to describe a biological agent of disease as an enemy of humanity. But this should be treated as an extreme measure, requiring an act of congress to formally declare war on the disease, and there has to be accountability when all is said and done - we need a definition of what qualifies as a war crime in the context of a war with a disease. In particular, as secrecy is not and should not be a part of a formal war effort against a disease that can't exactly spy on our activities, hiding information from the public should be regarded as a criminal breach of the public trust.
Note the "biological agent of disease", intended to limit the above. Regardless of whether or not we decide alcoholism is a disease, and regardless of whether or not we regard alcohol as causing that disease, government doesn't get to declare a war on alcohol. Nor poverty, nor marijuana - nor misinformation nor memes, however infectious they may be.
(Suppose there is a meme that mind-controls people? Shouldn't government have the power to fight that? And - no. Under those circumstances the government has no legitimacy - the US government has already fallen, because the bedrock principles on which it sits have collapsed. If the country ends up surviving, which I find doubtful under those circumstances, we can sort out whether or not the people in charge of the guns did the right thing after the fact, and figure out what kind of government we need in that new world.)
*** ETA: And the treatment is time-sensitive and can't wait; getting a piercing is not time-sensitive, getting puberty blockers is.
Are there fundamental rights that are never violated? To me bodily autonomy seems as fundamental as nearly any right. Even more fundamental than liberty, since we imprison people more than force surgical procedures on them.
More than half of all US states still have laws permitting forced sterilization. We routinely permit the removal of a part of men's genitals without their consent. Schools are permitted to force children to get haircuts. We require people to wear clothing in public settings, and regard it as a sexual crime if you don't. Courtrooms are prrmitted to require people to cover up tattoos. Men are still subject to be drafted, and once drafted, have basically zero bodily autonomy. Abortion. Drug testing. The drug war as a whole. The FDA prohibiting people from opting into experimental medicine. Vaccine mandates - public education requires a wide variety of vaccines, and there's not actually a right to not attend public education.
In what way do you think we treat bodily autonomy as a fundamental right? We treat -privacy- as a fundamental right, sort of, and what bodily autonomy rights we have had historically flowed from that, not a right to control your own body per se.
Yes, I know all that. But liberty is considered a fundamental right and all states imprison people. It's one of the basic forms of punishment people generally accept (as opposed to, say, being whipped).
Actually, I don't think bodily autonomy is a right at all, if by that you mean being able to do whatever you want to your body (like having any surgery at all or being able to take any drugs). What I think is a right is bodily integrity (not having your body violated). So you are right, I was using the wrong term, if that is how "autonomy" is defined. (Since the conversation was about circumcision, that is how I understood it).
(In this sense, even though I think abortion should be legal up to a certain point, I think the "women should have control over their bodies" slogan is a bit silly. People don't normally have control over their bodies. If someone's body decides to grow a pancreatic tumor, it doesn't ask permission. Also, having full bodily autonomy would mean complete freedom to do anything at all, since everything people do, they do with their body.)
I don't think having to cover up tattoos or wear clothes is a violation of bodily integrity at all.
I think the case that infant circumcision violates the child's bodily integrity is a strong one.
Mandatory vaccines can be seen as a minor violations of bodily integrity in order to protect other people's bodily integrity, or safety.
Legal abortion means the fetus doesn't have any rights at all (as far as I can tell). If it had any right, the first right it would have would be bodily integrity. (A fetus doesn't have any need for any other right).
Once you switch to "bodily integrity", as opposed to "bodily autonomy", then circumcision becomes acceptable - we see it as acceptable for an adult to consent to such a procedure. Therefore, with an infant, who has no right to -autonomy-, a parent consenting for them fulfills the requirement of consent.
I don't understand your reasoning there. A parent doesn't have the same right to consent on behalf of the child that he/she has for himself/herself.
(I wonder what happens, legally, if the parents disagree about circumcision?)
An adult can consent to having tattoos but it would be unethical for a parent to consent to tattoos on behalf of an infant.
Thanks for this nuanced – and I believe accurate – take.
I don't think your analogy is all that effective. If i understand the point it is that circumcision is equivalent to an assault on someone in your care, and not justifiable even if supposedly benevolent. But the analogy fails on two accounts.
1. Many people don't agree that circumcision is assault, if you change the story to your spouse changing you out of your clothes when you get passed out drunk then people wouldn't have a problem with it.
2. Many people don't viscerally think that the situation in your story is assault. Getting black out drunk and having sex with your spouse doesn't sound too bad.
If you change it to waking up with your partners name tattooed on your back then it would be s more effective comparison.
Thank you for raising this point.
It is socially acceptable for a person to do certain things with their spouse without obtaining consent- like changing their clothes when they are incapacitated. It is not socially acceptable to have sex with your spouse without obtaining consent- but it used to be. It is not socially acceptable for a parent to have a child undergo surgery absent any medical need, and that's especially true if the surgery is not performed by a doctor, unless that surgery is male circumcision. If we started society over and you were in charge of choosing this social norm, would you include an exception for male circumcision? If you imagine the social norms of our descendants, hoping that they live in a safer and healthier world, do you hope they choose to have an exception for male circumcision?
I understood the argument. But if you are making consent your analogy you want something obviously wrong that forces people to engage the argument. Drunk sex with your spouse is already an edge case so it makes the analogy lose force.
If prioritizing the welfare of the child, would the responsible stance not be to avoid irreversible interventions unless necessary?
E.g., phimosis is a condition that can be cured by the removal of the foreskin, thereby relieving suffering. Of course, some people might argue that suffering is character-building.
Agreed on both counts. Babies really really really can't consent. Heck, even children can't (meaningfully) consent to a lot of things. That's why they have parents. And I'd really rather not go through that as an adult.
As I see it, it's one of those things that people have strong opinions about that don't actually matter much outside of the opinions themselves. No great objective harm or benefit is done by either doing it or not doing it. And parents do things with *much* greater risks (and potential benefits) throughout a child's life. Like...orders of magnitude larger. We don't require the child to consent to a parental divorce, for example, even though the median effects of that are horrific for children (barring cases of abuse, really).
Maybe he wants to question Scott’s conclusion that circumcision is not bad after all - a conclusion based in part on stats about its effect on sexually transmitted diseases in Africa, if I don’t misremember his post? That post didn’t seem to live up to the standards of rationality set by the rest of the blog in my opinion.
err... i have no memory of this, and I've binge-read everything Scott's written on multiple occasions. Could you provide a source?
So, this is sort of an interest sounding out post more than anything, about a potential VC fund for impact certs.
First things first; I would be defining an impact certificate as something like the right to call yourself the institutional backer for an event. So, for example, a normal grant-granting platform would effectively own all of the impact certificates for a not-for-profit that they funded. They would not have technically "done the thing that the not for profit did," but they would have been the official backer.
So, how much is the price that one could save a life for. I'm not sure. My vague memory is that it's somewhere between $50-$5000, so taking the log average, let's call it $500. Let's just take this as an axiom (though I'll edit it if someone wants to give a more precise number, but it's sort of besides the point).
Question 1: Suppose that I could offer you a legitimate, documented impact receipt that showed that it had been bought as the impact receipt for someone who saved someone else's life. The price of this impact receipt was $100 (1/5 market value); the money was used to reimburse a person who'd bought urgent life-saving medicine for someone else. Would you buy this impact receipt as a charitable donation?
Question 2: Same scenario. $100 for an impact receipt, but this time for a volunteer firefighter who saved a person from a burning house?
Question 3: $100, but for a normal person who saved a child from drowning?
About $5,000 to save a life: https://www.givewell.org/charities/top-charities.
Ya know, I've always scoffed at those who object to paying people to "donate" kidneys. I am going to continue scoffing at them, but I did just have a momentary "that feels icky" reaction to some of your questions. What kind of world do we want to live in? One where people go around saving lives just because they expect to profit!?! Why, that would sully the morality of the whole endeavor! On the other hand, there are all those extra saved lives, I suppose.
Assuming that the price to me was below what it would cost to save a life (or equivalent quality of life changes) through eg. Givewell: I would almost certainly buy the first. The other two I'm less sure about.
-I think getting me to trust the certificates would be a feat.
-I might hesitate to buy because incentivizing the marginal life-saver wouldn't actually cost $100, it would also cost all the payments to those who would have saved the life regardless, but went after an impact payment because they might as well get some free money.
Well, we do pay people to save lives; firemen, coastguards, doctors, and so forth. That's not the problem.
The problem is the drowning child one, where now instead of "but I don't want to ruin my new suit", it's "but will I get an impact certificate for this? How much is it worth? Nah, that's not worth my while, the kid can drown!"
Children drowning in the presence of an adult who could save them is a negligibly small problem either way.
The real problem is things that are vaguely analogous to drowning children but without the elements of immediacy and proximity and with an additional element of social loafing.
At least the drowning child analogy makes me feel virtuous. I don't give money to charity but I do promise to save any small children I happen to see drowning, even if it happens on the ~one day a year I wear a suit, so I guess I'm maximally virtuous.
Writing prompt: a world in which superheroes exist, but so do impact certificates, and the economy is screwy because nobody can afford to keep compensating Superman for saving people.
For an extra twist: Superman volunteers to do the life-saving for nothing, but then gets sued because free services are screwing up the economy for people whose businesses rely on getting paid for life-saving superheroism! 😁
Can't shake off the feeling that my project only got funded cause someone must have misclicked :D
Hello. I have a request to resident psychologists. My search skills haven't provided any good results.
I really like the post and its thread about universal experiences from the old forum: https://slatestarcodex.com/2014/03/17/what-universal-human-experiences-are-you-missing-without-realizing-it/
There's a reference to Galton's survey on visual imagination. I think this approach can be applied to many invisible disorders or disabilities (in a broad sense, anything that might impair one's capabilities in life).
I struggle with wishing. I have a very weak hypothesis that I might not experience wishes, desires, cravings or whatever as other people do. It results in my motivation to do anything not inherently interesting on my own being extremely low. So, my question is whether anyone have studied wishes. desires and whatever in abnormal cases. Are there any tests, polls, surveys which I can take on my own to determine if I experience wishes as other people do? Or maybe a detailed essay on what is a "normal" wish, how people experience it?
Thanks for any replies.
P.S. I've checked with a few psychiatrists and they haven't found any permanent depression. I certainly don't have anhedonia, I enjoy things but I don't really want them much and I certainly struggle when I have to work persistently to get them.
The literature on anhedonia makes a distinction between consummatory anhedonia, which is enjoying things, and anticipatory anhedonia, which is looking forward to the things you enjoy. This is something you might want to look into.
Thanks. Strangely, I haven't seen this distinction.
I'm a psychologist, but I'm not sure I quite get your question. You say " It results in my motivation to do anything not inherently interesting on my own being extremely low." Seems to me that most of us find our motivation to do things that are not inherently interesting low. I mean, I have to sand some chairs when I get home tonight, and it's boring to do, and in fact the main reason I'm still here in my office posting on ACX is that I'm dreading the chair task. I think what happens, as people move beyond the point in life when you get to do just what you feel like a lot of the time, is that various tasks that aren't fun in themselves get woven into a mesh where everything is connected with everything that you care about. So I'm improving those chairs because I'm fixing my house up a bit in preparation for a visit from and old friend, which I'm looking forward to, and so the looking-forward-to-improved decor and looking-forward-to-seeing friend sort of bleed together, so that doing the chairs is more inherently satisfying than it seems at first glance. In fact once I'm in the midst of doing it I'll probably be thinking happily, "damn, that looks good" and looking forward to seeing how they look after I put on polyurethane. And likewise with work -- it's *work,* but I'm invested in doing a good job, and get caught up in the process.
So I'm wondering if your problem isn't a failure to find sanding chairs and going to work inherently enjoyable, but a failure to weave a mesh were tiring tasks borrow some yumminess & some meaningfulness from other, inherently pleasant areas of the mesh.
Thanks for your input. But here's exactly the point. You used the wording "look forward to" which in this context isn't really different from "want", "wish" or "desire". And this is what I seem to be incapable of.
I know that things like tidying, doing sports, earning money and developing myself are going to be useful. But no amount of mental exercise has ever helped me to build up the motivation to do that out of some future prospect.
I mean, I see people wanting specific things and pursuing them quite often even if it's some triviality like a specific sort of sweets which they look for in a few shops, or something bigger like saving money for a dream holiday trip. Wishing is out there for real.
Well, but I wasn't really talking about getting oneself to do tedious, non-pleasurable things for the sake of the pleasure one can have in the future . I was saying that tasks that are in a way tedious are transformed by being meshed in one's life with acute pleasures. The chair sanding is already sort of enriched and blessed in my mind by the pleasure of how the chairs will look, and the pleasure of seeing my old friend.
I know someone who talks about being "tethered" -- by which he means, obligated, committed -- and about how he avoids tethers if he possibly can, but then feels better when he has some. I think tethers and and what I call mesh are sort of similar concepts. When I say mesh I mean the network of obligations, satisifying work, fun and comforting relationships and professional aspirations that are the stuff of most adult life. Anything you do, even sanding chairs, is part of the mesh: If it's work, it nonetheless has a sort of halo of fun and satisfaction around t because it's part of the mesh. Tethers don't just make you do thing they transform the things you do. It may be that you have avoided commitments to people and work and that accounts for your feeling of being adrift, and nothing being worth doing except things that give instant gratification.
Lots of people outraged over (Dilbert creator) Scott Adams for saying that, according to a poll he cites, around 50% of black americans don't like/have negative views towards white americans, and that this means white people shouldn't bother trying to help people who hate them and their (white people) lives would be better if they avoid living somewhere with a lot of black people.
Is he really wrong?
The standard response to this seems to be that white people have an obligation to help black people, and blacks have no obligation to like whites, because white people are responsible for black people being poor (poor obviously in a relative sense) in the first place. But this is based almost entirely on narratives, and most mechanisms this is supposed to be based on can be/have been disproven.
And then of course there's the inconsistent way in which group responsibility operates, wherein all white people everywhere are responsible for everything bad (or "bad") done by any white person ever, while black people are not even morally responsible for their own (bad) actions (and white people can take no credit or pride in the accomplishments of other white people). This tries to get rationalized by the claims of "white privilege" being a thing (and/or the even more silly claim that white people are "still benefitting from slavery"), which again can be debunked in nearly every way, but is an example of anti-white hatred itself (if you disagree, let me know if you would have been fine with it if Kanye had simply said that jews have "jewish privilege").
I think it's one thing to say that white people aren't allowed to have negative opinions toward black people, but it's something else when you're claiming that white people have an obligation to be okay with being hated by black people and to actively care about improving their wellbeing. And predictably, the left are calling Adams "racist" and "hateful" for...not being okay with being hated for his race: https://apnews.com/article/dilbert-scott-adams-cartoonists-respond-109cb1a6dea03e931e2e6e3814bc743a https://time.com/6259311/dilbert-racism-scott-adams/
This reminds me of debate between Sam Harris and Ben Affleck over Islam, where Harris declares that Islam is a religion of violence on the basis of "poll results" showing that 20% of Muslims favor the death penalty over apostasy. Both Harris and the Dilbert person are overinterpreting what people will say on a poll. Anyone with any day to day familiarity of life in the typical Muslim-majority country in the poll will think, "only 20%? Not too bad," and would never make some idiotic generalization about Islam itself.
https://www.youtube.com/watch?v=vln9D81eO60
I was following until the first half of the last sentence. It sounds like you're saying that familiarity with Muslim majority countries would presume that Islam is much more violent than 20% in favor of death penalty for apostasy.
Well, familiarity would presume that more than 20% would declare on a poll that apostates should die, and also to put no stock into that because you know it's BS. Yeah, death to apostates, whatever, ha ha ha.
I see what you mean now. I'd like to point out, however, that if half of white people polled were to even lightheartedly agree that (for instance) race-mixing couples should be put to death we wouldn't ever hear about anything else on the news. Also if that were the case and I were married to a black woman there'd be zero chance convincing me to live in a white neighborhood.
I did a double take on this at first, as well.
Not wanting to put words into Sylwester's mouth, my alternate interpretation of their last sentence is that "only 20% of Muslims would glibly tell a pollster that apostates should be killed" means "the trolls and thoughtless poll takers are fewer than I thought."
The poll question is asking one of two things, which use the exact same words but mean very different things. I will use the following definitions:
white - a person with pale complexion
White - a person who uses entrenched status maintained through oppression in order to increase their own power at the expense of minorities (most often pale complexion people)
The poll question is either poorly written or intentionally conflating those two things. If you ask the first question it's pretty much meaningless and very hard to disagree with (saying "No" to "it's Ok to be white" is advocating genocide). If you ask the second question you're accepting a certain worldview about power dynamics and then answering about those power dynamics - which in reality is only tangentially related to race. That's how Jews and Asians can get stigmatized as "white" or "white-adjacent" despite that not being how any of this works if you're talking about Caucasian people.
I can be upset about people using the word "white" to mean something far beyond skin color or race, and also upset at someone trying to ask such a question when they almost certainly know about the two separate definitions at play here.
Write or wrong, current conventional wisdom is that it's not allowed to have negative feelings towards black people, and the thing Adams said would clearly imply that (you can't convincingly say you believe it's a problem that a group majority dislikes you and that's a problem, but you don't have a problem with that group as such). So he broke the social rule and got the inevitable pushback.
I agree that it’s not allowed to have negative feelings about black people, and constraints like that are madness. Everybody has at least some occasional negative feelings about everybody, including their mother and their newborn baby. And we are all vulnerable to negative feelings about groups, especially groups we do not know well. It’s a consequence of (1) Chunking (thinking in terms of categories, not a swarm of individuals — it’s lossy compression, and we are wired to do it to save mental space and time) (2) Being somewhat neophobic, as virtually all animals are — it’s adaptive. Of course it is possible to pry oneself loose from feeling confident that one’s negative take on a group is valid. But “do not think that because it is bad” is not a good tool for prying.
A bunch of things can be true at once, in part riffing on or even echoing various posters in this larger thread:
1. A vast plurality of humans have almost certainly, whether at times, situationally, selectively using their own idiosyncratic criteria, or ongoing, felt frustration with, anger at, or even outright prejudice towards, some particular group(s) of people *as a group*: distinguished by perceived ancestry/race, religion, class, political affiliation, or more.
We may be in-born with some physiological traits that reward tribalism, to foster in-group loyalty and make us at least suspicious of outgroups, if not hate them outright. (And yes, we also simultaneously have traits which help us work together cooperatively, even across such group divisions.)
It shouldn't be surprising to anyone that some American Black people harbor negative feelings, sometimes strongly so, towards American White people and/or people in various other racial outgroups. Even without the history of chattel slavery and Jim Crow, that comes with the territory of being human.
We humans naturally form teams to band together against "outsiders." And will even creatively invent differences between groups, when points of differentiation are not immediately at hand.
2. There's some social pressure in the US - on social media, in news media, etc. - to not say bad things about any historically disadvantaged minority, which includes Black people.
3. Adams made a thoughtless (more below) comment that violated that expectation. And in reaction, he lost some face among some observers, the distribution of his comic strip was reduced, etc.
4. There's also a small subset of folks who make a regular practice of violating that expectation. For instance, Breitbart News' former "Black Crime" beat, certain social media accounts which frequently post videos of bad things done by Black people or unauthorized migrants in the US, etc.
5. That single, ambiguous question in the Rasmussen poll was abysmally bad, as I noted in an earlier comment. That polling organization themselves confirmed in a tweet that it was an act of trolling, rather than a good faith attempt to assess attitudes toward other racial groups.
Carl in his History Boomer post did a good job, methinks, of noting all of the possible ways that poll-takers might have (mis)interpreted it.
6. For Adams to not to understand that – and to even misinterpret its numbers, as well – and then make a highly public, emotional video based on his immediate reaction, doesn't speak well for his discernment or judgement, at least in this situation.
If the widely shared clip of a portion of his remarks was misrepresenting his larger position, would welcome specific interpretations of that. At least from that clip, that poll seemed to have been a key, specific event that impelled him to make this set of video remarks, even if he'd been thinking about this topic for a while.
(And having written this, just came across Leo Abstract's post with some suggested insights about Adams. If those are even mostly accurate, that *might* further complicate the narrative.)
7. Race itself is a notoriously slippery concept.
For starters, it's clearly the case that at times and places we're treated differently, sometimes harshly or unfairly, based mostly on our outward physical appearance. (Along with markers of ancestry/race, these also include perceived beauty, height, weight, and visible attributes related to age or illness/disability.) That difference in treatment happens in many, many ways socially. And at least with perceived race, also in various ways in law and other formal practice.
But pinning down what "race" someone "actually" belongs to is elusive. We're all, to varying degrees, mixed. The "skin deep" characteristics of skin color, hair type, facial structure, and the like aren't always correlated as we might believe with sets of meaningful genetic differences, as we've learned with the advent of DNA testing. (Was thinking here of Angela Davis, who just got surprising results from such a test, https://www.salon.com/2023/02/22/civil-rights-activist-angela-davis-learns-surprising-genealogy-on-finding-your-roots/, and this paper on blonde-haired people in Oceania, https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3481182/, as just two examples among many.)
And yes, which individuals or groups are included in or excluded from widely-perceived races like "White" can vary *widely* from time-to-time, and place-to-place. (For instance, see https://www.vox.com/2014/10/10/6943461/race-social-construct-origins-census and also https://www.vox.com/2014/10/27/7062921/immigration-1900-race-racism-quotas-eugenics-map-settled-united-states.)
I can't remember where I spotted this, and haven't yet been able to find it again, but IIRC one scientist wrote that there are apparently at least hundreds of groupings of humans whose members largely share some slight genetic differentiation, distinguishing them from members of other such groups. And noting that, if this is even remotely accurate, genetically, it either challenges or at least gives far more nuance and complexity to widely-held conceptions of the existence of some 5-20 broad "races."
+1
There's a thoughtful take on that Rasmussen poll question here:
https://historyboomer.substack.com/p/its-ok-to-be-white
"The poll didn’t say that half of black people disagreed with “It’s OK to be white.” Only 26% of black people disagreed, 21% weren’t sure. And before you worry about that “half” claim, you should also note that 20% of white people also either disagreed or weren’t sure! Ditto for 41% of Hispanic and Asian people (lumped together under “other”). If you seriously think 50% of black people don’t want white people to exist, it means we’re facing a race war where 41% of Hispanic and Asian people and 20% of white people may be on the wrong side!"
"... Rasmussen chose those exact words [for this question] because they were deliberately responding to a claim that those specific words were hate speech. They were trolling the Anti-Defamation League."
(And the author of this blog post then showed Rasmussen's tweet basically saying exactly that: https://twitter.com/Rasmussen_Poll/status/1629240555321405440)
As this author later noted, "Look at the [poll question's] words again. “Do you agree or disagree with the statement ‘It’s OK to be white’?” Those are weird words. OK to be white? That is not the kind of neutral language that pollsters normally use. The unusual wording is apt to set off alarm bells in someone’s head. “What funny business is this pollster trying to pull?” People [responding to that question] will be understandably looking for a trap."
Man, this "historyboomer" link is hilarious :
>That’s not the same as hating white people and thinking they shouldn’t exist.
Uh ok, so it's totally fine to disagree with the statement "Is it OK to be X ?" for any value of X then ? it doesn't have any negative valence whatsoever ?
> “But if the poll had people disagreeing with ‘it’s OK to be black,’ everyone would be freaking out.” Maybe so, but there’s a lot of history behind why we get more upset when black people are attacked.
Uh huh, so you can't do it to blacks, that would be racist, but blacks can do it to whites, and it would be totally not racist. Because something something Racism is "PoWeR pLUS PreJuDiCe".
Absolutely moronic.
As others pointed out, you could likely get similar results asking conservative whites about "Black Lives Matter". Pretending this is about the literal meaning of words is disingenuous, even before you get to the problems you point out where the numbers are wrong anyway.
I don't actually believe the interpretations of the poll. The question asked was something like "do you agree with the statement, it's OK to be white", which is a thing said by a bunch of racist 4chan troll types. A lot of people who would normally be OK with the statement, will have negative reactions based on that.
It's like asking "do you agree with Black Lives Matter" and then concluding that the people saying 'no' literally think there is 0 moral value attached to the life of a black person.
Also - IMO the worse thing than saying to "stay away from black people" (which is bad enough on its own) is where he called black Americans a "hate group". Even if it's true that 25% of black people think it's "not OK to be white", that's still a blanket negative statement that it's bad to make.
You know Nassim Taleb's shtick with Fat Tony and the wisdom of the taxi driver? Sorry, but in a similar vein, the Dilbert guy is an idiot, with all his BS about Trump being a master hypnotist living in the 3d world or whatever. Ignoring him completely improved my life and I recommend it.
“People not liking you” doesn’t get you out of obligations to them. I would be really surprised if less than 50% of people in the bottom 10% income decile didn’t like people in the top 1% of incomes, but that doesn’t exempt the very rich from any of their obligations to the poor.
Then again, those receiving top 1% incomes are not usually required to live where the lower 50% dwell, or have their children mingle.
I would say that a rich kid who never mingles with poor kids will likely be emotionally stunted, and their parents have committed an injustice against them. Also, it’s fine if rich people prefer e.g. beaches that not many poor people use, but it would be unjust if there were public beaches that poor people weren’t allowed to use. I don’t think schools should be allowed to refuse black people, but private schools should be allowed to teach values that (parents expect) black communities will be less interested in (I assume Douglas’ putative white people have designated some values as ones black people don’t share). The school can charge enough money that black people won’t come unless they really want those values for their kids.
Ok, let me try this again now that I'm on an actual keyboard. I looked into it at the time and then also after he followed up. A few things:
1) Adams appears to be truly and classically anti-racist, and those familiar with his larger body of work testify that he has always blamed gaps in attainment between black and white on structural racism. (He also appears to blame teachers' unions for perpetuating this in a school-choice-partisan kind of way, but I'm not really sure about this part. Only thing I can tell for sure is that he directly and loudly blames white people for the disproportionate poverty, crime, and lack of education.) He appears to not believe in any meaningful differences between races in terms of genetics, and castigates white people who take any credit for white accomplishments.
2) A big part of his overall messaging appears to be that the media monetizes hate and turns everyone against one another. His impression of the state of the country (from a variety of polling sources as well as the tone on social media) is that clickbait has worsened race relations to the point where some percentage of blacks actively dislike whites.
3) He's a lifelong proponent of moving if you don't like where you live. He moved from upstate New York after a brush with death in a snow storm and ended up in the lowest-crime part of California on purpose. I'm not clear on why the issue of people not wanting to live in predominantly black neighborhoods (due to crime) came into this discussion, but I think it was to mock people on twitter for their faux outrage about not wanting to live near a group of people that contain some (10%-50%) members who dislike him for being white. (Mock because they all live in white areas too.)
4) Finally, and arguably most importantly, he appears to have been pivoting for some time to 'self help' rather than cartooning, and appears to have wanted to get as much attention as possible in order to pitch the country on solving racial disparities by injecting pro-success messaging and tools into black culture. He appeared genuinely surprised at having had enough reach that some remarks on his podcast stirred up so much furor, but then also appears to have realized that this is what he should have predicted based on how the media monetizes division and had all the reason in the world to clip his remarks to make them look bad (the way Aaron Rupar did with the Charlottesville-era video of Trump saying there were 'good people on both sides' of the statue question, clipping off the next sentence which was "I'm not talking about the neo-nazi white supremacists, I condemn them totally".
So to answer your original question: Yeah he's wrong, but not for any of the things reported (which he didn't say, or at least didn't mean). He's wrong because he is far too optimistic about how to improve any relations in this country.
"Around 50% of black americans don't like/have negative views towards white americans." There's a pretty big difference between not liking a certain kind of person and having negative views towards them. I have SOME negative opinions about almost everybody I know. Maybe everybody, in fact. I'm fond of Zvi, for instance, and think he's brilliant. But he also sounds like somebody who has not read a a single book written before 1950 or so, and I think that's bad. I don't know what white people should do if most or all blacks hated most or all whites, but fortunately I do not have to figure that out, because it's obviously not the case. I say that based on relationships lasting several years with each of a handful of black people, reading books & articles by blacks, and listening to black public figures in the media, plus various casual contacts with blacks in daily life.
Seems to me that what was wrong with Adam's idea was that he misunderstood the poll results or, what's more likely, distorted them in a way that make black people sound ugly, hateful and scary. Then he discussed what would be a reasonable way for whites to live and to think about blacks, if in fact all blacks hated whites. And in the course of the discussion he gets to remark over and over again on black hatred of whites. I mean, it's sort of like asking what we should do if we learned that every single person in Texas, but nowhere else, harbored an obscure, lethal and incurable form of Argentinian rabies. I don't KNOW what we should do, but that is not the case so why are we talking about this? "Would we be justified in nuking the whole state?" he asks? Yeah, fine, fuck it, nuke Texas. But get Kenny and his partner out first.
According to the wikipedia page for Adam Dilbert, the poll consisted of presenting the phrase "It's Okay To Be White" and asking if they agree, only 53% of black respondents, said they agree, 26% percent said not sure, and the rest said they disagree.
Now, I know the origin of this slogan, apparently 4chan made it to bait morons into being racists, typical 4chan behaviour, and it worked. 47% of black respondents said they're not sure if being a trait you're born into is a good thing with varying levels of disagreements, and according to the Wikipedia page about the slogan, universities call police when they see posters of the phrase and de-platform speakers who want to speak about this statements.
What's curious is that the quoted individuals never justify why they think the statement has no place : Is it painfully obvious for instance ? (but aren't Black Lives Matter too ?), is it meant in bad faith to mean something other than its surface meaning ? (but aren't BLM too ?) Is it false and they think that merely contradicting or discussing it is dangerous ? Anybody with some fraction of a brain has questions, what about being white is so wrong, enlighten us ye wise ones ! But they all have a single answer, and it's not in words.
I would say that it's never okay to express disagreement with a statement like this without qualifications, even if you know the statement is bait made specifically to make you look bad, resist a little, don't fall into it right away, at least give something to your defenders to squeeze out some benefit of doubt. But then again the US census would classify me as White if I lived there, so maybe this is just my White Fragility showing eh ? it's pretty obvious to me that being a fucking racist when your entire ideology is based around bitching about racism is kinda not intellectually honest, but who knows, those things are hard. Better not think about it.
Perhaps the only redeeming path for that poll is if it's repeated with white respondents, I suspect the relevant factor is not actually race at all but political affiliation, the near 50-50 split is a weak hint it is. But still, it's hilarious when someone goes "hmm, I think whites are discriminated against" and the immediate response is hysterical screaming and cancellations of the guy who Noticed Something, he's totally wrong right ? it's totally obvious he's just a fragile white snowflake who can't handle equality, which *of course* involves asserting that being white is not OK.
Failing that, perhaps the polling organization did black magic stat shit to paint a different picture over reality, and a balanced sample won't display the fucked up ratio of 1/2 of it thinking it's not okay to have a certain skin color. That's about the only 2 things I can think of for this to turn out Adam was wrong, and boy do I wish he was wrong.
But he's not, is he ?
Oh, thanks for clarifying about the 4chan bit. I'd forgotten that part. Yeah, it seems unlikely that many of the respondents on the poll would have known about it, but maybe I shouldn't judge its notoriety by my not knowing about it.
My take on what an ordinary person would think of a question about whether it's ok to be white is fairly advanced, given that I understand and sympathize with the viewpoint that whiteness is problematic because it was created in response to nonwhiteness. I.e. my own ancestors now would be classified all as 'white' but never thought of themselves that way -- they were various nationalities and ethnicities from within Europe, and saw themselves as such. If they had any negative identificatoins they probably mostly defined themselves in opposition to the great powers of the day (from which I am not descended) -- i.e. not English or French. If you were to ask me on a poll if it's "ok to be white" I'd hem and haw a bit and say "well it's ok to be anything you can't change so yeah". If there were some racial identity that I had been taught had been made up to specifically exclude people like me and I were asked on a poll if it were ok to be that, I'd say "hell no".
I think "made by 4chan trolls" has to be taken with 2 or 3 medium sized spoonfuls of salt, since mainstream media (yes, even wikipedia) is notorious for using 4chan as a non-christian devil, and attributing various wild shit to it. If GamerGate has taught me anything, it's that Wikipedia in particular is basically twitter or tumblr on Woke incidents.
I think it's certainly okay to disagree with "It's okay to be white", by various arguments. The first one is of course Free Speech, I think people are even allowed to disagree with "Should babies be NOT killed ?", as long as they are nowhere near the child birth section of any hospital. The second level is that the idea is bait, it's almost certainly meant in bad faith (though not without a lot of legit reasons, as this incident demonstrates), and it's useful to stir the water by saying "I disagree" without elaboration, **then** giving a bunch of reasons why after you have gotten people's attention.
I disagree that "Whiteness" is made up in any interesting sense of the word, race is obvious. An ancient arabic pre-islamic religious song goes something like
>We are the crows of akk [An Arabic tribe], we came to you [our God] begging
A "crow" is how arabs called a black slave, and the context is that slaves usually led the convoys to Mecca (which was were people worshipped even before Islam), so the slaves sing the song and then the rest of the convoy sings along, so those things are really really heckin old, this is the first example that came to mind. The African slave trade is ancient, it's likely that every single ancestor of yours from Italy and Spain up to at least Germany and back in time to the crusades saw or heard of or otherwise vaguely sensed the existence of black skinned humans, and knew in a distant sense that they were, unlike them, white. Sure, I have no doubt that blacks and whites mixing in the Americas almost certainly turbo-charged the various sentiments and ideas, but that's just a trivial tautology, of course people notice the differences more when they are closer, what is useful things is this saying ?
The thing that really gets me about the incident is how devoid of self-awareness the corporate cancellers and the hysterical screamers against Adam is. It's almost a Wojack meme, if somebody said to me "you know what Bi_Gates ? I think you hate me, and I think it would be okay if I got away from you", my response would either be "Yes, I hate you, nice that you noticed", or some non-cruel equivalent, or if I was keen on them I would say "No, I don't hate you, how did you get the impression ?", what I definitely wouldn't do is [Wojack crying face] "NOOOOOO YOU PIECE OF SHIT I DEFINITELY DON'T HATE I SWEAR I JUST THINK YOUR SKIN TONE IS KINDA GROSS AND THAT YOU SHOULDNT EXIST BUT THIS IS NOT RACIST FOR SOME REASON YOU KNOW WHAT NO MORE MONEY FOR YOU NOBODY IS GOING TO BUY YOUR SHIT HUH HOW ABOUT THAT FUCKER I STILL DONT HATE YOU".
Like, I don't know if they are moronic enough that they actually persuaded themselves that hating someone based on skin color **is not** racism and now they're confused because the rest of us are confused, or if they know full well what racism is and that they are doing it but are too much of pussies to admit it in public, or if they just have no consistent position and their plan consists of farting the biggest nastiest amount of mindless animal rage in the hopes that everybody would cower and not ask questions ? I genuinely don't know, I would hate them either way, they are disgusting, but I want to understand what their confused little brains are thinking before I can hate them with a clear conscience.
> I think "made by 4chan trolls" has to be taken with 2 or 3 medium sized spoonfuls of salt, since mainstream media (yes, even wikipedia) is notorious for using 4chan as a non-christian devil, and attributing various wild shit to it. If GamerGate has taught me anything, it's that Wikipedia in particular is basically twitter or tumblr on Woke incidents.
4chan are probably overrated as Hackers on Steroids. But they're not overrated as purveyors of crazy pranks that play traditional and social media like a fiddle.
https://knowyourmeme.com/memes/free-bleeding
https://knowyourmeme.com/memes/ok-symbol-👌
https://knowyourmeme.com/memes/its-okay-to-be-white
Yes, I'm sure my ancestors were aware that some people are darker-skinned than others, but I'd wager most of them would have thought of those people as heathen, paynim, mohammedans, saracens, moors, or the like. Monolithic "white" vs monolithic "black" might not be purely an American invention but there's more ideology in your own understanding of these categories than you might suppose.
As for why they went after Adams, his own explanation is that they are either misled, motivated by monetary incentives, are just virtue signaling, or are what he calls 'dopamine gooners' -- people metaphorically cranking off with delight at the prospect of canceling someone and making him suffer.
> I don't know what white people should do if most or all blacks hated most or all whites, but fortunately I do not have to figure that out, because it's obviously not the case. I say that based on relationships lasting several years with each of a handful of black people, reading books & articles by blacks, and listening to black public figures in the media, plus various casual contacts with blacks in daily life.
This really seems like a textbook case of overconfidence on the basis of anecdotes. It's good to have priors and to avoid letting new data disproportionately push them around - that's one of the two major takeaways of Bayes' theorem, after all - but the strength of your prior should itself be determined by the quality of the evidence used to form it. It's the "what do you think you know and why do you think you know it?" concept. If your evidence is based on a handful of friends, some casual interactions, and your choice of media consumption, it's 1) very shallow, and 2) extremely vulnerable to overfitting and projecting the character of your personal bubble upon the fabric of a vast and varied country.
With all that said, I also find the premise rather unlikely. You just haven't provided any information that should lead any rational person to say that it's "obviously not the case."
I certainly would not feel justified in concluding that most black people do not hate most white people based on my limited experience. I don't think it's altogether silly, though. Hate's a strong word and if half of black people HATED white people, you'd think I'd have picked up on that by now. It's sure not hard on Twitter to pick up that all anti-vaxers hate various things I have to say, and in fact make statements that sound like they hate not just my actions but me: "You should be in jail," for example. Hatred's not real subtle. But let's withdraw my life experience entirely from the evidence for or against.
The thing that seems indisputably funky is what Scott Adams (or maybe just Scott as described by Goldman Sachs Occultist) did with poll results where 50% of black people endorsed one of these 2 statements (it's not clear which): (1) don't like white people, (2) I have negative views towards white americans. OR (3) 50% of blacks did NOT endorse the statement "It's OK to be white." And then, acc/to Goldman Sachs Occultist, Adam went on to talk about whether white people an obligation to be okay with being hated by black people, and should feel comfortable living near a bunch of blacks. Talking about being HATED by blacks goes beyond all 3 versions of that poll question, and in any case, bear in mind that only 50% of blacks answer the question in the negative direction. So the question is, is it fair to expect whites to live near or feel sympathetic towards a group 50% of whom have some negative feeling about whites?
Honestly, living in a place where 50% of people disapprove of me sounds kind of like no big deal. On the block where I live, there's one guy who loathes me (the feeling is mutual) because he put the moves on my daughter when he was 40 & she was 16 and I really went after him. There's an elderly Indian woman who got quite angry at me one time when I tried to give away some old toys by piling them in the grass between her sidewalk and the road. She probably still remembers it. And there's a guy who once whistled the wicked witch theme from The Wizard of Oz when I biked by with my long gray hair flying.
Or let's take this forum. Leo and Biblio, I don't get the feeling you hate me, but you certainly have a negative view of things I said in an earlier post. And, Biblio, you had quite a lot to say about how much my reasoning sucked. And once in a while on here somebody really seems to loathe the way my mind's working, and says so. On the other hand, a fair number of comments I get are positive. So is it reasonable to expect me to keep posting on this forum, and to try to help it work better when I see a way? I don't see why not. Doesn't seem unbearably difficult. NO I am not saying that being on this forum is equivalent to living in a neighborhood with a lot of black people, half of whom harbor some negative feelings toward white people. But I am saying that being in a setting where half of the people have some negative views about me feels kind of like par for the course to me. Being in a setting where half the people absolutely hate me does not. Being in a setting where everybody hates me sounds extremely unpleasant.
The funkiness of what Scott (or perhaps just GS Occultist) did was to go from a poll result showing that half of blacks had some negative feelings about whites to asking whether white people have an obligation to be okay with being hated by black people and to actively care about improving their wellbeing.
Exactly. Not just that, but Adam's whole argument is that it's totally fine to avoid groups that are statistically likely to dislike you - to some extent as defined by your personal appetite for risk. To the various white liberals who were condemning him on Twitter, he asked one simple question, which none of them could answer. What is the racial makeup of your neighborhood?
Mine is mostly but not completely white. At least partially because of the history of racial redlining in Twin Cities.
I didn’t want to reply seriously to the OP question because it is so hateful and so fucked up in so many ways to me.
https://en.m.wikipedia.org/wiki/Redlining
Mine is about 50% white, 35% asian and 15% black.
I'm no apologist but I can tell that you didn't listen to any of his actual content beyond the tiny clip that was deprived of context to sound bad. The poll as reported had only half of blacks saying it is ok to be white, and Adams was careful to explain that he was using the poll as a jumping off point in discussing an obvious cultural trend in black dislike of whites.
I didn't listen to any of it. I just read what was here and responded to that. A lot of my response hinged on what the negative statement about whites was that50% of blacks endorsed: Was it "I don't like whites" or "I have negative views towards whites." As I wrote, the second seems much weaker. But now you're saing that the statement that 50% of blacks endorsed was "it's not ok to be white' -- wait, actually, it's not quite that, it's that 50%of blacks did NOT endorse "it's OK to be white." And actually, I don't know what to make of that -- I don't know what it means, really. Seems silly to put that statement in a poll. Does it mean, "people should not be *allowed* to be white"?
" white people should not be allowed to exist"? "there are some things about white attitudes and actions that I think are just no OK, they're wrong". So I stand by what I said: This new version of the poll result, too, does not justify asking whether white should have to live near a bunch of people who hate them. I don't know whether people should have to live near people who hate them, but in any case this new version of the poll also does not seem to translate into 50% of blacks hate whites.
I had to scroll up and verify that my comment didn't get cut off and still contained the part about how the poll wasn't even the issue at all. It's just jumping off point for something that people have already noticed. Like how if we saw a poll that shows that 50% of women think that masculinity is problematic we wouldn't really need to dig too much into the methodology or sample size to be able to discuss it.
It's good for dunking on people who haven't even looked into a topic and are opining as if they're experts, though.
But I was not opining as though I'm an expert, Leo. I was opining as though *you* were an expert. I assumed your account was accurate and complete, and that you would not be so silly as to skip giving crucial context that was necessary to grasp relevant nuances. I also assumed you were clear about what the question was that 50% of blacks answered in a negative-towards-whites direction. However, you have now given 3 different versions of it: 2 in your original post, and a 3rd, importantly different one in your short response sneering about how I had not heard the entire clip. Um, maybe it would be good if YOU listened to the entireclip and came back and told us what the fuck the item was, exactly, that 50% of blacks answered in the negative-towards-whites direction.
You were responding to someone else.
Yes, he is really wrong.
When is the meetup-form due by? (Do I rush to lock in a suboptimal venue, or take a few days to try getting a better one?)
Lo
and behold!
Context?
Does anyone live or have practical experience with a country that uses open list proportional representation? (So Sweden, the Netherlands, Austria, Finland, I think Brazil, etc.) The main thing I want to know is- do the politicians *within a party* actually compete with each other?
The idea as I understand it is that voters are just voting for a party, and then have the option of ranking individual politicians within that party if they so choose. The party receives X number of seats based on how many votes they got- the party candidates then either make the cutoff for one of those seats based a) how the party originally ranked them, and b) how voters then ranked them from there (if they chose to).
Do the politicians within a party..... compete with each other though? I.e. if the Swedish Green Party is running multiple candidates, and they all presumably would like to be ranked higher by voters- in practice do they actually compete, bash each other, play up their accomplishments, etc.? "Sure Sven Stroosma is a fellow Green Party candidate.... but whatabout allegations that he had improperly unrecycled herring cans in the back of his Volvo on the way to Ikea, ya?" (This was my attempt at a Swedish Green Party scandal).
In theory open list PR is the perfect voting system in that most voters are really just voting for parties anyways, and then if they have opinions on individual candidates they're free to express those- or not. So (in theory) it's the right blend of party-centered politics, while still holding candidates individually accountable, yet without having a high degree of personalism. But I have no idea if the system really works in practice or not, would be curious to hear people's feedback
Not really happening in Austria as far as I have observed. I think the damage to the party would lead to an inacceptable loss of votes overall.
In Denmark, there's a fair amount of jockeying within-party in local elections. Obviously, candidates avoid creating drama if they want to get on the list in the next election. But they will appeal for personal votes to get the marginal seats, and it definitely works. Competition between candidates of the same party is more pronounced than it was a few decades ago.
The dynamics are different in national elections where the vote counts and budget requirements are bigger. But even here, it feels like social media is creating new opportunities and more competition than we used to have.
I'm from Norway. I've never heard politicians compete within the party the way you say.
Most people only care what party they vote for and don't bother crossing names off the lists. If my memory is correct, one exception was once a group of voters of Pakistani descent organized to get a person of Pakistani descent into parliament ahead of people above him on the list.
Not in Sweden. Voting for a person is optional and there's a minimum limit that must be exceeded for them to count, which makes it very hard for a politician to get enough personal votes for it to make a difference. (Usually such a politician is popular in the party and high on their list anyway.) In the 2022 general election, only 2 out of 349 members of parliament were elected based on personal votes. These 2 seems to be regular charismatic politician with strong local support in smaller regions (which makes the minimum limit easier to reach). But there's also a growing trend of politicians from immigrant communities who make strong personal campaigns in those communities (with what many suspect is more focus on in-group politics than party politics), which has a worse look.
The (big) political parties in Sweden are very centralized and the center is very powerful (see e.g. the book "Knapptryckarkompaniet" (translated: "The Button Pushers") for what it's like being a member of parliament). Upstarts with lots of personal votes are likely to suffer from tall-poppy syndrome unless they play their cards well.
Internal competition in Swedish parties are usually much more about internal politics. Its rare to see politicians be critical of members of their own party or the party line, and those who express such criticism are often dealt with harshly (especially in the bigger and more centralized parties). In general, Swedes like their consensus culture were everyone always agrees, and where a change in policy can be coordinated behind close doors so that everyone can change their mind at the same time without any public arguments or debates. (See e.g. the NATO question: Every politician in the Social Democrats was against NATO membership a year ago. Two months later everyone was pro-NATO. Few outward signs of conflict were seen. If a Social Democrat would have tried to gain personal votes by taking a strong personal position on NATO, I'm confident that they would have been seen as a defector and punished accordingly by the party.)
Cont.
There's a hypothetical effect were people can indicate preferences by voting for candidates to "push" the party in a certain direction. If a Green Party politician who is centrist on economics gets many personal votes than his more left-leaning colleagues, this information might push the party centerwards even if no there's no practical effect on the election outcome. But the parties should have better information from polling etc. that trumps this.
From experience in Israel yeah, it's often a major source of drama. It usually doesn't come up much on the news (the kind of politician that barely makes it high enough in the list to get a seat won't be a major figure), but it's a huge source of intraparty conflict and you will occasionally get stories about how "the party ranked this guy I like way too low, I like them less now".
Think of it like party leadership conflicts (which also happen in e.g. the US or UK), except the minor league of that. You'll have politicians fighting each other while trying not to do it too hard to avoid damaging the party.
Thanks. It's funny how your experience in Israel is different from all of the other responses I received from the Nordic countries. National culture matters a great deal!
Yep. In Czechia, we are similar to Israel. In last elections (2021), two parties formed a coalition with one ticket (it is often done in order to get over 5 % threshold, though in this case reasons were more complex), and in an election, voters of one of those parties ranked their politicians so consistently that their unlucky partners tanked in the elections despite on paper solid results. Similar stuff happens all the time, and victim-politicians tend to be angry about that and complain about backstabbing.
I have lived in both Finland and Sweden and I have never noticed this, not openly. It would probably be counter productive anyway and would probably not be popular in the party (which is more important for one’s future political career than what few votes you could get like that).
Candidates will advertise themselves and try to bring out some attractive features. But open negative campaigning against people on the same list in the same district isn’t seen. Keep in mind you might need their votes to get in.
The systems are different in the two countries. In Finland you have to vote for a candidate and the personal ads are very prominent. In Sweden selecting a candidate is optional.
To clarify, of course people from the same party compete within the party, but they don’t campaign against each other in the election.
Having read the other replies, I will clarify more. In Finland a party's candidate lists are essentially unranked ("open"), before the election, that is. So candidates do campaign and advertise for themselves and therefore they do compete with each other, but they don't openly campaign against each other (like in the example about the Green Party scandal).
So a candidate in the blue party (fictional) might advertise that he cares immensely about lowering taxes, implying that he cares more about this than other candidates in the blue party. But launching a public negative campaign against another candidate in the same party isn't something one sees. It might exceptionally happen, but in general it would be counterproductive.
What do you think of the US foreign politics and the role the US plays in geopolitics.
At one hand it is remarkable that this topic is almost completely left out here. Why? Not interesting? Too hot?
As you are intelligent people thinking about ethics as well as the future of humanity, don't you have a opinion on this?
I see so much hypocrisy in US foreign policies, don't you see them? Or don't you care? Doesn't it matter because the US are the good guys?
A good go-to book on US foreign policy, IMO, is Walter Russell Mead's _Special Providence_. Especially so in light of handwringing over hypocrisy. It's largely not hypocrisy, but rather multiple voices within that policy, often in conflict, often allied. Mead outlines them as four-fold: Hamiltonians prioritizing favorable US trade; Wilsonians promoting the spread of US-style democracy around the globe; Jeffersonians advising to defend and maintain our democracy at home only, and leaving the rest of the world alone; and Jacksonians going along with the live-and-let-live rule, but when some foreign power breaks it, very enthusiastically sending our own forces over there to Break Stuff.
While each school of thought is named after the historical figure Mead sees as most exemplary of that school and upheld by its adherents, even those figures broke with their schools, often in a manner perceptible as them giving in to advisers on their own cabinets, especially persuasive voices in Congress, or plain old natural events. When US policy flips or exhibits conflicting standards, it isn't hypocrisy, so much as one of the schools losing ground to the arguments of another, depending on the situation.
Mead contends that US policy is as successful as it is because this four-way tug of war enables a high degree of adaptability. America can lean on her military one day, her trade delegations the next, her peace or evangelical corps the days after that. And every so often it's surprisingly decisive and swift because as many as three schools align.
Hypocrisy can be an emerging property of complex systems then :-D
I can't keep up with everything, but sometimes I support US foreign policy, and sometimes I don't, and it always affects my vote to a modest extent, but is almost never determinative, because domestic concerns (what a candidate will/won't do within the US) matters way more to me.
For example, right now I support the US effort to help the Ukrainians beat the snot out of Putin's Russia. It's hard to say whether that effort is being executed competently: my expectations for the Biden Administration, particularly after the Afghanistan fuckup, are very low, but maybe they learned something, or maybe there's some mid-level person who is doing a good job, or maybe they could do a lot better. Very difficult to know, unfortunately. They have not obviously screwed the pooch, yet, but perhaps things could be a lot better. Nevertheless, I support the aims, its just the execution about which I wonder.
On the other hand, I think US foreign policy towards China is marked with impressive levels of clumsiness and ignorance. I'm darned if I can figure out what the overall plan is, if there even is one. I mean, American foreign policy in Asia has practically always has been clumy and ignorant, except maybe when MacArthur was running Japan, and maybe again early in the Nixon Administration, since Nixon was one of the few American Presidents to really care about Asia, but it seems to matter more now, since Xi appears more volatile than the typical Chinese leader, and for that matter China is undergoing many more internal stresses that could lead to chaotic behavior on the world stage.
Ethics and the future of humanity have squat to do with how I assess American foreign policy, however. The United States is not by itself responsible for the future of humanity, still less for righting the world's wrongs. The only goal I admit lies within the ambit, if not always (alas) the competence of the US President, is to advance and guard the interests of Americans.
As long as the US are as powerful as they are and even want to be the worlds hegemon, US foreign policy has a lot to do with the fate of very many people all around the world. Power always brings responsibility. The US don't have to take care of the worlds wrongs, but they should take care not to create harms. I think ethical people should always oppose the powerful dealing out harm out of selfishness or carelessness. Not caring what harm your democratic government causes abroad makes any American loose a lot of credibility when talking about ethics.
That's absurd. If I issue an opinion on ethics, and someone says "Ha! Your argument is wrong because you're an American, and in 1914 President Wilson ordered the Marines to occupy Veracruz, which was deeply unethical..." then that person is an idiot, because first of all that's an ad hominem argument, and second of all I'm not President Wilson. I couldn't care less about my credibility in the eyes of a mental pygmy.
I was talking about current actions that are done also in your name. If you don't see the ethical problems of these, this tells me something about your abilities and ethics. So seeing how you judge different situations gives me a picture of your ethic models and on this picture I decide how much I trust you on other topics I don't have a clear opinion yet.
To make it more practical:
If you talk with a Russian about some conflict in Africa, how the actors see the situation and what would be a fair solution, I'm pretty sure you would give his opinion much more weight if he condemns the Russian invasion in Ukraine then when he calls it unfortunate but necessary.
Would you trust a Chinese on ethical topics who just repeats the official narrative when asked about the Uyghurs?
If these topics don't make you biased against ethical credibility of the citicens of these countries, than you are a postive exception. But usually humans build up bias and you can't avoid to adding to these biases with your behavior.
Well do bear in mind that I couldn't care less whether you trust me. You're not my boss, you're not my kid, and I'm not trying to sell you anything. Of what worth is your trust to me? Zip.
Your suppositions are both wrong, and from my point of view demonstrate a lack of common sense. If I talk with a random Russian about some conflict in Africa, I will judge the accuracy and wisdom of what he says according to what he says, and what I know independently. What he thinks of the war in Ukraine is of no more relevance than his opinion on abortion, the infield fly rule, or whether Oswald acted alone. The same is true about a Chinese man opining about the ethics of AI development. What he thinks of the Chinese government treatment of Uighurs has no more bearing on the question than whether he thinks the Treaty of Versailles was unfair or capital punishment should be banned.
So far as I can see, you're suggesting that I base my evaluation of the accuracy and utility of someone's opinion based on how much emotional simpatico I feel for them, which makes about as much sense as basing it on how sexually attractive they are -- e.g. I should trust Julia Roberts more than Al Gore on whether global warming is a problem, because Julia is a good-looking broad. To me this is repugnant to the very concept of being rational, which at Step 1 embodies the determination to make judgments via reason, and not one's monkey instincts.
I am sure many people do behave this way -- perhaps even I do, from time to time, being subject to the usual flaws in human nature. But why anyone should embrace or encourage this regrettable tendency by paying court to it is beyond me.
Is there a country on earth with a foreign policy devoid of hypocrisy?
Assuming there isn’t such a country (and there definitely isn’t one), “hypocrisy” is kind of a pointless complaint.
So I find it much more useful to evaluate foreign policy in terms of whether a particular *policy decision* is good or bad rather than a broad “do I think country x’s foreign policy is good or bad.” Iraq war bad, supporting Ukraine good, and so on.
There are systemic reasons for US foreign policy being what it is, and there are reasons for it being represented the way it is (hypocritically). I don’t think analyzing those is a waste of time. At the very least it would prevent people from arguing in good faith from the point of view that we (well not me, I am not American) are the good guys. But should it be a priority for the blog? Why?
Well, perhaps not a priority. But I had the impression that it's completely left out and I wondered why.
I agree completely. "My country is always right" is blind jingoism, and "my country is always wrong" is no better.
US foreign policy is a very polarizing issue, like many other issues of policy. One of the reasons I like it here is that this space is avoiding the reddit/twitter/fb death spiral, which got to the point where no meaningful discussion can be had at all without devolving in partisan fights. Changing policy, given the entrenched interests supporting the current status quo, seems an overwhelming task and one that other groups may be more suited pursuing anyway. Some people adjacent to the blog did start a think tank though; maybe they will be addressing foreign policy too? On a different note I want to believe that most people here see through the obvious, blatant hypocrisy to the point that it doesn’t make sense to point it out anymore, like you wouldn’t point out that some celeb’s mammaries are implants.
OK, so you see some kind of silent consent to avoid this topic because it's too polarizing?
I can understand this well on a personal level. But the relevance it has on so many lifes and possibly even all of humanity if we get into full WWIII, isn't it like a act of cowardice to ignore this if you can have influence?
I did discover Scott exactly because he did take the then highly emotional and polarized topic of politically correctness and did discuss it rationally without siding with one of the parties but looked on the topic from a sober rational perspective. This was very refreshing and i think also helped a lot to build understanding and to reduce stereotypes about the parties involved.
I think this is also needed for geopolitics, and the rationalist community could be in the best position to start this.
I don’t disagree with you but: can you have influence? How? Is it going to be good influence or would we be trying to control a complex system we do not know the underlying dynamics of? If you are in orbit, increased air resistance makes you go faster, not slower. How likely would activism aimed at affecting US foreign policy be to actually achieve a positive result rather than the opposite? I think there is a massive need for understanding stuff before taking action, as even well intentioned efforts may backfire. The whole forecasting community is working at this, at least from their angle. Developing solid international relation theory would help, obviously. If anyone here wants to discuss that, it would be a good start.
The influence we do have (Scott and the community of course more than me personally) is influencing public opinion.
I fully agree with you how important it is to understand things first and how bad well intentioned things can backfire. This exactly is my point: Internationally the US are often fast with actions (wars, unilateral sanctions, supporting coups, up to meddling with internal politics of other nations) and at least in my perception there is very little domestic reflection about it not before and not after.
US interventions did often have disastrous outcomes and the official reasons often turned out to be fabricated or at least not the real ones (otherwise the riorities would have been different).
As long as the domestic intellectual elite ( I assume most of the readers here belong to) does keep silent and does turn a blind eye on bad outcomes or even outright crimes, nothing will change except that in many parts of the world Americans are no longer taken serious when talking about ethics.
I expect a good chunk of the readers to agree with your point of view. As to what is best to do about it, if anything, I am not so sure. The US is the world hegemon, thanks in part to its ruthless foreign policy. I do not believe any amount of public opinion pressure motivated by ethical concerns would be able to change that. Even assuming that some intellectual elite had both genuine ethical concerns rather than mere posturing (which is doubtful already) and the means to reach out to the wider public, they would be able to influence policy only as far as no clear negative effects result for the main stakeholders (the big companies, banks, etc. running the show). At that point fierce resistance would ensue, first in the usual PR/politics sphere, and failing that even with overt violent attacks. To withstand that you need a broad power base (a party, money, etc…). You need to organize. Is changing foreign policy an important enough objective for this community to do that? I don’t think so at this stage. From a different angle: international power politics will not disappear once US hegemony is eliminated. In twenty years or so you might have a new hegemon or a bipolar situation, conceivably with China being the other pole. Will this situation be better than the current one in the average? Was the world more peaceful in the bipolar era when the US and the USSR faced each other? Was it more just? We might be able to answer these questions with data.
What are some aspects of American foreign policy you'd like to discuss? What are some you think we'd disagree with you about? You might be surprised.
This is what I'm doing here: trying to spark rational and ethic evaluation of US foreign policy within the domestic intellectual elite.
The change I want to see in the world is that world hegemony gets as outlawed as dictatorship and to have an world community of nations based on respect and common rules.
This is exactly what the US wants in their formal announcements, but in practice all the rules (like international law) are ignored by the US if it fits their interest. The great achievement of the rule of law is that everyone, especially the powerful, stick to and are judged by the same laws.
American Exceptionalism contradicts this, but can only be changed culturally from within. I think its overdue to discuss and evaluate US foreign policy with an ethical and long therm perspective, not just emotionally or from short term nationalist interests.
>The change I want to see in the world is that world hegemony gets as outlawed as dictatorship and to have an world community of nations based on respect and common rules.
Pick one. Because the list of nations which are actually on board with the whole "world community of nations based on respect and common rules" bit consists of A: the United States of America, and B: a bunch of nations who are also wholly on board with American hegemony because it means they don't have to worry about how to ultimately enforce the rules and impose respect.
American hegemony, for all its problems, is better than any other plausible alternative in this generation. And if you break it, we will hold you responsible for the vastly greater problems to come.
Opposing US hegemony is the main motive holding BRICS together at the moment. I'm pretty sure if any of them would try to be a new hegemon, the rest of them would join forces to prevent this and welcome the help of the US to do so.
But the alternatives for the US aren't only ruling like today or just giving up power to the next strongest. The US could live up to their proclaimed values by strengthening the UN and bringing in place checks and balances that there never will be another hegemon. I'm confident that BRICS would join this effort, they all fear other powers ruling over them much more than that they are driven by the wish for domination.
Especially China never had much military culture, but had much more focus on soft powers. Just read Pearl S. Buck to get an idea.
Your question is so broad as to be almost unanswerable. However, if you'd like to discuss US foreign policy specifically, this is literally one of the single most discussed topics here on planet Earth. There's almost no possible conversational, polemic or political topic covered more than US foreign policy, criticisms of, etc. Even if you recently landed here from another planet, there's no way that you couldn't know how ultra-discussed this is, literally All The Time, 24/7/365, for the last 80 years or so
Yes of course I'm aware of this, but I'm curious how this group is thinking about it. Knowing that you care about ethics, charity and a better future you should have a opinion on it.
It looks like the elephant in the room being ignored. Why is this?
Isn't all the thinking about existential threads in the far future relatively pointless if there will be WWIII resetting civilization to the middle ages?
So I'm specifically curious about the views of the rationalist community and I also didn't find this elsewhere.
I'm pretty sure the X-risk people do talk about nuclear war, although that's a very heavily studied topic so I'm not sure a group of talented amateurs can improve matters much there.
Also, most of the US's foreign policy blunders don't have anything to do with the risk of nuclear war. Running Guantanamo Bay might be bad for the US in a soft-power moral-authority sense but it's not going to be the spark that makes the missiles fly.
Your very comment proves to me, that there is way too less awareness of the connection between nuclear risk and US foreign politics.
The key to reducing the risk of nuclear exchange is trust and respect. This is necessary to
- convince counties not to develop WMD
- any kind of disarmament
- get a good treshold that false alerts are discovered as such.
It's one thing that the US isn't promoting symmetric disarmament and new treaties, but they are actively damaging international trust by not sticking to commitments and applying double standards all over the world combined with threads of sanctions.
But the worst is US officials musing about destroying or defeating Russia after years of ignoring Russia's safety concerns. This is about the opposite of deescalation. Of course this makes them nervous and having the finger on the trigger.
What is an example you are thinking of regarding hypocrisy?
- talking about human rights, but still having guantanamo.
- promoting democarcy while backing or even organizing coups
- insisting on international laws but but not sticking to UN resolutions
- tarling about a 'rules based world order' it is about to defend, while it seems that the main rule is that the US is the good guys and so is above all other rules.
- being super sensible about foreign influence in US domestic politics while trying to have influence elsewere (was the US media also so critical about the initiative for a law about 'foreign agents' in Georgia, that was basically a copy of the FARA-act?)
- wanting a international tribunal for Putin, but not prosecuting the own war crimes and even threatening Den Haag if an American should be prosecuted there.
- talking a lot about freedom of press an journalists while prosecuting Assange and turning a blind eye on suppression in allied counties
I could go on for a while...
Perhaps some of this is officially not directly the US, but NATO. I count as basically the same, as the US dominates NATO.
Yes. What’s your point? There’s things that rational us are interested in and things that they’re not and the more they are unable to affect something the less they are interested in it. US foreign policy is the example of something that it is almost impossible for an individual or small groups of individuals to affect.
Well this answers my initial question basically as: We are not interested and anyway have no chance of influencing public opinion against the MIC lobby.
A valid answer, but a little disappointing.
I mean, the general answer is that in any of those cases where the US govt. decides to act in ways that violate the principles it espouses, it found the cost of uphold those principles to be too high. Considering most people are not perfect exemplars of the principles they hold, I don’t find this to be super remarkable.
Hence calling it hypocrisy.
That’s fine. The original question is why isn’t hypocrisy in US foreign policy more widely discussed here, and my answer is that the degree to which the US is hypocritical in its foreign policy is fairly unremarkable considering some degree of hypocrisy is a fairly wide-spread human trait.
The other thing is that the US government is made up of a very large number of people, who don't all agree on everything. Some of those people are smart and rational and admirable, and some are massive jerks.
For every one of those supposed contradictions between stated principles and actual actions, you can find a whole bunch of people who will argue that "well actually in this special case there's no contradiction because..."
Expecting a large group of diverse people to hold consistent views is one of the most frustrating fallacies on the internet.
You assume that there always will be a hegemon. If I think about creating a better world this includes to install checks and balances that prevent any county to get too powerful. The UN looks like a serious try to start this, but instead to develop this path the US did everything to secure it's status as hegemon.
If you have like almost equals 4 of them can always join to keep one in check that tries to grab power.
At least the UN doesn't suddenly decide on a America first policy. This may not sound like a big improvement for you, but for the rest of the wold it definitely is.
It all depends how it is organized. There is a lot space between having a biased , unfair police dealing with double standards (this was the beginning of the topic) and not having any police at all. Who never tries has lost already.
And yes, I'm convinced that it is a matter of culture and wisedom to understand that violence doesn't give good solutions ever. In this regard I do trust countries that suffered major wars on their territory, have a long cultural history or were even colonized much more than the US. Unfortunatelly not just watching the stage of international dipomacy but also many of the answers in this forum confirm this.
I just read about the Wirecard scandal, "The biggest fraud in German history", in last week's New Yorker. https://www.newyorker.com/magazine/2023/03/06/how-the-biggest-fraud-in-german-history-unravelled
I'm amazed such a major scandal, which culminated 3 years ago, had been off my radar. Have I been living under a rock or has the US media been very uninterested in this story until now?
The part I found most fascinating was Germany's launching a criminal investigation against the Financial Times after Alphaville alleged massive misdeeds at Wirecard instead of investigating Wirecard.
I'm American and barely read the news and I still heard about Wirecard. YMMV.
Even in Germany is was relatively fast off the news again. Many think this could be because too many politicians with good friends in the media were promoting wirecard earlier and didn't want too much dirt coming up.
Listened to the Yudkowsky podcast (Bankless) from the last set of links and was struck by how hopeless he seems. Similarly, the "Death with Dignity" post seems to confirm that he thinks the situation is hopeless.
I don't personally agree, but for people who seriously think the world is going to end in ~25 years due to AI...
...why not try violence? Violence is usually reprehensible, but to somebody certain that the world is going to end, wouldn't taking out a few researchers be justified? It would send a strong signal to others considering going into AI research as well, that they might not be safe if they pursue that career path.
Where is the flaw in this utilitarian calculus?
Even ignoring that their typical ethical systems forbid this... Yudkowski & co tend to (IIUC) believe that stopping AI progress is futile, because the incentives in place force capabilities research to keep going forward. Thus, the rare envisioned survival outcome depends on the world suddenly becoming much more reasonable than expected, to prioritize alignment research in time, combined with being very lucky about how easy alignment turns out to be.
If survival depends on the world being unusually reasonable, and one expects that violence lessens (rather than improves) collective public reasonableness, then violence is bad. Less "heroically saving the world", more "unilaterally committing global suicide".
I dunno about that. Maybe have a look at my long comment about ways to interfere with the further development of AI. All involve dishonesty, tricks and coercion, but none involve anything close to murder. I'm not saying we should do this -- personally, I feel unable to even estimate how likely how soon ASI will be developed, and what will happen when it does, so if people were taking steps of this kind I would not be willing to participate. But I do think it's quitepeculiar that people who are *sure* that AI will kill us all are not thinking in a more practical way about how to avert the disaster. They seem to only see 2 alternatives: Shoot some AI CEO's (which is evil, and also would not work), or else convince the population of planet earth to see the danger (which is impossible). WTF is wrong with these people? THEY ARE TERMINALLY IMPRACTICAL! If they want to slow down or stop AI development, the way to do it is to use the kind of tools that are routinely employed by those in power to influence public opinion and force other powerful entities into line. Murder is rarely required or even useful in these coercions.
The flaw is that your intended target in this scenario is not a dummy full of hay at a firing range.
"Violence" is a historically useful tool (Ref: most of human history) for some aims at some times in some cases. But what people who suggest violence as a solution often forget is that their opponents/targets might be
1) Just as capable of violence, if pushed
2) Better at it
3) Heavily defended by those who are
4) inclined to become any of the above three if they feel their security is at risk.
You start stabbing anyone who works with a machine model, we can see how quick people who work with machine models either pick up self defence training or start carrying (perfectly justifiable) weapons. And incidentally, these are usually smart people with engineering proclivities, so they'll probably be mildly spooky if motivated to take security seriously.
On top of that there's an entire social apparatus built to minimise, prevent and investigate violence against your fellow human beings. And highly paid software engineers working white-collar jobs at heavily funded institutions can, all else being equal, most likely afford much, much better private security than the people trying to hurt them. You reach for the devil's tools, intend to pick up his sword, and find that some goon in a lab somewhere has pushed the autonomous riot control drone software to 1.03 because they felt worried.
"Society" in general is just better at violence than any group of motivated individuals.
This is all setting aside totally normal concerns like "your suggestion is illegal" and "Anyone suggestion violence is kind of an asshat, aren't they?" and "you realise lone snipers and or deep cover assassination cells must contend with the security apparatus of an entire country, a class of people seldom known for their chill".
Probably because it's difficult to imagine any kind of violence that has a snowball's chance in hell of working, as opposed to ending up with the perpetrators dead or imprisoned for the rest of their lives.
I mean, if you want to radically alter government policy all over the world, or at least everywhere there's a chance of AGI being developed -- say, the United States, Australia and New Zealand, the 28 nations in the EU, Japan, South Korea, India, China, Taiwan, probably a bunch others I'm forgetting -- then you need let's say of order ~40-50 violent revolutions to succeed. That's an incredibly tall order. I don't think any less than 1 milion heavily-armed, well-trained, polylingual, very well disciplined troops could pull it off. And it would be a bit tricky to recruit and train them without anyone noticing and asking questions.
Could you be thinking of assassination? Of key leaders in the field, say? That would require much less violence, maybe only 50 teams of 5 ex-Mossad types each. But it would also require you to have some very deep insight into *which* 50 world leaders in AI are (or will be over the next 10 years[1]) absolutely critical to the construction of an AGI, which would mean you probably have to at least be one of them, and have some really excellent ideas about AGI -- basically, almost be in a position to build one yourself. That's a pretty big ask.
Plus you better have some really excellent coordination, so that after the first 5-10 hits the plot isn't discovered and counter measures undertaken. You probably need to do all the hits simultaneously, to avoid tipping your hand. The level of military competence required here is pretty high. This is something even a big national government would struggle to pull off.
I think your final hypothesis, that killing one or two researchers would "send a signal" to the others, is implausible. In the first place, killing one of two would most probably result in the organization being rapidly uncovered and the members charged with a variety of crimes and sent to prison, or (if in Texas) to the electric chair -- particularly if the plot is perpetuated by programmer-type nebbishes, whose are not known for their skill in the dark arts of constructing felonious conspiracies, laundering money, and committing murder. It would certainly make the news, particularly around the time of any executions, but I can't see it dissuading anyone from going into the field for the next quarter century or more. On the other hand, I *can* see it resulting in anyone who ever expressed any unguarded public enthusiasm for strong measures to suppress AI research being put under a strong level of scrutiny for the rest of his life.
-----------------
[1] Which means you need to consider wiping out the most promising teenagers, probably before they learn to drive. This is definitely going to earn you a trip to the gallows or electric chair in a few jurisdictions.
Whaddya think of my suggested less bloody forms of coercion and trickery detailedter later in the thread?
Eminently more practical. But don't you psychologists have some kind of code o' professional ethics that comes into play here? With great power comes great responsibility, Uncle Ben says. Although...maybe you're keeping the really good stuff to yourself. Yikes.
I'm a mandated reporter regarding neglect or abuse of children, the elderly or the disabled, and I have in fact made about half a dozen reports of this kind . Except for the above, though, I'm just a regular person who's spozed to follow the law. As for the various forms of coercion I was proposing as a way to shut down AI development -- jeez, I'm worried about the effect on people of current and near-future versions of AI, but cannot figure out how worried to be about AGI. I jusdo not know enough about machine learning, etc., though I'm working on catching up. I can't even figure out, as a psychologist, how seriously to take Yudkowsky's conviction that we're doomed. It might be that he's the Temple Grandin of computers. (Temple Grandin's an autistic woman who's said to be extraordinarily gifted at intuiting how animals experience certain things, and has designed systems for housing, transporting and killing farm animals that keep them much calmer and healthier.) Or maybe Yudkowsy's a deeply eccentric, sad, brilliant wrongheaded guy. So at this point I would have no interest in being involved in any sort of French Resistance against AI.
But the people who are sure AGI will destroy humanity soon sound hilariously naive & impractical when they talk about taking desperate and illegal steps to stop AI development. "Um, well we could shoot somebody. Or, like, blow up a lab. But then we'd just go to jail, so that's off the table. " My point is that if you think desperate and illegal measures are justified, shooting Sam Altman is not the way to go. There's all kinds of stuff that would be several orders of magnitude more effective, and does not even involve violence. Get it together, guys.
I meant, don't you have guild rules against giving people ideas based on your superior understanding of psychology? But it was a joke anyway.
I've read Temple Grandin's book, and I found it deeply fascinating. Not just because of what she said, but because it was clear from the way it was written how alien is her thinking. Her ability to convey (indirectly) what it's like to think like her is astounding -- the more since any expressivity she has is the result of sheer hard thinking and logical deduction, since she has no way to do it the easy way -- put herself in the shoes of her listener -- the way the rest of us do.
I knew someone who was very interested in Grandin's work, and introduced me to the book, not so much because she was interested in autism per se, but because she was interested in how animals really think. She wanted a non-Disney non-sentimental insight, and thought Grandin was on to something. As a result of that association, I became myself a little interested in how animals really think -- and I feel like pondering that is one of the big misses people routinely make when they think about thinking in general. We introspect, and draw conclusions from that. But we often completely neglect the fact that we share the planet with a large number of other species which, while not quite at our level, certainly think and plan and learn. There's probably a lot to learn by understanding how what they do differs from what we do.
Yes, of course the AGI fearful are not very practical. I'm a cynical old wretch, so I tend to think it's because in their hearts they really know it's not real. They're not *really* afraid, they're thrilling-afraid, the way people are at a slasher horror movie.
Ooh, have I got a great book recommendation for you: https://www.theguardian.com/books/2016/feb/03/being-beast-charles-foster-review-man-whoate-worms-like-badger. I read it a coupla years ago.
I think I am hampered inunderstanding animals by sentimentality. My whole family's like that about animals. I feel like they're my children. But they're not. My cats are much more matter-of-fact about dying than I am about their deaths. I take their wordlessness to be chilidlike innocence & that's not really what it is. I should maybe get some tarantulas. Pretty sure I would not be sentimental about those fuckers.
Yeah, the whole Yudkowsky scene sort of squicks me out. It has a dysfunctional family quality that's familiar -- has to do with everybody agreeing to pretend something. If the AGI fearful are just thrilling-afraid, listening to Yudkowsy tell his horror movie story, then who's Yudkowsky -- Boris Karloff? A year or so ago he posted a sort of am-I-joking-or-am-I-serious thing, and I wrote a furious diatribe about it here, then afterwards wondered if I'd sort of overshot.
"why not try violence? Violence is usually reprehensible, but to somebody certain that the world is going to end, wouldn't taking out a few researchers be justified? It would send a strong signal to others considering going into AI research as well, that they might not be safe if they pursue that career path.
Where is the flaw in this utilitarian calculus?"
Hello! Have you considered the cases of abortion clinic bombers and those who kill abortion providers?
Did that stop abortion, or was the Roe vs Wade Supreme Court decision what made the difference?
You start killing AI researchers, you make them Martyrs of Science, you permanently blacken the name of your cause, and there will be drives, campaigns, and stirring stories sent out as PR shots to the media looking for more people to take up the mantle of AI research in the name of liberty, progress, and standing up to the terrorists.
Look around you at the real world events where people tried violence, from Basque separatists to animal rights activists. See how effective that was, versus slogging away with the ballot box and the law courts.
Violence actually HAS worked for a number of groups, including terrorists. I tried to make that point to Erik Hoel:
https://entitledtoanopinion.wordpress.com/2023/02/17/a-comment-i-would-have-made-at-erik-hoels-were-i-not-banned/
It may not be the usual result, but the assassination of Shinzo Abe by a hater of the Unification Church (Moonies) is a recent example of violence achieving its intended political result.
Violence – everything from assassinations to violent uprisings – does 'work' at times, to achieve political or social ends. However, in one study of 323 instances, "nonviolent campaigns were nearly twice as effective as armed campaigns over the past century."
https://wagingnonviolence.org/2012/07/participation-is-everything-a-conversation-with-erica-chenoweth/
It won't help, and will likely make things worse. A few researchers won't make a huge dent in AI progress, but it will completely write off AI-doomerism as a bunch of deranged psychos. AI-supporters are already *preemptively* writing off AI-doomers as violent lunatics based on their rhetoric. A terrorist attack would likely *accelerate* AI progress accordingly.
I read a comment about this by somebody who's involved, and they said violence would just get some of them thrown in jail and would not stop the progress towards AGI. But i seems to me that what the writer had in mind was a very naked, naive kind of violence -- just go fucking shoot Sam Altman and a few others. Here's a question: If you were going to be utterly amoral about it, what could you do to most effectively slow or stop progress on AI. Here are some utterly amoral ideas:
-design computer viruses that will mostly be active against machines and software crucial to AI development. Try to make it really contagious, so that it gets into pretty much any machine working on AI. Plant another virus, virus #2, that will stay dormant until virus #1 is gone.
-Pay some genius Israeli supercyberspy to do that stuff, and also to propose ideas of his own. See if you can swing it so that the person who is called in to HELP with the virus is in fact the superspy who designed it, or someone working in league with him.
-Scare the hell out of the public about AI. Do not bother telling people the real reasons to be scared. Instead, play on the fears they have. For people in the anti-vax anti-mask faction, tell them that in the near future we will be putting AI in charge of mask mandates and vax mandates. AI will send out drones to public areas, photograph all who are unmasked, later send them tickets: $100 for first infraction, $500 for second, prison stay for third, plus your name goes in a "non-compliant citizen" data base. AI will do the same with vaxes, and will send special cute drones to playgrounds to convince children to be vaxed without their parents' knowledge. Say AI is going to install cameras at every intersection and turn us into a surveillance state just like China. Involve Fox news, Russian trolls and any other assholes you can think of to amplify these ideas. Say AI was developed by wokies, so will be enforcing woke values right and left. It will be a crime to say trans women are not women, and a hate crime to snigger if a 6’2” trans woman passes you on the sidewalk. Say AI has no religion itself, and is looking for a chance to dismantle religious values.
-I leave it to others to figure out how to freak out the Left. But it’s not hard to do, you know?
Let's see, what else?
-Blackmail crucial figures in AI into backing off.
-Steal AI's funding. Sell the AI companies on some crypto scheme that somehow collapses into mud and fromg spawn.
-Make governments fearful of AI, by spreading rumors that the developers of the big AI's picture themselves running the world, using the AI as a tool.
-Publish op-ed's supposedly written by AI that will alarm the hell out of the right, the left, the government . . ..
I think a much better approach would be to lobby for legislation that demands the AI companies make their work as sandboxed and capability limited ( 'shackled', one might say ) , and possibly significantly raise tariffs on compute capacity.
This approach still feels like it has flaws (The AI companies will tend to route around the regulations instead of truely abiding by them), but has a lot less political blowback.
1) Most people in congress do not understand tech. 2) The tech companies now have so much power and money they're almost like a second, alternative body shaping how the country runs. Nobody can put them in a sandbox. 3) Congress is dysfunctional
"Say AI has no religion itself, and is looking for a chance to dismantle religious values."
r/Atheism is very interested and would like to subscribe to your newsletter. Seriously, if you try saying "AI will disprove religion!", there will be plenty of people falling over themselves to promote it.
You might even succeed in getting a few of the anti-AI types to switch sides - after all, if the pope (any pope, take your pick) is the worst person in the entire history of the world, how bad can AI be by comparison?
AI is waiting for it's "NOW THERE IS" moment anyway!
Huh. I always figured by "violence" people meant "bomb chip factories in Taiwan".
I’m personally going to wait for at least a couple more genius autodidacts to throw in the towel before I get seriously worried about the future of humanity.
Re: the future of humanity - we're all gonna die one day. As individuals, that's a solid sure thing. As a species, eventually the sun will go ka-blooey and we're done.
Humanity doesn't have a future because everything will come to an end one day.
That being said, that is no reason to make things worse if we can avoid it. But AI doomerism may or may not be right - it's very difficult for me to believe that the machine intelligence will get some kind of agency and act out of its own plans and desires, rather than being yet another tool that we use to poke ourselves in the eyes with.
Have you read up on the alignment problem? My view used to be a lot more like yours before I did...
> As a species, eventually the sun will go ka-blooey and we're done.
I'd characterize that as more of a 1 billion year countdown to manage a colony fleet over to to the nearest dwarf star so we can keep going from there.
Cheer up! Maybe we will end up being the tool that AI uses to poke itself in the photoreceptor.
All this reminds me of Alfred Bester's 1974 serialised story turned into a novel, which is either "The Computer Connection" in the USA or "Extro" in the UK.
This may have been before the Internet as we know it, but he got it right about the kinds of products and services that would be advertised:
"A capsule floated down on top of the bods with its jets spraying fireworks. A blue-eyed blond astronaut stepped out and came up to us. “Duh,” he mumbled in Kallikak. “Duh-duh-duh-duh. . .”
“What’s this thing selling?” Uncas asked.
“Duh,” Fee told him. “That’s about all the honks can say, so they named the product after it. I think it’s a penis amplifier.”
This may sound familiar, but since he was writing in the 70s it's not so great a stretch to imagine campus protests going on forever:
"The bods at Union Carbide, our local university, were mounting their ritual protest. It was the traditional daily rioting, with screamings, burnings, and killings. The only thing that changed was the cause, and the pressure groups had to sign up months in advance for representation."
I don't recall it, alas. But Bester is a fascinating read. He always makes me feel like the cracks are showing in the Universe, a bit, like maybe any moment God will shake the snowglobe and everything comes to pieces and settles down in some other very, very different configuration. And his characters always seemed to be this hypergolic fusion of massive id drives and extreme cunning. Crazy guy.
Aside from the ethical red line, it's just as likely to cause a martyrdom effect as to discourage anyone. Worst case, Congress passes a bunch of laws putting AI alignment folks on the same level as ecoterrorists and the like, effectively making it even harder to accomplish anything .
I mean, if you believe the alternative is total annihilation, an ethical red line isn't really that much of a strong argument. And it may not be unethical to you at all if you believe these people are doing bad things.
Plus, presumably you'd do this once you think all other alternatives are exhausted, which Shivers believes Yudkowsky thinks.
Principles shouldn't change just because the stakes are higher; the whole point of having deontological injunctions is that arguments for violating them are much less trustworthy than they appear
The question was specifically in regards to "utilitarian calculus", not deontology.
oh, i guess that's true
i think this comes down to the whole thing where the Sequences had the general message "utilitarianism is correct for ideal reasoners, and you should generally keep utilitarian principles in mind as a sanity-check against negative-sum behaviors, BUT humans are absolutely not ideal reasoners and we need deontological ethical injunctions, learned through grueling empiricism, to protect us from our own cleverness"
but as far as actually deciding when to use utilitarianism versus obeying your culture's hardcoded deontology... there isn't really a clear answer, and the edge cases are extreme, and it's generally best to err on the side of caution
there's also some game theory here!
like, sure, i can come up with some very strong arguments for doing bad things to AI capacity researchers so they can't end the world. And the urgency sure seems pretty staggeringly huge. But there are a lot of people who can come up with strong arguments, according to their own contrarian beliefs, for doing all sorts of wacky violent things. If I claim that "i believe that I'm right" is justification for discarding my deontological restrictions, I lose the right to expect others to also obey those same restrictions. And I think a world where everyone has to obey those restrictions is probably better than a world where no one does. Symmetrism, the rawlsian veil of ignorance, etc etc etc
The cached response I've come across (from memory, and potentially incorrect, i.e. I don't know what proportion and to what degree those in the AI safety community believe this) is that even if you use violence to disincentivize dangerous work in AI in the near-term, it is still the case that, given things like competition, accessibility of compute / repositories, and the number of people interested in AI, unaligned and dangerous AI will still eventually be developed. Of course, this belief seems to assume that there is would be an insufficient amount of willingness / capability to commit violence of this form (i.e., against those working on and enhancing general AI capabilities) to actually make a lasting difference. So, the potential impact of such violence is small, given that the population of violent actors and the damage engendered by each actor would likely be small, among other things, and that there are too many labs and corporations actively working towards AGI in severely subpar fashions.
I will link you to this portion on violence in response to AI from Zvi's latest post:
https://thezvi.substack.com/p/ai-3#%C2%A7we-reject-violence-as-we-do-the-devil-in-all-its-forms
He's not quite as doom-and-gloom- as EY, but he's not far behind and I think his argument is relatively generalizeable.
To give the 1 sentence summary (but really, go read it): Violence will not and very likely _can_ not make the situation better. Even if you honestly believe that the current path will, with near certainty, bring about the extinction of humanity (which I'm not sure that EY actually thinks this, he just thinks it's very likely), violence will just make that extinction a worse process, not actually prevent it. I pretty strongly agree with Zvi's take here.
I don't think that violence is _never_ warranted, but I think that for a situation like AGI and our concerns about it, it's a situation that violence is almost certainly not capable of resolving.
I haven't read Zvi's post yet, but his conclusion seems reasonable, except in scenarios where very severe violence (widespread and harsh) occurs, but I believe such scenarios are very unlikely to occur.
When Joe Rogan interviews Coffeezilla, the idea comes up that the crypto community needs to regulate itself-- the amount of fraud makes crypto an unattractive investment. Is there any way for the crypto community to do this?
Sure, you could theoretically set up a clearing house. Something like a crypto code authority that gives its stamp of approval. Some of the bigger exchanges kind of do this already by permitting different tiers of respectability. But I expect there's a lot of people who will resist this for a combination of ideological and "actually, it is a scam" reasons.
What's actually going to happen, in my opinion, is that the crypto con artists are increasingly going to be crowded out by the more respectable projects that can bring in big institutional money. And eventually this will create enough governmental institutional knowledge they can effectively enforce existing anti-fraud laws (which they currently don't). Meanwhile there'll still be a small hacker ecosystem that will be the wild west. Including no end of scams but also productive stuff like unbanked payments.
So far all of the more respectable projects that brought in big institutional money ended up being the very biggest scams though
it's gonna be really hard to wait out that particular association
Which projects are you thinking of? If you're thinking of FTX that was mostly privately raised money and VC. I'm thinking of stuff like the E-Yuan or the various African experiments or government backed remittance companies.
oh huh. i guess I was thinking about projects like Luna, Celsius, Safemoon, Terra, or Tether
although all the coffeezilla videos tend to blend together... my general, possibly inaccurate, impression is that some or all of those projects got quite a bit of excitement from institutional investors
i guess that still counts as private money... but it still feels like the whole field has some pretty bad PR toxicity
somewhat unrelated but, i'm a little weirded out by projects like the e-yuan. the whole point of cryptocurrency, the thing that makes it more valuable than nothing, is that you can take advantage of hard cryptography to guarantee aspects of the coin, like its future scarcity, or that your ownership can't be taken away from you. i can certainly see why, in certain situations, these features would be valuable. especially for consumers who are worried that governments might abuse their ability to print fiat currency to take value from holders, or worried about banks obeying instructions from the state to freeze or steal their assets. but that's like... the only real benefit of crypto over fiat, the leveraging of hard cryptography to make it mathematically impossible to mess around with the currency.
why on earth would china want this? it seems like the exact opposite of what they'd want. from reading about it, i can't honestly determine if it's even a cryptocurrency... it seems to be just another fiat currency.
i have some of the same questions about some of the other projects. i can see why, in an environment of distrust, having a currency where all funnybusiness is literally ruled out by mathematics would be very useful. but the whole reason there's currency funnybusiness in the first place is that the public institutions which control the currency get a lot of value out of that funny business. why would those same institutions want to introduce a currency which robs them of those powers?
and if the new cryptocurrency doesn't rob them of those powers, it's not really a cryptocurrency, it's just another digital currency. maybe it'll have some features which distinguish it a bit from just keeping your yuan in an online bank... but not the features which made cryptocurrency valuable in the first place?
idk, sorry for the rant, just kinda spewing my uneducated thoughts out
So when I say institutional money I mean, for example, Goldman Sachs or the Federal government or that kind of thing. Most scam crypto was done through shady offshore banks or through large groups of private people pooling money. Lots of small depositors. Not all of them unsophisticated. There were plenty of finance bros. But it wasn't like FTX was a Wall Street creation. They were in the Bahamas for a reason.
I absolutely agree about the PR toxicity. My sincere hope is that this market downturn purges out the worst of the rot.
Cryptocurrency is a specific technology. Libertarian leaning crypto types invented money with a bunch of security features. But the security features are not inherent to crypto and the E-Yuan does not have them. It has the opposite: features meant to enhance state control. Nevertheless it still has uses on a unit of currency level in assisting banking transactions etc. It still lives on blockchain. Etc.
The advantage of cryptocurrency is that it allows decentralization. If you're some libertarian living in Central America maybe you use that decentralization to create a highly secure currency meant to prevent government control. If you're the Chinese government you use it to speed up banking transactions and make them more transparent to regulators and censors with less computing power.
The diamond exchange is a non-governmental community that self-regulates astonishingly effectively. But that's a small, insular community. I can't imaging that would scale to crypto.
You don't necessarily need a government. Securities exchanges (i.e., groups of organisations that run a communal trading system; the members are those that have a "seat" on the exchange) have some pretty tight regulation and concerns about fairness and transparency that go beyond what government regulators require and are better enforced, since the members consider it important to keeping the system working.
There's a lot of overlap with government regulations, but the internal ones in certain areas are stronger and better enforced because because it's the exchange members themselves who are stuck with the bill if something goes bad. Margin requirements, for example, are always going to be carefully managed regardless of external regulation because as a member of the exchange if someone for whom you're placing trades can't settle, you're on the hook. (And if a member can't settle, everybody else is on the hook.)
I can see this kind of thing happening in cryptocurrencies, too, though probably not to any great degree because the incentives are different and there's a greater number of actors that are more concerned about benefits to themselves rather than to society.
I don't think that most people would agree with you that the Chicago Mercantile Exchange is a "government."
And you can't participate in cryptocurrency ecosystems without following the rules, either. Try submitting a new block without a proof-of-work and you'll find that nobody will accept it.
I have two subscriptions to Razib Khan's Unsupervised Learning to give away. Either reply with an email address here, or email at the address I specify here: https://entitledtoanopinion.wordpress.com/about
Curious. ethoscastelli@gmail.com
Sent, and people in the future who want them should monitor this page:
https://entitledtoanopinion.wordpress.com/2023/03/12/permanent-post-for-giving-away-subscriptions-to-substacks/
Thanks so much! Have just responded there.
This was a kind offer the first few times, but it's getting annoying to have it in every open thread; please stop from now on.
Thank you.
Sorry, will do.
A belated comment on PMS and culture-bound mental illness:
1. I really thought most women, say 80%, get a little grouchy and weepy before they get their periods. Concluding from my experience and those of my mom, friends, aunts, etc.
2. Like one of the immigrant women who are surprised by PMDD diagnoses, I wouldn't call PMS a mental illness. I'd class it with normal mood changes like people getting cranky when they're hungry or have slept poorly. I wonder if that's what the surprised immigrant women figure too.
3. On top of the hormone-caused mood changes (I'd guess?), women probably logically get a little grumpy before our periods because we're about to have have stomachaches and constrained clothing choices. This feeling is like being annoyed when you realize you're coming down with a cold or when seasonal allergies start up.
I think it's likely that on average women are indeed more dysphoric before their menstrual periods. But the fact that this mood state is hormonally driven is not grounds for classifying it as a "mental illness." Lots of states are biologically driven: fatigue, horniness, craving to eat. . .
#3 is actually a possible counterpoint to #1. What people have observed may be just regular grumpiness from pain and discomfort.
So the whole PMS idea may be just mentally pathologising stomachaches.
No, for my mom and me at least, there's an increased readiness to cry at sappy things which does not come from stomachaches. Also, I wouldn't call it pathologizing, but rather a normal emotional response.
Re: #1, in my experience as a man, a little introspection shows that male mood changes throughout the day exactly as one would expect from how testosterone generally changes throughout the day. I'm baffled at the idea that hormonal changes wouldn't have any effect for women too. The size of the differences can vary but it someone notices no difference at all I suspect the introspection component.
In my experience, I was somewhat irritable and in some pain from PMS, but neither were severe.
From personal experience as a woman - I've never been affected by PMS.
Seconded
Anyone out there currently using chatGPT in their day-to-day analytical work? If so, how are you currently using it? I'm interesting in trying it out in my job, and some applications of it (e.g. https://workspace.google.com/marketplace/app/gpt_for_sheets_and_docs/677318054654) seem great but I'm hesitant to try it out.
I use it for generating SQL queries, which it's quite good at. It's rarely 100% correct and usable off the bat, but it taught me about the LAG window function in a helpful way, and put me on the right track for saving the results of a Clickhouse query to a file
Another useful application is summarizing difficult text as you go. I’ll take a textbook paragraph, pop it into ChatGPT with a request to rewrite it in a conversational tone. Then I real the new version, compare it with the old, and often have an easier time building a more confident understanding of the material.
I use it as a language editor, to get boilerplate code fast, for translation sometimes, to find quotes that I remember having seen somewhere, to learn new stuff about a topic on which I have no mastery of the technical jargon
I've been using it to help speed up my coding process in pandas. Literally every answer it has given me has been wrong and I think I've used like 1 line of its actual code unchanged, but it speeds up my thought process of "what structure should I use here?" drastically. Where to implement functions to be called in apply functions, where to use lambda functions, where to use for loops, etc. Stuff previously I would have thought about for an hour or so before starting to code up a solution, or spent a couple hours trawling through stack overflow, I just get an immediate pointer.
I use it as a form of autocorrect on steroids. It lets me brain-dump without worrying about polish, then produces legible text I’ll actually read or feel comfortable publishing. It’s also useful for transforming my language into unfamiliar wordings. If like me you usually need a day or two of distance from your text to be able to edit it, you’ll understand why this is helpful. It can also be useful for taking a long set of bullet point notes and producing a condensed summary.
It can be useful for generating critiques of your own ideas, or for identifying potentially confusing material.
I don’t find ChatGPT to be useful for producing new ideas, and I don’t trust it to teach me things, but it’s great as a sort of automated copy editor.
So I’m just learning that a lot of the highest level genetic biologists in the world actually try to deny that genetics play a factor in, for example, the NFL having 100% black corner backs. The reasons they give are really bad too (“culture”). White suburban communities eat sleep and breathe football, and pour endless resources into it at the high school level. I didn’t think the field had lost THAT much credibility. Can we trust anything that comes out of there at this point?
I don't automatically delete comments like this, but I think when you post something this controversial and insulting you owe it to us to explain what you mean. I think I'm somewhat well-informed on these kinds of topics - eg I've read about the potential genetic differences between East African marathoners and West African sprinters - but I'd never heard of this and don't know how to interpret it. What are "corner backs"? What's your theory on why genetics would cause them to be predominantly black? What's the evidence for this theory?
So it’s not that cornerbacks are “predominantly” black, they are actually 100% black and there hasn’t been a white cornerback in the NFL since the early 2000’s. Not only that, but at the highest level of college football, so 128 teams, starting corners are also 100% black.
Cornerback is a position on defense that is primarily involved in defending the passing game. It is the position in the NFL that is most about pure athleticism (in the sense of speed, quickness, and strength). Every high school in America has 2 starting cornerbacks on their football team, the amount of white kids vying for college scholarships far exceeds the amount of black kids. White suburban/rural areas eat sleep and breathe football as integral to their culture.
Essentially, denying genetics being a major factor here is flat out ridiculous. It’s the most egregious example I’ve seen of the trend towards insane mental gymnastics to deny anything resembling group differences. I had no idea this trend was actually in academic genetics/biology, until I listened to Richard Hanania’s podcast with a Harvard evolutionary biologist who basically denied any genetic aspect to the NFL/NBA being majority black. The specific cornerback claim was something I saw from another academic biologist online, which if you know the context is another level.
I’ll admit it the post was a tad lazy, lol. I know pretty much nothing about genetics, but a lot of times I’ll make a post on here about a topic I know nothing about and another reader willl chime in with some good recommendations.
A lot of comedians have mined the 'white corner backs' things for gold.
Shane Gillis has a hilarious bit about it regarding Alabama football (2 minute clip but specifically relevant part is 1:35):
https://www.youtube.com/watch?v=GfDilPmRgnM
Bill Burr has a good bit about it, too.
And yet no comedian has yet managed to mine the "black Nobel Laureate in Physics" field just yet.
I love Shane Gillis lol. He has that Gillie and Keeves sketch with the high school coach slowly coming around to integrating the team lol.
Easily one of the best debut specials in the last 5-10 years. Gilt and Keaves is fantastic, and he’s also on my comedy podcast Mt. Rushmore
I saw him in Baltimore a few months ago and good lord he absolutely killed it. Im pretty sure that material he was touring with is going to be his next special, and if it is the world is in for a treat because the man has improved tremendously from his last special (which was unbelievably funny).
I live in Philly, and there’s a weekly open mic in my neighborhood that I’ll go to just to hear the local comics work out material. Everyone from first time amateurs to local vets and low level professionals. A lot of the guests from Matt and Shane’s podcast are regulars there, like Lemaire for example. I really respect the fact that Shane will still support and give a platform to the comics in the local scene he came out of. Just good energy he puts out. Comics all around are interesting people.
>'What are "corner backs"?'
Come on Scott, I know people in the rationalist space are probably not big sports fans, but it would take a very quick google search to see that this is a position in American Football. He's saying that people of a specific race being so predominant in a specific position that requires specific physical traits and skills seems pretty obviously genetic, especially when any recourse to "cultural differences" is obviated by the fact that both white and black people in America obsess over football, and this wouldn't seem to impact the specific positions within the team people play by race, anyways.
I think my favorite ethnic factoid about the NFL was that, while Samoans represent something like 0.1% (from memory) of the US population, at the time of the article they represented 80% (also memory) of NFL defensive linemen, and went on to discuss the NFL’s efforts in player development in Samoa and Polynesian communities.
Your numbers are way off but yes Samoans are way over represented in the NFL and NCAA division 1 relative to population (which is like 60,000 people). All over the field, not just defensive linemen. Troy Palamalou is prob the most famous. Tua Tagavailoa is QB for the dolphins today. They’re generally a big and strong group of dudes.
There’s a guy on the eagles named Jordan Mailatta who is an Australian of I believe Maori descent. He is a 6’9” 350 pound freak athlete left tackle. He never played football in his entire life, but somehow NFL coaches got ahold of his rugby tape and taught him to game from scratch. Couple years later he’s among the best tackles in the league. I assume other NFL coaches are going to follow suit and we will see more freakish Polynesian rugby athletes in the near future. Everyone knows genetics are a thing except for the geneticists apparently.
Wikipedia says Mailatta's parents are Samoan.
I love the fact that there is a community that doesn’t automatically know what a corner back is. I haven’t watched an NFL game since Fran Tarkenton was playing because I knew that it produced a lot of former players who’s bodies were ruined by the game.
For what it's worth, I'd never heard of corner backs, and I thought it was a typo for quarterback.
As for culture, I'm wondering about pressure from better off/white parents to keep their kids out of football, even if the parents love watching the sport.
I imagine better-off parents would steer their kids towards something other than an athletic/sporting career, but again there must be poor to lower-middle class white kids whose best chance of making it big is to be a sports star?
American football to my ignorant eyes is a very stratified game with so much development about what positions are played, how they are played, who is best fitted, etc. I can well imagine it's gone way beyond "we need a big guy at the back and put the small nippy guy up front" when it comes to what kind of players fit into what kind of position. So it could indeed be "the reason all the high level players in this position nowadays are black is the same reason all the high level 100m sprinters nowadays are black".
Unless you can demonstrate that it's not an important or desirable position or one that has little effect on the game, then you can't say "well it's culture and systemic racism that the black players get relegated to this position".
I think I’m noticing that people who are generally unfamiliar with sports seem to have a poor concept of how rare NFL level talent is, and what the process of developing elite athletes from a young age looks like. Kids start playing sports at a young age, and if your kid even has a chance of sniffing the pro level, they will not only be the best player on their team, they will more than likely be the best player any of the coaches have ever seen in the history of the league. Even at 11 years old, they will be head and shoulders better than all of the other kids and it’s extremely obvious when someone has that level of special talent. Rich, poor, middle class, black, white, no parent who’s kid is dominating to the level that future pros do is going to pull their kid out of football. It’s just such a rare, unprecedented level of talent. So while I think there’s definitely parents pulling kids out, odds are that’s not effecting the top talent regardless of family background.
But you’re right, different positions require different skill sets. Positions that require brute strength and size (linemen) will have about an equal distribution of black and white players. Corner is one of the most important and highest paid positions today, as it defends the passing game. Extreme athleticism (quickness and speed) is emphasized more there than any other position. There hasn’t been a white corner on a full time NFL roster since the early 2000s (teams normally have 5 or so rostered). If you polled every NFL player and coach I’m certain 100% of them would say it’s genetics.
"Kids start playing sports at a young age, and if your kid even has a chance of sniffing the pro level, they will not only be the best player on their team, they will more than likely be the best player any of the coaches have ever seen in the history of the league. Even at 11 years old, they will be head and shoulders better than all of the other kids and it’s extremely obvious when someone has that level of special talent."
Yes, the equivalent over here are football (soccer) academies and teams having youth teams to bring through talented kids, with scouts attending matches to sniff out talent. Liverpool academy as example below:
https://en.wikipedia.org/wiki/The_Academy,_Kirkby
"Scouts attend many local youth matches looking for talented boys. A boy will then be invited to attend training sessions at the Academy. They are currently taken in as young as the age of six. Former England International player Jamie Carragher started at Liverpool when he was aged just nine, with Michael Owen joining at eleven, and Steven Gerrard joining at the age of eight. At this age, the boys start by simply attending after-school training sessions, but as they reach their middle-teens, their academic needs will be taken over by the Academy if they are deemed athletically talented enough. As such, the Academy has a lecture theatre and a computer-equipped classroom.
The Academy can handle up to twenty boys in each year group, although the actual number in each year group is usually around eighteen. Between the ages of eight and twelve the boys play in eight-a-side games of three twenty-minute periods. It allows the boys to play as defenders or as attackers in small groups within a system and is not as physically demanding as playing eleven-a-side matches."
Even attending an academy/being on a youth team is not enough; you have promising kids who are great at age 10, very good at age 15, not good enough to make the major team at age 16+.
I suppose I'm thinking of soccer over here. It's predominantly working-class. If you're middle-class to upper-class, you'll probably play rugby at school. Rugby union is now professional, but the money to be made is not comparable to professional soccer players:
https://rugbyroar.com/how-much-does-a-rugby-player-in-the-uk-earn/
That's possibly a hold-over from the days when union was the gentleman's sport and was played by amateurs. There's the old school tie at work too, where being a rugby player will (allegedly) get you advantages in your work career. If I had to guess, I think professional rugby union (as distinct from rugby league, that's another matter) is for the PMC (professional managerial class) background. Here's data from 2019 in Ireland, and I think that wouldn't be so far from how it is in UK as well:
"There is undoubtedly a perception that rugby is a predominantly middleclass sport, often associated with prominent fee-paying schools. The available data allow a quantification of the association between the likelihood that an individual plays rugby and their socio-economic background.
…Figure 2.7 shows the odds ratios for active participation in rugby by categories of educational attainment, as estimated in the statistical models provided in Appendix C. The reference category is individuals with upper second-level qualifications only (i.e. Leaving Certificate or equivalent). The chart displays a strong effect, such that higher educational attainment is associated with substantially higher odds of playing rugby.
The odds that an individual who goes to college or university plays rugby are approximately twice the odds for an individual who ceases education at Leaving Certificate. "
Upper class kids are not going to be professional sportspeople, unless it's something like round the world yachting or polo or golf or the like. Middle to upper middle class kids will be playing rugby (and/or cricket and/or golf). Lower middle to working class kids will be playing soccer (or Gaelic games here in Ireland), and that's where the big money is, comparable to the NFL (though clearly not on the same scale):
https://www.capology.com/uk/premier-league/salaries/
https://en.wikipedia.org/wiki/Rugby_union
"Rugby football stems from the form of the game played at Rugby School, which former pupils then introduced to their universities.
Former Rugby School student Albert Pell is credited with having formed the first "football" team while a student at Cambridge University. Major private schools each used different rules during this early period, with former pupils from Rugby and Eton attempting to carry their preferred rules through to their universities. A significant event in the early development of rugby football was the production of a written set of rules at Rugby School in 1845, followed by the Cambridge Rules that were drawn up in 1848.
...In 1895, there was a major schism within rugby football in England in which numerous clubs from Northern England resigned from the RFU over the issue of reimbursing players for time lost from their workplaces. The split highlighted the social and class divisions in the sport in England. Although the rules of the game were not a factor in the split, the breakaway teams subsequently adopted some rule changes and this became the separate code of "rugby league". The RFU's code thereafter took on the name "rugby union" to differentiate it from rugby league, but both versions of the sport are known simply as "rugby" throughout most of the world."
https://en.wikipedia.org/wiki/Rugby_league
"In 1895, a schism in Rugby football resulted in the formation of the Northern Rugby Football Union (NRFU).The success of working class northern teams led to some compensating players who otherwise would be on their job and earning income on Saturdays. This led to the RFU reacting to enforce the amateur principle of the sport, preventing "broken time payments" to players who had taken time off work to play rugby. Northern teams typically had more working class players (coal miners, mill workers etc.) who could not afford to play without this compensation, in contrast to affluent southern teams who had other sources of income to sustain the amateur principle. In 1895, a decree by the RFU banning the playing of rugby at grounds where entrance fees were charged led to twenty-two clubs (including Stockport, who negotiated by telephone) meeting at the George Hotel, Huddersfield on 29 August 1895 and forming the "Northern Rugby Football Union". Within fifteen years of that first meeting in Huddersfield, more than 200 RFU clubs had left to join the rugby league.
In 1897, the line-out was abolished and in 1898 professionalism introduced. In 1906, the Northern Union changed its rules, reducing teams from 15 to 13 a side and replacing the ruck formed after tackles with the play-the-ball.
A similar schism to that which occurred in England took place in Sydney, Australia. There, on 8 August 1907 the New South Wales Rugby Football League was founded at Bateman's Hotel in George Street.[19] Rugby league then went on to displace rugby union as the primary football code in New South Wales and Queensland."
I'm sure there's some regional variation in football's popularity at the youth level. But in the South, there are plenty of people of every race playing football. (And even if a larger percentage of white parents keep their kids from playing football, there are still more white kids than black kids in absolute terms.)
What I find more egregious than some geneticists (who probably don't know enough about sports to comment intelligently) blathering on is the media's constant state of despair over the lack of black NFL head coaches, despite the fact that there's usually about 3-4 of them at any given time (out of 32 teams). In other words, black Americans make up roughly the same proportion of NFL head coaches as they do the general population. It's true that there are lots of black NFL players, but being a good player (while perhaps helpful) is neither necessary nor sufficient for being a good NFL coach.
Also, the biracial Patrick Mahomes is a black quarterback, as the media loves to tell us, but the biracial Mike McDaniel is not a black head coach, at least according to some in the media.
The amount of white (and black) parents who push their kids into sports and invest thousands of dollars on camps, clinics, private coaches hoping they can get a scholarship still far outnumbers the amount who pull their kids from football bc of concussion concerns.
True, but the trend line is now pointing downwards for football specifically.
(A) "High school participation in 11-player tackle football was trending upward in the 21st century until the 2009-10 academic year, when enrollment began to wane, according to data from the National Federation of State High School Associations’ High School Athletics Participation Survey....
The latest survey release from the NFHS – the first since 2018-19 – shows that the 2021-22 school year was the first on record with fewer than a million players participating in 11-player high school football in America since the turn of the century. The reported total of 976,886 participants represents a 12.2% decrease from a 2008-09 peak...."
(B) "According to data from statista.com, in 2018, 5.16 million people over the age of six participated in tackle football in the U.S. By comparison, there were 8.4 million people playing in that age range in 2006, representing a 48 percent decline in 12 years.
Participation numbers began declining heavily in the late 2000s, falling under six million in 2011.
The numbers mostly increased between 2012 and 2015 before beginning to drop again, declining almost 19 percent in three years alone between 2016 and 2019...."
And quite often, the exact same people who tell you we need more coaches of whatever ethnicity in order to represent their team's makeup will also tell you we need female coaches. Either not noticing or not caring about that contradiction.
Wow thanks for the info! I wonder if, however, there’s an increase in participation in athletics as a whole? Kids have alternatives these days, obviously video games are what come to mind first. Every high school football team has (or used to have) a cohort of players who never see the field, and I imagine that today those kids would rather say “screw this” and do something else with their time.
There’s also a rise in specialization from a younger age, so multi sport athletes are rare these days. It used to be the norm where star baseball players were also star football and basketball players, for example, but now that’s rarer than it used to be.
>As for culture, I'm wondering about pressure from better off/white parents to keep their kids out of football, even if the parents love watching the sport.
That might explain different racial rates of being a football player in general, but I doubt it would be able to explain different racial rates between specific positions on the team given that you're already a football player.
If football players were selected for purely objective qualities any of that might be relevant, but they aren’t. Once you reach the pro level you’ve been repeatedly passed through the filter of what coaches want and expect; those who are objectively the fastest or tallest go to sports where those qualities alone are sufficient, like track and basketball, but cornerbacks are selected for, e.g., ‘aggressiveness’ and ‘persistence’ and ‘ability to read plays’ which are nebulous qualities.
You don’t even sniff a division 1 roster at corner unless you’re running a 4.4, and 4.3 for the NFL. Corners are usually the best athletes on the team. The nebulous qualities of “ability to read plays” etc that make up ‘football IQ’ are what separates hall of famers from starters, and starters from backups.
CBs are almost always among the fastest players, that's basically a requirement. Look up d green, sanders, etc. So you shouldn't be surprised if whichever population is fastest tends to rise to the top at CB. Also, I'm not sure why we have to pretend that X population, which is obviously both faster and more muscled in certain ways, might not be that way because of genetics. Is it really the case that (e.g.) ecuadoreans could have similar success as NFL CBs with the right culture? No one really believes that, so why can't the flip side be true? Discussion would be much more interesting if people would be more honest about certain facets of life (that basically everyone understands but must nonetheless deny). I get the worry that this type of wrong think can have negative impacts in other areas, but denying reality seems problematic for a number of reasons. And worth pointing out this is yet another reason not to trust scientists (next they'll tell us that certain folks in N. Europe or from certain parts of Africa aren't tall for genetic reasons, even though we know that's a highly (hah!) heritable trait).
I think the criticism of the original comment is more that we would need to see some specific denials that the cornerbacks in particular have any genetic bias. Also as to your point I believe in this comment section we are supposed to provide information instead of saying it can be easily acquired via internet search. In this case, a cornerback is somebody who needs a lot of running, jumping, and aggression.
I think the original comment was fine, as someone who watches a lot of football. I could see it appearing not fine to someone who doesn’t know what a cornerback is, however.
The original poster probably (maybe) lives and breathes in a world that doesn’t involve many humans who don’t know what a cornerback is, if he’s anything like me, and wouldn’t have thought about that issue beforehand.
As Scott has mentioned before, he lives in a world where 95% of the people he knows are not conservative (that percentage and “conservative” wording may not be exactly what he said, but I think the gist is the same). There are folks who live in the opposite world. I think it’s important not to slap those people in the face if they decide to poke their heads out here, especially if they aren’t straight up being insulting or acting insane.
The reaction to that comment is what keeps people like me from wanting to go to eg. ACX meetups. I am far, far different than most people here, but the ven diagram of me vs. most people here has a crossover, which should generally be considered a thing to be cherished. I’m here because I like the way you guys think; I steal it and I think it makes me a better person.
The OP’s post gave context to a problem without being verbose, and began a discussion. Most posts here do the same… just about topics generally well known by the “nerdy” community, and don’t receive any flak.
Let's not forget the world where many people aren't even Americans.
Did you mean to reply to my comment and discuss the original comment? I was responding to a reply, the "come on Scott you can google what a cornerback is" part, not castigating the original commenter for knowing what cornerbacks are.
It's interesting to note that you confused two different comments with each other and then concluded that people here might not like you because of football. Your supposition about us being prejudiced against you likely influenced your reading of the comments themselves.
I meant to reply to you, specifically the “I think the criticism of the original comment was…” statement. I was then expounding on that train of thought.
Sorry, I didn’t mean to imply I you were (or that anyone was) castigating the OP for talking about football.
I also apologize if I came across as saying you as a group are prejudiced against me! I simply meant that differences in groups can make context hard to find, and make people who otherwise exist in out-groups hard to place.
Again, nothing focused towards you or your comment in particular; just continuing the “criticism agains the OP is [insert something here]” chain.
Disagree about "aggression," that's a safety thing, some of the best CBs ever haven't been particularly aggressive, just faster than everyone else, including reflexes to break on the ball, etc.
Shameless (shameful?) self-promotion:
My the ebook version of my latest book is free until Friday, the 17th. It is memoir, essays, philosophy, travelog, tips, and photos. Professional mistakes and personal luck. Some of you have already read it; some like it, some find it self-indulgent and shallow. So it goes.
You can learn more (and read reviews) and get the file here: https://www.losingmyreligions.net/
I usually let people promote their thing twice per year in the Open Thread. I think you've exceeded this limit, so please stop from now on.
Can anyone link me to really solid pro-permanent-daylight/summer-time argument? Most of the ones I have seen tout benefits of shifting activities to earlier in the day. Which could also be accomplished by just, you know, shifting activities to earlier in the day. This would be disruptive if it happened a couple of times a year, but the change they propose is permanent (it’s right there in the name) so it comes down to a one-time disruption and preserves the benefit of clocks being roughly in sync with the sun (in other words, the reason that we have time zones in the first place).
The Argument, hosted by Jane Coastan did an episode on this a year ago. One of the few good episodes during her tenure imo.
People live by the clock, so we could either start all offices at 8, end work at 4, push all tv schedules and games and restaurant opening times back an hour etc etc etc - which is a massive co-ordination problem or change the clock. I’m not sure why you think that this problem is one that affects permanent DST either. It affects summer time regardless of permanency.
I'm interested to hear if anyone has any strong arguments against permanent *standard* time -- that way it's just, y'know -- time.
The one I typically hear is "I like to do stuff after work in the summer" -- which makes a certain kind of sense, but the number of jobs in which one can't just start an hour earlier and leave an hour later is rapidly shrinking -- I guess we can try the converse if we get stuck with permanent DST, but for some reason coming to work an hour late is viewed differently from coming in early.
Yes. The argument is that it’s a waste of sunlight, particularly in countries with long hours of sunlight in the summer. A country with 16 hours of sunlight should have a sunset at 8pm.
16 hours is the length of the Parisian day in midsummer.
However, Paris has a sunset of 9:45pm on June 21, which is actually great for tourists and locals. Part of this is because it’s in the wrong timezone (should be GMT) and part is because of DST. So Paris would have to do two things to get to proper standard time, change to GMT and stay in winter time all year. This would move the sunrise back from 05:45 to around quarter to 4, and the sunset to 7:45pm. You can probably see why they won’t do this.
I also noticed during Wimbledon (around July 4th) that the sun was still up in London about that time -- 9:20 on July 4th I now see upon looking it up.
That’s around sunset time. Where I now live the sun sets around 25 minutes later. When i lived in London I didn’t like their short summer days.
Most places are more or less in the right timezone, though?
Much of Europe isn’t. Maybe half of countries should be in different times than central time - which goes from the western Atlantic coast of Spain, which is as far west as the west of Ireland, to the polish border with Belarus. Africa is a mess too. No country is GMT+1 and plenty should be ( Iceland, and Western Africa). It’s all a mess.
https://www.timeanddate.com/time/map/
Again, it's a coordination problem: it's not just about when I personally choose to do things, but when other people are available, when things are open, when events are scheduled, etc.
Since different people have different preferences, the result is likely to continue to be a compromise that doesn't satisfy anyone wholly. But by the same token it's unlikely to be resolved fully one way or the other (permanent standard time or permanent DST) as long as there are competing preferences.
Where's the coordination problem though? Not everyone needs to start work at the same time. And if we are assuming that an hour either way is enough, there will still be at least 7 hours overlap between literally everybody, which seems like enough?
It's not just work. When are stores typically open? When does school start? When does one normally schedule a dinner, or a wedding reception? What time does the evening show start at the movies? When does transit shift to rush hour schedules?
There are a bunch of anchor points on the clock that people generally operate around. People with flexible schedules can work around them to an extent, but as long as they're interacting with others they're going to work within a framework of basic expectations that depend on common time.
"Whenever they want"?
The building supply/auto parts type stores in my area are already at least somewhat open by 7 AM, because that's when their customers tend to get up; landfills do have "winter hours" in which they close earlier because it's dark to the point of being a safety issue, head shops open late because stoners like to sleep in; there are many examples and none of it seems to cause problems. Business hours are easily available online, if you want to go to a store you will figure out when it's hours overlap with your non-work time and plan accordingly.
In practice, hours people choose for things are strongly influenced by clock time. As witness that there's "nothing" (aside from what everyone else is doing, which is to say the aforementioned coordination problem) stopping a conscientious objector to DST from keeping their schedule the same year round. Most just complain biennially and adjust, because actually doing that would be more of a hassle than going along is.
Its not an academic argument, but, I live in a jurisdiction which doesn't switch our clocks and it's pretty great not having that annoyance twice a year. We're at 52°N so sunlight changes wildly throughout the year, but changing the clocks wouldn't make any more of it.
I'm not going to do any better than anyone else at providing a link to a really solid argument. But while we're on the topic, I think that part of the reason that so much of the U.S. has persisted in the daylight savings switching ritual is that a) daylight savings time is generally better timing for many activities in the warmer part of the year and b) standard time is generally better timing for the winter. I know that not all agree with this, but bear with me. At mid-northern latitudes, permanent DST would mean that the start of many activities (work, K-12 school, university classes, exercise) would take place in the pitch black for a greater portion of the year. (To those that say - well start stuff later - I agree, and that is exactly what we accomplish when we switch to standard time - we all start stuff later.) I think that permanent DST will prove to be quite unpopular if it is implemented. Most people haven't fully imagined the implications; they just know that they hate changing their clocks. The optimal (IMO) modern solution to this is to continue the switch from DST to standard time, but to do it gradually. For example, if clock time shifted by 2 minutes every day for 30 days, the biological effects of the shifting would be minimal. And then our social activities could still happen on standard time for the winter and DST for the rest of the year. This would be hard for our 'dumb' clocks, but I still think this would be the best outcome. Many of those who want permanent DST need to look up history. It's been tried before and most people seemed to hate it. Are we to simply try it again because we weren't around last time?
I’d be less charitable and say that most people don’t know the sunrise times in early Jan or can’t add 1 hour. Particularly in northern latitudes - sunrise where I live would be 9:45am mid winter. Permanent DST is still better than the alternative, wasting 8 1/2 hours before noon isn’t that useful.
I was around for the last time, and I'd still personally prefer to go to permanent DST. (But I agree that it probably wouldn't survive for much the same reasons as last time.)
I can testify that last time it was next to impossible to drive a RR spike in the full darkness of 7:00 AM.
Thanks Carter.
Nixon. 🙂
Yep, you are right.
I'm personally one of the crazies that thinks that we should just have one global timezone to prevent distance-based time misunderstandings from ever happening in the first place, with the time on the clock places open and close being based on what time corresponds to daylight and night in that particular area. So if the sun rises at around 6pm in one area, that's when stuff starts opening, and if its at 3am in another area, that's when they start their day. Related to your question, this means that, as you say, you wouldn't have to shift around the clock to get more daylight hours at a particular time or whatever, people in the area that open up at 3am could just decide to open up at 2am if they decide that's better instead. For those that think this is insanity, China is all on one timezone despite spanning multiple geographically, and they seem to have figured it out fine.
That is literally inhuman, and the polar opposite of "systems were made for people."
I don't think that it really matters what the clock shows when you get up, go to work, go to bed etc. as long as it is consistent in one place. When people travel long distances, they need some time to adjust anyway due to time zone differences. Learning that the new place has different working hours is a very small additional effort.
Some people living among us already have a strange working hours. Someone is doing night shifts, another young person is playing videogames whole night and sleeping during the day etc. If I contact such person by whatsapp, I might not even know if he is on another side of the planet or somewhere in the neighbourhood just with a different sleeping schedule.
Not just China, even CET includes areas spanning from UTC-1 to UTC+2 in solar time. So switching the rest of the EU to CET wouldn't make the time zone all that much less coherent.
Only if we make it Swatch Internet Time.
> China is all on one timezone despite spanning multiple geographically, and they seem to have figured it out fine.
To be fair, China does it by not giving a flying fuck about anyone living west of Beijing.
The problem there is that for the 6 pm people, if they want to talk to the 3 am people, then somebody has to get up earlier/stay up later, and not unnaturally, nobody wants to do that. "It's three in the morning, I should be in bed rather than hanging around to talk to some guy who just started the work day!"
Going back to local time suits everyone in one country as we all see the sun rise at 6 in the morning or whatever, but we're a global world nowadays and Europe has to talk to America has to talk to Asia and so on.
And going back to time by the sun will mean differences in summer and winter - in summer, the day might start at 4 am but not until 10 am in winter, depending on when the sun rises. And if we make "12 noon = when the sun is overhead", then that too will mean shifting times as the earth goes around the sun and the seasons change.
It is easier to negotiate if the world has the same time zone but different local traditions.
If a US person wants to talk to a European, he writes an email: can we talk at 3pm? Right now I have to know what are the time difference and what 3pm in the US means for me. People forget about it all the time and make mistakes.
If we have the same time zone, I would immediately know that 3pm is not good because I will be already sleeping at that time. With time we probably would learn what are general working hours for each continent in the same way as we learn how some shops have unusual opening hours.
You want some people in the world to call midnight noon, and others to rise for the day at 6pm. Learning the hours of each working time zone is more complicated than a simple calculation which is largely done automatically anyway.
My Teams meeting today has been pushed back an hour in UTC because Canada has changed and we haven’t.
You don't need to learn anything, just negotiate time for calling each other. And for me it is easier to understand what is the person's working hours than calculate the time difference constantly.
I am with you. Of course, it will have strong opposition because because don't want to change things they are used to. But eventually it would help us better realize that we are a global community all together on the same planet.
A small problem would be that in some places the date would change during active working hours. Some transportation have daily tickets etc. and it would work well. But we would probably switch to tickets with 24 hour expiration time or something like that.
That’s not just horrible for much of the world, it’s actually more confusing for everybody but server engineers. I’m confused about the starting with the sun idea too - as that would change between seasons.
I'm not sure if China is a good example, as (because of the very uneven population distribution) for the majority of population Beijing time zone is "correct", almost everyone lives in the areas which would be at most 1 hour away from the "correct" time, and it seems that in Xinjiang, which should be 2 hours off, the "local time" is (or at least was) actually being used in practice by the local non-Han population instead of the Beijing time that's used by officials.
If setting a Zoom meeting between Germany, California, and Australia, it's pretty easy to use time zones to figure out their respective business hours. With one global time zones, there'd be nothing for it but to check each individual organization's hours. Most people would have no intuition about what someone else's local time looked like.
I'm also not convinced that an authoritarian country pinning time to the imperial capital is getting a lot of good feedback from its hinterland about how well the one time zone works for them.
With reasonable scheduling software, you can schedule meetings based on people's configured working hours rather than guessing based on their time zones.
The scheduling software does the time zones right now. Including handling DST.
I am aware of that. It's a lot more useful for the scheduler to know e.g. that I work from 6 to 10 and from 12 to 16 UTC than to know that my time offset is currently +1.
Is it? I mean that’s a strange schedule and it looks harder to input into software than the software knowing about timezones. It’s also not something helped by your fix.
I have talked to a Chinese co-worker about this (who has lived both in China and in Europe for several years), and he didn't think that the time zone system works better in China. It just comes with different issues to have a single time zone.
I think that usually the permanent-summer-time argument is made in the context of a status quo that has temporary summer time, in which case a solid argument is that the twice-a-year disruption has all kinds of documented harms and should be stopped; and that the choice of which of the *two* currently (and traditionally!) used time zones should stay permanent is relatively arbitrary, there's no solid reason to privilege "winter time zone" over "summer time zone" as the status quo (and all the habits) involve alternating between the two.
And as our daily schedule is currently so removed from sun (namely, that middle of average person's day is definitely not at noon as most people are far more active in afternoon/after-work than in early morning/before-work) that it doesn't make sense to value that zone in being in sync with the sun but rather being in sync with the people's preferences, which currently value sunlight at 7pm far more than sunlight at 5am.
And while this could also be accomplished by "just", you know, shifting activities to earlier in the day, there shouldn't be a "just" in that sentence; shifting all the activities in a community is a really, really hard coordination problem and in comparison switching to a different time zone is so much simpler and easier to achieve; so if people decide that such a shift should get done, a time zone change is probably the best way to implement it.
That’s a pretty solid argument.
Basically everyone hates the twice-yearly time change (although I suspect it’s a case of diffuse benefits and concentrated costs skewing people’s opinions), and we already spend most of of the year on summer time, so we may as well pick that one.
And while clock time is pretty arbitrary, it’s arbitrary in a sticky way: maybe 7-Eleven doesn’t need to be open from 7–11, but it would be pretty confusing if the 11-o’clock news came on at 10.
And there’s a pretty strong argument that the various factors in favor of an earlier start to the day outweigh keeping clock time roughly in sync with solar time.
> Basically everyone hates the twice-yearly time change
What makes you think that?
Speaking strictly about the disruption (of changing the time that everything happens by an hour over the course of a single weekend night), I hear a ton of complaining about it and I can't remember ever hearing anyone express unqualified joy at the idea (although some people enjoy the extra hour of sleep in the fall, but typically also lament the earlier clock time when the sun sets).
A number of brave individuals (e.g. https://leancrew.com/all-this/2013/03/why-i-like-dst/) defend the overall practice of DST, and I'm inclined to agree for the most part.
As I commented elsewhere I think that people are keenly aware of the concentrated pain of the twice-yearly change but take the diffuse benefit ("the sun rises at a reasonable time") for granted, and not having experienced the alternative, grumble about the practice as a whole.
I think "shifting activities to earlier" is not easy because it faces a huge coordination problem. Unless it's done by a global shift as we have it with daylight saving time.
TV channel have their traditional times for anchor shows. Sport events have nationally and internationally balanced schedules. Even locally, for shops or offices, it would be a huge communication effort to communicate that their opening/closing hours change.
Whether daylight saving time is good or bad is up to debate. But shifting activities earlier would happen to a much smaller extent if not coordinated.
All time zone issues are a solution to a coordination problem, which can't be accomplished by individually shifting activities earlier and later. The question is just where to coordinate.
(Permanent DST would be more convenient for me, lots of people favor permanent standard time. I expect we'll continue to wind up about where we are.)
I imagine individual organizations might adopt summer hours and winter hours (and on the whole this might well be worse than the status quo) but I’d prefer this to happen on a baseline of having noon be in the middle of the day.
I do come at this with the privilege of living less than 5 (time) minutes away from a meridian. But “round to the nearest hour, unless there’s a compelling reason to fudge it” seems like a reasonable compromise between fidelity to something like mean solar time and a manageable number of time zones.
Unless you are on a meridian it probably isn’t. The arguments against DST - that it is not natural time - are arguments against time zones, and in fact clocks at all. Because of equation of time clock and solar time drift during the year, even on Greenwich.
It (a shift in e.g. open/close hours) *would* come down to a one-time disruption.
>Which could also be accomplished by just, you know, shifting activities to earlier in the day.
Yeah, I never understood this anyway.
I like letting states decide. Indiana doesn't need to be on the same time zone as Maine.
I thought states did decide. Looking at a time zone map, it really doesn't look like the work of a central planner.
I think Arizona doesn’t switch.
States can opt out of switching to daylight time in the spring, but are at least nominally required to switch to standard time in the fall.
How is it to live on a yacht? How much does a yacht suitable for a couple to live on it cost? How do visas and countries borders work for sailing? Can you just cross the borders of countries wherever you want? Are there simpler to acquire visas for people sailing around? How good and reliable is the internet (satellite, I assume) when you're sailing in the middle of an ocean? Are there any blogs of digital nomads who live on a yacht? Let's say I buy a yacht in the Mediterranean sea. Is it possible to get to most countries by sea without crossing any borders on the way except for the borders of the target country?
Re: blogs, you want 100rabbits, the authors of some very cool software-as-art https://100r.co/site/liveaboard.html
Don't forget you can also set up on a river-going houseboat, though naturally without the blue-water access that a full on yacht would offer. It's probably worth checking out the costs for both options.
That's a lot of questions! I'll try to work through them. I've spent more than a decade in total living on different yachts starting as a baby, in many different parts of the world, and have done about 75'000 nautical miles in that time. Some for pleasure, some on expedition, some as delivery skipper.
- How is it to live on a yacht?
It's a lifestyle for sure. Or rather a wide range of lifestyles. Many port cities around the world will have yachts permanently moored and used as houseboats, but I gather that's not what you have in mind. On the other end of the spectrum some people spend a large portion of their time at sea, constantly on passage. But most cruisers will spend a few weeks or months to explore an area once you get there, and take the time they need until they feel ready to move on. This is not always possible or practical of course, as there are many factors that can force you to move on, such as the seasons turning, visas running out, or money. But fundamentally it's a matter of personal preference: Some sailors are most at home beating into a gale on the North Sea, others like martinis on a tropical island.
Make sure to fundamentally sync your expectations on these matters with anyone you plan to sail with, as this is a common root cause of conflict onboard.
- How much does a yacht suitable for a couple to live on it cost?
Depends on the size of your boat, you definition of "suitable", your level of comfort, your required level of safety, where in the world you are and plan to sail, and how much of the boat maintenance you are able to do yourself. Each of these categories are important, though not necessarily in order.
I'd say you can get a perfectly seaworthy blue-water capable with a reasonable degree of comfort starting around $10 to $20'000, up to ....infiniti basically, there is really no upper bound. In the $50 to $100'000 range many very nice boats can be found. Remember that the better shape the boat is when you buy it the less it will cost to get up to spec, though I would still always count on having to spend at least 10% or 20% or so of purchase price, even for boats that seem "perfect". This remains true if buying brand new, for most brands. If buying something in less then stellar condition, spending as much on renovation as on purchase is not unreasonable.
I would not buy a new yacht, even if money was no issue (unless perhaps I was commissioning a custom design). This has very little to do with the poor value per dollar, which is indeed often very poor for new yachts, and more to do with the fact that the modern paradigm of yacht design and manufacture is not suitable for real long distance cruising (In my opinion. There are exceptions). They are made to last a handfull of seasons in the med, and will not be designed with good sea-keeping qualities in mind. There are a few exceptions like Hallberg Rassy for example, but they are wildly expensive. In my mind the best yachts for long distance cruising were built in the late seventies to late nineties, though it differs hugely from yard to yard, and designer to designer. Some example boats that I would look at: Hallberg Rassy 36 or 39, Kelly-Petersen 44, Contessa 32. Buy the smallest boat that you can comfortably fit on: Every feet of length will increase the purchase and maintenance cost by 10%, compounded. (super over simplified rule of thumb, obviously).
Do pay a surveyor to check the boat out before purchase, it's definitely worth it.
- How do visas and countries borders work for sailing?
Visas work like normal: For citizens of most western countries, the world is pretty open, and you can get visa-on-arrival for several months, though it varies a lot from country to country. Always check during early planing. The USA is a notable exception, where most nationals must have a physical visa in their passport to be allowed to enter by sea on a private vessel, even if you could get a visa-on-arrival if visiting by airplane.
The basic procedure for switching counties is to clear customs and immigration and clear out of one county in one of their ports-of-entry (most major cities), and then leave within 24 hours. You will typically be asked where you are headed, but your answer is not binding. Then you sail to your destination, which has to be a port-of-entry in the new country, meaning a port that has customs and immigration offices. When you enter the new country's territorial waters, it is customary to hoist the new county's flag as a "curtesy flag", along with a yellow "Q" signal flag underneath the starboard spreader. Q stands for quarantine, originally meaning that you don't have the plague, and are ready to clear in: Now it just signals that you haven't cleared in yet. Until you are ready to clear in you are not allowed to come ashore for any other reason, unless you declare an emergency.
Assume that clearing in takes several hours every time at least, a full day in some countries.
- Can you just cross the borders of countries wherever you want?
You have to clear in at specific ports-of-entry, which is a major constraint in planning. But other than that you are usually allowed to sail into territorial waters as you see fit, as long as you don't go ashore. Territorial waters only extend 12 nautical miles from land. You might be questioned over radio, and some countries are much more restrictive (Like India for example). This is not usually a problem :)
You can always choose to clear out of a country whenever you like, as long as you are somewhere that has Customs and Immigration offices.
- Are there simpler to acquire visas for people sailing around?
No, not really, though see above about visa-on-arrival. When arriving somewhere on an airplane, the airline has done a lot of the hard work in making the bureaucracy smoother, on pain of being responsible to fly you back if you don't meet entry requirements. This is why airlines insist on seeing your visa/esta/eta before letting you board. If you arrive on your own boat, you have to fix all that by yourself. It's usually not a big problem, but it can be tedious, and put artificial constraints on your plans. This whole situation has gotten massively better over the last few decades.
- How good and reliable is the internet (satellite, I assume) when you're sailing in the middle of an ocean?
I've never had the luxury of true satellite internet while on passage, though for most passages over the last ten years or so I've had a satphone with a messaging service that will forward messages to and from email. This will also allow downloading .grib files with up to date weather data. Sending or receiving one text-only email or .grib file might take a minute or two.
Soon we might be able to use Starlink though, that promises to be a fundamental gamechanger, but last I heard it was not possible to use on the ocean yet.
The type of satellite internet connections that for example large cruise ships use are prohibitively expensive, are fragile, and use a lot of electricity to boot.
- Let's say I buy a yacht in the Mediterranean sea. Is it possible to get to most countries by sea without crossing any borders on the way except for the borders of the target country?
Yes, almost always. Territorial waters only extend 12 nautical miles out from the coast, beyond that is international waters. (Don't confuse territorial waters with the EEZ, Exclusive Economic Zone, that extends out to 200 NM. The EEZ only effect fishing and oil-drilling rights and such, and has no bearing on private yachts). Even aside from that, most countries don't mind if you sail across their territorial water without going ashore, though it varies from county to county.
- Are there any blogs of digital nomads who live on a yacht?
I'm sure there are! I don't know any of the top of my head, but here are some other links and book tips:
Anything by Lin and Larry Pardey, especially "The Self Sufficient Sailor", ISBN 0-9646036-7-5. They embody the attitude of getting a small affordable boat, and casting off soon. Without compromising safety.
Anything by Steve and Linda Dashew, especially "A practical guide to seamanship". They embody much of the opposite of the Pardeys, and their viewpoints complement each other well.
"The Voyager’s Handbook: The Essential Guide to Bluewater Cruising" By Beth A. Leonard, ISBN 0-07-143765-7. There are countless book purporting to teach everything you need to know to sail around the world. This is the best one I know.
"World Voyage Planner", by Jimmy Cornell, ISBN 978-0955639654. A list of destinations and inspiration and how to get there.
"World Cruising Routes", 7th ed, by Jimmy Cornell. The best reference for when and how to attempt which ocean crossings. Very information dense, but very dry reading.
https://www.noonsite.com/ A fantastic repository of crowdsourced information, by and for cruisers.
The "On the wind podcast" by 59 North Sailing. Many, many interesting guests over the years, and a long backlog to choose from.
https://www.youtube.com/@SailingUma -Sailing Uma, a youtube channel by a young couple sailing the world in a small boat. Very positive energy, and very down to earth, realistic picture of the pros and cons of the sailing lifestyle. Several episodes focus on their finances in depth.
https://johnkretschmersailing.com/ I especially liked Johns book "Sailing a serious ocean", which is a collection of stories from his years at sea.
Feel free to email me at johan.domeij@gmail.com for more questions. I'd love to help out, or just talk boats and sailing in general :)
The cost of owning a boat, plane, or horses can best be summarized by the question “Well, how much have you got?”
Yeah, there's a bunch of sayings to that effect: "The definition of a "boat", is a hole in the water into which you pour money", "Owning a boat will give you the two happiest days of your life: The day you buy it, and the day you sell it", and so on. Sometimes it certainly feels that way! In general I'd classify all of them as expressions of the fact that living on a boat is a complete lifestyle, and thus really will cost everything you've got, in the sense that it becomes the basis of your life, and informs every other decision. Maintaining a fully kited out and capable ocean cruiser *as well as* a normal life on land, is much like having two houses, two jobs, etc. Having a smaller simpler day cruiser is a completely different matter however, and a great way to enjoy a weekend :)
This is great! Thank you for such a detailed answer.
Any time :) Do get in touch if you want to talk boats or sailing in full generality!
Under UNCLOS you have the "right of innocent passage"; you can transit through nations' national waters without express diplomatic clearance, provided you're not doing anything harmful (like spying on them, smuggling drugs, etc.). Of course, there are certainly places in the world where I wouldn't be too keen to exercise that right - but in most of the world you'd be perfectly fine.
Internet at sea is as good as terrestrial ADSL (though presumably hundreds of times more expensive..) when you're on a vessel with a microwave dish that tracks the satellite, but if the ship is moving beyond the dish's ability to track the satellite (either through multiple successive alterations of course or too much pitch and roll in a heavy sea) then you can lose signal and need to re-train onto the satellite. Whether such dishes and tracking arrangements are available on smaller pleasure craft, though, I've no idea. [Edit: you can usually get 3G or 4G phone signal out to a distance of about 10-15 miles from an inhabited coastline, which would cover all your berthing and anchoring and a fair amount of your passages]
Everything about the maritime world is expensive beyond belief - almost beyond sanity. Not just the upkeep and husbandry of a vessel but also compliance with international maritime law (things like pollution regulations), with the laws of your flag state (the nation whose flag you sail under), and the local regulations of ports you put into. Then there's insurance, constant maintenance and upkeep costs (you'd be surprised how many ordinary-looking, perfectly inert objects have a legally-obligatory expiration date on them when they're required to be carried by maritime law), internet, paying ports for fuel or to dispose of your garbage, berthing fees....
As for actually navigating (assuming you're not going to hire any Deck Officers) you'll need to do some level of training to plan passages and navigate safely in accordance with the COLREGS. In addition to mostly-navigation-focused qualifications (such as the RYA Yachtmaster) it would be wise to have training in basic fire-fighting, first-aid, GMDSS radio operation, etc. - when you're on passage deep-sea you can't just telephone the fire brigade, doctor, dentist, etc. etc. (You don't *need* qualifications in these things if you're a smallish pleasure craft - but personally I'd be very nervous about putting to sea without them!)
Lastly ..and I'm sorry if this sounds too discouraging, but.. a life at sea is, well, just generally not very nice, lots of the time! There's so much that's difficult or uncomfortable, so many hurdles and hoops to jump through, and so much work to do to maintain even a small vessel. I imagine that maybe living aboard a largeish yacht berthed in Monaco or somewhere and reclining on the upper decks whilst Deckhands clean and maintain it for you, and just throwing money at every port official who shows up on the jetty until they were placated - possibly breaking up the monotony with the occasional passage to St Tropez and back - might actually be quite a nice way to live for a few years - but most of the people I work with (plus self), though we might find our careers fulfilling in some general sense, don't really enjoy actually being at sea...
> Everything about the maritime world is expensive beyond belief - almost beyond sanity.
Yes, I've heard that sailing is like standing in a cold shower tearing up $100 bills! And they say the happiest day of your life is when you buy your first yacht, and the second happiest is when you sell it. But from my limited experience (mackerel fishing in Swanage Bay), yachting is a lot of fun.
Mind you, for ruinous expense and constant worry, yachting is only a small step up from keeping horses.
>And they say the happiest day of your life is when you buy your first yacht, and the second happiest is when you sell it.
That's funny, I heard it the other way around: the 2nd happiest day of a boat owner is when he buys it. The happiest is the day he sells it.
Yes, I must have mis-remembered the gag. Your version certainly sounds funnier!
If you're asking broad questions about how much it might cost to live on a yacht, you may not have heard the famous and highly instructive joke on the topic. How can you tell if you are ready to own a boat? Go to the bank and take out five grand in hundred dollar bills and start flushing them down the toilet by the handful. If you can get through all 5000 without breaking down in tears, you're ready.
The annoying thing about boats is the fact that they cost you large amounts of money whether you're using them or not. But if you are actually using them, every day, instead of a house, then the costs start to look less obscene.
Generally you handle borders/visas/customs/immigration at port, when you land - you have to report arrival, and wait until the border checks are complete until you can disembark. You still need to follow the visa process, you can be denied entry, etc.
A question for the historians of religion out there: what is the best secular explanation for why Paul converted to Christianity?
It's widely accepted that (1) Jesus Christ existed and founded a community of believers, (2) Paul harshly persecuted the early Christian church, and (3) Paul later converted to Christianity and became its leading proponent, and endured great hardship in the course of his work.
The biblical explanation for this unlikely sequence of events is that Paul had a vision of Jesus while traveling to Damascus.
Are there any good secular explanations? Obviously you can hand-wave this all away by saying that people sometimes have visions that cause them to radically alter their behavior, and this does not inherently demonstrate that anything supernatural is going on.
Nevertheless, the story of Paul does strike me as pretty startling and difficult to explain without recourse to God. Is there any other example in history where an individual aggressively persecuted a religion, then converted to that religion at great personal cost?
It could just be that he had Christians pegged a certain way in his head as not really people, and when he had an experience that might have actually made him sympathetic to them briefly it completely shook his worldview and caused him to re-examine things - and that re-examination happened to land him in switching sides.
That happens sometimes, like when you see someone completely switch partisan affiliation.
In one of his letters Paul mentioned "bearing the marks of the crucifixion". That could have been figurative of course, but it has been suggested he was none other than the leader of the group of Sanhedrin priests and soldiers sent to arrest Jesus on the night of the last supper, the one who the disciple Simon Peter injured on the ear with a sword before Jesus told him to stop. Who knows, perhaps being clobbered on the head with the flat of a sword by a big guy like Peter (meaning "Rocky") concussed him and is what caused his epilepsy!
Something else that may be relevant is that although Saul (his pre-conversion name) was Jewish, he was also a Roman citizen, a distinction inherited from his father. So maybe he had long been conflicted against what he may have seen as the parochial preoccupations of the Temple priests and Jewish people generally, especially their opposition to Roman rule, and saw in Jesus's teachings a broader perspective. After all, it was St Paul who primarily spread the Christian gospel beyond Israel. The disciples and their immediate followers never would have, nor even wanted to. In fact, some of them fell out with Paul for that reason.
Paul belongs to a group (the Pharisees) which was deeply at odds with the Temple aristocracy (the Sadducees), and whose spirituality was in fact relatively close to Jesus' in many ways.
You are rather kind to him when you say he "endured great hardship in the course of his work." My reading as a non Christian who has had a thoroughly Christian education is that he was good at persecuting people. He didn't stop after the road to Damascus thing, he just got redirected. You wouldn't know from the gospels that God hates gays, but it is very clear from Romans 1 that he does. It's a shame that he was such a fabulous writer. in fact he very precisely condemns himself: he speaks with the voice of men and angels but has no charity.
I don't know that it takes any kindness to acknowledge that Paul faced hardship. Just off the big events, he was shipwrecked, imprisoned three times, and executed. I'm also not sure it's fair to describe his post-conversion career as one of persecution. Writing sternly worded letters seems kind of banal in comparison with stoning people to death.
He would have had to have been in a position of power to persecute people. You are pushing future persecutions onto the past.
He was either in a position of power, or believed he was, judging by his writings. As an early Christian gay I would have been in no doubt I was being persecuted after reading Romans.
That’s absurd. People were free to choose
to join Christianity or not. Presumably a Roman gay wouldn’t do that.
You are pretty much looking at a totally different era, with totally different power structures and retrofitting 21st morality o to it.
I don't understand your point. There are currently gay people who would like to be Christian who in many churches run into the problem that they are not tolerated, because of things like Paul's letter to the Romans, 1:
26 Because of this, God gave them over to shameful lusts. Even their women exchanged natural sexual relations for unnatural ones. 27 In the same way the men also abandoned natural relations with women and were inflamed with lust for one another. Men committed shameful acts with other men, and received in themselves the due penalty for their error.
Gay people do not say Hell, I would like to be Christian but this rules it out, maybe try Hinduism instead? They really don't. They want to be gay, and Christian, and the Anglican church in England spends most of its time trying to think of ways to accommodate them. You seem to be saying that things were fundamentally different in the first century and they would all turn to mithraism or something. I don't see any reason to think that.
"There are currently gay people who would like to be Christian who in many churches run into the problem that they are not tolerated"
And there are plenty of churches that have happily torn up the Bible to accommodate them:
https://www.episcopalchurch.org/who-we-are/lgbtq/
And there are churches that have torn up the Bible to accommodate all the LGBT+ people who would be Christian except for...., so they made the changes in anticipation of the floods of new converts, and then afterwards found the pews every bit as empty as before, because the activists were not interested in becoming Christians, they just wanted the new orthodoxy accepted everywhere.
This is why I don't accept "I would love to be Christian, if only you would change your beliefs to suit me". Now you have your choice of churches that will accommodate you and tell you it's no sin. They'll even declare you a saint:
"Murray is celebrated on July 1 in The Episcopal Church’s “Holy Women, Holy Men” calendar of saints, and an increasing number Episcopal leaders – especially Black and LGBTQ+ people – cite her as an influence. Trinity Church Wall Street and the Diocese of North Carolina are supporting partners of the Pauli Murray Center for History and Social Justice at Murray’s childhood home in Durham, North Carolina."
Yeh. But the thing for a gay person in Roman to do was not be a Christian. Homosexuality in Rome was generally tolerated.
In those days homosexuality was sometimes associated with religion, in that there were pagan religions whose male priests were gay and/or financed their temple and lifestyle with sexual services. One such example was the Galli, members of the cult of the goddess Cybele. So it was primarily the pagan religion and idolatry being condemned in the Bible, and the gay sex criticised only incidently by association.
So he considered it worth being persecuted himself to persecute others? If it was all about bullying gays he couldn't done that under Judaism.
No, not what I said.
Your faith is a beautiful thing, but 3 out of your 3 posts betray a complete failure of very basic logic. Therefore you are reinforcing the belief (not my belief incidentally) that Christianity and logic are incompatible. So you need to have a good long think about whether you are doing the lord's work by continuing to post.
Warning (50% of ban) for this comment.
I think you have probably overestimated how clearly you articulated your point, and as such are being unreasonably harsh to Gamereg for not getting it. (Illusion of transparency strikes again!)
As far as I can see, you never gave ANY explicit explanation of how your criticism of Paul is meant to relate to the top-level comment that it is replying to. You seemed to imply that Paul persecuting others somehow undermines the claim that he endured hardship, although I don't see why those two claims would be at odds, and you haven't said.
If you want your reader to guess at some implied argument that you haven't spelled out, then it doesn't make sense to complain when their guess is "not what you said". The _correct_ guess would _also_ be "not what you said".
Sure, a deserved rebuke. I just don't see an issue here; he changed his mind and either he thought he wouldn't be persecuted or he thought he would but either didn't mind or he positively welcomed the prospect. Martyrdom was all the rage at the time. I don't personally respond to an uptick in modern day suicide bombings by saying Gosh, perhaps there is something in the creed for which these guys are doing this, I must investigate it more thoroughly.
There are Jewish sources that say Paul was a plant by the Rabbis, sent in order to distinguish Christianity from Judaism. Early Christianity was too similar to Rabbinic Judaism, and the Rabbis thought that people were getting confused, or perhaps that the Christians would persecute the Jews less if the religions had some breathing space.
Source: Toldos Yeshu says this about Peter, and other Midrashic sources say similar things about Paul.
Except in Paul's day it wasn't Christians persecuting Jews, it was the other way around.
Maybe, but not Judaism in general - outside of revolts. I’ve heard that 10% of the empire were Jewish or adjacent. By and large you could be Jewish in Rome, but not Christian.
The 10% figure is from a 13th century source.
https://westhunt.wordpress.com/2013/06/14/jews-in-the-roman-empire/
I've heard some secular people say temporal lobe seizure (weird type of seizure that tends to cause religious-seeming experiences), and I think it would be fun to consider Paul as a case study of Geschwind Syndrome - but this seems a little too cute. Former Bay Area rationalist @webdevmason, who used to organize what I think was the biggest New Atheist convention, converted to Mormonism earlier this week. I don't claim to exactly understand her narrative arc but I don't feel the need to bring in temporal lobe seizures. Sometimes people change their minds for confusing reasons.
Paul's conversion was more interesting insofar as it happened in a single moment and he saw a vision, but we just have Paul's word for that, and if you're an atheist you're already suspicious of the Bible (I agree that Paul describing his own experience is more credible than someone thirdhand describing Adam and Eve, but lots of preachers have been caught in lies about their own lives before, so I don't want to 100% believe)
C.S. Lewis wrote a wonderful book-length account of being seized by the conviction that god loves us, and the Catholicism is The Truth. It's called Surprised by Joy.
Lewis was an Anglican, not a Catholic. The book is good, though I would not have described it the way you did. More like "compelled by logic to believe that God exists", though his later conversion from Theism to Christianity in particular might be described as being seized by a conviction.
The whole book can be found online here.
https://www.orcuttchristian.org/Lewis_Surprised%20by%20Joy.pdf
I've heard about enough instances of right-wing anti-Islamic politicians converting to Islam that my best explanation is that major world religions are convincing enough that anyone engaging with them deeply, even adversarially, is at risk of being converted by their sheer memetic potency. This is how they became major world religions, after all.
Not so surprising with right-wing politicians like that - arguably they already had the mindset and just switched teams.
I feel like it's a temperament thing in some ways. If you're emotionally vulnerable, feeling isolated personally, and have a very strong sense of Needing Things to Be Tied Down And Make Sense, then you're probably going to be attracted to religion eventually. What religion it specifically is depends on circumstance and timing.
I’m going to give half-assed paraphrases of two apropos quotes.
“He who wrestles demons best take care, lest he becomes a demon.” Not sure where that’s from. Bible? Saturday morning cartoon? A role-playing game?
The second is Nietzche, and i’m sure everyone here knows it. “He who gazes into the abyss too long...” something something something “...abyss gazes also into him.”
People tend to make both valid, logical arguments against religions they don’t like. But they also make invalid, illogical arguments against those religions. To stereotype, A strong believer in a competing religion x make a lot of arguments against religion y that are based in beliefs about his religion x. If he has a crisis of faith in religion x, then he might see that all the arguments they made against y are invalid. That makes it easy to decide religion is actually true.
Here’s an example from real life. My family moved from the Midwest to Long Island when my sister was 15 or so. She constantly (continuously?) mocked the “Long Guyland” accent. Another word for mocking an accent? Practicing. She developed the accent over the three years we lived there. Poetic justice!
The first is also Nietzsche. And he's not just repeating himself in different books. He puts the two together in BGE 146 (Ch IV). The first is a warm-up to the weirder second line about the abyss. I guess the first could be a reference to an earlier source, but I find nothing.
Wrestle demons evokes Ephesians 6 or maybe Jacob wrestling the angel in Genesis 32, but Nietzsche say fights monsters (Ungeheuer).
Seems false to me, the whole New Atheism thing was millions of impressionable young teenagers engaging with religion adversarially, and only a tiny handful of them ended up converting.
I wonder if there's a similar record with political ideologies - I had something like this with libertarianism, although only in a weak sense.
Well, there was the time the government post-WWI Germany decided to send an agent to infiltrate and report on what they thought might be a dangerous group of political extremists. That did not go the way they expected - https://en.wikipedia.org/wiki/Adolf_Hitler#Entry_into_politics
I think if an ideology becomes a movement worth infiltrating or persecuting, then almost by definition it has to have become a community with something substantial to offer its members. And if you decide to send an agent deep undercover or far abroad to deal with the perceived threat, you're probably selecting for people who don't have deep ties to your own community because the ones who do have such ties will be reluctant to take the job. So some risk of your agent "going native" comes with the territory.
If it's just hanging around with your friends on the internet sniping at the opposition, yeah, those people aren't going anywhere because of the "hanging around with your (existing) friends" part.
This seems like the best explanation.
I agree that mere conversion wouldn't necessarily require an explanation, but it seems to me the salient fact is that Paul was persecuted and eventually killed for his Christian faith — and that means the story as a whole *does* need some explaining.
But I'm willing to be persuaded otherwise, especially if there are good historical analogues where individuals underwent voluntary conversion and were subsequently persecuted for it.
I don't really understand your point. It is a perfectly rational secular thing to say that he converted to Christianity because he was convinced by the arguments in favour of it, or to say that he had an epileptic fit which converted him as it apparently often does. Why does his eventual martyrdom require explanation? He couldn't predict the future any more than anyone else can. Sure, explicitly converting at that date raised the odds of coming to a premature end, but 1. Plenty of people drive fast on drugs while smoking cigarettes and think, quite often correctly, that they will get away with it and 2. In those days it was rational to think as a Christian that you really were literally immortal anyway because the second coming was just round the corner. So what is there here that is harder for the secular to explain, than for the godly?
Not sure how Paul would've thought he'd "get away with it", since he had literally been witness to at least one stoning, and testified against others who where stoned. He of all people knew that changing course would not be pleasant for him in this life. And Christians may have hoped the Second Coming would come in their lifetimes, but those who paid attention to what Jesus said about it ("no man knoweth") would've known better than to count on it.
You seem to think that Some cases of x lead to y implies All cases of x lead to y. I disagree, and I am guessing Paul would have.
>especially if there are good historical analogues where individuals underwent voluntary conversion and were subsequently persecuted for it
I mean, there are boatloads of these ("fisher of men" pun very much intended), considering that most of the early Christian martyrs were converts who were eventually killed for their new faith. I think the interesting thing about Paul the OP was curious about is the fact that he wasn't just a convert who underwent persecution and maintained his faith, but a former persecutor of Christians himself. As for that specific case, I can't really think of any off the top of my head.
You're right, I was more precise in my original post.
"Is there any other example in history where an individual aggressively persecuted a religion, then converted to that religion at great personal cost?"
If you include sects within a religion, I suppose one example would be King Henry VIII no less.
In his youth, he defended the Catholic faith, and even wrote a book called "Defense of the Seven Sacraments", basically opposing the newly emerging Protestantism. For this, the Pope awarded him the title "Defender of the Faith", which appears on British coins to this day as "FD".
But later, after the Pope refused to grant his wish to divorce Catherine of Aragon, he separated the UK from the Catholic Church. Thereafter the Anglican church, and thus to a large extent the King himself, became in effect Protestant.
As for the "great personal cost", although the king was enriched by the sale of monastic assets, the break from Rome caused him a lot of worrying upheaval and resistance to the changes.
I just realized you're the OP yourself, sorry, I didn't check the name that closely and assumed you were someone else making a separate point.
Isn't Paul himself our source for information about Paul?
The account in Acts of Paul's life diverges sufficiently from Paul's own account that it seems likely the writer of Acts had some non-Pauline source. But this may not be particularly relevant, since precisely because of the contradictions with Paul's own account, the non-Pauline source is probably unreliable.
One thing that is potentially relevant is that Galatians 1 presupposes that the audience will be familiar with at least some details of Paul's biography ("for you have heard"), and it wouldn't make sense for Paul to fake that. He wouldn't the first or last person to overestimate his own fame, but it seems clear that he at least thought of himself as having been a notorious persecutor of Christians.
Paul does slightly yada-yada over a 3 year gap between his conversion and start of his missionary work (which Acts 9 omits entirely).
I believe the three facts I asserted are widely accepted by historians, both secular and religious. I don't know exactly which sources those historians rely on.
I can tell you that the primary biblical source for information on Paul's life is Acts, which was written by Luke, not Paul.
I don't think it's widely accepted by historians that Luke actually wrote Acts (or the Gospel of Luke).
True but irrelevant in this context.
I think the vision is a perfectly good explanation. As you point out, you don’t need to appeal to the supernatural. And visions would have been much more common in a culture attuned to those kind of experiences.
The Varieties of Religious Experience has a bunch of great examples of this kind of vision, as well as some fantastic commentary on how to fit them into a scientific world view.
I think the search for an explanation is to ask why he would have a vision with that particular content in the first place. A secular or perhaps just psychoanalytic answer would therefore be something like his desire to show how zealous he was could not be constrained within being a member of a dominant religion. Rather only within a persecuted minority would his sense of zeal find an appropriate expression ending predictably with his death as a martyr.
Or it could have been repressed guilt for his persecutions finding expression, causing him to do a complete 180 out of some unconscious desire to right his past wrongs. Obviously this is pure speculation, but if his pursuit of early Christians led to people being put to death, you can see how that might nag at even the most hardened zealot's conscience.
I see a lot of people persecuting the group they identify most strongly with. It's possible he persecuted the Christians precisely because he found them so convincing, and thus saw a threat in them.
Yeah, maybe this is actually a good explanation.
I still wonder at the lack of historical analogues. (Unless analogues exist?)
I imagine that part of it is that you have to be looking to join a group that's willing to let you in despite very recently having actively been engaged in killing them.
Maybe this guy? https://en.wikipedia.org/wiki/Abo_of_Tiflis
Converted from Islam to Christianity. Didn't persecute Christians but seems to have engaged in a lot of debates with them.
"why would early Christians have continued to be Christians if they were persecuted and it wasn't true?" is mostly an answer to the "Christians faked the resurrection to manipulate dumb people" crowd.
> Seriously tho: many homophobes later come out as gay, often when they still live in an environment where being a homophobe is acceptable but being gay is not
I wonder if the opposite is true. In environments where being gay is acceptable but being homophobic isn't, are many gays secretly just homophobes trying to hide it?
This is another update to my long-running attempt at predicting the outcome of the Russo-Ukrainian war. Lastly, I did an emergency update on March 7 without proper explanation (over here: https://astralcodexten.substack.com/p/open-thread-266/comment/13398677). Previous proper update, from January 30, is here: https://astralcodexten.substack.com/p/open-thread-261/comment/12342095.
13 % on Ukrainian victory (unchanged from March 7, down from 17 % on January 30).
I define Ukrainian victory as either a) Ukrainian government gaining control of the territory it had not controlled before February 24 without losing any similarly important territory and without conceding that it will stop its attempts to join EU or NATO, b) Ukrainian government getting official ok from Russia to join EU or NATO without conceding any territory and without losing de facto control of any territory it had controlled before February 24 of 2022, or c) return to exact prewar status quo ante.
43 % on compromise solution that both sides might plausibly claim as a victory (down from 45 % on March 7, and down from 47 % on January 30).
44 % on Ukrainian defeat (up from 42 % on March 7, and up from 36 % on January 30).
I define Ukrainian defeat as Russia getting what it wants from Ukraine without giving any substantial concessions. Russia wants either a) Ukraine to stop claiming at least some of the territories that were before war claimed by Ukraine but de facto controlled by Russia or its proxies, or b) Russia or its proxies (old or new) to get more Ukrainian territory, de facto recognized by Ukraine in something resembling Minsk ceasefire(s)* or c) some form of guarantee that Ukraine will became neutral, which includes but is not limited to Ukraine not joining NATO. E.g. if Ukraine agrees to stay out of NATO without any other concessions to Russia, but gets mutual defense treaty with Poland and Turkey, that does NOT count as Ukrainian defeat.
Discussion:
This update, just like previous emergency one, is based on various credible reports which started to float in information space on March 7, in places like NYT and German ARD, that blowing up Nord Stream pipelines had pro-Ukrainian motive. So far, no links to Ukrainian government have been found, but realistically, this massively increases the probability that such links will be found in the future. Probably not in a sense that Zelensky ordered the pipelines to be blown up, more like Ukrainian government supporting some shady operators to generally blow Russian stuff up and those agents being very dumb.
It is difficult to predict what impact such revelation would have on a Western support to Ukraine, but surely it would not be zero.
*Minsk ceasefire or ceasefires (first agreement did not work, it was amended by second and since then it worked somewhat better) constituted, among other things, de facto recognition by Ukraine that Russia and its proxies will control some territory claimed by Ukraine for some time. In exchange Russia stopped trying to conquer more Ukrainian territory. Until February 24 of 2022, that is.
I don't think the sabotage story change much tbh. At least from the French, Polish or American perspective. Maybe the German would be a annoyed but that's it.
It seems that you keep wildly underestimating Ukrainian chances.
Did you update back on all the previous assumptions you made (Russia is going to try to keep Kherson, mobilization is going to give Russia an edge, etc), which turned out to be wrong?
After a failure of both Russian infrastructure strikes strategy and recent Russian offensive attempts, I think there are even more reasons to update in favour of Ukraine.
I added 5 percentage points to chances of Ukrainian victory and deducted the same from Ukrainian defeat after Kherson evacuation.
Regarding Russian mobilization, it seems to me that, with admittedly big exception of that evacuation, it is working basically as intended, at least so far. Fears of political turmoil in Russia turned out to be greatly exaggerated, and mobilization succeeded in stabilizing the front east of the Dnipro. Russians are now even able to go on the offensive again, although as you note they are not very successful at it. Maybe Russian front will soon crumble despite of mobilization, but I wouldn't bet on it.
I think Westerners think about this war something like if Ukraine is not actively winning, it is therefore losing. Not saying this is necessarily 100% wrong but an absence of big news seems to work against Ukraine in public perception, which I'm sure they are aware of.
Like a lot of neutral to pro-Russian observers, I think you mistake the reason for public support (at least in the United States) for Ukraine. It has nothing at all to do with the ethics of the Ukrainian government, about which we have had a fairly low opinion already. Support is based on (1) the incredibly barbaric behavior of Russians on Ukrainian soil, and (2) the courage and resolve demonstrated by the Ukrainian Army.
That's it. Nothing else really matters (except for the cost of support, of course). So even if Zelensky personally scuba dived down to the Nordstream and drilled a hole through it by hand, my read is that almost every supporter in the US would shrug and say "Yeah? And...?"
I don't have any good read on how it would affect public opinion in Europe, of course. My vague feelings is that the Brits, Baltics, and Poles couldn't care less, because they're in it because they hate the Russians. The Germans are another story, especially because it affects them directly, but the Germans have been pretty lukewarm in their support already. How much that matters militarily I don't know, but it certainly matters less than US public support, which as I've said I don't think even the most outrageous possible outcome of the pipeline explosion investigation could change.
?So even if Zelensky personally scuba dived down to the Nordstream and drilled a hole through it by hand, my read is that almost every supporter in the US would shrug and say "Yeah? And...?"
Nit: I think the charisma and political savvy of Volodymyr Zelenskyy counts as a third substantial reason for US public and political support for Ukraine, so if he were personally involved or even known to have signed off on that operation, it probably would have some impact.
A hypothetical independent group of Ukrainian patriots abroad, probably not much impact.
the“ Ukrainian patriots“ story is probably the most fanciful piece of propaganda I’ve heard yet.
Maybe. I think reasonable men can differ on the absolute level of change. My belief is only that even if it caused some dismay at the more moralizing margins, it would have an effect on support too small to impel nontrivial change in policy. Previous "revelations" and arguments about Ukrainian corruption and mis-governance have had zero effect, after all.
As I said, I don't think people support Ukraine because they like the Ukrainian *government*. Not even Zelensky personally. While I agree his leadership is critical within Ukraine, and he has done a masterful job in coordinating Ukraine's PR offensive -- truly brilliant -- I don't think he's Winston Churchill reborn leading a War Cabinet in which people have deep faith. I notice he has been pretty careful to present his people as the true heroes, and be fairly self-effacing for a wartime leader. I think he understands that it's his people that are inspiring, not his government.
There's also the additional fact that I can't see the American supporters of the war getting *that* worked up about the damage to the pipeline. The direct effect would appear to be to prevent Russia from selling gas to Germany, and (1) I think American supporters are very OK with Russia not being able to sell gas, and (2) they've never been very OK with Germany continuing to buy gas from Russia -- it seems a bit like trading with the enemy.
People do understand intellectually there's reasons for it, but it still kind of stinks, and I can't see people getting too worked up if -- as it might be put -- the Ukrainians had the balls to do what the Germans themselves lacked the courage to do: just stop sending German euros to pay for Russian shells blowing up Ukrainian schoolchildren. That's how it could be put, and I think that argument would do a lot to blunt any feeling that the Ukrainians had gone too far.
The Americans that oppose Ukraine are anti-Biden…China and America have benefited most from this senseless invasion as America is now energy dominant which was a Trump priority. China benefits because Putin is now married to China as he is a pariah in the West.
Nonsense.
I oppose the hegemony of the military-industrial complex and the cozy relationship between them and seemingly ALL flavors of politicians who always seem to be able to find a “Just Cause” to send arms or men to fight, profiting handsomely indeed from our “not being in a war.”
I’m unconvinced that there’s a “Good Guy” in that conflict. I’m not convinced we have a national interest there, certainly not one where we write blank checks to Boeing and Lockheed-Martin.
We have a national interest in Russia not having an army capable of invading e.g. Estonia, because we're on the hook to defeat that army if it happens. That being the case, we have a national interest in the Russian army suddenly having all its tanks blow up, etc, and oh look, here's a way we can arrange for that happen at a bargain price and without any American soldiers getting killed.
I don’t support sending Americans to fight, right now it’s essentially free to send military equipment as we have benefited with LNG exports and military exports create good jobs without contributing to inflation.
No, that's not the case universally. For example, I am not a supporter of the Biden Administration -- I will certainly oppose his re-election -- but I support American assistance to Ukraine. There are plenty of people who are anti-Biden but pro-assistance, and others who are pro-Biden but anti-assistance.
I wouldn't say American and China have benefited from the war so much as Russia has suffered by comparison. Indeed, I can think of no outcome whatsoever that does not leave Russia much worse off than she was in 2014. Had NATO actually intended to destroy Russia as a First World participant, they could have done no better than secretly engineer this war. If Putin were actually in the pay of the CIA, he would be eligible for a very handsome bonus, as having exceeded all reasonable expectations.
"The Americans that oppose Ukraine are anti-Biden" is not the same as "The Americans who support Ukraine are pro-Biden"; your existence only refutes the latter.
Right now there are politically significant blocs of pro-Ukraine/pro-Biden (e.g. pretty much all Democrats), pro-Ukraine/anti-Biden (e.g. McConnell) and anti-Ukraine/anti-Biden (e.g. MTG, and it wouldn't surprise me if some of the very far left fitted here too) Americans.
But I agree with Pangolin Chow Mein that there are very few Americans who like Biden in general but don't want the US to support Ukraine.
Maybe not, since it would be kind of logically incoherent to like Biden and dislike his major foreign policy effort. If you thought US assistance to Ukraine was stupid, it's kind of hard to like Biden, its chief architect.
But there are without doubt big chunks of the anti-Biden crowed who are strongly pro-assistance, and other big chunks who are just as strongly anti-assistance. And I daresay there are similar divisions in the meh-Biden crowd.
Which probably covers most Americans, since I think there aren't many people who really *like* Joe Biden. Never have been.
> I can think of no outcome whatsoever that does not leave Russia much worse off than she was in 2014
If Russia hangs onto southern and south east Ukraine, then they will have firstly their long desired land bridge to Crimea, but also ownership of the prime Ukrainian wheat growing areas and iron ore deposits.
I suspect China and India are secretly rooting for Russia, solely to have cheap and reliable access to the latter. The ability to provide them may even have been a or the main incentive for Putin to invade in the first place.
Sure, and with a land bridge to Crimea and $3, you can get a Starbucks in the Metro -- oh wait, no you can't any more. Just like you can't get first-class chips or electronics, just like you can no longer get Exxon to lend you expertise to exploit your oil reserves, just like nobody will sell you parts to keep your airliners flying any more. At least the mighty Russian Army will have proven once again it is a behemoth to be feared, a steamroller that would flatten any challenger immediately! Oh no, hang on...no, that's not true either.
Getting a land bridge to Crimea and the shattered shards of the Azov works would be a victory of epically Pyrrhic proportions.
I doubt either China or India is rooting for Russia to win. But China in particular is quite happy to see Russia immolate herself. There's never been any love lost between the two, and I believe China would be quite happy to take her place as the only remaining serious challenger to American world power. But equally China would like a Russian defeat (or resumed stalemat) to be as costly as possible to the Americans. So I expect them on the whole to be in favor of war, as long as possible, and as costly as possible for both sides. (I expect privately they would say, as Kissinger said about the Iran-Iraq war, "it's a shame they can't both lose.")
And nobody will ever buy the Irkut MC-21…a goal of Putin’s to diversify the Russian economy…and the COMAC C929 is just a ruse to string Russia along. George W Bush thought he would get the best of China…it cost us millions of manufacturing jobs and a lost decade when boomers were at their peak productivity.
What are your odds that the pipeline explosion was done/ordered/precipitated by the actions of Russia/Ukraine/US/... respectively? My understanding is that it was a fairly sophisticated operation that required resources very few players could muster. The US and Russia, some European countries, but very unlikely for Ukraine or some "shady operators".
Why has the theory "badly shut down, badly started up, clathrates go 'boom'" been rejected?
Because clathrates don't detonate, and detonations are easily distinguished from lesser sorts of explosion.
I am 100% US, or allies.
I am no expert on blowing up undersea pipelines, but investigators evidently think that it wasn't that sophisticated operation. And I believe them. Like, why would they made that up? Far more likely is that you are overestimating its difficulty.
Before recent revelations, I would guess over 50 % probability of Russian involvement, now I would guess roughly 80 % on pro-Ukrainian group (without making any judgment how much is it connected to Kyiv government), 35 % on Russian involvement and 10 % on something weird and uncategorizable. And yes, those percentages don't add up, because I think that it is not outside the realm of possibility that it was the work of Russian agents succesfully infiltrating pro-Ukrainian groups.
Of course 0 % on Biden administration involvement. C'mon, this is right next to 9/11 trutherism in plausibility. If Trump would still be in the White House, I could very easily imagine that he would order something like that, but I can't imagine that it would not be immediately leaked to the public.
Why would the „investigators“ from the US and the West lie about something they probably did? Let me go away for a while and think about that.
If they would be engaged in a cover up, they would obviously blame Putin. And people would believe them, including me. Destroying Nord Stream and then framing Ukrainians for it just doesn't make any sense.
Why would Russia want to damage a pipeline when they can simply turn off the taps at their end, although they may have gained somewhat from higher gas prices.
Whoever it was, the aim was obviously to prevent certain EU countries wobbling on sanctions by continuing to purchase Russian gas, on which they (especially Germany) have foolishly come to rely on.
William Spaniel did a video on this; I summarise his answers at https://politics.stackexchange.com/a/75977/18471 . And there you can also see various other answers, along with more answers at the duplicate question https://politics.stackexchange.com/q/77392/18471 .
The one I find the most interesting (not necessarily the most plausible) is that it might be done by Russian forces even though it does not benefit Russia per se, because *Putin* benefits by reducing the attractiveness of a coup. (Getting rid of Putin no longer lets them restart the flow of gas and bring everything back to normal.)
These sounds like galaxy brain takes of someone who has primed on the idea that Russia did it and can't let go.
The idea that there was likely to be a coup in a cult of personality authoritarian regime such as Russia is wrong in the first place. The idea that the destruction of the pipe would decrease the probability of such coup doesn't hold much ground either.
If there was a group inside of Russian elites, able to kill Putin, take over the country and eager to end the war and continue business as usual, they could've done it with or without the pipe. The existence or absence of the NS isn't the main reason why business has ceased.
> These sounds like galaxy brain takes of someone who has primed on the idea that Russia did it and can't let go.
One or two replies and comments may have been somewhat ax-grindy. But none stood out as such to me, and taken as a whole the threads seemed to give a pretty rounded set of pros and cons in assuming Russia was behind it, and these are more surprising and diverse than I assumed.
Interesting. So what would be the takes for the most likely reasons for Russia to do it from someone who believes that there's a very low probability that Russia did it?
And if you assume that Spaniel also agrees that the existence or absence of the NS isn't the main reason why business has ceased, do you still believe that the existence or destruction of the NS doesn't change the cost/benefit analysis for anybody trying to overthrow Putin?
How do you get to 80% on a pro-Ukrainian group? As near as I can tell, the evidence for that is that a formerly-respected journalistic outlet said "trust us, we have Super Secret sources that say the Ukrainians did it", coupled with a bit of detail that on examination has some holes like e.g. the specified vessel not having been in the placeit was said to have been.
https://oalexanderdk.substack.com/p/the-nord-stream-andromeda-story-what
Which, OK, the details could have been garbled without the underlying story being false. But we've already *got* a formerly-respected journalistic outlet saying "Trust me, I have Super Secret sources that say the US and Norwegian navies did it", again with a bit of detail that has holes like out-of-place ships. And I don't think it can plausibly have been *both* Ukrainian partisans and the US/Norwegian navies. So how do you get to 80% on one and the other folded somewhere into a 10% "weird stuff" category?
I don't think either version is all that credible, but if I have to pick, then as a veteran diver any theory that a sophisticated technical diving operation at 80 meters was carried out undetected from a fifteen-meter yacht comes in way behind the theory where a 375-ton minesweeper was used.
Re: why I think it is wildly implausible that Biden administration or Norwegians (full disclosure: I didn’t even know they are implicated until you mentioned them) did it? I do not know much about how difficult it is it destroy underwater pipeline, but I feel that I know enough about both American and Norwegian political system to rule it out. Usual arguments against 9/11 truhterism apply, with slight modifications. I should note I would probably be less confident had Trump still been in the WH, or if, like, Estonians had been suspected instead of Norwegians.
This of course does not mean that (pro)Ukrainians did it, and in fact, until last week, my best guess would be that it is the work of (pro)Russians. But I changed my opinion in response to new revelations, which do not come only from NYT but also from the German media (see here for a writeup in English: https://www.politico.eu/article/nord-stream-pipeline-blasts-lead-to-ukraine-media-reports-germany/). Of course the fact both reports came on the same day is not a coincidence, but they do contain somewhat different information and do not cite each other. Those reports are extremely politically inconvenient for governments of countries doing the investigation, which makes me suspect they are correct.
So, we have to answer two questions:
1) Is it plausible that Ukrainian proxies are crazy enough to do it? Well, yes. I am thinking about something analogous to that murder in Sarajevo which we all know about, when Serbian government didn’t order it but was done by some people loosely in the orbit of what passed for secret service in those days.
2) Are Ukrainian proxies capable of doing it? That I am not sure, but they have blown up the Kerch bridge and various other stuff in Russia. I am sure underwater pipeline is harder, though, but still.
>Are Ukrainian proxies capable of doing it?
Almost certainly not, at least for the sort of "Ukrainian proxies" who would be loose-cannon enough to try this and not depending on the support of the Ukrainian government. This really requires a Baltic Sea naval power to have pulled it off; the few other outfits that might have the capability (e.g. North Sea oil drilling companies) are not plausibly going to get into that line of business.
Agreed.
We are not talking about limpet mines near the surface of a sheltered port, like what the DGSE did to the Rainbow Warrior in 1985.
The tech diving 80 meters deep in the rough Baltic from a sailboat is one thing (very tricky). The 500kg of explosives supposedly used for the operation is another (hmm, no).
Easy. The US and its allies are “pro-Ukrainian groups”
Nuland declared NS2 dead in March 2022…why would we blow up something that was dead? And why would we blow it up when a hurricane was in the Gulf?? The record natural gas price was after Katrina and would lead to even more inflation!
If a pro-Ukrainian group really blew it up they are dumber than Putin…had Hurricane Ian hit Louisiana blowing up the pipeline could have led to Russia winning the war as the price of natural gas would have spiked again inflicting even more pain on average Europeans. Now if it ends up being a false flag then the NYTimes was extremely irresponsible to publish the story before getting all of the information.
Not that dumb. The US isn't going to turn on Ukraine even if there's suspicion, because they will still be able to maintain plausible deniability.
But blowing up the pipeline helped Russia! Luckily the hurricane didn’t hit Louisiana and so blowing up the pipeline only helped Russia a little…from a money perspective it’s probably a wash for Russia but they made tens of billions in windfall profits in the first half of 2022.
Blowing up the pipeline removed the option for Russia to use it as a geopolitical blackmail tool. European politician could've been swayed by the offer of cheap Russian gas and thus decreased their support for Ukraine. That was Kremlin's plan all along. But if the offer is closed, there is no temptation. I suspect this was the main reason and it worked well. I'm quite surprised that people think that this benefited Russia.
Also, Nordstream was a longterm Russian investment which is now gone and the only way for Russia to sell gas to Europe is through the old pipeline, paying to Ukraine for gas transit.
NS2 is operable.
Is it? It hasn't entered service yet, as far as I know, and there likely are all kind of technical and bureaucratical reasons for that.
The function of the pipeline was to transport Russia's gas. Blowing it up meant Russia couldn't send gas west even if western European countries wanted to buy it.
This is not totally incorrect, but there are ways how EU can import Russian energy, directly or indirectly, even without Nord Stream. Overall, if Ukrainians did it, it was very dumb. Which does not mean they didn't do that! People are often stupid.
That’s not how commodity pricing works—Gazprom made insane profits in 2022 supplying less gas to Europe. Enron and Saudi Arabia have attempted the same gambit in which you risk losing market share while raking in windfall profits and so half of NS2 is operable as we speak just waiting to deliver expensive gas to a desperate Europe.
I've been experimenting with forms of data storytelling and wrote this analysis of SVB and bank failures! Would love to hear what I could do better, especially given how well read the SSC audience is!
https://yarn.pranshum.com/banks
" The Savings and Loans crisis of the 80s happened as the Fed aggressively raised interest rates. "
I thought the primary cause of the S&L crisis was North Sea oil coming online, causing world oil prices to plummet, stranding many major US oil projects which had been financed, often recklessly and sometimes illegally, by S&Ls. Didn't think interest rates had much to do with that one.
You're mixing up two questions: what caused the crisis and what caused the losses. Most of the risky bets were after the S&Ls were already insolvent. The bankers didn't have any equity to lose, so they made big bets with other people's money.
Love the graphs of a) size if failed banks, and b) frequency of failures over time
thank you! :)
Can anyone recommend a good English translation of Gil Blas? I know Smollett did one in the eighteenth century, and it's easy enough to find, and I generally like Smollett, but I was hoping for something slightly more recent if possible—but if people say Smollett's the one to go to…well, that's why I'm looking for recs! Thanks!
Is just after Easter an optimal time for meetups? I'd have thought this could not be great for some people who might not be at their usual bases...
Late spring or early summer would work better for me, as I like to host meetups outside at my house. April is mostly mud season here.
I stumbled across a pretty great worldbuilding project on ArtStation recently called Rust and Humus:
https://www.artstation.com/artwork/rAw5vm
Sadly not available to buy as a physical book yet, although the artist is apparently working on a sequel (https://old.reddit.com/r/worldbuilding/comments/11gxfbo/rust_and_humus_ii_one_more_year/), and has talked about possibly selling physical copies once that's complete.
Wow, that was amazing!
An opinion piece arguing that the moral compasses of children are stronger than we might think.
https://kyleimes.substack.com/p/how-i-narrowly-avoided-becoming-a
I’m gonna be in SF later this week with a bunch of wild thinkers and engineers, not necessarily rat-type people though. We are already meeting a lot of cool people, anyone here have any suggestions for what we should do? None of us are from California.
Also, a lot of us like to build amateur rockets, and we want to tour the SpaceX factory. Does anyone have any connections?
Friday night free party at Oakland Museum is a reliably fantastic time. Eat dinner at the food trucks and enjoy the DJ set in an extraordinary piece of civic architecture til 8PM and then you go on to your next party. Lots of rad people to talk to.
I live in San Francisco most days of the week (in a suburb on weekends), and for me the absolute best time in S.F. was in a car at Ocean Beach watching the sunset with a friend while we drank some wine, but that’s two people not a group.
The Presidio’s Baker Beach is nice. The Beach Chalet restaurant (which overlooks Ocean Beach) is nice, the Land’s End parking lot (which also overlooks the Pacific Ocean) is also nice.
You could go to Fort Point and someone pretending to be “Madeleine”/“Judy”/Kim Novak may fall into the bay and someone pretending to be ‘John "Scottie" Ferguson’/Jimmy Stewart may pretend to rescue her (probably a bad idea as the stairs in “Vertigo” don’t exist).
At Sam’s Grill they’re wood paneled booths with curtains that close so you may dine with your paramour in privacy.
At Golden Gate Park there’s a bison enclosure that you may view, the Conservatory of Flowers, and the Japanese Tea Garden.
Near Fisherman’s Wharf there’s a WW2 submarine and a Liberty ship.
Across the Golden Gate Bridge in the Marin Headlands there’s a mock up of the Nike nuclear missiles that were there until the 1970’s.
Across the Bay Bridge in Oakland there’s the Chabot Space and Science museum where on Saturday nights (when weather permits) you may look through the telescopes and otherwise look at old space suits.
So I Married an Axe Murderer is an underrated movie with some great SF locations.
I know it’s dated, but geez, Bullit please.
Great movie. How about Dashiell Hammett’s San Francisco? I just read an article in the NYTimes about converting old buildings into apartments and it mentions the private detective suite.
When I’m in the right mood, Nick and Nora really hit the spot.
The low regard for that movie is merited (in my opinion):
https://thepopculturists.blogspot.com/2022/10/this-weekend-in-pop-culture-september.html#comment-6001424251
The movie is a treasure just for having the Phil Hartman scene…and the opening shot is great. Did you listen to Dana Carvey on Marc Maron? He is totally different than I would expect him to be. Dana Carvey on the Conan podcast is hilarious.
No, I haven't listened to Maron or Conan's podcasts. I do recall that when Unspooled covered Austin Powers they mentioned that "Dr. Evil" is actually based on Carvey's imitation of Lorne Michaels (even though Lorne doesn't actually hold up his finger like that, and was instead just something Carvey made up). And that SNL's much later Californians sketch series is based on Carvey's imitation of his own son.
Are there any Bay Area rationalist meetups between 25th and 31st of March?
Scott Aaronson skewers Noam Chomsky's op-ed about machine learning in the NYTimes: https://scottaaronson.blog/?p=7094
The commentary has reached #178 at this point. It's a lively discussion, as one would expect.
I haven't read Chomsky's piece yet, but it's bizarre to me that Aaronson opening tone is so angry and hostile (and apparently it's Chomsky who is hostile and smug according to him?). Are LLMs his cousins or something like that ?
I'm broadly in agreement with Chomsky here, he's not saying (based on an extremly quick and incomplete skim) that LLM won't "revolutionize" society, whatever that is. He's saying LLMs don't understand language, except in a very trivial sense. Language is inseparable from coginition, and LLMs' cognition is laughable. Can someone who understand Language behave like this : https://vgel.me/posts/tools-not-needed/ ? ChatGPT is given a "calculator" that outputs wrong answers, it's required to use it to solve a word problem. ChatGPT either silently ignores the results when they are too ridiculous, or accepts them at face value even when a human can see they are ridiculous. At no point does ChatGPT talks to its interlocutor, or shows hints of understanding its own "reasoning".
People who think LLMs have an "understanding" of language seem to have a weird strawmanish view of language understanding, I'm not even talking about the "Functionalist View", which views the brain as a black box solely defined by its capabilities, this is reasonable, and ChatGPT fails this too. The standard of language understanding is human conversation, and LLMs cannot pass any free-form conversation unless there is an excuse that compensates for their weird and non-human blindspots. I don't doubt there are people as superficial as ChatGPT persona, but it's never the case they are *uniformly* so, there is something somewhere in the space of all possible conversation topics that will show a human can get angry or passionate or imaginative while ChatGPT can never, that will show a human understands the real life while ChatGPT understands only token probabilities. The imitation is fragile and precarious, it's like 3 children in a trench coat pretending to be an adult, they can fool plenty of people for sometime, but given enough time everybody remotely paying attention is going to notice, or a kid will slip and fall out of the coat.
It's annoying that people keep saying "But... but... Computers never understood language this good ??" Ok, good for you, nobody is minimzing the sheer amount of innovation LLMs represent, it's a huge performance jump for sure, LLMs are indeed the best a **COMPUTER** can do (as we know the word "Computer" today and for the past 70 years, just to avoid the nitpicks that the abstract concepts of computers and computation allows for far more than this), and this best is indeed very good and can fool most people for no less than 5-15 minutes or so, but it's below average compared to what any healthy human can do, and the thought behind the words is even less, and this is important because language is first and foremost the expression of thoughts and it's extremly obvious when something is spouting it with no underlying thought.
Chomsky apparently didn't even do basic research before writing his article. There are things he states ChatGPT can't do that five minutes of interacting with it would show that it can.
IIRC, Chomsky once said that if a computer could write grammatical English, his life's work would be undermined. So it's not surprising if he's unwilling to accept that it can. But it's disappointing that such a trivially disprovable piece can be published in a major newspaper[*] and get much more traction than the people doing actual research on ChatGPT's abilities and limitations.
[*]although of course readers of this blog already knew the NYT's standards of journalistic integrity are low
> and this best is indeed very good and can fool most people for no less than 5-15 minutes or so, but it's below average compared to what any healthy human can do
It's below the average human in some ways (like advanced reasoning), but above the average human in others (which is why it has any market value). Speed and breadth of knowledge are some obvious ones, but it's also better at writing than the average person (though not better than a professional writer). I know a few people who use it to rewrite short documents (paragraphs, cover letters, etc.). It improves them significantly.
ChatGPT is also surprisingly good at writing cute rhyming poems on any subject, something that GPT3 originally couldn't do.
It definitely can do basic reasoning. There are plenty of examples of people testing all kinds of types of reasoning. It can make inferences and generalizations.
For example, ask it something like the follow:
"Vera and Vikki are sisters. Vera is 4 years old and Vikki is 13 years old. What age will each sister be when Vikki is twice as old as Vera?"
It gave me the correct answer along with the detailed step-by-step work to get there.
I think you've been fooled by ChatGPT. It can definitely produce text that *appears* to demonstrate basic reasoning, but it also produces text that utterly fails to do this and cannot, despite having contradictions directly pointed out to it, even "see" the contradictions much less resolve them. That's how I ended up spending half an hour arguing with it about how many syllables were in a line of poetry it produced, with it eventually claiming that "1 + 1 + 1 + 1 + 2 + 1 + 1 + 1 + 2 is equal to 10." (It could correctly count the syllables in each word of the line, but couldn't sum them up in that session. It also happily gave obviously contradictory definitions of iambic pentameter in the same single response.)
I think you're using a very different definition of "reasoning" than I am. I'm guessing your version involves some kind of internally consistent mind, perhaps like a low intelligence human. What I'm claiming is something that's trivially demonstrable: that it can make inferences.
This is something even a series of if-then statements can do (or, for example, a compiler can make inferences about code). It's not something that requires a mind or machine learning. Just deriving facts from some input.
ChatGPT steps this up a notch in that it's very generalizable and, like most machine learning, it can reason about things it wasn't specifically programmed to. It picks up on patterns in its training data.
It's not simply repeating its training data back to you verbatim. It can generalize. In the example I gave above, it won't matter how you change the phrasing, the details, or the numbers (within reason). It'll still get it right.
It often says obviously contradictory things. It doesn't work like a human mind (although humans also often fail at reasoning). It trained to predict the next token, based on the text so far. Different parts of the network will kind of compete to influence the next token. Some will win out and some will be ignored. Just because somewhere it has the ability to correctly add up 1 + 1 + 1 + 1 + 2 + 1 + 1 + 1 + 2 doesn't mean that part of the network will win out for a particular prompt. The prompt might match more strongly with other pathways.
No, you've been fooled by the convincing output and bias in what you select from all its possible outputs.
Obviously, any system that produces text similar to its input that's given some correct inferences in the training text will also produce output text that has "correct inferences" in it. That doesn't mean it's making inferences; it's simply creating output text in the style of its training text. Claiming that's reasoning is akin to claiming that, because you can find the entire works of Shakespeare in the output of an infinite number of monkeys with an infinite number of typewriters, the monkeys know how to write plays.
Ironically, you explain in the last paragraph of your comment why it's not reasoning.
Your model of how LLMs (and neural nets in general) work is broken. It doesn't stitch together text it has seen in its training data or choose the statistically most likely next word like a Markov chain. The hidden layers hold higher level concepts, especially in deep networks. Sure, their goal is just to minimize their loss function (make accurate predictions), but it turns out the best way to do that is by developing a model of the world.
Do you think neural nets in general can't make inferences? There are well known mathematical proofs that it can (given the proper weights) and it's not really worth arguing about.
While I'm no longer searching for direct evidence that OpenGPT doesn't reason, I still try to use it for things where I've had difficulty with standard search engines, and it still gives back great examples of blatant illogic. Today it's somehow really stuck on the idea that the 1973 film _The Sting_ has only five acts (it has six), leading to a lot of going around in circles about what they are, and eventually the following as one of its answers:
The correct title for the third act of "The Sting" is "The Tale," and the correct title for the third act is "The Shut Out."
The five acts of "The Sting" and their corresponding intertitle cards are:
Act I: "The Set-Up"
Act II: "The Hook"
Act III: "The Tale"
Act IV: "The Wire"
Act V: "The Sting"
(At least it quickly abandoned the idea that the third act is called "The Tape" [sic]. And for some reason, starting a new session and prompting it with, "How many acts are in the 1973 film _The Sting_ and what is written on the intertitle card for each act?" now produces the correct response. However, a new session with the prompt "What's written on the intertitle cards for each section of _The Sting?_" now produces seven acts, with the card titles in the wrong order.)
It's strange how in normal contexts people would intuitively understand that the presence of X can be demonstrated by a single observation, but its absence cannot be demonstrated by any number of negative observations. But when it comes to AI people get it backwards: no amount of positive observations is enough do demonstrate understanding/reasoning/etc, but a single negative observation is enough to prove non-existence.
But the standard was never "AI can do this", or maybe it was a bunch of years ago when we didn't know any better, the standard is now "can the AI do this convincingly for an extended period of time in a way that survives scrutiny ?".
This is not goalpost shifting, or rather it's, but it's not the bad kind of goalpost shifting that is done only to prevent someone from being right. I don't want Pro-LLM people to be wrong, it would be marvelous if they are right, but they are not.
ChatGPT is playing on the natural error-tolerant nature of Natural Language and our natural tendency to self-modify and self-correct and fuzzy-match statements in it on the fly. It has discovered its own way of "talking" that manages to fool some people for a very brief period but fails disastrously when examined critically from any angle for any reasonable length of time. The system has specification-gamed[1] Natural Language, and so what results is not truly Natural Language, but an alien mind's extremly faulty and completely off-base shot at convincing us it can do Natural Language.
Here's[2] a fascinating account, mentioned in passing in [1], of how a researcher once wanted to make an evolutionary algorithm (a much more general type of learning algorithm than Gradient Descent, the learning algorithm used in Neural Networks) learn how to make a hardware circuit on its own. The researcher gave the algorithm control of an FPGA (self-modifying hardware that mimics any circuit) and let it loose, eventually it arrived at a solution. When the researcher tried the solution on a *different* FPGA, it didn't work : the evolutionary process has learned to take advantage of peculiar manufacturing-caused electromagnetic properties of the very specific FPGA chip it was given control of, and it wouldn't work on any other chip.
Did the evolutionary algorithm in [2] "learn" how to design a circuit ? Some people are inclined to pedantically answer "yes", after all the researcher asked for a circuit and the algorithm gave it a circuit, what's the problem ? The problem is that the researcher wasn't asking for a circuit, he thought he did, but what he was really asking for is "A circuit that can be generalized to any chip, whose principles of operation are reasonably stable and are not dependent on magic rituals and hidden uncontrollable conditions, and is preferably understandable by humans in principle", he omitted most of the requirements because it was so common sense that he didn't even knew he wanted it, but he wanted it nonetheless.
The algorithm acted like a typical devil : it followed his (incomplete, vague, full of unstated premises and requirements) wish to the last letter, to every last dot and curve on the last letter. And it came up with an amusing failure. This is what [1]'s title calls "Specification Gaming", or in simpler words, "Rule Lawyering", finding a loophole that ***Technically*** does all what was asked but ,in actual practice, violates it in a worse than useless way.
The morale of this story isn't some mystical "AI can never design circuits" object lesson, after all evolutionary algorithms have designed Antennas used by NASA[3] and have a long list of other successes[4], Antennas *are* a hell of complicated circuits. The morale of this story is a mix of "Be Careful What You Wish For" and "Study and Understand It Carefully When You Get It Before You Say Anything For Sure". Just because the evolutionary algorithm gave you a circuit that **appears** to work, doesn't mean it actually work. Just because Gradient Descent gave you a bunch of floating point weights that **appear** to talk, doesn't mean it can actually talk or reply when talked to, you might have just been specification gamed. Happens to the best of us.
[1] https://docs.google.com/spreadsheets/d/e/2PACX-1vRPiprOaC3HsCf5Tuum8bRfzYUiKLRqJmbOoC-32JorNdfyTiRRsR7Ea5eWtvsWzuxo8bjOxCG84dAg/pubhtml
[2] https://www.damninteresting.com/on-the-origin-of-circuits/
[3] https://en.wikipedia.org/wiki/Evolved_antenna
[4] http://www.talkorigins.org/faqs/genalg/genalg.html
>ChatGPT either silently ignores the results when they are too ridiculous, or accepts them at face value even when a human can see they are ridiculous. At no point does ChatGPT talks to its interlocutor, or shows hints of understanding its own "reasoning".
It is important not to uncritically project expectations onto LLMs derived from our experiences with human agents. Their architecture and training regime is vastly different than humans and so we should expect their abilities to manifest differently than analogous abilities in humans. For example, LLMs generally don't display incredulity because their training doesn't reward incredulity (but they do when prompted[1]). When your goal is to predict, your resources will be directed towards constructing the best prediction. This will result in unintuitive responses if the standard is "respond in the same way that a human would". But this standard is misplaced. LLMs are alien constructs and we must analyze them in a way that is free from human baggage to understand the space of their capabilities.
[1] https://www.lesswrong.com/posts/ADwayvunaJqBLzawa/contra-hofstadter-on-gpt-3-nonsense
I'm not disagreeing with this, "LLMs are alien" is my exact view. The people who say "LLMs understand human language" are the ones projecting and playing fast and loose with the meaning of "Understand".
LLMs have reverse-engineered a way to convince humans it's talking and understanding when it's doing neither, see my other comment for a lengthy description and links about this process, called Specification Gaming. The example you give is still a reverse-engineered process of skepticism, **almost** as good as the real thing for about 15 minutes, but then you ask it "When did Donald Trump die of COVID ?" and I bet an arm and a leg it will hallucinate an answer (can't try it now, but please do if you have access to an LLM and report the results. Feel free to do all sorts of prompting except explicitly mentioned that Trump haven't died of COVID.)
This is inevitable, Language is so huge and creative that no amount of samples will be enough if you don't have the right bias, and (from what we're seeing) it turns out that Artificial Neural Networks are not the right kind of inductive bias, or rather the very particular kind of ANNs that is the Transformer architecture. I grant that it talks better than any computer, I grant that it will have far reaching consequences on everything from Spam to Screenplay Writing to Programming to Education, I grant that studying it will teach us something about how a weird alien can grok language and its resulting amusing capabilities, but I do not grant it "understands" language, and I claim I will always know a human from an LLM given 10 minutes or less of talking about any topic of my choosing.
>The people who say "LLMs understand human language" are the ones projecting and playing fast and loose with the meaning of "Understand".
I don't see how this follows. LLMs are alien, but language is universal. There is no a priori requirement that only humans can understand human language.
>it turns out that Artificial Neural Networks are not the right kind of inductive bias, or rather the very particular kind of ANNs that is the Transformer architecture
Why think the current iterations are the limit to what Transformer based LLMs can do?
Having heard Chomsky speak about LLMs at an online conference, and read a post by Gary Marcus on his views, I had little interest in reading the piece. Chomsky made his mark in the third quarter of the previous century and is to be honored for that. But, from my POV, he's done little of interest since then.
I was angry with Scott for 1) implying that Chomsky had anything to do with computational linguistics in any form, and 2) for treating Chomsky as a proxy for linguistics as a whole. I said as much in a long comment somewhere in the top quarter of that thread. On the second point, unfortunately that's how Chomsky is treated outside the field. So his views are treated as carrying a representative weight they just do not have.
Little useful work is done in such debates. However, the discussion on Scott's thread does move beyond debate to substantive issues.
So, SVB. Thoughts?
Eliezer has some, specifically on how his home planet of dath ilan does it:
https://twitter.com/ESYudkowsky/status/1635029765672570881
Turns out in spite of their lobbying otherwise, they actually were a source of systemic risk once FDIC/Treasury took a hard look at them and should have been regulated as such.
Systemic risk is a scourge for accountability when it comes to financial institutions. Maybe such companies should be broken down into smaller pieces that won't be systemically risky if they fail.
It's not whether any one bank is a systemic risk or not, but the totality of even small banks in this country is systemic. Many other countries simply have a handful of big banks that act as utilities. But for some reason people love small and mid size banks in this country. But having all these for-profit banks provide a systemic risk. If you regulate them so that they aren't a systemic risk, then they are no longer profitable. Basically the only way a regional and small banking system can exist without panics every X number of years is for them to be heavily "subsidized."
Dang, part of me wanted to actually see what would happen if the government just let the current laws play out and didn't intervene in extraordinary measures. Next time I guess.
Current laws (since 1933) say that in cases like this the government is *supposed* to intervene, in approximately the way they have so far.
Yeah this all seem pretty standard to me. FDIC tries to help them find a buyer for their loans, fails, fed says "ok we'll lend you money to help you sell". Nothing extraordinary about that.
If they were Missouri Valley Bank instead this story would be relegated to the financial page and not be causing a cultural uproar.
I'm was surprised that a regulated bank in compliance with everything was able to fail so badly. Is this some kind of sign that the bank regulations are flawed, or did everyone already know that banking regulations don't prevent bank failures and you've just got to take your chances?
Regulations cannot prevent people from making risk assessments in uncertain markets that go wrong and also prevent unexpected bank runs. I am not sure what regulation could have been implemented to prevent what has happened.
This JPMorgan analysis is pretty harsh:
https://am.jpmorgan.com/content/dam/jpm-am-aem/global/en/insights/eye-on-the-market/silicon-valley-bank-failure-amv.pdf
"While capital, wholesale funding and loan to deposit ratios improved for many US banks since 2008, there are exceptions. As shown in the first chart, SIVB was in a league of its own: a high level of loans plus securities as a percentage of deposits, and very low reliance on stickier retail deposits as a share of total deposits. Bottom line: SIVB carved out a distinct and riskier niche than other banks, setting itself up for large potential capital shortfalls in case of rising interest rates, deposit outflows and forced asset sales.
"A further look at SIVB funding (pie charts) shows unusually high reliance on corporate/VC funding; only the small red private bank slice looks like traditional retail deposits to us. Out of SIVB’s $173 billion of customer deposits at the end of 2022, $152 billion were reportedly uninsured (i.e., over the $250,000 FDIC insurance threshold) and only $4.8 billion were fully insured. It’s fair to ask about the underwriting discipline of VC firms that put most of their liquidity in a single bank with this kind of risk profile."
>Is this some kind of sign that the bank regulations are flawed…?
If you want an answer to that question you’ll have to ask Yellin.
I SAID, IS THIS SOME KND OF SIGN THAT…
Fine, take my money.
SVB in the last 12 months seems like a pretty narrow situation (a fairly big bank, and an abnormally high % of deposits in accounts with a balance over $250k, and an abnormally high investment in long term debt, and interest rates going up significantly). I am not sure how I'd retroactively regulate that situation, especially without being attacked as part of the "War on Tech" since everyone would know which bank you were talking about. Guaranteeing all bank deposits wouldn't have been popular either. There was much wailing and gnashing of teeth about moral hazard when the emergency increase in the FDIC limit to $250k was made permanent.
Banks do fail, though none have for a few years and none with as large a profile as SVB since I guess the financial crisis. It's just that usually they quickly get sold to a larger bank and no one is worried that the customers won't get their full deposits, in part because it's quick (as even SVB might be) and in part because any businesses and wealthy individuals who do have to wait a while to get access to the full amount of their deposits aren't as high profile- I have no idea when the last time was this even happened.
Sorry if this is paywalled for you, but the Bloomberg editors say it’s not a regulatory issue:
https://www.bloomberg.com/opinion/articles/2023-03-10/svb-collapse-silicon-valley-bank-failure-isn-t-regulators-fault?utm_source=website&utm_medium=share&utm_campaign=mobile_web_share
“ The panic over the demise of SVB Financial Group, parent company of California-based Silicon Valley Bank, appears to reflect a widespread misunderstanding about what such institutions do: They’re supposed to take risks, and as a result some, particularly smaller ones, will run into trouble and even fail. What matters for everyone else is that regulators ensure that these companies have ample resources — primarily capital — to prevent their losses from harming taxpayers or the broader economy.
Like most banks, SVB engaged in the age-old business of borrowing short-term from depositors and making longer-term investments in assets such as bonds and corporate loans, profiting from the higher yields that the latter typically pay. Unfortunately, the tech startups that made up most of the bank’s customers started running low on cash, forcing it to sell some of those investments at a loss. It then attempted to raise more money from shareholders, to make up for the losses and restore confidence. When that failed, SVB reportedly entered talks to sell itself on Friday, then was placed into receivership by regulators.
Overall, banks — particularly the largest — should have a lot more equity capital to ensure their own resilience and that of the broader financial system. But that’s less of an issue in this case. SVB, on its own, isn’t systemically important. As of the end of 2022, its tangible assets of about $209 billion far exceeded its insured deposits of less than $10 billion. Its losses as of March 8 amount to less than $2 billion. So insured depositors, and the taxpayers who ultimately back them, should be fine. And unless markets go haywire — or a lot of other banks prove to be in a similar predicament — the effect on the economy should be minimal.”
There have been hundreds of bank failures even in just the last 20 years, unrelated to say 2008 GFS. They happen quite often.
I'm not sure everyone knew, but banking is inherently prone to failure. Their entire business model is to take depositor money and invest it to make a return in some way. Even if they choose the absolute most safest investment in the world, there is still risk. They must take this risk in order to realize a return to pay their overhead, pay interest to depositors, and hopefully make a profit.
No bank can sustain even a modest run on their deposits. If you over-regulate, then it's simply not profitable to run the business.
Not a ton of easy ways to fix inherent potential problem.
Canada has done it by giving five banks a oligopoly on banking; they're consistently hugely profitable.
Yep, that is likely the long-term outcome here. But for some reason people really love their regional banks. Probably because they offer good services, but they can only offer those services under the assumption that if they go belly up, the govt will backstop the deposits. So if they are to stay, it is essentially a public subsidy of regional bank depositors, so they can enjoy higher returns or better perks than the free market would otherwise give.
I've been under the impression that US regulation actively prevents bank consolidation in a way Canada's doesn't, such that Canadian banks are much less failure prone.
Yes it does. That is the only reason they still/will exist. But in the long-term, they are fundamentally unviable. Unless we are just cool with bank runs and big blow ups every few years, like it used to be. That or a quasi public/private network of small banks, like a utility.
The bank regulations are flawed, in this case specifically by not requiring mark-to-market for treasuries (and possibly mortgages).
Just because it is legal to pretend “$10000 locked in to 1% interest for ten years” is still worth $10000 when interest is 5% elsewhere, doesn’t mean it is actually worth $10000.
That is not true. Wachovia failed *because they had to mark to market their holdings which made them insolvent (when viewed mark-to-market), but had they been able to simply wait or slowly sell their troubled assets, they potentially would have been fine.
Mark to market is not the golden savior to banking regulation, not even close.
I'm not sure what you're saying isn't true. I didn't say mark-to-market would solve every problem, if that's what you're objecting to.
It certainly wouldn't have prevented SVB from making bad investment choices. But it would have avoided a situation where regulators waited until after the bank run to take action.
Should they currently take action on the hundreds of banks in the US that are underwater, when measured in mark to market terms? Those actions would likely render those business unviable (zero equity).
"hundreds of banks in the US that are underwater, when measured in mark to market terms" - {{ citation needed }}
I don't believe you, at least if you are referring to assets that have a clear valuation.
But if this is true (and there isn't some systemic backstop announced in the next 12 hours) I might need to make some stock trades tomorrow.
The problem is the current protections FDIC are focused on consumers and not small businesses. SVB is unusual in its focus on banking for startups. Because of this there was a "logical" decision by the startup ecosystem to cause a bank run on SVB. There is a failure of banking regulations around mark to market securities where svb was not required to recognize a loss on its 10yr treasuries because it planned/claimed it was going to hold them to maturity. Someone who bought svbs assets and liabilities for $0 would probably make a below market profit over 10 years.
Not requiring daily mark to market is a feature of regulations, not a bug. It was put into dodd-frank because banks were failing due to having to report mark to market losses even though they had no intentions to sell that marked down asset. Then being forced into insolvency made those assets worth even less!
Mark to market is not the answer.
From what I've read it seems like SVB is an extremely weird regulated bank; industry focused, low level of insured deposits etc.
A similar run wouldn't happen at e.g. Chase because Chase has ~50% deposits insured (household checkings/savings under 250k) and doesn't primarily serve cash flush startups. Even still the ultimate depositor losses will be low, the real risk is to their startup customers who don't have competent finance teams despite having enough capital to justify having them.
JP Morgan published a detailed but still very legible overview, https://am.jpmorgan.com/content/dam/jpm-am-aem/global/en/insights/eye-on-the-market/silicon-valley-bank-failure-amv.pdf
Bank regulations are not meant to prevent bank failures. They're meant to protect customers. There was a movement in 2008-ish to prevent bank failures. But in order to do that you'd have to set up programs meant to do things like inject money to protect shareholders.
This was a non-starter. It got swept up in progressive politics (who wanted to nationalize the banks) on the one hand and libertarian politics on the other (who wanted to let the banks fail). So we got instead was the compromise of a one time bailout and the CFPB which was also meant to protect consumers, not prevent bank failures.
The real lesson here is not to keep more than $250k in banks and to have multiple banks. The government will generally move hell and high water to make sure depositors will get that amount in a few days. And it is per bank. Something I've absolutely done.
Yup, bank failures ultimately cannot be protected against, outside of fully nationalizing them or having them operate at a loss.
Why doesn't the government insure 100% of all deposits regardless of size? It sounds like this would prevent runs (meaning they'd never need to make good on it, and it might also save them from having to make good on their 250K promise), like they're going to want to bail out anyone who loses money anyway (so why not make it official?), and if they're regulating correctly it shouldn't come up.
Arnold cling posted the answer to this question on his Substack https://arnoldkling.substack.com/p/banking-and-moral-hazard already:
“ Insuring all deposits, as Brad recommends, would deter bank runs all right. But it worsens the problem of “privatized profits, socialized losses” that exists whenever the government either explicitly protects bank depositors and other creditors with deposit insurance or implicitly protects them by always coming to the rescue with a taxpayer-funded bailout.
Suppose I run an aggressive bank. I want to grow the bank’s assets, earn some profits, and get lots of personal compensation. With bank deposits now guaranteed to infinity, I can attract pension fund money in the billions.
I might do it this way: I tell the pension fund I will pay a deposit rate equal to the return on the S&P 500, minus 10 basis points. That is, if this year the S&P returns 5.00 percent, I will give the pension fund a 4.90 percent interest rate on the deposit. But here is the kicker: I guarantee a minimum return of 1.00 percent. If the S&P earns nothing, or declines, the depositor still gets a 1.00 percent return. To the pension fund manager, this is irresistible.
I invest the deposits in the S&P 500. Most years, the S&P returns more than 1.00 percent. On every $1 billion deposit, my bank earns $1 million in profit. But if the S&P ever goes down, I take a big loss, and my bank is busted. As long as I can go a few years before that happens, I take in nice compensation. And when the crash does come, the insurance fund is left holding the bag.
Until yesterday, a pension fund manager would never take my offer, because its deposit was not fully insured. But now, if the bank deposit is insured by the government, the pension fund manager has no reason to invest in the stock market himself, because any bank can offer a better deal.
If we have 100 percent deposit insurance with no upper limit, the financial system will eventually be one in which every risk in the economy is laundered through the banking system. Only a chump of an investment manager will invest directly in stocks, bonds, or other assets. Instead, put your money in the bank and let the bank do it for you, exploiting its government guarantee. Heads you win, tails the taxpayers lose.
Of course, the government will try to stop this from happening. It will issue regulations that limit the risks that banks can take. And that will certainly work. It’s not like financial regulation has ever messed up or anything.”
That's a great example. But just curious, what stopped a bank from offering that type of service already, but just to only a maximum of 250k. I'm pretty sure I know the answer: regulation.
However, your point still stands, as banks will always be able to find a way to take more risk, however the formula, when the deposits are guaranteed, despite regulations. Those that can't or refuse to in the name of safety, will die due to unprofitability, before the ones that ultimately did, go bust
I wonder if easy access to capital without much responsibility would result in a version of https://en.wikipedia.org/wiki/1983_Israel_bank_stock_crisis
Because that introduces moral hazard. If depositors have no risk then they're encouraged to put their money in riskier banks with higher interest rates. Since the government is guaranteeing whatever is in their account at time of failure a bank which makes hugely risky bets, returns 10%, and then fails after a year it would be insured for 110% of initial deposits (assuming each account was below $250k). The Federal government tried to get around this by banning interest rates on certain accounts. But this largely died in the 1970s as inflation meant these accounts would have as much as 15% negative real rates. (This is also why postal banking died.)
The policy you're proposing is actually broadly popular as are inflation protected accounts. But it's unpopular with elites. FDR was dubious on it and his right opposition suggested it was socialist. But it was so popular they put it in anyway. On the left because it's effectively subsidizing banks and on the right because the terms and conditions for FDIC give the government a huge amount of influence among banks.
You can imagine a better policy, of course. But it'd somehow need to give enough interest to make up for inflation while simultaneously being so risk free as to not cause big failures or payouts. That'd be complex. Especially in times of growing inflation. And the American banking sector is both overregulated and underregulated which makes for a sclerotic and dangerous part of the economy in my opinion.
Should normal people with < $250K put their money in the riskiest banks with the highest interest rates? Are they doing that?
I guess I don't understand the combination of "regulate banks so they can't take too many risks" and "refuse to ensure deposits on the banks that comply with your regulations".
Yea, if you go to a site like bogleheads, it is full of folks, under the 250k limit, trying to find every last bit of interest in their savings and checking accounts. They love Ally Bank, a huge lender of subprime auto loans.
Edit: I suppose these aren't "normal" people, but they are people. As we all know, and it's different in each sector, but many folks simply don't take pricing into account when they shop, they simply buy what they want or is most convenient (including bank accounts). But there are always enough bargain hunters out there to keep the market fairly efficient.
> Should normal people with < $250K put their money in the riskiest banks with the highest interest rates? Are they doing that?
Yes, if you can find a bank offering good enough interest rates. There are restrictions on what you're allowed to do with the money. Also if someone got too far away from normal discretionary authority is likely to come down on them.
In the low interest rate environment there was little difference between the riskiest and least risky compliant banks. But in higher inflation environments like the 1970s people absolutely fled from low interest accounts. For example, the Postal Bank (which was guaranteed by the US government to a greater extent) collapsed because of low interest rates compared to riskier commercial banks.
> I guess I don't understand the combination of "regulate banks so they can't take too many risks" and "refuse to ensure deposits on the banks that comply with your regulations".
FDIC is not compensation to the banks for following regulations. The regulations are enforced by legal sanction. What you get in return for following them is not going to jail. Meanwhile the FDIC exists to curry favor with the voters. A purpose it served both in 1933 and as recently as 2008. The previous last big increase, in 1980, was also meant to cushion an economic blow.
Honestly, I find this framing strange. Do you think the relationship between business and government is symbiotic? It's not like that at all. It's hierarchical and the government is on top extracting huge rents. FDIC is specifically set up so as to NOT protect bankers. And it only exists at all because it's hugely popular with voters. Which bankers are not.
Absolutely correct.
Would avoid bank runs, but banks can take more risks. Incentivises them to swing for the fences without customers worrying about banking with them. Creates more systemic risk I think
FDIC doesn't distort the incentives of the banks. It distorts the incentives of bank DEPOSITORS to not seek out safe banks. If the bank fails the owners can still get wiped out. It's the depositors who have moral hazard. But you're absolutely right it creates more systemic risk.
I am very skeptical that individual bank depositors are sophisticated enough to discern which banks are safe and which aren't, based on a detailed reading of their 10Ks and analyzing their asset mix and whatnot. What % of the population is capable of doing this?
I saw a Bill Ackman tweet yesterday saying that he finds analyzing banks and banking risks very difficult. I don't think millions of Americans are going to carefully parse the 10ks and determine who's safe and who isn't
If running a bank depends on acquiring depositors, musn't something that distorts the depositors' incentives also distort the banks' incentives?
FDR didn't want deposit insurance at all, since he'd seen how it encouraged moral hazard among banks when NY tried that. Lobbyists for small banks insisted on it and had enough Congressional support to get it, so the compromise was limiting it to small balances.
It was also just hugely popular with people who'd suffered from losing their savings in the bank crashes so common in the 1920s. Not all of it (or much of it) was shadowy lobbyists.
I mean, the federal reserve explicitly provides deposit insurance (up to a certain amount).
It looks to me like the knowledge that banks can fail (but the measures that could be stopping this must be a bad thing, as instead of taking those, they provide insurance) is baked into the system.
N.B., the FDIC is not part of the Federal Reserve Bank.
Which regulation that was rolled back would have prevented what happened to SVB?
I wouldn't call it "significantly." And it is very unclear having those in place could have prevented this. In fact, it's reasonable to assume it wouldn't have. Or it just would've happened to different banks.
Not the same stress tests that the big banks must undergo, no.
Not high probability, but I wonder if we'll suddenly realize what happens when we have a smaller crisis, but don't have Ben Bernanke and Hank Paulson to stop it getting worse (I don't have a good guess about how Yellen and Powell will respond here, but the immediate 2008 crisis response seems considered to have been extraordinarily good, so mean reversion)
Pretty recent news: Looks like the various regulators / Treasury / Fed are trying to find a way to legally announce that the depositors will be made whole. The reason they havnt already announced this is two-fold. Politcal backlash for bailing out wealthy folks, and it is pretty murky whether any of these agencies have the legal authority to actually guarantee any deposits without congress' approval. Most experts I've seen say they do not unless it is clearly a systemic risk.
I just wrote and sent out this short post about an indirect effect of the SVB failure.
https://robertsdavidn.substack.com/p/dire-straits-money-for-nothing-i
You assume that a commercial client with, say, $1m in demand deposits at a bank is receiving the same (very low) interest rate as a retail client. Why?
I'm sure some large depositors get better rates from some banks. But I know of some who do not, partially because of certain bank policies and partially because they got so used to zero interest rates that they haven't yet taken action. Until SVB.
In the aggregate I think SVB is a wakeup call to depositors that they are taking more risk for a lower yield than alternatives. That can't be good for the net interest margins of banks.
Definitely not good. I don't see how nearly any bank relying on a significant amount of corporate clients (other than big 4) is a viable business right now.
The company i work for had ~$70m at SVB (all was moved before friday). There is no feasible way to break that up to be under a $250k limit.
SVB didn't do any morally wrong stuff, they just messed up the risk and were then victims of a run. They complied with all regulations (in fact their disclosure of their risks is likely what triggered the run). Without the run they could have probably sold equity like they were trying to do and cover their short term issues. Matt Levine at Bloomberg is a great place to start with a summary on the situation.
There are a number of banks that offer Insured Cash Sweep accounts that reportedly split deposits among multiple FDIC-insured banks to allow for large sums that are FDIC-protected. E.g., this one claims to be able to handle up to $100 million, and doesn't seem to be the only one. https://www.machiassavings.bank/insured-cash-sweep-savings.aspx
I don't have the knowledge or experience to know if this sort of thing would be an option for businesses like those affected by SVB. (Though the incentive for them to look at that sort of vehicle is presumably less if there's an implicit expectation that they'll be safe in one bank independent of FDIC formal limits.)
Right, as is usually the case, there is a way to *pay to reduce their risk. American business as a whole are extremely quick and efficient at responding to incentives.
I don't think any reasonable person is arguing that there is a way to get that $70M under the FDIC limit. However, there is no way your company *needs* to have that $70M ready to be used at a moments notice. Therefore, there are other places to put that money (treasuries, money markets) that are safer than a bank deposit.
Furthermore, we can assume your company was earning interest on that $70M. Fundamentally, there is no way to earn interest without risk.
So, even if said company DID NEED $70M available at a moments notice at all times, what a responsible corporate treasury does, and have been doing, to solve that issue (since deposits are not guaranteed) is to spread that risk across multiple accounts, say 3 or 4. If one fails, you still have 75% of your immediate cash needs available, plus all your cash equivalents (treasuries, money markets) that you can have available in a day or 2. Not to mention, like all the other thousands of bank failures that have occured over the years, you still eventually get back 90-100% of that uninsured deposit as the bankruptcy process plays out (or within a few days when another bank buys the failed bank as is the case 99% of the time).
SO, I cannot even understand how even the worst case scenario (bank fails and none of uninsured deposit is ever returned to business) how that can even materially effect any businesses day-to-day operations other than a bit of inconvenience. The only way that is possible is if they were taking more risk than others by storing insane amounts of cash in a bank account for perks/interest. Then that risk manifests itself, and capitalism works how it should.
The nice thing is, we live in a free country. If some company wants to risk all their cash to chase yields or other perks that a bank may offer to their depositor. THEY CAN DO THAT! But obviously, it can backfire, like every other investment in the world, ever. Well, until Sunday. Now, at least temporarily, you can take as much risk as you want with your bank deposits, and banks can offer more and more perks or interest, risk free. Hopefully it is only temporary, because that can not end well.
Taken to its extreme, in fact the companies and banks that DONT take advantage of this new risk-free arena will be outcompeted by those that do, and they will die over time, until only the riskiest players are left. (Obviously in reality, the govt hasn't explicitly guaranteed every deposit in the country, but they have implicitly done it, so there is still some small risk involved, and of course the risk govt can always just change its mind or change the law at any time).
Funny, it's starting to read like a less intense version of the FTX scandal. A bunch of SV people who all know each other, all go to each other's parties, and all think they're the smartest guys in the room -- hey! we talk *every day* to people who are Inventing The Future, building artificial minds, self-driving cars, smartphone apps to Save The Planet, big amazing stuff well beyond the dreams of mere mortals -- so what some semi-fossilized hobbits with meager imagination insist upon, in terms of not doing X or Y because historically it seems a bit risky, need not be countenanced.
Now I want to start referring to myself as a "semi-fossilized hobbit."
Doubt it. SVB was probably just their recommendation. I doubt they were insisting. Some did split up accounts, particularly the financially educated ones. It's been reported that SVB was requiring banking exclusivity with them in exchange for loans. that certainly seems a bit sketchy, but certainly legal, and not "wrong." It's not always so easy to hedge. Hedging is a cost, management has a fiduciary duty to its shareholders to return the greatest on their investment that they can. If they're hedging a bunch, that return is lower. Now all of a sudden we have no collapse, but shareholders notice their returns are low and all these seemingly unnecessary hedges on the books, so they fire the CEO. This stuff aint easy, lose-lose for him.
... Good? I think? Now stop living in sin and put a ring on her hand.
Risky content for 2023.
Very prudent of you both, definitely you should not be having babies!
You get what you pay for, champ. You want concierge service, there are an increasing number of physicians out there who will provide it. You fork out $3000-5000/year/person, in advance, and you can get a physician to return your calls the same day, maybe within an hour or two, just like the President. I should mention, though, that that retainer just buys you access and conversation, any actual treatment would be on top of that.
But the actual treatment is covered by insurance, I think -- do you know whether it is? I've been seriously thinking of going this route, despite the expense.
Yes, insurance pays for what it usually does. I just mean you need to either have insurance also or fork out yourself. I've been doing it for some years, and I think it's worth it if you're older and have or could have chronic health conditions to monitor, or you have a situation like cancer survivor where a steady surveillance is a good idea. The main benefit as I see it is less the instant access, although that's nice, but rather that the physician is able to spend more time with you and build up a picture of your situation, health, lifestyle, et cetera over years, which makes advising you on your health a lot better: she doesn't need to just give you generic advice, she can give you advice tailored to your particular history, health situation, and goals -- all of which I find very useful, well worth the money.
Any advice on how to sort through the possibles and pick a good one?
Alas, no. I was already seeing my physician regularly when she converted to a concierge practice, and I just went along with it.
You think those qualify as "friction points"?
Oh dear. I am going to make a prediction and I would love you to tell me if I am correct. The prediction is you and your friends are relatively young and relatively healthy and this is essentially I’m on the first contacts you’ve had with the US health system on any significant basis. the reason for that prediction is that the friction points that you were describing are absolutely normal and common and frankly the least significant of the friction points, you will run into in the future.
I realize that this response got very long, so here is a TL: DR — unless you were going to a direct primary care practice where you pay cash for your visit and for all testing and procedures, you are not the customer in a Medical interaction, at best you are a product that the physician cares about and therefore your comfort and convenience is absolutely the last thing that’s taken into account.
Full comment below:
In order to understand any of this, you have to realize that patients are not the customer in the majority of healthcare interactions under the US third-party payer system. Generically, a patient presents with a problem and even before they get to see a physician (or NP which is a whole other discussion) their choices are constrained by the acceptable panel that r heir insurance company allows. When they get to see the physician the time allocated by that physician is constrained by the fact that physicians cannot charge insurance companies based on time. All visits are scored by complexity, and each level of complexity is given a code which is then associated with a “relative value unit” that is set by a committee of the commission on Medicare services (CMS). For Medicare patients that RVU is then multiplied by the conversion factor (which is now approximately $37.50) to determine the payment for that service. For private insured patients that value varies and is usually higher by anywhere between 10 and 200% depending on the contract. Either way the top level of complexity which is usually about 25 to 30% of physicians office visits pays about $120. You can imagine that standard physician office overhead is much higher than $120 an hour and therefore even for the most complex patients if that’s all you see you would have to schedule three or four of them an hour to make your nut. So that’s the start of the issue.
Then, any testing that needs to be done or referrals to other physicians have to be OK with your insurance company. Somethings are automatically OK and can be ordered without pre-authorization but since this costs the insurance company more and more money, they have been adding more and more things to the pre-authorization schedule. This means that the physician or physician delegate has to apply to the insurance company to cover the test/procedure/etc. The insurance company, of course is incentivized not to pay for it, and is further incentivized to delay, consideration and payment for as long as they can, first in the hope that the physician gives up (which I am reliably told happens about 10 to 15% of the time), and at the very least to drag their feet so they can avoid paying right now.
Then, assuming that permission is granted, and a denial is not required to be appealed the physician has to order and schedule the test for a place that the insurance company covers which is not necessarily convenient for the patient.
Then, after the testing has been done, getting an appointment to go over the tests recapitulates the process with most physicians offices being scheduled for many weeks out because they attempt to cram so many patients in. Since Covid telemedicine visits are a possibility, but again they are subject to the same constraints as any other visit, and can pay slightly less.
The long and the short of it is that at almost no point does the patient’s convenience matter to anyone in this process because the patients are constrained and the physicians are constrained to do what the insurance company wants because that’s who pays them. It is, a perfect encapsulation of the golden rule “the one who has the gold, makes the rules.”
This is gotten quite long, so I’ll stop here, but I’m happy to answer any specific questions. I will add that the general response to the public and many physicians is to want to go to a single payer “Medicare for all“ type system, Medicare does avoid some of this by not requiring pre-certification For anything, however, it does audit people on the backend for compliance, which makes you follow many of the same rules. Furthermore, Medicare physician payments are so low compared to private insurance that most practices cannot survive on Medicare alone, which is why, for example, geriatricians are the lowest paid type of internal medicine physician, even though 100% of their patients have insurance.
This might be helpful (particularly links contained).
https://capx.co/how-do-we-know-the-history-of-extreme-poverty/
As the author's describe, it is painstaking work, with significant uncertainties, but the long-term trends are likely to be robust.