THE JOYFUL REDUCTION OF UNCERTAINTY

OT115: Oberon Thread

This is the bi-weekly visible open thread (there are also hidden open threads twice a week you can reach through the Open Thread tab on the top of the page). Post about anything you want, but please try to avoid hot-button political and social topics. You can also talk at the SSC subreddit or the SSC Discord server – and also check out the SSC Podcast. Also:

1. Comment of the week is John Schilling, explaining the good-cop bad-cop relationship between state courts and state legislators.

2. From now on, I will be deleting comments that say “first!”, whether or not they also say other things. Come on, people.

This entry was posted in Uncategorized and tagged . Bookmark the permalink.

Leave a Reply

631 Responses to OT115: Oberon Thread

  1. ing2 says:

    Your mission, should you choose to accept it, is to prevent crime in Gotham City — both supervillain crime and mundane bank-robbers-and-muggers crime.

    Batman has been trying to do this by terrorizing the populace and punching individual muggers. Many people have made the argument that this is not efficient. Batman is a billionaire; there must be a better way to prevent crime than what he’s doing.

    What, specifically, is that better way?

    Assume that your available resources include billions of dollars and fancy technology but no actual superpowers. (see: https://scifi.stackexchange.com/questions/141593/do-batman-villains-typically-have-super-powers)

    Immoral or unethical solutions are technically permitted, but only if you have a way to avoid Batman identifying you as a supervillain and punching you a lot. Batman is very, very good at punching individual people he thinks are supervillains, so you probably don’t have a way to avoid this.

    • dick says:

      Are we fighting “Joker puts LSD in the reservoir” supervillain crime, or just regular robberies and frauds and such?

    • ing2 says:

      Ideas:
      * The violent crime rate in Gotham City really is absurdly high. Check if everyone has moved off of unleaded gasoline and lead paint. Check if there’s lead in the water. Let’s just generally do some blood tests on criminals and test for bizarro forms of poisoning. (Apparently there’s a villain named Scarecrow who specializes in mind-altering chemicals; does he have an under-the-radar henchmen group that’s messing with the water supply?)
      * Do a survey of captured violent criminals and find out what sort of environmental stresses they’re under. If they’re in poverty, we can roll out a basic income. If they’re jobless, we can attach the basic income to make-work jobs: street sweepers, infrastructure maintenance, whatever keeps people too busy to pursue a side job as a bank robber.
      * How are people getting recruited for criminal activity? How are people getting hired as henchmen? Is there an underground website somewhere we can hack? Are there Shadowrun-style “fixers”, operating as intermediaries between villains and henchmen, that we can arrest? Let’s run some sting operations.
      ** Is somebody going to argue that being a “fixer” isn’t very illegal? Let’s change the laws and make it super super illegal.
      * Do we have a recidivism problem? If so, why? Are prison sentences long enough? Are people getting broken out? There are lots of reasons IRL to not want more prisons, but if Gotham City has that many violent criminals, Gotham City can build more prisons.
      * How much surveillance should we be doing? England apparently has quite a lot of video cameras; we should do the same.

      None of this touches on the actual rate of insane supervillains wearing absurd outfits.

      * Let’s start by implementing better OSHA standards for toxic mutagenic waste storage. No more open vats, please.
      * Let’s also politely ask Batman to stop doing the “frighten everyone into not being criminals any more” thing — that seems to cause a lot of this. Instead of the “masked vigilante will punch you” PR campaign, let’s try something more like “this is our city and we’re all in it together”.
      * Of course, this won’t stop all supervillains. Supervillains right now seem to be operating on contrariness: they identify something Batman doesn’t want them to do, and then they do that just for spite. So, let’s make a decoy.
      ** We’ll make a giant statue of Batman looking stern something that won’t aggravate anyone’s existing neuroses. Let’s make a giant statue of Superman saving an airplane, actually — that won’t aggravate anyone other than Batman. Let’s put it out on an island, somewhere there won’t be any collateral damage if it blows up. Then let’s make a big deal about how the statue is really important to us and we would be super sad if anyone blew it up. We’ll also put a bunch of security around it. This way, supervillains will have an outlet for their antics that won’t kill dozens of people.

    • RavenclawPrefect says:

      Assuming we’re just combating regular crime (or that the supervillians will quiet down once there isn’t a superhero for them to fight): Offer really good monetary incentives for all the residents of crime-ridden neighborhoods to move to Metropolis; replace those areas with nice parks or something. Once the base rate is down, use existing police resources to punish existing crime with high consistency and short-term severity. If you’ve got funds left after that, run some RCTs on educational interventions in different neighborhoods and stick with whatever seems most promising.

      If we still have supervillians: get Batman to do his thing somewhere else, and hope they follow?

    • Eric Rall says:

      Brainstorming options:

      The Ra’s Al Ghul method: destroy/depopulate the city, thus preventing all further crime there. This will probably fail due to Batman punching me.

      The Lord Vetinari method: use my money to buy influence in the city government (possibly including running for mayor) and encourage a policy of tolerating reasonably well-behaved organized crime on the condition that they help stamp out poorly-behaved and disorganized crime. E.g. let the Penguin run some card rooms and protection rackets on the condition that he keeps a lid on the muggers and helps take down the Joker. I’ve got an uphill struggle selling Batman on tolerating this, but I might be able to keep him busy by having Commissioner Gordon continually point him at the Penguin’s less-cooperative rivals, Jonathan Wild style.

      The Bloomberg method (very loosely inspired by the stop-and-frisk campaign): again, use some my money to gain political influence. But this time, work directly against crime making use of my money and gadgets to perform illegal/unethical surveillance. Probably the best way to obfuscate this (and escape unfriendly attentions of both Batman and the ACLU) would be to privately fund “detective agencies” (staffed with licensed private investigators) to wield my surveillance gadgets. They can take complaints to investigate from crime victims (often on a pro bono basis, or on an income-based sliding scale) or consult with the police, and use the illegal surveillance to figure out who did the crime and how. If they turn up evidence that would be admissible in court, hand the case over to the police. Otherwise, let Batman know we found someone for him to punch.

      The Ron Paul option: take over the city government and repeal all laws under my jurisdiction, thus defining away as much crime as I can.

      The Low-Hanging-Fruit option: buy Arkham Asylum, fire Harley Quinn and the Scarecrow, and do a massive security upgrade project.

      • gbdub says:

        Vetinari approach. Gotham is a crap sack world and doomed to some (high) level of corruption. Vetinari is an expert at recognizing how corruption can be managed successfully.

        Vetinari also has demonstrated the ability to corral extremely talented, but in some ways very naive / dumb crime fighters into effectively serving his ends without realizing they are doing so. That’s key for managing Batman.

        He needs to get Bruce Wayne married off to a strong willed, down to earth but respectable wife. Then make him a Duke city councilman to keep him too busy to do much Batmanning until there’s a Batproblem that Vetinari could subtly steer him toward to Batsolve (all the while making Bruce think he’s putting one over on Vetinari)

        • Civilis says:

          Vetinari also has a penchant for “persuading” a number of the more intelligent and wise criminals to work for the good of his city. I’m sure we can find some post in Gotham’s government for the unique talents of someone like Edward Nygma.

          On the other hand, our DCU Vetinari might be more effective at stopping crime in Gotham if he was in Amanda Waller’s position rather than mayor of Gotham…

    • sammy says:

      Some of these policies seem useful:

      https://www.vox.com/2016/2/15/10981274/crime-violence-policies-guns

      -The fancy technology Batman has seems like it would help with preventative policing efforts. Generally increasing police presence while improving checks on police power/corruption would be pretty cost effective.
      -Strict gun control (though this may not be very cost effective)
      -Investments in local economy and new businesses especially focused on providing jobs to young men with HS diploma or less, to compete against criminal/supervillain organizations for workers/henchmen

      Gaming the crime metric (quasi-unethical?):
      – lobby local government, police to change sentencing and labeling of crimes (e.g. decriminalize drug use)
      – bid up housing prices/gentrify neighborhood
      – provide homeless shelters outside the city center, effectively moving a portion of crime elsewhere (providing shelters also seems to reduce crime)
      – buy up lots of housing and stores in the city and turn it into an office park; a city with lower population has less crime

      I think its best to leave punching bad guys to Batman while most of his money should go to improving policing, improving the local economy and lobbying the government to reduce alcohol, guns and poverty in the city.

    • Nornagest says:

      Immoral or unethical solutions are technically permitted, but only if you have a way to avoid Batman identifying you as a supervillain and punching you a lot. Batman is very, very good at punching individual people he thinks are supervillains, so you probably don’t have a way to avoid this.

      I think I’m safe as long as I don’t give myself a weird name or wear a weird costume.

    • Simulated Knave says:

      Re the supervillain crime – stop imprisoning them five feet away from the damn city when they’re found criminally insane. Put them somewhere that’s actually HARD TO GET OUT OF. That alone would make a huge dent. Batman, frankly, solves supervillain crime fairly well. It’s just that they keep escaping again.

      Street level crime…it depends what’s causing it. Cleaning up city corruption helps. Cleaning up the city and urban renewal will help. A strong, effective justice system would be a big help. Job opportunities. The usual array of things that actually work, and which we don’t do because we either don’t or say we don’t have the money.

    • ManyCookies says:

      Man the Mission Impossible franchise has moved in a strange direction as of late, these crossovers are getting weirder and weirder.

    • Bugmaster says:

      How big is Gotham; i.e., what is Gotham’s annual operating budget ? If the budget is on the order of billions (or even 100M) of dollars, then Batman’s wealth will just be a drop in the bucket.

    • aristides says:

      I’m normally against the death penalty, but in a world of super villains escaping prison regularly, an expedited death penalty process would be helpful. One appeal, and that’s it.

    • The Nybbler says:

      I’m going to leave the supervillain crime to Batman; he’s good at it. So I’m going to turn Gotham into a police state to deal with the regular criminals. Basic plan is “Lights, Camera, Action”. Gotham is a dark place, and I’m going to literally light it. More streetlights, more building lights, keep lighting it until the local observatories scream and then light it some more. No dark corners for criminals to hide. Lights will be very expensive because they’re going to have to be hardened against bullets and any easy vandalism.

      To take advantage of this lighting, it’s going to take cameras. Cameras everywhere, fixed and some drone and vehicle based. And an army of people to watch the feeds. These will need to be somewhat distributed; no single point of failure should be able to take out more than a small fraction of cameras or viewers, because otherwise it’s a supervillain-magnet.

      Then action — police, also distributed throughout the city, ready to stop a crime or at least capture the criminal. We’ll probably need more jails too. Regular ones, in addition to building a better Arkham as others have suggested.

      Anyone in any of the policing organizations who suggests that Batman is an affront to their authority or otherwise an enemy of the city shall be assumed to be a supervillain-minion and dealth with accordingly.

      In addition, we will also be sealing and/or demolishing abandoned property, especially large industrial properties suitable for the headquarters of a criminal enterprise. If for some reason accidents keep happening when we try to do this, assume supervillain interference and call the Batman.

      • silver_swift says:

        > Lights will be very expensive because they’re going to have to be hardened against bullets and any easy vandalism.

        Make sure there is a system to remotely disable any individual lights for a short duration and give Batman access to that system. It will save money from not having to replace lights due to batarang related incidents as often (those things hold explosives, bulletproofing isn’t going to stop them).

      • Vorkon says:

        I’d suggest something like China’s proposed fake moon project, but that sounds JUST close enough to a supervillain plot that you run the risk of Batman punching you.

      • Callum G says:

        The issue with technological solutions is that with supervillans around they too can be corrupted (especially if you’re putting in a backdoor for someone like batman). Control of the entirety of Gotham’s surveillance systems would be a nice target for a batman supervillan.

        I think it’s a good idea, especially the lights part. Just maybe not the first thing to be done.

    • MrApophenia says:

      The issue with Gotham isn’t street crime. That’s a symptom. The problem is the truly staggering level of institutional corruption. The Mayor’s office is basically an appendage it organized crime. The police were too, until Gordon took over – and even he has only been able to mitigate the problem, not eliminate it.

      How do you solve massive, institutional corruption at a systemic level? Man, I don’t know. Bruce Wayne spends a lot of money and influence backing anti-corruption political causes and candidates, which seems like a nice start, and it has worked out pretty well with folks like Gordon. (There is, admittedly, one very notable exception. Or two, depending how you count it.)

      I do think robust anti-poverty measures are a good idea, too. You know, like the ones funded by the Thomas & Martha Wayne Foundation.

      People give Batman too hard a time about only punching criminals, come to think about it. He only spends a small portion of his resources on bat-shaped weaponry, the bulk is going to large scale charitable work and reform efforts.

      • Bugmaster says:

        One sure-fire way to do this is to take over all organized crime in the region. Rupert Thorne, Carmine Falcone, the Penguin and a variety of other villains (most of them of the mundane, non-super variety) have tried this over the years, with mixed success. Generally, they get knocked out by their rivals, or punched out by Batman, at which point Gotham once again descends into total chaos.

        • Conrad Honcho says:

          Perhaps that’s Batman’s real problem: when he’s punched out a criminal, all he’s really done is create a power vacuum. Since no one can punch out the Batman, perhaps Batman himself should take over organized crime in Gotham, and manage it in the least anti-social way possible.

      • Randy M says:

        When, canonically, did all this corruption show up? I was a Marvel comics buyer in my day, but I took in most of the Batman video fare; I don’t recall the high levels of corruption in the Adam West show or the animated show of the ’90s. The Micheal Keaton movie I don’t think had too much institutional corruption. Then the Penguin/Catwoman movie had Penguin become the mayor and Catwoman work for a corrupt executive.
        Corruption was obviously a big theme in the Dark Knight, though, and in the current Gotham series.

        • Walter says:

          I dunno exactly when it showed up in universe, but I wouldn’t be surprised if it was after the big crime drop that happened a while back.

          Like, back when it seemed like crime would just get higher and higher, reefer panic, etc, it was entirely plausible to have Batman fight crooks forever. Crooks are the natural order of things.

          But later, when cities got super safe, Gotham city didn’t change. There is still a thug in every alley. Which is weird now. Why don’t the cops stop them? Well…they are corrupt. Unlike other places. That works.

          Nothing to support this, just a just-so story that came to me when I heard the question.

        • MrApophenia says:

          I know it was a big part of Frank Miller’s Batman: Year One in 1987 – Batman and not-yet-Commissioner Gordon dealing with a Gotham City overrun with corruption. That comic was hugely influential on Batman from that point on, so it is possible that is the origin – but I know Batman comics also got darker throughout the 70s and 80s, which is before I was reading them (or born) so it may be earlier than that.

          • Le Maistre Chat says:

            My superficial understanding is that Batman comics divide into four periods:
            In the Golden Age (1939-49?), the Batman initially fought organized crime, mad scientists, and the odd vampire-werewolf. His first super-villains were created after about a year, once Robin appeared and Early Installment Weirdness like using guns ended.
            In the Silver Age (~1950-1968 for this character), the Depression Era organized crime and first wave of super-villains were taken away under the influence of psychiatrist Frederic Wertham (Seduction of the Innocent) and Gotham became a bright, successful city from where Batman and Robin had zany adventures. At first these were transparent rip-offs of the better-selling Superman stories (they fought aliens, got a canine sidekick, etc.) but later the super-villains were reintroduced as thieves whose schemes never killed anyone. This ‘second Silver Age’ was the basis for the Adam West show.
            The Bronze Age was inaugurated by Denny O’Neil right after the TV series was cancelled. New villains were introduced, like the eco-terrorists Poison Ivy and Ra’s Al Ghul, and the classic ones were rested in order to be brought back with darker, grittier characterization. I think it was around 1975 that corruption appeared as a theme for the first time since the Golden Age, with newly-created crime boss Rupert Thorne holding elected office. By the time the Bronze Age began, it had been established that the Golden Age and Silver/Bronze Bruce Waynes were parallel universe versions of each other: Golden Age Gotham had a two decade run where Bruce Wayne married Catwoman, hung up the costume, and kept Gotham safe through a combination of huge charity programs and holding office while raising a daughter.
            In 1986, Frank Miller wrote The Dark Knight Returns, coincidentally in the midst of the editor-in-chief declaring a total reboot of the DC Universe. DKR was considered so superior to anything that came before that Miller was told to rewrite Batman history from “Year One”, and his version featured an absurdly corrupt Gotham for the Batman and Lt. James Gordon to struggle against. Also Catwoman was re-imagined as a ex-prostitute (whores whores whores). Elements of this new status quo found their way into the second Tim Burton movie, but didn’t really take in other mass media adaptations, being as they were aimed at kids and all.

          • Nancy Lebovitz says:

            For a moment, I thought there was a character in Batman named Frederic Wertham (a joke about the real person) who was taking the supervillains away.

            I bet it would have worked as parody.

            This is as good a time to mention that Ada Palmer and Cory Doctorow are teaming up on a history of censorship.

          • Le Maistre Chat says:

            @Nancy:

            For a moment, I thought there was a character in Batman named Frederic Wertham (a joke about the real person) who was taking the supervillains away.

            I bet it would have worked as parody.

            … yes, that would have worked considerably better than disappearing without explanation. 😀

      • ThrustVectoring says:

        Anti-corruption causes have a tendency to be co-opted by corrupt forces in the wrong cultural background. When corruption is the only way to actually get anything done, anyone who is squeaky clean is usually powerless enough to be ignored. What remains is a charge that you can find documentable evidence for charging whoever you like, so the corrupt officials in charge of deciding who gets prosecuted can hand out corruption charges to whoever isn’t playing ball.

        • Bugmaster says:

          Agreed. As I see it, corruption in Gotham is a systemic issue that cannot, in principle, be fixed (barring some sort of comic-book mind-control magic). One might be able to channel it into more productive courses, to some extent, but ultimately Gotham will never become Metropolis (or even San Francisco).

          Sadly, there are parallels to this in our own world. Countries with high levels of corruption pretty much remain corrupt throughout all of their history, barring cataclysmic events such as mass genocide. No amount of intervention (again, short of extreme violence) will help, because if the people really want to be corrupt, they will always find a way.

          • John Schilling says:

            because if the people really want to be corrupt, they will always find a way.

            No, it’s worse than that. And also better. In a corrupt system, even if the people don’t want to be corrupt, they will find it each individually preferable to be corrupt anyway, because they will be surrounded by people who will coordinate to make sure they don’t upset the (rotten) apple cart but also to reward them if they go along with it, and there’s an intractable coordination problem in getting them all to shift behavior simultaneously.

            In an honest system, even if the people do want to be corrupt, they won’t find a way to expand beyond the most petty forms of corruption without an extreme risk of being turned in by one of the apparently-honest people around them. The coordination problem here isn’t quite as intractable as in reforming the corrupt system, but it’s close.

            Broad corruption is probably the minimum-energy state, but both broad corruption and very low corruption are stable equilibria. The middle ground very much isn’t, and that’s what you need to traverse in any reform. Making that happen typically requires outside intervention, or a religious or quasi-religious movement from within. In Gotham’s case, I’d go for outside intervention from the hopefully not-corrupt Federal government. Or possibly Superman.

          • Bugmaster says:

            Ok, but realistically, what is the federal government and/or Superman going to do ? Superman absolutely can knock out each and every one of the corrupt officials in Gotham — at which point, he will end up knocking out the entire government staff. Now what ? Who is going to staff all the empty offices ?

            You could recruit people from the local population, but they’re all corrupt, as well. The median Gotham citizen doesn’t think, “dang, I sure wish my local police officer was honest”; he thinks, “dang, I sure wish I could become the local police officer, so that I could collect bribes like he does”. You could bring in an entirely new staff from the outside; but I doubt the people of Gotham are going to go for it, and Batman might even start getting punchy over this blatantly unconstitutional takeover.

          • cassander says:

            there are ways you can root out corrupt systems, but they are slow and require root and branch reform.

            You pick one area of gotham government, say, the meter maids. You fire everyone who works there, and then rebuild the department from scratch making it clear that corruption will not be tolerated in this one little corner of the world. A councilman comes by asking for a job for an in-law? Sorry, not in the meter maids, but maybe somewhere else. A campaign donor wants a sweetheart contract on their uniforms? Nope you can give him a leg up bidding on the Gotham PD uniforms, but not the meter maids. A ward leader wants a ticket fixed? A speeding ticket sure, but never a parking ticket.

            Basically you build up an isolated little island where corruption is not tolerated, then hope and pray that that culture is strong enough to survive when your tenure ends. And you get away with it by picking areas that aren’t a direct threat to powerful institutions that benefit from corruption. Only once you have lots of those islands, and people are starting to remember that low corruption society is possible and desirable do you move to take on the big players.

          • MrApophenia says:

            This is kind of what Gordon was trying to do with the police. There wasn’t much he could do about City Hall, but he can slowly but surely replace everyone in positions of power in Gotham PD.

            One thing he actually canonically did do is start bringing in cops from outside Gotham to run things, too. Major Crimes is run by Maggie Sawyer from the John Byrne Superman run these days, and a lot of the cops on the force were Metropolis imports – that’s how Crispus Allen got on the Gotham PD, too.

          • Bugmaster says:

            @cassander:
            Firstly, any strategy that relies on hope and prayer is probably going to fail. Secondly, mafiosi are not known for their tolerance of resistance. If you tell them “sorry, your nephew/wise guy/obvious plant can’t be a meter maid, try somewhere else”, then their answer isn’t, “curses, foiled again”. Rather, their answer is, “those are some nice kneecaps you’ve got there. Shame if something happened to them”. And they can make it happen with impunity, too — because they have the support of literally every institution, and a large proportion of the citizenry… Minus the meter maids.

          • cassander says:

            @Bugmaster says:

            Firstly, any strategy that relies on hope and prayer is probably going to fail.

            Agreed. It’s the best strategy you have, but it’s not a good strategy.

            Secondly, mafiosi are not known for their tolerance of resistance. If you tell them “sorry, your nephew/wise guy/obvious plant can’t be a meter maid, try somewhere else”, then their answer isn’t, “curses, foiled again”. Rather, their answer is, “those are some nice kneecaps you’ve got there. Shame if something happened to them”. And they can make it happen with impunity, too — because they have the support of literally every institution, and a large proportion of the citizenry… Minus the meter maids.

            While corrupting forces are powerful, they aren’t all powerful. A mayor/commissioner/what have you isn’t a puppet, at worst, he’s a favor broker. He will have some agency, and can impose some costs on people for taking certain actions, even if that cost is just doling out benefits to a different set of corrupt interests.

            On top of that, no one has an interest in corruption for its own sake. People want money, power, perks, etc. If you can provide those things from other sources, they’ll probably be fine, as long the institution in question isn’t one they have a vested interest in controlling, like the police.

            the combination of those two factors creates space for action that can be used to carve out corruption free areas. The worse the corruption the smaller the space, but some is better than none.

          • Bugmaster says:

            @cassander:

            On top of that, no one has an interest in corruption for its own sake.

            While that is true, corrupt people are interested in allowing corruption to continue; and allowing an oasis of honesty to flourish is not conducive to that. This is why mafiosi tend to respond to any sustained disobedience with disproportionate force. They are not going to break your kneecaps just to make sure you personally comply; they are going to do so to send a message to everyone around you: “see, this is what happens when you do not comply”.

            Due to this — as well as due to rational self-interest — overwhelming force is very rarely necessary. Non-compliant people who refuse to pay protection money or grant little favors are usually told, “look, if you snub me on this one thing, I won’t be your friend anymore, so I won’t be able to help out your aunt Mildred when she gets sick again”. This is followed by, “hey, remember Johnny McPizzaChef ? Remember how his shop mysteriously burned down with his entire family in it a couple of years a go ? Well, it’s a shame his friends weren’t around to prevent that from happening.” Very few people can stand up to that sort of pressure.

          • Civilis says:

            In order to get corruption out of a city like Gotham, you’d need an Untouchable law enforcement force that is immune to corruption or threats.

            The Justice League is immune to corruption and threats, but isn’t a law enforcement body; it can’t prosecute crimes and arrest people. Any superhero is going to be able to stop the mob from shaking down a target once, but few superheroes are powerful enough to stop every shakedown attempt. If law enforcement is corrupt and there’s no real penalty for a failed shakedown, the mob will eventually succeed enough that corruption is the default state.

            Commissioner Gordon’s Gotham PD is a law enforcement body, but, even if he can effectively purge the corruption from the ranks, the police are still answerable to the city government.

            Fortunately, some versions of the DCU suggest that there’s an alternative law enforcement body that might serve, and there’s real world historical precedent to back it up. Most petty criminals in the DCU are terrified of the supervillains. The supervillains in the DCU are terrified of the Joker. Who’s the Joker scared of? The IRS.

            This suggests that to reduce crime in Gotham, Batman should devote more of his time to study forensic accounting.

          • Tarpitz says:

            Someone already made that film.

          • acymetric says:

            The Justice League is immune to corruption and threats

            Citation needed?

    • Squirrel of Doom says:

      > Many people have made the argument that this is not efficient. Batman is a billionaire; there must be a better way to prevent crime than what he’s doing.

      To me, this is obviously a Division Of Labor argument.

      As a billionaire I, and my organization, can handle financial, logistic, political, technical and scientific support for crime fighting.

      It is not rational that I also personally engage in the combat operations part of the enterprise. For one thing, if I’m killed, the whole thing os over, and crime has won. You can thus also see this as a Single Point of Failure argument.

      But even if I somehow always survive, this detracts from my ability to work on all other aspects. The logical choice is to hire a small cadre of highly skilled and motivated crime fighters who can use my tech, costumes and support, and focus 100% on The Fight, while I and my billions focus 100% on support. As they die off, they are replaced, and we have a sustainable strategy for long term crime fighting!

      • Bugmaster says:

        For one thing, if I’m killed, the whole thing os over, and crime has won.

        Well, no, this isn’t generally true. Most corporations have a very clear line of succession in case their CEO is killed or simply retires; a competent CEO would ensure that his potential successors are likewise competent (although, perhaps, not excessively ambitious). Thus, if he does get killed, the company operations should recover after the initial dip in performance.

      • idontknow131647093 says:

        This also assumes that the crime is caused by something that a billionaire can fix in some way other than by dressing as a bat and scaring people. Maybe Gotham (like most fictional cities) is truly a cesspool populated by the descendants of Sodom and Gomorrah. Not that I find this model 100% true, but it actually fits what I see as data showing things like charity (public or private) never working very well at all. Also law enforcement strategies aren’t all that different from NYC to Chicago, but Chicago has much worse crime/capita.

    • nameless1 says:

      Finance a political change for giving mandatory testosterone blockers for everybody who committed a violent crime. While testosterone is not the only or primary reason of it, it is a necessary link in the chain and fairly easily blocked element. Without it, people become risk-avoidant and may fantasize about assaulting someone but will not carry it out. I am not demonizing testosterone, I think it plays only a role in crime by encouraging risk-taking, which is a good thing for e.g. entrepreneurs. But it is the simplest way to chemically make violent offenders meek.

      • albatross11 says:

        Use my lobbying muscle to find ways to eliminate public housing and poverty programs within Gotham. Justify this with plausible-sounding rhetoric about “a hand up, not a hand out” and promoting individual responsibility. Work on gentrifying all the bad neighborhoods so that poor people simply can’t afford to live there anymore. Close down the existing housing projects while making sure the poor have nowhere else to go within the city.
        Lobby, give campaign contributions, etc., but don’t violate the law and don’t do anything violent, to avoid bat-punches.

        At the same time, lobby extensively for generous public housing and poverty programs in other cities. (“Metropolis: The City That Cares.” With a food pantry on every street corner, subsidized housing for anyone who needs it, and generous social welfare programs for single mothers, the unemployed, long-term disabled, etc.). (It’s safer to do this crap in Metropolis, because while Superman is more powerful than Batman, he’s not as good an investigator. Also, Superman can produce essentially infinite resources, so he can fund Sweden-on-Lake-Michigan much more easily than even a billionaire like Bruce Wayne.)

        Run extensive public ad campaigns (perhaps from a shill opposition lobbying group) noting that while Gotham heartlessly leaves its impoverished people in want, just a short Greyhound bus ride away in Metropolis, no beggar goes hungry. As a publicity stunt, have your shill opposition group start offering busfare and relocation cash to the poor and improvident of Gotham, to move them to a more benevolent regime where they will be happier and better off. Spend a lot of resources selling the PR image of Gotham as a heartless enclave of wealth and hard-heartedness, and Metropolis as a soft-hearted land of milk and honey.

        Encourage Batman to keep punching criminals, both supervillains and normal villains. Back him up with police who come down so hard on small-time criminals that most such folks decide to move somewhere a little more relaxed about the whole knocking your teeth out with a police baton thing.

        Perhaps also start subtly hinting in your media influence operations about the idea that when you’re a supervillain looking to make your name, Gotham is really the minor-leagues. I mean, Batman’s cool and all, but he’s just not in the same league (*ahem*) as Superman.

        • ing says:

          If you deport all your poor people, who’s going to be the janitor and the convenience store clerk and the taxi driver and etc?

          You could answer “well, we’ll have middle-class people do that”, but those jobs don’t pay well enough to stay middle-class.

          I suppose if you get a bad enough labor shortage, low-end wages will rise. But the price of goods will also rise, and it’s not super clear what that does to the economy overall, but I’d want to make sure someone had seriously thought about it before implementing.

          • albatross11 says:

            Plenty of places manage to be very nice, and manage to fill their required janitor slots, without having any subsidized housing nearby.

      • Bugmaster says:

        Mandatory drugs in the water supply are exactly the kind of thing that Batman’s fists itch for.

    • honoredb says:

      In our universe, I think endowing a non-profit private prison could be an effective intervention–most of the incentives causing U.S. prisons to be terrible come from either politics or profit motives, so a prison that was semi-insulated from both would be free to be non-terrible in ways that reduce recidivism.

      In Gotham, you presumably have the same dynamics, but part of the terrible is being a cardboard prison for supervillains. A well-funded non-profit private prison would be free to provide education and employment opportunities for the regular criminals, while doing ridiculously expensive things like exiling supervillains to a Martian colony via a magic portal or something. I’d hire Batman as a Batconsultant for the details of that part, both to reduce the escape rate and to ensure he has buy-in so he doesn’t end up punching me.

      • albatross11 says:

        I’m pretty skeptical about this. I mean, we have for-profit and government run prisons, and as best I can tell, they tend to have basically the same set of pathologies. This suggests to me that the profit motive isn’t the source of the pathologies.

        My not-that-informed guess is that a lot of what’s wrong with our prisons is upstream of the actual prison system, and so fixing the prisons can only do so much to make things better. (I do think the prison system should be fixed to be more humane and less focused on extracting pennies from inmates’ families, but I also think that will only go so far.)

        Here’s my guess: I think the fundamental problem here is that most people who get locked up in prison are awful people who do awful things, and whose very presence imposes costs on everyone around them. By the time you get to prison, you’ve usually gone really far down the path of nasty awful behavior that hurts other people–some people end up in prison after doing one bad thing, a few didn’t even do the thing they’re in prison for, but most of the folks in prison have been committing crimes and victimizing people around them for many years.

        Also, we lock up a fair number of people who are seriously mentally ill. We have decided that we’re too humanitarian and libertarian to forcibly commit crazy people to asylums much, so instead we lock them in prison or let them freeze to death in a cardboard box. It sure seems unlikely that the right way to deal with criminals and with seriously crazy people is the same, so probably the prisons aren’t handling this well. Another big group who tends to do some jail time (though usually not serious time in state or federal prison unless they were dealing) are drug addicts. It’s hard to see how a couple years in prison for forging a prescription is actually going to help you kick your opioid habit, but that’s what we’ve got. My impression is that nobody has a great way of treating drug addiction, especially not when the people you’re treating are there because a judge ordered them to be there. Prisons surely aren’t going to handle this too well.

        • Walter says:

          I remember reading an expose on a prison that was particularly bad, and the embedded reporter was basically like “the problem with this prison is that there aren’t enough guards, and we are barely trained and paid peanuts”.

          Like, profit margin or getting reelected (depending on if this is private or public) both means that taking one more prisoner is always a good idea. You are either getting paid for the new body, or have just saved money you can spend on something your constituents care about more than ‘coddling felons’. Underpaying guards is same thing, ditto for not hiring lots of them.

          Based off this one article, I’d expect that if we doubled the number of guards in jail, and paid them more like secretary wages instead of grocery bagger wages the prisoners might have a better time.

          • cassander says:

            California pays its guards a fortune and has a lot of them, and they have the worst prisons in the country. Paying more only gets you a better class of employee if you have sensible procedures in place for hiring and promoting good people. Most civil service systems, almost by definition, do not have that, and giving them more money won’t improve the quality.

            On top of that, according to the bls, the average prison guard makes 42k a year, but that doesn’t count benefits, so we’re already at more like secretaries and less like baggers.

          • albatross11 says:

            Prison guards in the worst prisons have genuinely horrible working conditions. If the pay is good but not great, I suspect you’re getting people who either had no other options, plan to make extra money smuggling in drugs, or who are taking extra pay in the form of opportunities for brutality and abuse.

          • cassander says:

            @albatross11 says:

            The guards at maximum security facilities, sure, but that’s not most prisons or most guards.

        • People interested in prisons might want to read David Skarbek’s The Social Order of the Underworld, which is about prison gangs. He never quite says so, but he clearly thinks they are on net a good thing, responsible for the sharp drop in prison deaths over the period when they replaced the previous informal norm system.

      • Furslid says:

        I don’t think non-profit fixes the prison problem. Profit or non-profit, a lot of the same incentives are still in place.

        There is a private scheme that I think could help a lot. Charter prisons. Like charter schools. A prison has certain requirements it has to meet. Prisoners can apply to serve their time in a private prison that meets their security level or serve their time in the regular prisons. These prisons can try different forms of rehabilitation.

        The prison model that I’d try would run as follows for minimum security. Start with a business that requires a lot of unskilled labor. Set up tolerable housing on site. (Think college dorms). When a prisoner who works well is released, let them continue in the same job or transfer to a similar job at a reasonable wage. This would make recidivism less likely.

    • benjdenny says:

      The question of how to improve Gotham has a necessary sub-question we have to resolve before we can fix anything, and it’s this: Why aren’t we Metropolis?

      Metropolis should, theoretically, have the same problems that Gotham has; It has endless villains that only one resident can defeat who are otherwise infinitely more powerful than the populace(and, because they have to be able to go toe-to-toe with Superman, harder to keep in prison and can kill more civvies faster). Yet, Metropolis is doing far better in terms of non-super street crime, and is a much brighter, happier, hopeful place.

      So then we look at the golden age of Gotham, a period when it was moving in a Metropolis-like direction, and we find there was a family of super-rich people guiding the area and doing good with money, and that this is the only thing that ever made Gotham stand a chance of improving. Metropolis, then, might logically have a person who serves the same function but is much, much better at it. When we check, we find out that this person is named Lex Luthor.

      After I figure out that Lex Luthor is probably the best hope for Gotham, all I need is tax credits and some sort of credible reason why Batman isn’t worse for Luthor than Superman(no laser eyes, actually in danger from anybody with better gadgets). If he moves in, we all do better.

      • The Nybbler says:

        As long as you’re up for regular epic yet never decisive episodes of Batman punching Luthor’s minions (err, employees) in the face, reaching Luthor, and finding that Luthor has some Very Good Reason Batman shouldn’t punch him in the face, OK.

        • benjdenny says:

          I’m sort of relying on:

          A. It’s a lot more trouble for Batman to even get to Luthor than Superman

          B. Batman is pretty pragmatic; the super-villains that Luthor organizes will be slightly more effective, but I’m guessing he’d trade that for Luthor’s apparent dislike of the unorganized nature of street-level crime.

          Basically I’m betting that as long as Luthor could get a foothold, Batman would probably trade things being harder on himself for things being easier on Joe Average.

          • John Schilling says:

            It’s a lot more trouble for Batman to even get to Luthor than Superman

            Yeah, but Bruce Wayne can just buy enough LexCorp stock to walk into any board meeting. Why are these two trying to punch each other out again?

          • The Nybbler says:

            Lex will certainly hold a controlling interest in LexCorp (or it may not be publicly traded). Or there will be some sort of poison pill that leaves any raider holding nothing but the bag, with Lex free to rebuild LexCorp trademarks and all. He’s evil, not stupid.

            Which doesn’t mean Wayne won’t be on the board; both men seem like the type to believe the adage of keeping your enemies closer.

          • ing says:

            “Pragmatic” is not on the list of adjectives I would have used for Batman.

            This is the dude who refuses to ever kill, or ever use guns, not even when fighting The Joker for the fifteenth time because he keeps escaping prison again and again and again.

      • Bugmaster says:

        There are two reasons for why Gotham isn’t Metropolis.

        First, Metropolis has Superman. Superman has a nigh-instantaneous response time to any act of crime. He has a nearly perfect arrest record, and he absolutely cannot be stopped by anyone who isn’t Lex Luthor (or possibly some sort of an alien from Apokolips). For all practical purposes, if you commit a crime in Metropolis for the purposes of self-enrichment, your chances of being caught are 100% +- epsilon, so crime literally does not pay.

        Secondly, Metropolis has Lex Luthor. Luthor practically owns the entire city, including both the legal and the criminal social strata. Luthor is deeply invested in his assets, and thus he enforces peace with an iron fist — because he won’t get a decent RoI if Metropolis becomes a Gotham-style violent hellhole. This means that Luthor ruthlessly suppresses maniacal super-powered terrorists before they have a chance to act (and be caught by Superman after the fact).

        • toastengineer says:

          Now wait a second, that’s not how Superman works. Supes is good, but an entire city has a whole lot of crime going on simultaneously, and he does take significant amounts of time off. At best crimimals will just find a way to only do crimes with Superman is off-duty.

          Superman going around stopping street crime personally is probably a pretty effective psychological warfare tool, but he cannot be the only petty crime crime prevention apparatus Metropolis needs or has.

          • Le Maistre Chat says:

            One of the better cartoons on the air when I was in elementary school was Superman: The Animated Series, which did an episode on the psychology of street criminals and Metropolis PD under the regime of “random Superman intervention.”

            This being the superhero genre, it was all in the context of Apokalips giving patronage to gangs in the form of superscience weapons to demoralize the citizenry, who were happy with the Superman status quo.

      • Vorkon says:

        Incidentally, when gbdub mentioned Vetinari’s approach above, and said:

        Vetinari also has demonstrated the ability to corral extremely talented, but in some ways very naive / dumb crime fighters into effectively serving his ends without realizing they are doing so. That’s key for managing Batman.

        I couldn’t help but think, “Hmm, I wonder if that’s what Lex Luthor has been up to over in Metropolis?”

    • Le Maistre Chat says:

      Tangential to my other Batman post, I’d like to talk about one of the best Batman stories I’m aware of: “To Kill a Legend” (1981).
      The high concept is that, if there’s a world where Thomas and Martha Wayne were killed a few decades before 1981, and a parallel world where they were killed in the 1920s, why not show a world where they’re just now about to be killed? So the Phantom Stranger, a Neutral Angel who goes around doing good deeds as penance, offers Batman and Robin a portal to this parallel Gotham. Robin doesn’t know if saving Bruce’s parents is the right thing to do, a qualm Batman brushes off unthinkingly. For him there’s no weighing consequences: he is just going to save the parallels of his parents. Fighting crime is just what he does.
      Dick Grayson’s utilitarian concerns lead him to a public library, where we get this bombshell.
      So this is an Earth whose history diverged from ours around the time of Gilgamesh: kings have never been the subject of heroic mythology, which led to people not having mythic role models. Just to twist the knife, the star Krypton orbits doesn’t exist. The utilitarian calculus is stacked so this world needs Batman more than any other.
      But… SPOILER ALERT:

      .
      .
      .
      .
      .
      .

      Rather than wallowing in tragedy, the story lets Batman knock the gun out of the killer’s hand and beat the daylights out of him while yelling “Not this time, you hear me! This time you’re not going to win!”
      And guess what? Saving two lives and saving a little boy from “being angry all his life” could not possibly be the wrong decision, as the epilogue reveals. Because “our” world has epics, because it has the tragedy of the Batman, this Bruce Wayne will be inspired to become his world’s first legendary hero, by the intervention of a weird masked figure in the night the cause of whose presence he can’t begin to comprehend. His life will be an incomprehensible mystery, accepted in gratitude.
      … it makes me tear up.

      • ing says:

        It’s weird reading this, because I kept thinking you were going to say: “…and without superheroes and supervillains the world was way better off! But then Batman ruined it all by superheroing, and other people started doing it and it was all downhill from there.”

        But the twist never happened. : )

    • J Mann says:

      Well, if I were Batman, and one of the best two or three non-superpowered people in the world at every conceivable skill, then …

      Step 1: As Bruce Wayne, endow a foundation for the re-education of thugs, gunsels, mooks, and henchmen and women of all sorts.

      Step 2: As Matches Malone, use my nigh-preternatural skills at cold reads, psychoanalysis, and motivational speaking to convince thugs to enter said program. If I can counsel between 2 and 4 thugs a day, I should be able to drain the swamp pretty well. Before you know, they’ll all be churchgoing community organizers.

      Step 3: Without a ready market in thugs, the remaining exotic criminals should be easy to spot. Reform who you can, punch those who remain.

      • Le Maistre Chat says:

        Step 3: Without a ready market in thugs, the remaining exotic criminals should be easy to spot. Reform who you can, punch those who remain.

        The problem with this step is that the post-Frank Miller consensus is that good storytelling = wallowing in pop psychology porn. Anyone Batman punches gets sent to Arkham Asylum, where reform is impossible and the doctors themselves may be psychotic (the Scarecrow, Harley Quinn). And just to make it Darker and Edgier, the city was built atop the grave where a Nyarlethotep-worshiping warlock was buried alive. Heck, I wouldn’t be surprised if Arkham Asylum was built atop Wilbur Whateley’s farm.

      • Nabil ad Dajjal says:

        This only works to the extent that access to thugs is a limiting factor on crime bosses and supervillains.

        If the next issue of Batman had the Joker leading a small army of clown-bots or mutant hyena-men instead of career criminals in makeup, readers might dislike it for being too Silver Age and goofy but it wouldn’t be unprecedented. It’s not clear at all in-universe that the criminal population represents a limit on supervillains’ ability to recruit henchmen.

      • ing says:

        I agree, but that leads to a less fun thought experiment.

        (“I blackmail Frank Miller into writing a canonical Batman story. In this story, Batman punches out the evil warlock guy from upthread, ending all crime in Gotham City!
        The story has an epilogue set five hundred years later in which Batman’s descendents look back on the Warlock-Punching event and reflect on how great it is that there wasn’t ever any crime after that.”)

        (“I write five million procedurally generated Batman stories. In these stories, Batman goes out and checks the city for crime, but there’s no crime today because he’s ended it all. My stories make up more than 99% of all Batman stories, therefore crime has dropped by 99%.”)

    • gbdub says:

      Batman must be stopped, or made to be less Batmanny.

      Basically, Batman is counterproductive, especially if you know that he’s Bruce Wayne and know he could be sharing his awesome toys with the Gotham PD instead of just being a cosplayer that beats up the mentally ill. Or just donating his wealth to endow a humane prison/asylum system with competent, non-sadistic management.

      Bats needs a lesson on traumatic brain injury – Gotham has a really high rate of criminal insanity, and I can’t imagine his approach, which seems deliberately designed to induce serious PTSD and CTE in every petty crook, is helping.

      The Gotham prison/asylum system is terrible. Not only is it apparently super easy to escape, but Arkham is a hellhole that makes the gulag look like a luxury resort while being run by corrupt monsters who are creating more criminals than they are curing. In that light, Batman’s no-kill policy is insane: he’s giving people mental illness and then shipping them off to be tortured. A bullet to the brain would be a mercy, not to mention eliminating the crimes committed by criminal masterminds after they’ve escaped for the fifth time.

      • Bugmaster says:

        Sharing Batman’s toys with the Gotham PD would be counterproductive, since the entire police force is utterly corrupt, save for possibly two or three individuals.

      • The Nybbler says:

        Batman must be stopped

        Uh, Mister Batman, I think we have a supervillain minion here.

        • gbdub says:

          Far from it. I suspect Batman is intentionally practicing “catch and release” with criminals because he loves punching so much. Which would be fine if it weren’t for all the collateral damage. Maybe we need to create a Bruce Wayne Memorial Villain Game Preserve so he can punch all he wants while Gordon focuses on effective policing.

    • pontifex says:

      This question made me giggle because realistically, almost everyone on SSC is a Batman supervillain. We would get one issue, tops, before getting punched.

      I mean, come on. “The Nybbler” wants to build a giant artifical moon. “Bugmaster” opines that “mass genociode” is the only cure for corruption. “Conrad Honcho” proposes that “perhaps Batman himself should take over organized crime in Gotham.”

      Bumbling fools! None of them understand the true solution, an orbial doom laser governed by an anarcho-syndicalist government. Now, hear me out on this one…

      • Bugmaster says:

        Heh, that reminds me of a dialogue snippet from some old webcomic (I forgot which one):

        Villain: Finally, Hero, I have you at my mercy ! Tremble before my foolproof plan !
        Hero: I’m not really worried. Your plans always fail due to being needlessly complicated, and also orbital.
        Villain: Silence ! As soon as my new satellite moves into position…
        Hero: See ? Orbital. I knew it.

      • The Nybbler says:

        No big artificial moon for me; that was Vorkon. I’m just going for ordinary state villainy.

  2. Philipp Bogatyrev says:

    I thought commenting “first!” was an exclusively Russian thing, lol.

    • doubleunplussed says:

      Back on early 2000’s fark.com it was annoying enough that they made a filter turning “first post” into “boobies”. So it’s at least that old and in at least that demographic.

    • RavenclawPrefect says:

      I’ve been a little surprised to see it in the SSC comments section too, since I would assume that at least the subpopulation which follows this blog with enough dedication to show up within minutes of a post submission is of a little higher caliber. But the ban is definitely welcome; the fact that such a fantastic comment space has managed to exist in one public place online for 5 years is something of a miracle, and anything that keeps it from deteriorating into the internet’s entropy of lowest common denominator sludge is great.

      Semi-relatedly, would it be worth imposing any restrictions on generic upvote comment type things, or at least constraining them to one thread? There’s some value in getting a feel for the community’s response to a post (and I imagine Scott doesn’t mind the positive feedback), but the fifth top-level comment that says nothing more than “great post” begins to drag on a bit. (Though I don’t think I have a great read on most semi-famous content creators’ preferences for feedback; I usually err on the side of “anyone I choose to send generic ‘I think your work is fantastic’ feedback to will for anthropic reasons be disproportionately likely to get a lot of that feedback, and thus probably has to filter it out to find important things” so I stay quiet unless I know with high confidence that a person isn’t getting much attention and I could be reasonably assured of not being an annoyance.)

      • christianschwalbach says:

        I wouldnt mind a restriction of upvote comments myself. It gums up the threads that I am trying to follow, unfortunately. As for the “First” commenting trend on this Blog, I find it irritating, as I do on other blogs, and would like to see it reduced. That being said, my pet theory on why people do it in the SSC context is something along the lines of a general excitement at finding a blog/discussion arena that fits , to a large extent, their ideas, viewpoints, or perhaps just a desire for reasonable discourse that seems overly absent on a lot of online resources. Excited people are more likely to post pointless but exuberant “First” posts, in my opinion. Of course, there could be the factor that those folks post likewise on all internet threads, demonstrating a childlike and immature nature……of course, thats my ever so unhumble conjecture.

    • arlie says:

      I think it’s kind of fun, actually, and am mildly sorry to see this policy change against it.

  3. Taymon A. Beal says:

    The monthly Boston SSC meetup needs a new organizer. If interested, please email or message me (or reply to this thread; I’ll try to check for replies but can’t promise I’ll quickly see them).

  4. deluks917 says:

    What norms should rationalsits have about ‘really bad arguments’. Say you are in a rationalsit space and someone really blatantly misunderstands your argument. A common example is that they aggresively point out something you explicitly considered or already conceded.

    For example say you are arguing that we support animal wellfare regulations. You explicitly admit that this will increase the cost of meat and that this cost will be felt the poor and people with medical conditions that require them to eat meat. However you argue that the net utilitarian impact is positive. Someone comments ‘that policy is ableist and hurts the poor!’ as if its a knock down arugment. How are you supposed to respond?

    Another example is that someone is being negative in a really uninformed way. A common example is that somene argues its impossible in theory to build a ‘general intelligence’. Hence we should stop worrying about AI risk. Regardless of the merits of worring about AI risk it is obviously possible to build a general intelligence. Humans exist! Even if someone refutes these dumb counteraguments they tend to drag down the level of discourse and throw cold water on useful conversations.

    How should rationalist spaces handle this sort of thing?

    • RavenclawPrefect says:

      Is the hypothetical offender here someone who just happens to have stumbled into a rationalist space, or a person who’s reasonably steeped in the norms and ideas of the greater rationalist diaspora but being uncharacteristically bad at arguments? In the latter case it seems like pointing out the fact that their argument sucks and doesn’t contribute to discussion would suffice, since it’s understood that there’s a norm against such things.

      The former case seems trickier, but I feel like a policy among bystanders/fellow commenters (in cases where “rationalist space” entails multiple rationalist-adjacent people together in some environment, IRL or otherwise) of joining in to point out that the counterargument is dumb helps reinforce a norm of “we’re better than that here” and makes it clear that the person is detracting from the group as a whole rather than just having an isolated debate with one other person.

    • However you argue that the net utilitarian impact is positive. Someone comments ‘that policy is ableist and hurts the poor!’ as if its a knock down arugment. How are you supposed to respond?

      Do you need to respond? Through the process of debate you have come to a point in which all facts are agreed upon, and the rational and empirical debate is over, and what remains is only the differing individual reaction to the facts. There’s nothing left to debate, as everything that needs to be exposed and examined as true or false has been. All that is left over is framing.

      You can frame it as “the net utilitarian impact is positive”, or “that policy is ableist and hurts the poor”. Both accept the same set of facts, which is that there is a trade-off going on. The difference lies in whether you can emotionally accept trading off poor people’s wallets for animal welfare. You either can or you can’t. There’s nothing to debate. The debate should be over. Both sides should walk away, and “agree to disagree” (and begin sharpening their knives).

      I guess rationalist spaces could have a norm about not getting overly-emotional as this is interpreted as a threat display and tends to lead to escalating outburts and grandstanding.

      Another example is that someone is being negative in a really uninformed way. A common example is that somene argues its impossible in theory to build a ‘general intelligence’. Hence we should stop worrying about AI risk. Regardless of the merits of worring about AI risk it is obviously possible to build a general intelligence. Humans exist! Even if someone refutes these dumb counteraguments they tend to drag down the level of discourse and throw cold water on useful conversations.

      This example can still be debated, as their argument that “general intelligence is impossible” has flaws, which you’ve pointed out yourself. The conversation can still continue if they accept this and make their next point “Okay, sure, but I don’t think general intelligence will be achieved any time soon”. Alternatively, you could get into whether “general intelligence” is a coherent concept.

    • ing2 says:

      Eliezer’s post, Well-Kept Gardens Die By Pacifism, argues that most communities aren’t aggressive enough about removing bad actors. The scenario you describe does not seem to require a ban for a single incident, but it’s something to keep in mind.

      Scott’s post, Against Interminable Arguments, touches on the problem of people who are negative in an uninformed way. What I take from the post is that You Really Should Have An FAQ — either a community FAQ or a personal FAQ. You can then reply to these sorts of messages with a request to check the FAQ.

      • baconbits9 says:

        I think Eliezer’s post is fundamentally off. What attracts you to a community at first is different from what attracts you 4 years in, and it is the older and more established members who are likely keeping out new blood, not the lone ‘fool’ (who is often not more foolish than they elder statesmen once were). Or it is the interaction between the fool and the elders.

        • Nornagest says:

          Eliezer’s post strikes me as a little self-serving in retrospect, but there’s a grain of truth in it. I’ve left communities because they started getting flooded with low-quality content. I’ve also left communities because the administration got too heavy-handed for my taste. But I’ve most often left communities because all the interesting people stopped contributing. Which implies, insofar as I’m representative*, that the most important consideration is to keep the interesting people happy.

          (*) Arrested Development voiceover: “He wasn’t.”

          • baconbits9 says:

            It is inevitable that some people will leave a community, and some of those are going to be the interesting people that made it worthwhile. You can do somethings to reduce or increase the speed at which they leave but you can only sustain with a constant injection of new interesting people. Nothing else will work, nothing will stop interesting people from dying, having major life changes or having a shift in their responsibilities at some point in the future.

          • Nornagest says:

            There’s a lot of overlap between what makes a community attractive to new interesting people and what helps a community retain existing ones.

    • Deiseach says:

      Someone comments ‘that policy is ableist and hurts the poor!’ as if its a knock down arugment. How are you supposed to respond?

      It may be that the person argues that because they haven’t understood your point. Or it may be that they wish to point out that you sound like you are saying “yeah I know and I don’t care because I prefer moo-cows and cluck-clucks to people” and are you really sure you want to sound like that?

      Responses can be along the lines of:

      (1) Yes I do want to sound like that because I do indeed prefer moo-cows and cluck-clucks to people
      (2) Try and do the “yuuuuge utility for mass numbers!” thing, where the reduction in suffering for 100 cows outweighs any loss of utility for 1 human or the likes
      (3) Say “thank you for pointing that out, my solution to that is – ” and outline suggestions to ameliorate the bad effects on the poor and those who need to eat meat
      (4) If you don’t have any suggestions to ameliorate the bad effects, consider if you really are saying “Yeah I know and I don’t care because I prefer moo-cows and cluck-clucks to people”

    • Plumber says:

      @deluks917

      “What norms should rationalsits have about ‘really bad arguments’”

      Please explain why my argument is bad in language that I can understand so I may learn.

      Thanks.

      • Robert Jones says:

        This is a bad norm, because, in the example stated, I have already expressed my argument as clearly as I can in the OP. I have no way of knowing what language you can understand. It is possible that if I rephrase I might have more success, but I’m just guessing. In a forum like this, somebody else will usually jump in with a different approach, and it’s probably better if I leave it to them. Of course if I was teaching a seminar, I would have to try expressing the same thing in a different way, but that’s a different situation and even then it only seems to work about 10% of the time.

        Essentially, your proposed norm places too much of the burden on the person making the comment and not enough burden on the people reading the comment, bearing in mind that we’re all participating broadly as peers.

        • Deiseach says:

          This is a bad norm, because, in the example stated, I have already expressed my argument as clearly as I can in the OP.

          Yes, it could be that your interlocutor is an idiot who didn’t understand what you were saying, or didn’t pay attention and only skimmed over it and jumped in with the first objection that occurred to them.

          Or maybe you did not explain your point as clearly as you thought you did, or it’s buried down in the underbrush of the long argument, or your counter-argument isn’t as convincing as you think it is. Why should I care more about the utility of a cow over the utility of a human? You’re asking me to accept or give in on a first principle upon which the rest of your argument depends (that animals have moral worth, or that we can stack up utility versus disutility, or that it scales by numbers so the suffering of 1,000,000 cows means that 100 humans who get sick if they try to live on a mostly-vegetarian diet just have to suck it up and deal with the ulcerative colitis) and maybe I don’t want to give in on that point.

          I think you have to, as the saying goes, “let every dog have one bite”. Your person who comes back with the slam-dunk “this hurts the poor!” may be an idiot or simply bad at presenting their counter-argument. If, after having repeated your point, they still continue to misunderstand you and keep repeating the same thing, then yeah, assume they are arguing in bad faith or are uninformed and either go to town on them or politely disengage as there is no further use in rehashing the same ground (depending on personal taste of how to react).

    • baconbits9 says:

      The norm (ideally) would be to pay little attention to people who make those arguments. Being a rationalist means that you to a large extent selecting the information you let through. You don’t (shouldn’t) cite flawed studies, you shouldn’t use rhetorical tricks to win an argument that you couldn’t win on facts, and you should try to pay as little attention to someone who is untrustworthy as a source of information as you can.

    • Nancy Lebovitz says:

      Sidetrack: Is it possible that people don’t have fully general intelligence?

      • ordogaud says:

        Depends on how you define general intelligence, at this point I don’t think anyone has a really rigorous definition on that. But in everyday use I’d say most people just use it as a reference to human or human-like levels of intelligence. From the context of the OPs argument that might be something they should confer with their interlocutor so they aren’t just arguing past each other with different definitions of general intelligence.

        • Nancy Lebovitz says:

          I was playing with the notion that there are aspects of reality that we just aren’t seeing and maybe can’t see.

          Could there be other mental tools as powerful as mathematics?

          • albatross11 says:

            That seems inevitable. As an example, I think game theory only goes back to the 20th century, and that’s a pretty powerful set of mental tools that nobody had until then.

      • Viliam says:

        People have limited memory. For example, I cannot imagine multiplying two five-digit numbers without using pen and paper. Yet, a simple machine can do that.

        I imagine that there are statements that can be understood only by spending the same amount of working memory at some step, so I would be unable to understand them, too. I imagine that some of those statements cannot be simplified. And again, a machine — or just a human with extra large working memory — might be able to understand them.

        I don’t believe in aspects of reality we couldn’t see in principle. We can’t see elementary particles directly, and we have particle physics anyway. (Okay, I can imagine a particle that doesn’t interact with the other particles; or a particle that is so rare in universe that we will never capture it in an experiment. But that isn’t inherently a human limitation; a different species would have the same problem.)

        But I can imagine irreducible problems that simply have too many moving parts. Those could be solved by a different species, or by a computer. Well, what can be done by a computer, can also in principle be done by a human following the script… but I suppose that the human would be unable to understand what the script actually does; they would understand individual steps, but not the large picture; and they would be only able to follow a script generated by the computer, not create their own script.

    • Robert Jones says:

      There seem to be two related questions here: (i) how should the OP respond and (ii) how should the community deal with it?

      The answer to first question is simple: you don’t respond. You have to trust that other people will be able to read your argument and the response and see that the response adds nothing. There’s no need for a further response explaining that it adds nothing. It’s likely to be a waste of time trying to argue further with the responder, because having failed to engage meaningfully with the OP, they probably also won’t engage with any subsequent response. If for some reason you really want to convince them (e.g. because they’re someone you otherwise respect), it is probably best to respond privately, rather than clog up a public forum with debating the obvious.

      If people are just deeply uninformed, then we can sympathise with their position (since everyone is uninformed on some topics), but we can’t have a useful conversation. I think ignoring them is still fine, but it might also be helpful to direct them to some specific resources where they could learn more about a topic.

      As for how rationalist spaces should handle it, I’m not sure. If it’s just a few bad (but good faith) comments, it’s probably not a problem, but if somebody leaves bad comments all over the place, something should probably be done, but banning somebody for stupidity seems harsh. Maybe cap the number of comments per user per day?

    • Robert Jones says:

      I’m more troubled by how to respond to really bad OPs. Sometimes somebody who I respect generally posts a link to something which is just terrible. Usually this involves some cargo-cult mathematics, because a lot of people are either too lazy or too bad at maths to read the maths attentively. If the argument goes wrong at one point, you can respond saying, “This is wrong because of this error,” but sometimes there’s just a whole series of incomprehensible steps, which are often not even wrong. What’s been done is so far from what ought to have been done that it’s impossible to say the result is wrong, only that the analysis should not shift one’s priors. Explaining it step by step would be tedious and doesn’t progress anything.

      I am currently tending towards leaving a comment like, “This is nonsense” or “This is really bad”. The problem is that comments like that pattern-match to comments which fail to engage with the OP. They also usually fail to prevent people from arguing pointlessly over something which adds nothing and I’m worried that I come over as an arse.

      • chrisminor0008 says:

        Have you ever heard of an example of engaging with a math or physics crank that produced a productive outcome? (Outside of a teaching/tutoring relationship where you’re being compensated to deal with them…)

        • albatross11 says:

          There’s a difference between a crank and just someone who doesn’t understand why the thing they’re linking to or the argument they’re making is goofy.

          • Le Maistre Chat says:

            Like somebody who takes the square-cube law to heart so much they decide that T. rex proves Earth’s gravity was different in the past because it didn’t walk on four elephant legs?

    • LadyJane says:

      For example say you are arguing that we support animal wellfare regulations. You explicitly admit that this will increase the cost of meat and that this cost will be felt the poor and people with medical conditions that require them to eat meat. However you argue that the net utilitarian impact is positive. Someone comments ‘that policy is ableist and hurts the poor!’ as if its a knock down arugment. How are you supposed to respond?

      At this point, the argument has been completed, in the sense that you’ve come to a point where you fully understand each other and agree on the actual facts, but still disagree due to an intractable difference of values. So in a sense, you’re not supposed to respond, because the debate has already served its purpose. You’ve both already ‘won,’ in the sense that you’ve both managed to express your views and your underlying reasons for holding those views in a clear and concise way.

      Of course, every disagreement over values is fundamentally a disagreement about facts (and vice-versa!), much in the same way that every curve is comprised of angular points and every angle is comprised of curving points, if you look really closely. So you could always try to understand why he prioritizes the plight of poor humans over that of farm animals, or make him understand why you prioritize the reverse, or both. But since people’s reasons for holding fundamental values tend to be complex, deeply rooted, and opaque (even – perhaps especially – to that person), I wouldn’t expect that to be very productive unless you’re ready for a very long discussion.

    • Murphy says:

      Whatever norm would need to account for downplaying.

      It’s fairly easy to mention the weakest possible version of counterarguments and “address” them to attempt to immunize your argument against the stronger versions.

      Person 1: [mentions in paragraph 13 section 5 sub sub sub sub text 3 that maybe just possibly there might occasionally be cases where the tiny number of people who are profoundly disabled and in absolute poverty might just possibly, maybe possibly might be slightly negatively affected by the possible rise in meat prices however on net it probably barely matters]

      Person 2: ‘that policy is ableist and hurts the poor!’

      Person 1: I already specifically mentioned that! Why are you violating the norms! bad faith! bad faith!

      • cryptoshill says:

        @Murphy – This is why I think there should both be a norm against statements like “that policy is ableist and hurts the poor!” *and* a norm against treating the first type of comment as an irrelevant critique. My response would be with a question “Ok, I thought I had addressed that – is there something you’re concerned about that means that you *still* have that reaction to my proposal?”. The opponent will either
        (1) – Have a legitimate response that was shrouded behind (norm-violating) reactions.
        (2) – Admit that they hadn’t actually understood your argument completely.
        (3) – Point out that you were actually strawmanning *them* by making a post like this, and didn’t understand *their* argument
        (4) – Clearly reveal themselves as not worth engaging with.

    • AG says:

      “I/X person has already addressed this argument, see it at this location (whether it be a comment above or an external reference.”

      It appears that “Read the sequences” was also a common refrain. Far better than a “I’m not obligated to google for you scrub” response.

      Assume that they merely haven’t heard of the existing literature, and treat it as an opportunity for education.

      If they continue and it becomes more obvious bad faith is involved, then “I will not be continuing this conversation with you.” will suffice.

      • Nabil ad Dajjal says:

        “Read the sequences” was, if anything, a worse answer than just saying to get lost.

        I did actually read the sequences and in all of the times I saw “read the sequences” posted as a reply to a question, I have never seen a situation where the sequences actually answer the question. It’s a non-sequiter.

        And even if you did encounter a question where the answer was in the sequences, why not just link to that particular essay? The damn thing would be a doorstopper if you printed it out and beyond that it’s arranged haphazardly.

    • Frangible Waterbird says:

      I often feel like I lock down when someone states something that so obviously didn’t pay attention to what I said… (or that shows they are ignorant of something that seems so obvious to me!)
      I just don’t know where to begin, because I am thinking of how great their shame will be at having that pointed out.

      I like the idea of other people than the OP being responsible for “jumping in.”
      Best, I think, is if it’s said by a person who is sympathetic to the person who is currently looking ignorant / silly / like they didn’t pay attn. to the conversation.
      If someone can say, “yeah, I was thinking the same thing at first, but it was already mentioned where the OP said ‘this will increase the cost of meat and that this cost will be felt the poor and people with medical conditions, etc.’ ” they are identified with the person who’s getting the sorta-reprimand-for-not-paying-attention.
      And also, in this way, it allows for more opportunities of genuine connection if it’s a newcomer who pops in.
      If you’re sympathetic to either them or their view, you’re more likely to have an actual rapport-building conversation, (I think) even if it’s a side-quest. (though I probably undervalue staying focused & value digressions more than most here!)

      Also, I’ve noticed that sometimes when I have just discovered a really interesting online community – or a really interesting conversation – I am -SO- excited about it or eager to talk to specific people there… that I am just looking for any excuse to say anything about anything!
      Not sure if it’s helpful to model people like that sometimes, but… this results in way lousier listening / less attention to details than what my baseline listening skill is at.

  5. Sniffnoy says:

    So, math/econ/game-theory question. Thought about MathOverflow but I figured, hey, we’ve got econ people here and this problem isn’t totally formalized, so why not ask here first. 😛

    Informally stated, the problem is this: What’s the fair order for two players to conduct stage-striking in? In particular, is it given by the Thue-Morse sequence, as I would expect it to be?

    Stage-striking, for those not familiar, is the procedure used in competitive Super Smash Bros. (and probably other games but I’m less familiar with those) to choose which stage to play the first game of a match on. There’s a list of possible starting stages; players take turns (not necessarily in strict alternation — hence the question) striking stages from the list, and the last one left is the one that the game is played on. With a list of 5 stages, say, players strike in the order P1-P2-P2-P1. What’s the fair order for more? (For some appropriate notion of fairness — more on that in a bit.)

    Offhand, it seems to resemble other fair-sequencing problems; and there are theorems to the effect that a number of these have Thue-Morse as the answer (see this paper for something of a survey; note that going by the author’s website this paper isn’t done yet). And it seems to me that, heuristically speaking, the same reasoning should apply — if some order gives one player a slight advantage, then following it up with the same thing but the roles swapped should mitigate that advantage. So I would still expect it to be Thue-Morse.

    However, the problem can’t easily be transformed into one of the existing known fair-sequencing problems, as best I can tell. For instance, instead of getting the sum of what you choose, you get the one thing that wasn’t chosen — and importantly, it only matters which stages were chosen, not who chose which ones, which is quite different from the other problems.

    Indeed, formalizing the problem at all seems a little tricky. If we assume that both players agree on the value of each stage (here “value” is in terms of the advantage it gives to Player 1 over Player 2, since this is a zero-sum game), then the order they strike in is actually irrelevant, so long as they get the same number of strikes, since you’ll end up with the median stage being selected regardless. On the other hand, if we imagine each player has their own evaluation of each stage, and no “true” value, then it seems (unlike in the fair-division problems mentioned above) the problem bcomes meaningless.

    So I think in order to formalize this we have to suppose that each stage has a true value, but that the players might mis-estimate it. That is to say, we have to introduce randomness into the estimates — which as noted in the linked paper is also necessarily for formalizing other variants. So you could imagine that each one estimates the value of each stage as a normally-distributed random variable centered on the true value, for instance (all independent and with the same variance, say).

    There’s also the question of what’s meant by “fair”. The notions from fair-division problems don’t apply here. Rather I think we have to judge fairness by closeness to the median, as the median is what you get when both players agree. We could judge this either by, how high is the probability that you get the median, or alternative, how close is the expected value to the median.

    That said, the fact that we’re looking for the median, which relies purely on order, suggests that maybe we don’t want cardinal values assigned to the stages at all. Perhaps the stages’ values, as well as players’ estimates of such, are simply some total order; and to model error in estimation, we say that orders with more inversions relative to the true order are less likely. (Not sure if number of inversions is the correct notion of distance here, but it seems a decent one.) And then presumably we want to judge by probability of getting the median, although you could still get expected values if you translated things back into “cardinal” values 1, …, n.

    So, yeah — it’s kind of a mess. What is the best way to formalize this? Having done so is there a unique answer and is it Thue-Morse? Thought I’d ask and see if anyone has any insight beyond what I’ve already said.

    • Tomminn says:

      Using the classic mathematical technique of “transform this into a solved problem”, we want to turn this into a cake-cutting problem, where we can use the algorithm “I cut, you choose”.

      For a given set of stages, player A chooses the sequence of stage-strikes, say P1-P2-P1-P2-P2-P1 for a 7 stage game. Player B then chooses whether they want to be P1 or P2 in this sequence. Player A is now incentivized to make the sequence of stage-strikes as unbiased as possible, which may intricately depend on the information you know about your opponent.

      The point is this. In the case that you and your opponent have perfect information about one another, the sequence of choices is irrelevant. All that matters is that number of choices each player gets, since you know what your opponent is going to pick. The only extent to which sequences matter is the extent to which you have uncertainty about your opponents preferences. In which case choosing last is best, because you have new information to inform your choice.

      If you have partial information about your opponent, you might find their n choices completely predictable, but their last m choices completely unpredictable. In which case, you would like them to make their first n choices after you, and their last m choices before you. If m=n=1, then you would strongly prefer to be P1 in the sequence, P1-P2-P2-P1. In fact, it’s conceivable that your desire for information is so strong that the sequence P2-P2-P2-P1 is less biased.

      • Sniffnoy says:

        I mean, you could also “solve” any of the other fair-division problems mentioned above by adding a similar layer of indirection — and yet they still have an actual answer in the form of the Thue-Morse sequence. Another layer of indirection is not what I’m looking for here.

        Also, remember that any details of the stages available should ultimately be abstractable away into the overall advantage it gives to player 1 over player 2 (in terms of probability of victory).

        • Tomminn says:

          Cool, well I think I’ve said all I want to say about this.

          In my mind, by the reasoning I gave in the parent I’m pretty convinced that the definition of an “unbiased sequence” is only meaningful relative to the extent players have uncertainty about one another’s preferences, and therefore it’s not uniquely defined.

          Good luck with it, and I’ll read the progress this thread makes with interest.

    • RavenclawPrefect says:

      Another model you could use: players have completely random preference orderings for the stages, and want to strategically pick stages hoping that their opponent will eliminate some they were hoping to cross off anyway. You want to minimize the difference in expected rank of the final choice as measured by each player.

      I’m not positive what the optimal strategy would be here as a player; the first thing that comes to find is “go down from your just-below-median choice, hoping your opponent takes out ones you dislike for you; once everything in your least favorite half is eliminated, repeat with the remaining strikes you have on whatever’s left (and so on as necessary).” I don’t immediately see a proof that this is optimal, but I’d assign at least 30% odds to it being such; it doesn’t seem like it gives your opponent any more information than it has to, and it’s optimal in the extreme cases of getting the first half of turns or the second half.

      • cryptoshill says:

        Something to note here – competitive Smash players have *tons* of information about each other. At the levels of the livestreamed tournaments – those same players have been playing each other over and over again for *years* – and stage selection is going to be affected mostly by “What character(s) do I know my opponent plays most often, and is there a stage that disadvantages that character without opening me up to a counterpick?”.

        Given that this is a known value, and there is only one truly important factor at top levels (platforms or no platforms), and only one stage totally without platforms – you could make this problem as simple as “Final Destination or No?” and give each player alternating choices every round.

        Of course – this is to solve the problem as it practically exists, and not the problem that has been theoretically presented.

        • Sniffnoy says:

          That also seems like a somewhat Melee-specific answer, when of course what prompted the question is Smash Ultimate’s potentially very large number of legal stages. 🙂 (And the legal stagelist for Melee isn’t expanding, that’s for sure!)

        • MNH says:

          “Only one truly important factor at top levels” is quite far from true. The size of the stage, the platforms’ heights, and the placing of the blast zones all matter–consider how much better Pokemon Stadium is for Fox vs Puff where uthrow -> uair kills off the low ceiling like 20% sooner than anywhere else. Or similarly how good for Puff Dreamland 64 is against any slow character, where the far blast zones protect her from dying to her own light weight (since she can recover from such a great distance) without hampering her since she most typically kills by edgeguarding.

          • cryptoshill says:

            @MNH – Ok so we clearly have some near-competitive Melee players here. Puff herself is an edge-case that is played well by fewer players – but I will allow that her presence makes stages more complicated than platform advantage (In fact – is there anybody else than HungryBox that routinely wins with Puff? From my perception he is the one carrying her spot in the tier list). I just meant to suggest that “the number of stages actually worth considering against a particular known player are a lot lower than naively expected”. A lot of the stages are functionally identical or their utility values for a given matchup are so small that the differences in pick order wouldn’t be particularly relevant. Smashers won’t commonly blame stage advantages for losses even at lower levels of play. I would expect each player getting even *one* ban in Melee would give each player almost 100% of the utility of a full ban order.

            @Sniffnoy – I was unaware that there were any Smash games made other than Melee, Smash Ultimate is an elaborate hoax. 😛

      • Sniffnoy says:

        Huh, you know, I ignored this at first, but I just realized this is basically the same as uau’s suggestion. This seems like a decent model, thanks!

    • watsonbladd says:

      So I’ll take a stab at it. The right notion is that each player ends up at least on what they would evaluate as the median in their subjective evaluation.

      Interestingly in Counterstrike: Global Offensive the map selection is done by teams alternatingly vetoing, or after two vetos each picks one, and the third is selected randomly as described in more detail which is different from Smash.

      • Sniffnoy says:

        Oh, that is a good criterion — one that gets rid of the need for any sort of randomness or trickiness in formalizing this. Don’t know why I didn’t think of that, instead mistakenly saying above that the problem becomes meaningless in this case.

        However, if we adopt this criterion, then I think we have to come to the conclusion that the order really doesn’t matter! Any order will accomplish this so long as both players get an equal number of strikes.

        OK, you’ve basically convinced me — this really does seem like the right formulation of the problem, and it has an easy answer, namely, striking order just doesn’t matter that much; Thue-Morse is not any sort of unique answer here. Hooray! (…unless of course someone presents a better argument to the contrary. 😛 )

        Interesting that CSGO uses a different system (or at least sometimes does). Of course I didn’t get into discussion of stage selection past the first game in a set in Smash, which is done differently, because, well, that wasn’t what I wanted to discuss.

        • uau says:

          Order can matter significantly. “At least the median” is just too weak a requirement that’s trivial to satisfy.

          Consider the case where where both players have independent random orderings for 100 stages. You let one player eliminate 50 stages, then the other player eliminate 49 of the remaining. Despite having less choices, the second player has a massive advantage. From the first player’s point of view, he eliminates 50 least preferred stages, and gets a random one from his preferred 50. The second player has 50% chance of getting his top choice (if that happens to be in the top half of the first player), otherwise a bit above 50% chance of getting his second choice, otherwise a bit above 50% chance of getting his third choice, and so on.

          So first player gets “random stage from your top 50”, second player gets “50% chance of the single best stage, and almost always within the top few”. Obviously this is a massive difference.

          • Sniffnoy says:

            That’s a good point. I guess it does matter after all. But I think we’re getting somewhere here! Because you’ve done something interesting here, which is, for each player, to regard that player as fixed and the other as random, and to consider it from that perspective. And then we can ask that the expected values (where here I’m assuming values are 1,2,…,n) be close.

            Of course the reintroduction of expected value raises the question of whether we should be taking a more cardinal approach. But I don’t presently see how to make sense of that in this new context — at least, not if we want to regard the other player as uniformly random. We might want to do something like I suggested above, where each player regards the other as random but biased towards agreement. Then you could make sense of it in a cardinal context.

            But that just seems like needless complication. This version of the problem, without that, and with uniform randomness, seems nice. Will have to think about this. Maybe there is a unique answer after all! And maybe it’s Thue-Morse. 🙂

    • ing2 says:

      I’d like to challenge the frame of the question. : )

      The process you already had, where people take turns striking stages and reach a median, was already super super fair — arguably it sacrificed a lot of fun for fairness, in that it decreased the variety of the game.

      Normal people who are playing a normal game will not care about fairness to this degree. They’ll say: “let’s choose a random stage!” or “you choose one, then I’ll choose one!”

      Very competitive people might care about fairness to a very very high degree, but this solution will not satisfy them: it will not be completely 100% exactly inarguably fair.

      If you encounter someone who is not satisfied with the fairness of taking turns striking stages, your next attempt at fairness should be something that uses simultaneous play to be provably fair.

      Have each player prepare a list of all the stages in order of their preference. Find the N such that letting each player drop their least-favorite N stages leads to one stage remaining. If there is no such N, find the N such that letting each player drop their least-favorite N stages leads to two stages remaining, and flip a coin among those two.

      Alternatively, set up a UI such that each player chooses a stage to strike simultaneously and in secret.

      • cryptoshill says:

        “You choose one, I’ll choose one!” is even common in the example game tournaments. It might even be provably *more* fair, at the risk of creating a small subgame where each player tries to guess the opponents preferences correctly. Unless of course, *substantial* information is gained about the opposing player in one round of play (which is unlikely at these levels). Something important to note is that in terms of winning competitive game tournaments – selecting the stage your opponent likes the least is more optimal than selecting the stage you like the most. Maybe “I go, you go” with the first chooser selected randomly is the best method?

    • sammy says:

      This seems like it could actually be a really general problem. It looks like a social choice problem where voters eliminate candidates instead of voting for their favorites.

      If each person has a VNM utility function over the possible stages (given their chosen character and opponents character) they would each want to maximize the difference in utilities, but ideally the striking order would minimize the utility difference (median). Put this way, it would seem like alternating striking (0101…) with a coin flip to determine who goes first (is randomization allowed?) would be best since their orderings over stages would be exactly opposite. For an odd total number of stages this would always choose the median, for an even number the first striker has an advantage.

      In cases with more people or non-opposed voters I have no idea.

      • Brad says:

        This immediately looked like a ranked voting problem to me too. I wonder if there’s a limited information element though in the back and forth–that is if information can be gleaned from the choices your opponent makes and that fed into further choices.

        • sammy says:

          That is a good point. Like you don’t know your opponent’s familiarity with each stage and your opponent could misdirect you with their striking choices. Seems like there is a paper to be written here about having priors about your opponent’s utility function, making striking choices adversarially, and what equilibrium that reaches.

    • uau says:

      So the starting assumption is that this is strictly a zero-sum game, where each player is trying to maximize the chance of winning? (No situations like player A leaving player B with a choice between a stage where A has the advantage and a stupid stage that everyone hates to play.)

      One way to get a more precise problem statement is to assume that the players have their own independent ranking for each stage, and learn absolutely nothing about their opponent’s ranking by seeing their choices so far (beyond what they already chose of course). This would give you the following problem:

      Players A and B both have a random ordering of the stages. Each time they get a turn, they eliminate the remaining stage that is lowest in their personal ordering. How should you distribute turns between them for both to have a similar distribution for the index of the last remaining stage in their orderings? Possibly you could ignore the exact distribution and only look at a single real-valued metric like expected index value.

    • AG says:

      Can this be applied to jury selection?

  6. johan_larson says:

    Here’s a link to a good talk about a Vietnam-era program sometimes known as McNamara’s Morons, a plan to fill out the ranks of the military by drafting men who had previously failed the military mental skills test. It didn’t end well. The speaker is Hamilton Gregory, who wrote the book McNamara’s Folly.

    https://www.youtube.com/watch?v=_J2VwFDV4-g

    • Sniffnoy says:

      FWIW, here is Gwern’s review/summary of that same book: https://www.goodreads.com/review/show/2333328079

    • cassander says:

      wrote the book McNamara’s Folly.

      How could the author pick just one?

      • johan_larson says:

        He’s saving the rest for a series of sequels?

        • cassander says:

          If so, he must be gunning for Robert Jordan’s record.

          • bean says:

            Well said. I think my internal comment on seeing the title and subject of the book was “I’m not even sure that makes the top 10.”

          • johan_larson says:

            I’d be interested to see that top-10 list.

          • bean says:

            1. Approaching every situation as if everyone involved except him was straining at the leash to start a nuclear war.
            2. Allowing the Army to steer us into an intervention in Vietnam.
            3. Combining 1 and 2 to fight a war without any intention of actually winning.
            4. Application of numerical analysis to military decisionmaking in the most boneheaded way possible.
            5. The TFX program.
            6. Cancelling the B-70.
            7. Shutting down missile defense work.
            8. Moving power to the SecDef’s office, and out of the hands of the services, who know what they were doing.
            9. Destroying an entire generation of procurement programs to try and keep the national fisc from feeling the effects of Vietnam.
            10. Giving power to lots of people who didn’t have the first clue about military realities. For instance, there was apparently a serious proposal to stop CSAR work because it was cheaper to just train new pilots. “I doubt they have a morale (some versions say moral) setting on their computers.”

            In fairness, my view is heavy on the procurement/policy side, and not likely to pick up the sort of stuff this talks about.

          • A Definite Beta Guy says:

            6. Cancelling the B-70.
            7. Shutting down missile defense work.

            I’m interested in the rest of the hivemind’s opinions on these two, if anyone has two cents. My impression is that both the B-70 and the 60s-era missile defenses would provide, at most, short-term strategic advantage and will be mostly pointless by the late 60s and early 70s. The B-70 seems to have been out of favor by the late 50s in favor of ICBMs and I think the only time it would have really seen useful service would have been Linebacker II (and I am guessing it would have been prohibitively expensive to fly B-70s for those raids compared to B-52s, though at least you aren’t losing as many to missiles).

          • Nornagest says:

            The B-70 is stupendously cool but I think it would’ve ended up being a bit of a white elephant. It didn’t have the bombload or the range of the much cheaper B-52, it would’ve been vulnerable to the high-performance SAMs that were coming out about then, and strategic nuclear bombers were starting to look a little obsolete by the early ’70s — that’s about when ICBMs started getting really good.

            I’m less certain about missile defense. They were really pushing the technology of the time, but Sprint was promising, and another twenty years of continuous ABM development would probably have gone a long way towards easing the current boondoggle. I get the sense they were canceled more for political than technical reasons, although the political reasons are somewhat compelling — ABM development still makes the Russians really nervous, for example, and it would only have been worse in the Cold War.

          • John Schilling says:

            If the United States had continuously maintained a missile defense capability from the 1970s to the present, it would have been mostly useless for most of that time, but we would have forty years of experience in how to do something resembling useful missile defense and we would probably have an operational system that had gone through a couple of upgrade cycles based on lessons learned. Instead, we have a system that was rushed into service as an experimental prototype a little over a decade ago and only patched since. It works about half the time against extremely limited attacks in tests, but we have people in positions of real power who profess to believe it will work 97% of the time in actual combat, and we may be urgently needing it sometime in the near future.

            The precise details of our present situation would obviously not have been predictable to McNamara, but the general principle that if there is a major threat to the United States we should probably not completely zero out our defenses against it seems sound. Also see every James Bond movie or cautionary-tale proto-technothriller where one rogue bomber or missile triggers Armageddon(*), and consider the wisdom of setting the defense threshold at “zero”.

            (*) er, Armageddon IV: this time it’s really apocalyptic.

          • sfoil says:

            bean, why do you think the B-70 should have gone ahead?

            The reasoning behind canceling it might have been wrong, but approving it would have diverted the Air Force away from the B-1, which was better at low-level penetration (and preferring this seems sound).

            If the USAF had the B-70 instead of the B-1, they probably would have scrapped the B-70 when SAC ended, giving them a high/low B-2/B-52 bomber force currently. Which doesn’t sound so bad, actually, but I don’t know if that can really be counted in the B-70’s favor.

          • bean says:

            Re the B-70, I think you greatly overestimate how good SAMs of that era were. One of the advantages of a manned bomber is that it’s able to fly a rather unpredictable path, so the missile needs a big performance advantage to hit. In terms of performance, the B-70 was about the same as the SR-71, and none of those were ever shot down, or even seriously threatened. I don’t have firm numbers, but anecdotal evidence suggests that it’s only in the last 20 years or so that SAMs have had even a halfway-decent chance against targets of that type. These days, people in the defense world often claim that speed is the new stealth. If that’s true today, it’s hard to see why it wasn’t in the 60s. Except that the word “hypersonic” wasn’t horribly abused back then.

            Re missile defense, I’m not sure that making the Russians nervous is a bad thing, particularly given how unreasonable their level of concern is. The level of missile defense that protects us from North Korea and gives the President more time to think in cases of a few missiles getting fired is very far from the level that makes it thinkable for the US to start an all-out nuclear war with Russia. The fact that this isn’t considered obvious is really confusing, and I can only assume it’s a combination of political attacks on the system by those who want to kill it and leftover Soviet-backed propaganda.

            Edit: Please note that in 1973, the city with the heaviest air defenses on the planet was not Moscow but Hanoi. And during Linebacker II, the BUFFs got through with fairly reasonable losses. Now, imagine the plane is three times faster and flying higher, making it much harder to shoot down.

          • bean says:

            @Sfoil

            I think the B-1 was ultimately a mistake (even though it’s very pretty, and an impressive piece of engineering). Playing in the weeds sounds great, but it moves the threat from “anyone with a big SAM system, which we can try to dodge or jam, and maybe fighters, if they’re fast enough” to “anyone with a rifle”. One of these seems strictly better than the other to me. Note that every time we go to war, extreme low-level penetration, which works OK in exercises, is soon discarded in favor of fighting at higher altitude where the threat environment is better. It happened in Vietnam, it happened in Desert Storm to both the vaunted A-10 and the Tornado, and it happened over the Balkans.

          • sfoil says:

            I think the concern with the B-70’s vulnerability to air defenses was less to do the state of the art in the 1960s than with trend lines. It might have been untouchable in 1970 and a sitting duck by 1980. As far as threats to the SR-71, if it had been capable of dropping bombs instead of just taking pictures, the Soviets might have tried harder to develop the capability to shoot them down.

            The threat to stealth aircraft, as I understand it, involves a difference in kind and not merely degree of detection measures. In fact, that might be a good counter to the SR-71 example: if the most dangerous aircraft were really fast instead of really hard to detect, more effort would have gone into defending against fast instead of stealthy aircraft over the last 40 years.

          • cassander says:

            @bean

            I remain somewhat suspicious of the ability of something flying higher and three times as fast as a BUFF to actually anything with 70s era technology. Granted that’s a lot less of an issue today, but no one was pushing the B-70 on the basis of the coming PGM revolutio. Tohit anything, it would have to slow down and drop lower, and that would have severely limited the advantages it provided.

          • sfoil says:

            That’s a very good point about the real-world effectiveness of low-level penetration. The problem isn’t someone with a rifle, but low-altitude point defenses around the target. And those are better than ever. They’re also going to become more ubiquitous since they can now plausibly intercept munitions launched from altitude rather than just hoping an aircraft flies in low.

            I’m not sure the B-52’s ability to penetrate Hanoi’s air defenses is a slam dunk however. As a whole the Soviet IADS had much more depth, and in particular the North Vietnamese had fewer and far less effective interceptors than the Soviets did.

            That being said, I still think canceling the B-70 was the right call. Choosing to go in low on an initial attack and guaranteeing that some bombers will get through and some won’t might be safer than betting that your adversary won’t be succeed in his dream SAM within the next decade (especially when the limits of the relevant technology weren’t understood well beyond “it’s getting better, fast”).

          • sfoil says:

            As far as low-altitude defenses being a nasty surprise, obviously the solution is better exercises. I suspect the problem is that SHORAD is the aerospace equivalent of “light infantry”: you get a lot of relative effectiveness out of pretty basic equipment if you train and prepare correctly, and are willing to accept a certain level of vulnerability that the United States isn’t comfortable with. American SHORAD has been terrible forever to the point that there’s little institutional understanding of what’s really going on, and we can’t correctly simulate an adversary who does know what they’re doing. I think the easiest solution is to run exercises against the Navy. Maybe this is already done, but I’ve never heard of it.

            @cassander
            The B-70 was competing with missiles that had a CEP measured in kilometers in order to hit targets the B-52 might not be able to reach at all, so I don’t know how much of a concern that really was.

          • cassander says:

            @sfoil

            The B-70 was competing with missiles that had a CEP measured in kilometers in order to hit targets the B-52 might not be able to reach at all, so I don’t know how much of a concern that really was.

            For the nuclear mission, it probably wasn’t an issue. I was thinking more about the conventional strikes the B-52 proved so good at. They weren’t particularly accurate either (actually, I’d be curious to see how they did compared to ww2 era bombers) but as a rule, the higher and faster you go, the less accurate you are with dumb bombs.

          • John Schilling says:

            The B-70 was competing with missiles that had a CEP measured in kilometers in order to hit targets the B-52 might not be able to reach at all, so I don’t know how much of a concern that really was.

            If the idea is that the B-70 is going to actually overfly defended Soviet targets and hit them with dumb gravity bombs, I don’t think that was going to be plausible against even late-60s Soviet SAMs like the S-200 once they were upgraded to deal with the B-70 threat, and I’m pretty sure the USAF knew that at the time. The B-70 would have been a plausible launch platform for air-launched ballistic missiles, but then so was the B-52. In that application, you’re trading the extra ~700 m/s of launch velocity you get with the B-70, against the extra payload capacity of the B-52 – which allows heavier booster rockets to add an extra ~700 m/s to the ALBMs if that’s what you need, but also the flexibility to carry e.g. more dumb bombs to plaster logistics sites in North Vietnam.

            The B-70 would have had a responsiveness advantage from its higher cruise speed, but once it was clear that the magic “zip fuels” were off the table and every strategic mission was going to require refueling from subsonic tankers, that advantage was greatly diminished.

            Magnificent as it was, I think cancelling the B-70 was the right call – but it should possibly have been followed up by developing an ALBM somewhere between SRAM and Skybolt to arm the B-52. And then the B-1 came along and blundered its way into being an extraordinarily useful aircraft in ways entirely different from the original plan.

          • sfoil says:

            The B-70 was a nuclear bomber first and foremost. I don’t know that anyone really cared about its performance in delivering high explosives at all. Obviously if a target was undefended it could just fly lower and slow down to achieve better accuracy, but those bombs dropped from 70,000 feet at Mach X were always going to be nukes.

          • A Definite Beta Guy says:

            If B-70s can be intercepted by S-200s, I learned something new from this thread. I always figured Mach 3+ targets weren’t really targetable until the last 20-25 years. I kinda figure Bean is of the same opinion (though I don’t really want to speak for him).

          • bean says:

            @sfoil

            The threat to stealth aircraft, as I understand it, involves a difference in kind and not merely degree of detection measures. In fact, that might be a good counter to the SR-71 example: if the most dangerous aircraft were really fast instead of really hard to detect, more effort would have gone into defending against fast instead of stealthy aircraft over the last 40 years.

            I think this is a bad comparison, though. Speed and altitude are hard to counter in a way that stealth isn’t. Improved sensors, different wavelengths, and better signal processing can defeat stealth. To defeat speed, you need a really fast missile, and propulsion technology just hasn’t been improving that much recently. So it’s more expensive to counter incremental speed than incremental stealth.

            @ cassander
            I’ll grant that conventional bombing would not have been the B-70s strength until PGMs showed up. Unfortunately, the Jenkins book had no details on expected accuracy beyond “goal of 1,500 yds, reduced to 5,000 if it’s cheap enough and easy enough to take care of.”

            @sfoil

            I’m not sure the B-52’s ability to penetrate Hanoi’s air defenses is a slam dunk however. As a whole the Soviet IADS had much more depth, and in particular the North Vietnamese had fewer and far less effective interceptors than the Soviets did.

            Again, speed and altitude. The MiG-25 was the only Soviet fighter that could hope to touch a B-70 in 1973, and it had only been in service 3 years, with all the attendant teething troubles that causes. And even it is only very marginal. Everything else is just useless.

            @John

            We are finally getting around to building those ALBMs. It’s just a few decades late, give or take.

            And I’m definitely with ADBG on the feasibility of shooting down a B-70 with even an upgraded S-200. Unfortunately, I don’t have the time to do the math on this. I’ve looked hard, and open sources are pretty quiet, but getting to speed and altitude isn’t trivial.

          • Nornagest says:

            If B-70s can be intercepted by S-200s, I learned something new from this thread. I always figured Mach 3+ targets weren’t really targetable until the last 20-25 years. I kinda figure Bean is of the same opinion (though I don’t really want to speak for him).

            The early Seventies versions of the S-200 top out at a speed of around Mach 7, with a range of 150 miles, and the 5NH6 radar feeding them targeting info can detect targets well past that — giving them a couple of minutes to acquire a lock and launch against a target moving at Mach 3 before it’s too close to usefully engage. That’s not a lot of time, but it beats the pants off the missiles that were the state of the art when the Valkyrie program began. And the Valkyrie, unlike the SR-71 that we got most of our real-world experience with these speeds and altitudes from, isn’t particularly stealthy.

            S-200 sites aren’t very mobile, though, and each site can only engage one target at a time — there’s six missiles to a radar in a typical deployment, but Russian equipment being what it is, they’d probably be salvo-fired against a single target. And maneuverability probably isn’t so hot at altitude. The Russians claim high kill probabilities, but who knows what kind of targets they’re thinking of with those.

          • bean says:

            The early Seventies versions of the S-200 top out at a speed of around Mach 7, with a range of 150 miles, and the 5NH6 radar feeding them targeting info can detect targets well past that — giving them a couple of minutes to acquire a lock and launch against a target moving at Mach 3 before it’s too close to usefully engage. That’s not a lot of time, but it beats the pants off the missiles that were the state of the art when the Valkyrie program began.

            Granted that the S-200 has at least a theoretical capability against the Valkyrie, but in practice, I don’t think it’s that big. Soviet SAMs were not easy to use, and the crew training was suspect, too. Given a fast-moving target with jamming that’s doing its best to throw the system off, it’s going to be very hard. And I trust SAC training a lot more than the PVO equivalent. And remember, the Russians lie about their equipment. A lot.

            And the Valkyrie, unlike the SR-71 that we got most of our real-world experience with these speeds and altitudes from, isn’t particularly stealthy.

            This is less true than most people think. North American wasn’t ignorant of stealth, and while they did focus on IR reduction, I assume that was for a reason.

            The Russians claim high kill probabilities, but who knows what kind of targets they’re thinking of with those.

            Going by their combat record, it’s airliners. Seriously, it looks like the S-200 has never had a single combat kill against a hostile target. Yes, I know that monkey models are less capable than the Russian version, and that the first combat use was in the mid-80s, but it’s been used many times, the Russians keep claiming kills, and the people they’re shooting at keep saying they’re crazy.

          • bean says:

            I tracked down a copy of the book, and started through it, and I’ll agree that it deserves to be in the top 10. I’d still rank it below his insane mismanagement of Vietnam, and his wrecking of the military procurement system. We still haven’t fixed the latter, although it looks like things might be getting better.

    • Deiseach says:

      I suppose this experiment does rebut all the arguments about “we draft the best of our men to go off and die in war and leave the idiots safe at home”, in that drafting the idiots as cannon fodder didn’t work so you do need a selective process to pick troops.

    • johan_larson says:

      A question worth considering is what options McNamara really had. If he really needed more manpower for the war effort, he could have

      – lengthened terms of service (angering those willing to serve, mostly the working class)
      – activated more Reserve and National Guard units (angering middle-class folks who had taken refuge in them)
      – started chipping away at exemptions, probably starting with bogus medical exemptions (middle-class again)
      – pushed to increase the size (and effectiveness) of the ARVN (a long slow process)
      – hired mercenaries ($$$)
      – leaned on allies to contribute more (hard)

      None of these options look great.

      If he had to bring on these low-ability men, he should not have drafted them as solders. They should have been drafted into some sort of auxiliary military corps where they could be deployed for low-expectations work overseas, but weren’t expected to actually fight. Most of these men could do useful work; they just weren’t up for the demands of actual combat. Heck, many of them had trouble getting through basic training.

      • John Schilling says:

        There’s also the one where we stop it with the nonsense about Exclusion Zones in wars, and tell the already adequately funded and no-morons-needed USAF to continuously bomb the crap out of anything that is threatening our beleaguered ground forces no matter what side of the imaginary line it is on, and tell the poor innocent civilians of North Vietnam that it isn’t just our allies that get to evacuate their children to farms in the middle of nowhere if their government insists on putting legitimate military targets in their cities.

        And then there’s the plan where, if we aren’t going to do any of the other plans, SECDEF needs to have the stones to tell POTUS that we aren’t going to win this war and it’s time to pull up and go home, resigning in protest if that’s not going to happen.

      • bean says:

        activated more Reserve and National Guard units (angering middle-class folks who had taken refuge in them)

        I’ve long suspected that this had very little to do with the decision not to send Guard/Reserve units to Vietnam. The basic problem is that they were still staring down the Soviets in Europe, and it was important to be able to keep those men in case the Soviets came west.

        And I’ll echo what John said. Linebacker II proved that we could bring them to their knees with air power when we really tried. Everything before that was stupid posturing games. Do what we did then in the mid-60s, and you don’t need a huge expansion of the army.

      • cassander says:

        Other than working harder to improve the ARVN, I don’t think any of those methods would have made things better. The underlying trouble was Macnamara’s underlying concept of graduated pressure, and the way it wasted US strengths. It wasn’t entirely his fault, he was giving LBJ what he wanted, but he also genuinely seems to have believed in what he was doing, and it was utterly foolish.

        @bean

        I’ve long suspected that this had very little to do with the decision not to send Guard/Reserve units to Vietnam. The basic problem is that they were still staring down the Soviets in Europe, and it was important to be able to keep those men in case the Soviets came west.

        The military might have had its own reasons, but LBJ seemed to think that it was imperative to keep vietnam from looking like a “real war” so he could keep the great society going. I’m not aware of any smoking gun quote from him on the reserves in particular, but that’s probably just because there didn’t have to be one, because it was clear to everyone that worked for him that that wasn’t in the cards.

      • A Definite Beta Guy says:

        I dunno, if I were a 1960s official, I might have done some of the same things. At this point we’re coming off a relatively unpopular and indecisive Korean War, and there is a lot of fear that any stupid little spat might become the next Cuban Missile Crisis. The West has done a relatively good job containing communism so it looks like we can have a relatively easy victory doing normal counter-insurgency. It’d seem doubtful to me that little North Vietnam is going to be willing to sacrifice people endlessly against one of the world’s superpowers.

        Some of the stupid analytics definitely seem like something I would have believed if I were a 1960s man.

        I most likely would have escalated more quickly, more rapidly. 1966? Yeah, if you’re drafting 400,000 Americans, it’s time to mine those ports and turn Hanoi into rubble.

      • Nancy Lebovitz says:

        Conscripting mentally incompetent people as soldiers doesn’t resemble a solution to any of McNamara’s problems. I assume if was a loss which wasted time, money, and lives.

        I’m also surprised there were more? so many? men with serious mental deficiencies rather than a larger number of men who were a little unqualified and who wouldn’t have done *that* badly. Or were all of those already drafted?

        • bean says:

          The book does point out that a fair number of Project 100,000 men were either slightly underqualified or tested poorly for reasons not linked directly to intelligence (illiteracy being the most common). And that even a fair number of the truly sub-par did OK if they ended up in the right job. (Someone with an IQ of 70 can do a great job as a janitor, but he’s a terrible infantryman.) It’s primarily an anecdotal book, though, and doesn’t really have the statistics to show what proportion this was. (To be fair, I’m not sure those statistics exist.)

    • Nancy Lebovitz says:

      Supposing you wanted to prevent people at the top from engaging in obviously stupid projects, what organizational structures would be needed?

      • johan_larson says:

        I don’t think any organization structure or process avoids stupid projects at the top. In the middle layers, you can avoid stupid projects by being very deliberate and allowing no exceptions. If you want to do something, you ALWAYS need a formal proposal, and you ALWAYS need a VP sponsor, and you ALWAYS need a pilot project, and you ALWAYS need a cost-benefit analysis from the Operations Research department. Of course, this regimen will also kill responsiveness and agility, but maybe that’s worth it if you are in a high-stakes domain. But it won’t stop the really really senior folks because they can always come up with some reason why the regular rules don’t apply and everyone knows that disobeying them means crumpling up your career and tossing it in the trashcan.

        I think the only ways we have found of keeping the most senior folks from doing stupid things are being very careful about who we put in these positions, making sure they don’t have perverse incentives, and having some sort of countervailing authority, typically the board. Except these days founders often have super-voting shares, so they both run the company and control the board, which is worrisome.

        • Nancy Lebovitz says:

          Scott Adams has said that while no boss recognizes themself as being like Dilbert’s boss, it’s quite possible to stop a stupid idea dead by saying it sounds like something out of Dilbert.

      • Nabil ad Dajjal says:

        Giving them skin the game, probably some form of ownership.

        People with a stronger business background than me should feel free to correct me here, but the way I understand it privately held companies do substantially better than publicly traded companies of similar size both in terms of their rate of growth and profit margins. The more distributed ownership of a firm is, the weaker the incentive for executives have to do what’s best for the firm’s long-term health as opposed to schemes which enrich themselves at the expense of the firm.

        This seems like the same principle that comes up a lot in discussing public choice theory. The incentive for a given shareholder in a firm or a voter in a polity to cast their votes wisely, or at all, diminishes rapidly the more distributed the benefits and costs are and the lower of a vote share they have.

        Now obviously property isn’t a panacea. Lots of privately held companies go under due to stupid blunders, and pretty much every monarchy which ever existed could point to a “bad king” who ruined or nearly ruined the country. But if I’m right, giving leaders more skin in the game should make these blunders occur less frequently.

        • Mr. Doolittle says:

          @ Nabil

          Two major confounders on your thought about publicly traded verses private companies, in regards to growth and profitability. One, companies built privately that become public (Facebook!) that will read as “public” but should mostly read “private” or a mix(?). Two, failed companies. If you’re looking at currently active companies, I would expect that you will have sorted out more of the private companies that failed than public ones, but have more mediocre public ones. We probably care a lot about failed companies as well as active, so pulling in that information would be important.

          To add, I think you are missing a couple of key factors.

          Centrally controlled companies (owner-operator especially) do have benefits if the small number of controllers make wise decisions, but they are also worse if they make poor decisions. Decentralized systems struggle with coherent plans and conflicts of interest more, but also don’t dive into failure as much.

          This is similar, to me, to the idea of a Philosopher-King verses a Democracy. PKs can make great decisions for a group, if the right person is the PK. Finding the right person, getting them in control, and especially how to transition to the next PK are hard questions with that approach. Democratic institutions often fail to be more than mediocre, but they can more consistently approach a problem without doing something that completely breaks the arrangement.

          Corporate boards and significant shareholders can act as a mitigating factor on bad decisions, as can creditors, CFOs, COOs, etc., rather than just having a Owner/CEO making all decisions. If you’ve got the best CEO ever, you’ll lose through moderation, but few companies do.

      • Erusian says:

        Could you give an example? I find obviously stupid projects are much rarer than people think. Everything everyone does usually makes sense from the decider’s point of view. Now, their point of view might be, “I sell concrete and this bridge to nowhere would make the government buy concrete.” But that’s self-serving, not obviously stupid.

        • Nancy Lebovitz says:

          McNamara’s Folly– drafting men with seriously sub-normal intelligence to fight in Viet Nam– is the example I had in mind when I raised the question.

          • Erusian says:

            Sure. So, Project 100,000 was not an obviously stupid decision in the context of its decision makers. The Project was not about manpower (it supplied, at most, 4% of military manpower and probably less). It was an anti-poverty measure, meant to bring in people who could not otherwise find employment easily. It was portrayed in these terms by the President, the Army, McNamara, and internal documents.

            Johnson’s incentive was to get re-elected. To this end, he started the War on Poverty. If that had negative externalities, these diffuse negative effects are less than the concentrated benefit of winning elections for him personally. McNamara’s incentive was to please his boss, who could dismiss him at will. This incentive got stronger as the war went worse and it became more advantageous for Johnson to do so. So he used the resources of the military to put in place a program that supported his boss’s priorities. Again, most negative effects were diffused through the army while McNamara got the concentrated benefit of keeping his job.

            Again, self-interested but not stupid.

          • Nancy Lebovitz says:

            The policy may not have been stupid for McNamara’s personal goals, but it was stupid for anyone who cared about the war, the military, or what happened to those conscripts.

          • cassander says:

            @Erusian

            1,000, or maybe even 10,000 would not have been an obviously stupid decision. After all, why not see if it works. but 100,000? that’s taking the idiot ball and running with it.

          • Erusian says:

            The policy may not have been stupid for McNamara’s personal goals, but it was stupid for anyone who cared about the war, the military, or what happened to those conscripts.

            Sure. Though with the conscripts, it was believed, at the time, that giving them opportunities in the military would help them advance out of poverty. This turned out not to be the case. But this is Moloch. (Am I using that term correctly?) The system is not an agent. McNamara is. Johnson is. It was stupid for the army. But the army doesn’t make decisions, Johnson and McNamara do.

            Now, the need to align incentives properly is very important for organizational health. But that isn’t preventing stupid decisions. It’s making sure that people’s interests are not different from the organization’s interest. Because most of the time, personal interests will win out.

            1,000, or maybe even 10,000 would not have been an obviously stupid decision. After all, why not see if it works. but 100,000? that’s taking the idiot ball and running with it.

            It was actually between 300,000 and 350,000, though a percentage of those were regular soldiers included as a control group. And why does having more make it more stupid? The more people in the program, the more otherwise unemployable people with jobs, the more poverty alleviated, no? And again, this was at most a few percentage points of total manpower.

            Also, note that by the end of the war many of them were working as janitors and other jobs that were deemed possible. It was realized they were terrible infantrymen and the system adjusted.

            Again, I’m not saying I like the program. My point is that it was not stupid in the sense of being irrational or ill-motivated. It was people following incentives to a bad end.

          • cassander says:

            @Erusian

            And why does having more make it more stupid? The more people in the program, the more otherwise unemployable people with jobs, the more poverty alleviated, no? And again, this was at most a few percentage points of total manpower.

            You’re trying something new that will literally get people killed if it fails. Not even something new, something that, until the day before, was considered positively unsafe. The smart decision is to test it out on a small scale first, we don’t roll out new drugs to hundreds of thousands of patience before clinical trials.

          • johan_larson says:

            The smart decision is to test it out on a small scale first, we don’t roll out new drugs to hundreds of thousands of patience before clinical trials.

            The program was also implemented poorly, whether the end goal was to help the war effort or improve the station of the men in question. Trying to turn them into regular soldiers by sending them to basic training was typically a waste of time. These men needed a slower ramp-up to build up physical fitness and a more instructional approach to firm up long-neglected mental skills, starting with basic literacy. This might have turned some of the men into competent soldiers, though at a hefty cost. Others could probably have done some good in low-expectations jobs; heck, some actually did. But there seemed to be a third group who were just too broken to be of use to anybody, whether by inherent inadequacies, or due to a couple decades of gross mistreatment.

          • Nancy Lebovitz says:

            If the video was correct, there was a belief that these men could be made adequately capable during basic training.

            Actually the techniques didn’t exist and nothing was tried.

            When it was clear they couldn’t manage basic training even with a lot of help, they should have been assigned easier work or discharged.

  7. We are having another South Bay meetup on Saturday, December 8th, starting at 2 P.M.

  8. Uribe says:

    AI Feelings – Do Androids Stub Their Toes?

    To me, existence means you experience something. You feel something. You can feel without being conscious. For instance, you can be asleep but feel the horror of a bad dream or the pain in your right foot. To experience these things, you must exist.

    I’ve read many discussions about AI “consciousness”, how it might come about, what it means, how likely is it. I don’t have any interest in consciousness. Thoughts don’t interest me, sensations do.

    How does a computer “feel” anything?

    • RavenclawPrefect says:

      How does a human “feel” anything? It’s far from obvious that a collection of neurons and glial cells ought to give rise to any kind of subjective experience, so it’s plausible at first glance that whatever kind of mysterious phenomenon happens in humans would transfer to the right kinds of AI minds. As far as I’m aware, most arguments for subjective experience in nonhuman entities proceed along the lines of providing reasons to suspect these properties of experience in humans to transfer. For instance, a high-fidelity emulation of a human, instantiated in some AI, would when given a simulated impact to their simulated toe give out a simulated cry of pain exactly as real as the cries of pain that you (presumably) use to infer that other biological humans stub their toes.

      • Uribe says:

        Maybe matter itself experiences being. Perhaps different elements experience being at different degrees of sensation, meaning perhaps carbon experiences a higher level of being than hydrogen.

        For instance, a high-fidelity emulation of a human, instantiated in some AI, would when given a simulated impact to their simulated toe give out a simulated cry of pain exactly as real as the cries of pain that you (presumably) use to infer that other biological humans stub their toes.

        It’s true that I can only infer that other humans experience anything, but I am certain that I do. A robot might pass the Turing test but that wouldn’t prove that it feels anything.

        • Aging Loser says:

          Maybe truly atomic particles (electrons but not protons) are aware/desiring beings (electrons want to orbit nuclei the way Aristotle’s stars want to orbit the depths of the world) but then awareness doesn’t reappear until you get all the way up the staircase of entities to living things with nervous systems.

          • Deiseach says:

            Maybe truly atomic particles (electrons but not protons) are aware/desiring beings

            From the last chapter of “Perelandra” by C.S. Lewis:

            “That Dust itself which is scattered so rare in Heaven, whereof all worlds, and the bodies that are not worlds, are made, is at the centre. It waits not till created eyes have seen it or hands handled it, to be in itself a strength and splendour of Maleldil. Only the least part has served, or ever shall, a beast, a man, or a god. But always, and beyond all distances, before they came and after they are gone and where they never come, it is what it is and utters the heart of the Holy One with its own voice. It is farthest from Him of all things, for it has no life, nor sense, nor reason; it is nearest to Him of all things for without intervening soul, as sparks fly out of fire, He utters in each grain of it the unmixed image of His energy. Each grain, if it spoke, would say, I am at the centre; for me all things were made. Let no mouth open to gainsay it. Blessed be He!”

          • Aging Loser says:

            Deiseach — I was just thinking yesterday about maybe rereading Out of the Silent Planet and Perelandra! (Or perhaps I was thinking about maybe nagging my son into reading the whole Space Trilogy after he finishes The Silmarillion; I can’t recall whether it was he or I that I had in mind as the reader.)

            I reredd That Hideous Strength a few years ago. I thought that Lewis let himself off easy by presenting the prole-boyfriend (who had just escaped from the prison where the convicts are being pharmaceutically mind-improved) as having been the perpetrator of very minor drunken hijinks that didn’t hurt anyone. If Lewis had really wanted to make his case he should have figured out how to make us hate the idea of pharmaceutically mind-improving a serial killer or mass murderer.

            Anyway, with regard to your quotation from Perelandra — it sounds like Lewis wants to have it both ways there — a particle “utters the heart of the Holy One with its own voice” but then again “has no life, nor sense, nor reason ….” So, atomic particles are sort of conscious except that they’re not at all conscious.

          • Nancy Lebovitz says:

            Huh– I thought the character in That Hideous Strength was in for a fairly minor crime, and Lewis’ point was that indefinite sentences were arbitrary and cruel.

            I may have missed it, but I didn’t think the prison was doing anything to affect the prisoner’s mind, they were just not going to release him until they decided he was rehabilitated.

          • Nick says:

            If Lewis had really wanted to make his case he should have figured out how to make us hate the idea of pharmaceutically mind-improving a serial killer or mass murderer.

            Isn’t that what A Clockwork Orange is for?

        • Enkidum says:

          It’s true that I can only infer that other humans experience anything, but I am certain that I do. A robot might pass the Turing test but that wouldn’t prove that it feels anything.

          Right. So if you’ve got a robot you can’t distinguish from a human, then you have exactly as much reason to think it has experiences as you have with any human other than yourself.

          You could push back against this by saying that there’s something about humans other than their behaviour which you think allows you to make the experience-inference. This is John Searle’s claim referenced in another response to OP – humans have nerves and other specialized meat, this specialized meat is what is necessary for experience, and so robots without the meat don’t have experiences, Q.E.D.

          But if you want to go the full Penrose route and argue that matter itself is conscious, then you’re going to have a very hard time convincing anyone that robots aren’t conscious, since they’re made of matter (and silicon has a higher atomic weight than carbon, for what it’s worth).

          Personally, I think both responses are terrible, though I have a little more sympathy for the Searle one than the Penrose one.

        • Viliam says:

          Maybe matter itself experiences being. Perhaps different elements experience being at different degrees of sensation, meaning perhaps carbon experiences a higher level of being than hydrogen.

          This is the kind of “mysterious” thinking the Sequences were supposed to dissolve…

          How would an atom “experience” anything? To experience requires to process data. Where does the atom store the data? How does it move them around?

      • Aging Loser says:

        I like the rule, which I redd in a book by someone named “Searle”, that feelings require nerves. Apparently jellyfish have something sort of like nerves but that isn’t quite nerves, so jellyfish would be a borderline case. It would be interesting if they are, because then we’d be able to wonder what it feels like to be a radially symmetrical entity.

        It’s hard to avoid feeling that trees are conscious, but by Searle’s rule they aren’t. I guess we’re inclined to imagine them as conscious because they look so much like theatrically posed human beings. In I AND THOU Buber using grooving on a tree as an example of having an I-You relationship with natural things. He doesn’t deal very satisfactorily with the question of how you can have a genuine (as opposed to imaginary) I-You relationship with an unconscious thing, though: “Does the tree then have consciousness, similar to our own? I have no experience of that. But thinking that you have brought this off in your case, must you again divide the indivisible? What I encounter is neither the soul of a tree nor a dryad, but the tree itself” (trans. Kaufmann 58-59).

        • zqed says:

          It’s hard to avoid feeling that trees are conscious, but by Searle’s rule they aren’t. I guess we’re inclined to imagine them as conscious because they look so much like theatrically posed human beings.

          This is one of the weirdest, most confusing things I have ever read on SSC. I am certain I have never met any healthy adults who attributed consciousness to trees or expressed any feeling that trees might be conscious (this includes many older acquaintances otherwise prone to “magical” thinking).

          In my brief experience working with children in primary education, many of them have the opposite default assumption: many of them take a long time to understand that plants are even _alive_, much less conscious.

          • Chalid says:

            On trees having consciousness, panpsychism is perfectly respectable philosophy which (the article claims) is taken seriously by “a significant and growing minority of analytic philosophers.”

          • zqed says:

            @Chalid: Panpsychism may be a respectable philosophical position (and I am somewhat sympathetic towards it), but it’s definitely not a “hard to avoid feeling”.

          • Aging Loser says:

            Well, there are the Ents and Huorns (and Old Man Willow) in Tolkien, and the Tree-Guardians in Maleficent (the movie in which Angelina Jolie gets her wings cut off), and the Mother-Tree in Avatar (the movie with the tall blue primitives called “Navi”), and in one of the sequels to Ender’s Game there are Ewok-type people who grow up to be tree-like beings, and for a while there was a poster advertising some kind of hard cider that featured a tree with a face in its trunk, and there’s a talking tree in the Legend of Zelda (a video game).

            In some novel by Virginia Woolf there’s a crazy WWI vet who wants to preach to the world that “trees are alive” (=conscious), but I guess his craziness doesn’t support my claim that normal people tend to think of trees as conscious … but then Virginia Woolf had to think of thinking of trees in that way in order to make her character think that way. And I redd that when a certain European poet (Stefan Zweig?) reproached Heidegger for his silence about the Horror, Heidegger advised him to go commune with trees. And of course there are “tree huggers.”

      • Tarpitz says:

        There is a reason they call it the Hard Problem.

  9. Harry Maurice Johnston says:

    Scott,

    Perhaps you could have a look at this Stack Exchange Science Fiction & Fantasy question about the end of Unsong? None of the existing answers have been very well-received, and your commentary would be interesting.

    • Conrad Honcho says:

      Ha. I was going to say “no, I think Harry Johnston there has the right idea,” but I’m guessing that’s you. That was basically how I took it. Thamiel was still created by God, as part of a divine plan to do good, so even Thamiel wasn’t really evil, he was a true consequentialist, doing evil (even an enormous amount) in order to ultimately do good.

      I don’t agree with that in our universe, but in Unsong it’s cute.

  10. Hoopyfreud says:

    Asking this in the CW-free thread explicitly to decouple this conversation from the things which, as-enacted, are the subject of this question. I think that the answers I suggest don’t locate me on a particular side of the culture war, and that people from everywhere on the political spectrum will have different responses. That said, if it gets too culture-war-y, or you think this effort at decoupling is not worth pursuing, I’ll delete this post and put it on the .25. Also worth noting that I am not a parent.

    What, to you, are the primary goals of parenting?

    I have come up with a list of 5 things, ranked in order from most to least important to me.

    1 – Giving kids tools to deal with/find meaning in life (philosophy, logic, dealing with grief, recognizing beauty)
    2 – Giving kids concrete knowledge and skills (reading, writing, arithmetic, campfire-building, cooking, accounting)
    3 – Making sure kids have good outcomes for them (not ending up homeless, paying for good schools)
    4 – Giving kids good values (promoting my own moral beliefs and encouraging them to adopt them)
    5 – Making sure kids produce good outcomes for me (caring for me in my dotage, providing me with grandchildren)

    I should note that 4 I’m skeptical of, and 5 I would actively try to avoid doing, insofar as it conflicts with anything above it.

    Feel free to rerank and/or articulate additional goals.

    • bullseye says:

      I don’t have kids either.

      I feel like giving them good values should logically be the top priority, because it serves society as a whole, but I also feel like I’d place my children’s well-being first.

      Giving them concrete knowledge and skills would be important to me mainly to the extent that it serves their own interest.

      It would never have occurred to me to list meaning of life, but that’s probably because I’m a bitter cynic. Anyway, it’s a subset of the children having good outcomes for themselves.

      I think the best way to get my kids to take care of me in my old age is to raise them as generally good people, which I should be doing anyway.

    • Ozy Frantz says:

      As a parent, one striking absence from this list is “making sure my kids have a happy childhood.”

      Kids are people, not clay to be molded into people. And how happy your kid’s childhood is is something everyone agrees that parents have control over.

      • Hoopyfreud says:

        I had a very unhappy childhood despite my father’s very best efforts, due entirely to bad fortune. I don’t feel that he failed to discharge his obligations. I think that’s probably why it doesn’t occur to me; I don’t feel that it’s possible to guarantee happiness to children, and that it’s better to help them try to make happiness where they can.

        • Evan Þ says:

          It’s not possible to guarantee it, but it’s possible to try, and there’re a whole lot of ways parents can help.

          For example, my sister hated whole-wheat bread as a kid. So, our parents got her white bread. There wasn’t any grand principle behind that, and in fact a couple pointing the other way (our mom’s a dietician and emphasizes healthy eating) – but they wanted to make her happy.

          • Hoopyfreud says:

            I think I understand this sort of thing to fall under (3), insofar as preference-fulfilment is a good outcome commensurable with the benefits of not eating white bread. I took Ozy to be talking about making sure the child has a happy childhood in a more meaningful/general/existential sense – delivering what Plumber might call a long chain of happy days.

        • quanta413 says:

          I think I agree with Hoopyfreud about this. My childhood was pretty unhappy although in a fairly ordinary way, and it’s hard for me to imagine my parents could have done better. I think if happiness follows from 1-4 great, but if it doesn’t then the parent did their best.

          But my belief here may be related to my belief that happiness itself doesn’t make sense as a life goal for adults either. I think of it as something you might get if you fulfill other goals. But you might not too.

      • Randy M says:

        I agree with Ozy here. The way I think of it is that while you can’t weight children’s immediate desires or opinions terribly highly due to lack of information and high impulsiveness (no ice cream every meal, no matter how much happiness it brings them while they are actually eating it), their feelings have equal weight to everyone else’s in the household.
        What you can do for their character or long term success you probably should (likewise yourself, for that matter), but in the meantime try to not frustrate or humiliate them; ask their opinion, explain rules, offer comfort, give them agency in as much as is reasonable for the development, consider their preferences, etc.

    • Le Maistre Chat says:

      Teach them how to think rather than what to think (which is what schools do), having faith that this will lead to the best outcomes (3 & 5).

    • Plumber says:

      @Hoopyfreud

      “….What, to you, are the primary goals of parenting?…”

      I have a two year old and a thirteen year old sons, and it’s different.

      Keeping them physically safe is paramount, after that giving the two year old a happy day (with the bad air quality due to the “Camp Fire” up north, it’s been taking him to libraries and Barnes & Noble instead of playgrounds lately).

      Unfortunately with the thirteen year old a “happy day” is no longer the goal as he has ambitions to go to U.C. and “study computers”, and now there’s little time for him to do anything except study, and his childhood is pretty much over.

      It’s very saddening, but with the odds of success so narrow, what’s the choice?

      Right now he’s on the computer trying to learn Spanish (knowing a foreign language is a requirement), and I know when I tried to learn a foreign language in High School (when I was a couple of years older than he is now) I was a failure at it (the only classes that I got consistently good grades in High School were history classes).

      No idea if were doing the right thing, I could just have him compete High School, practice doing arithmetic fast, along with spatial relations and “mechanical aptitude” tests, and have him become an apprentice plumber like I did, but when I last took him under the house to see the piping he wasn’t interested.

      I don’t think we will ever know whar the right thing to do is for sure.

      • Nornagest says:

        Studying a foreign language is a requirement, but I guarantee that no one on the UC admissions board is going to try talking to your kid in Spanish.

        Which is a good thing considering how much our schools suck at teaching it. I took four years of German in high school, and got good grades, but while I could probably order a beer and start a fight in Berlin now, I’d have trouble doing much else. (These days, I hear my old school teaches any language you want as long as it’s Spanish.)

      • Elephant says:

        … with the odds of success so narrow, what’s the choice?

        I disagree that the “odds of success” are small. Rather the opposite — your kid is very unlikely, assuming some degree of conscientiousness and drive, to be starving or unemployed. As a father of a 13 year old myself, I encourage you not to let him think that making himself miserable now is necessary for the future.

      • Evan Þ says:

        I’m normally very reluctant to generalize from my own experience, but since you’re so convinced that your oldest son’s childhood being over is a sad necessity, I’ll comment: I hardly studied programming before my junior year of high school, was a complete failure in foreign languages till then, and still got into a good college and graduated with a computer engineering degree. It’s possible.

        Yes, I spent a lot of time studying my junior and senior years; no, I don’t think I could’ve done it without a really good computer science teacher; no, I’m not sure how much the specifics of my personal story generalize.

        So, for stuff that I’m pretty sure does generalize: For Spanish, nobody cared about more than the check-mark on my transcript; if your son isn’t excited about studying it, I recommend not pushing just yet. (If he is excited, what really helped me was good old-fashioned flashcards and repeated writing of vocab words.) For math, most of us software developers really don’t need calculus or higher-level algebra per se, but studying it is valuable for the sake of logic. For computer programming, it took me a semester or so to really get used to wrapping my head around the logical flow of how a computer runs a program – your son should definitely get that before college; it’ll take a whole lot of time at the start (I went to every office hour my teacher had; it helped), but it gets a lot easier after you figure out how to think in that way. If you want more specifics, I’m glad to share.

        But… his childhood isn’t over. There’re good schools that don’t require kids to marathon through the rat race.

      • Unfortunately with the thirteen year old a “happy day” is no longer the goal as he has ambitions to go to U.C. and “study computers”, and now there’s little time for him to do anything except study, and his childhood is pretty much over.

        There should be some things that both contribute to his getting into U.C. and are fun. Some examples:

        There was one school where our home unschooled daughter was accepted (and offered money) where we had no legacy connection. Their admissions person told us that what blew them away was her list of 400 books she had read. All or almost all were read for fun.

        Computer programming is fun and should be well within the ability of a thirteen year old with access to a computer. Including with his application a couple of ingenious and entertaining programs he had written would be a sizable plus.

        The World of Mathematics is a several volume book of essays on various bits of mathematics and math history. It’s fun and interesting, and if your son got interviewed his knowledge of that sort of stuff would be of some value.

        I’m sure there are some things your son will want to do just in order to get into college—our kids spent time preparing for the SAT math exam, math not being a subject of much interest to them. But a lot of it should be stuff he likes doing, both so he won’t be miserable and because you learn better if it’s something you actually want to learn, not just something you are supposed to learn.

        Why U.C., which I assume in your context probably means Berkeley, in particular? It’s a good university, but there are others at least as good.

      • ana53294 says:

        There are several ways to make language learning more fun:

        Find something you want to do in your target language. Here is where English learners have an advantage over learners of other languages, because a lot of very obscure hobbies only have enough following in English.

        Travel.

        Fall in love. I know a lot of people who fell in love during the Erasmus exchange program and ended up learning some obscure European language.

        Of course, number 3 is probably not a good idea for such a young boy, but the first two options are still there.

        There are also ways to make language learning more effective, so even if it’s miserable, it takes less time. Spaced repetition using flashcards is one of the most effectie ways of learning vocabulary (although it’s not pleasant).

      • Winter Shaker says:

        If he wants to actually learn to be able to communicate in Spanish… there is a vibrant internet community based around the task of optimising language learning, and while Evan Thorn’s suggestion of flashcards for vocab is a popular one (provided you use a sensible spaced repetition system), there is a vocal contingent that says you are better off spending the bulk of your time reading and listening to material (or better yet, simultaneously reading and listening to the same material) which is just at or slightly above your level of understanding, and which is actually about stuff you’re interested in. Ages ago fellow commenter Onyomi recommended a site called LingQ which is pretty good for that, and which has helped me with Dutch and Finnish*, albeit it’s a paid subscription service, and luckily Spanish is one of the best-served languages for English-speaking learners, so you should be able to find books of stories for beginner learners that have matching audio – I’m currently making my way through one of Olly Richards’s books for Italian, which seems pretty well-constructed, but I know he has Spanish as well.

        For more on the ‘comprehensible input’ idea, look up Stephen Krashen, who seems to be the leading exponent of this position (although not himself a particularly active member of the online polyglot community)

        Once you have actually got good enough at listening and reading Spanish to get the gist of most of what’s going on, you still have the task of learning to speak it, for which either paid tutors or volunteer language exchange partners are highly recommended … but, apart from making the initial effort to wrap your mouth around the pronunciation of the new language, the recommendation is that it is not particularly worth the bother starting speaking practice until you have internalised enough of the language to have a rough idea of what you want to say without having to look up every other word.

        *Come on, your kid is learning Spanish when he could be learning Dutch or Finnish instead? 😛

      • arlie says:

        Poor kid. I don’t have any idea how low his chances are, in the current economic situation and the one he’ll face throughout his career. I wish I could tell you that you’re being unrealistic, but I just don’t know.

        In my childhood and young adulthood, the door was wide open in general, and if you had computer talent and interest, people would reach out and drag you through it. I graduated from college with negligible debt, and went straight to a job that paid pretty close to what my (factory worker) father had been being paid only 4 years earlier. (He never talked about how much he earned, but I got a look at one year’s financial aid forms.) And then raises and promotions arrived pretty much annually for at least the next five years. (Farther progress took work.) Some of my colleagues in the field didn’t even have college.

        What I hear these days is that all you get with a basic college degree is massive debt, and a chance to compete for minimum wage jobs 🙁 I think that isn’t as true for software engineering, but I don’t know anything about your son’s level of talent or interest. If he’s just in it for practical reasons – i.e. that’s where good jobs seem likely to be – then it’s harder than if he loves the field, or if he has the specific talents that make it easier. (I had both.) But the better paying computer jobs, with opportunities to advance (other than maybe into low level management) generally require a master’s degree these days, and that requires even more time, and even more debt.

        So I happily drifted, studying whatever interested me, basically following my personal “north star”, and the results were excellent – but I don’t know that it’s wise to advise anyone else to do the same, in current decades.

        Good luck to him, anyway. And if he wants to learn Spanish well, the sooner he’s speaking it with real native speakers, the better. IIRC, your location should make it easy to meet native speakers – OTOH, my experience as an English speaking child in Quebec trying to learn French was that the native French speaking children didn’t like Anglos, so hanging out with them was a non-starter, and the same problem may apply for him too.

      • Zeno of Citium says:

        I’m an engineer – not a programmer, but I have programmed from time to time. At the moment, if you can use a computer with above average ability, that’s a huge boon in the job market. If you can program, you can write your own ticket. I expect the demand for programmers to outstrip supply basically forever – most people can’t simply learn to program. It’s not that hard, but it requires a certain mindset.

        Is your son interested in learning to program? That is, not in the career path programming brings, but the pure joy of filling a document with strange symbols and having a computer program come out? If so, I highly recommend he pursue that as his fancy drives him. If he’s only interested in the career path, all he really has to do to is get a degree in computer science. A degree in computer science from even a decent university – much less UC Berkley – will get his foot in the door in industry. I’d suggest taking programming classes in High School, aiming to take AP computer science or whatever the CA version of that is – a High School class that counts as college credit. Also take as much high-level math as possible – trig, algebra, and calculus. Calculus isn’t directly useful, but basic calc requires mastery of algebra and is a good way to make sure he develops that.

        It’s been a long time since I applied to college, but if you’re looking at a school that’s less selective than the Ivies (i.e., not Harvard, Princeton, Yale, etc.) good grades and and good SATs are all you need. Don’t bother with extracurriculars unless he actually likes them (and on the other hand, if he actually likes Theater or Track or whatever, go for it even if it won’t help admissions to schools, life is not a series of checking boxes). No one learns anything in High School Spanish classes, if he isn’t actually interested in foreign languages I’d dissuade him from spending time studying them and suggest he just struggle through two years of Spanish class since a lot of colleges want a foreign language. On the other hand, if he’s one of those lucky people that genuinely likes learning foreign languages, that’s great! Tons of work for someone that knows Spanish in the US too, and plenty of opportunity to speak it in CA.

        Being a programmer is a pretty good lifestyle. You don’t have to lift anything heavy, engineers tend to be smart, interesting people, the pay is great, and the skills are widely transferrable. If your son doesn’t feel any specific calling to any particular career and just wants to be able to make enough money to do whatever he wants in his off hours, or until he decides on something else, programmer is a great choice. Programming is also fun!

        For programming specifically, I’d recommend learning Python if he hasn’t chosen a language yet. Python is a very friendly language, it’s fun to program in, and it’s widely used – possibly the most common commercial language besides Java. Java is currently the industry standard but probably won’t be when your son gets to college. It’s worth learning, but it’s a bit fiddly and not a great first language. At some point, if he’s serious about programming as a craft, he needs to learn C++, but that can come later. If he just wants to get paid, Python as an introduction and then college will teach him whatever the industry standard is in 5 years. Learning more languages is better, but keep in mind his entire career will be learning a new language every 5-10 years. Learning a new programming language is actually pretty easy – a good programming can get up to speed in a language in two weeks and be useful in 1-2 months. Specific languages go in and out of style as we develop better programming languages, so the important thing is the understand the principle of programming, how to think logically, and the basic structure of a programming language.

        I learned Python from *Learn Python the Hard Way*, at https://learnpythonthehardway.org/ . The Python 3 book is $30 and also has some videos with it, apparently. The great thing about programmers is that they love showing off how smart they are, and there’s no better way to do that in the general programmer community than teaching people, so there’s tons of other tutorials out there – just search “how to learn (programming language)” and try some out. Stack Exchange is a good place to search for help – https://stackoverflow.com/ . Kahn Academy or EDx likely have some free courses as well. Once he’s ready, he can sharpen his skills and help out some good projects by contributing to open source projects – https://www.firsttimersonly.com/ has some resources.

    • WashedOut says:

      To raise children that are able to help themselves and others, or, independence + good citizens.

      Happiness and material success are great when they come around, but they won’t always be around and it might will be a long time between drinks. What’s important is that they have the set of capabilities to improve their own life, persist in the face of inevitable suffering, whilst being minimally reliant/dependent on others. From there you’d be free to create your own fun/happiness.

      I might also add a secondary goal of learning from parents’ mistakes and doing a little better than they did.

    • baconbits9 says:

      1. Having a good relationship with them throughout the rest of my life.
      2. Get them to a point of self sufficiency where they see that as a responsibility of their own.

    • Everyone wants to indoctrinate their kids with their values to some extent. Otherwise we would be just as happy with a kid that wanted to genocide Jewish people compared to a normal kid.

    • Nicholas Weininger says:

      Parent of 6 year old here.

      1. Keeping kid on track to develop the endowments, upon adulthood, to choose his own path effectively. This means: good physical and mental health; reasonable level of knowledge and basic skills (diminishing returns set in quickly here); sense of agency, self-worth, self-mastery.

      2. Maximizing probability that kid will leave the world better than he found it. This is where instilling values come in, and in my case there is more than a dash of noblesse oblige: my kid will be a multigenerational child of privilege and reasonably likely to end up in positions of power over others, and my responsibility to those others is to try and make sure he doesn’t abuse that power. So: lessons in generosity, considerateness, peaceableness, respect for consent boundaries, not being entitled or bossy.

      3. Happiness of childhood, as Ozy commented.

    • arlie says:

      Hmm – inserting 0 here.

      0. Keeping them alive, healthy, and happy, to the extent practicable.

      That’s kind of your #3, but different enough that I think it matters. And it easily expands to avoiding major potential effects on their adult lifespan, health, etc. And ‘happy’ doesn’t mean that bad things never happen, etc.; that’s both impractical and undesirable, IMO.

      p.s. No kids here either.

    • Aging Loser says:

      I agree with Plumber, above, that each day your main goal with a small child is simply to keep it alive until the end of the day, and your secondary goal is to keep it cheerful. I think that it’s important for as-yet-childless people to be aware of this, because they tend to psych themselves out of having children by imagining that being a parent is a tremendously technically difficult mission that requires lots of training and special knowledge. In fact, a twelve year old can do it, and in many places (for example, Boro Park, Brooklyn) barely adolescent or pre-adolescent kids seem to care of their toddler-siblings for most of the day. You just keep the child out of traffic, stick food in its mouth from time to time, occasionally change its diaper, and tickle and bounce it — that’s all you have to do.

    • Enkidum says:

      I’m not sure about the order, but like you I would avoid 5 (I am a parent of two teenagers, for what it’s worth). I am comfortable with 4, and explicitly reference it to them trying to produce good outcomes for other people, because I am a utopian socialist kind of guy.

    • Randy M says:

      This reminds me of a discussion I started here a year ago.
      Personally from my own point of view, my priorities as a parent are:
      0-Do not inflict trauma, abuse, or neglect of any sort (this is really baseline, easy to avoid, and anyone who fails here should be judged harshly)
      1-Raise children who do not abuse or take advantage of others.
      2-Children are responsible; as adults are not (net) burdens on others (barring some mental or physical disability which they are lucky enough not to have), act with integrity, behave honestly, rectify mistakes, etc.
      3-Children carry on, to some extent, the faith of their parents.
      4-Children are on net happy; know how to set goals & delay gratification; express gratitude; have perspective and are not envious; are resilient and can cope with stress and setbacks (I consider these behaviors are requisite for long term happiness or satisfaction)
      5-Children marry and have a child or children of their own with parenting at least as good as they got. They marry wisely and their marriages endure.
      6-Children maintain good relationships with their parents and each other
      7-Children are healthy and have habits and knowledge that maintains their health
      8-Children are actively pro-social; friendly, help the needy, protect the innocent, etc.
      9-Children are cultured and learned
      10-Children are financially successful

      Exact rankings may shift a bit with more consideration. And then, regardless of how important a goal is, my ability to effect it varies a lot too. And they are interdependent; happiness probably requires some level of financial success, for example.

      • Aging Loser says:

        It seems impossible to help a child get into the right groove for (5 — marrying, having kids) and (10 — being financially successful) because everything’s changing so quickly. It seems as though you have to use websites to date anyone and get jobs and probably those are never the same websites from year to year and the things that you have to type on the various lines of the various forms in order to level up probably also change from year to year. There are probably higher-level websites telling you how to find the appropriate lower-level websites and deal with them correctly, but those higher-level websites are probably very quickly obsolete as well.

        • Randy M says:

          It seems impossible to help a child get into the right groove for (5 — marrying, having kids) and (10 — being financially successful) because everything’s changing so quickly.

          Maybe! That’s part of what I was alluding to by pointing out that the ranking is by how important it is rather than how achievable it is.

          Theoretically effort expended should be some function of the importance and the efficacy.

          There’s got to be some people who meet face to face, these days, but even after arranging hook-ups on the hook-up app of the 2020’s, interpersonal skills and values are going to effect likelihood and permanence of marriage.

          Regarding financial success, that involves educational achievement, setting goals, choosing careers wisely, making a good impression, being reliable, etc., which will be things a parent can train or at least advise on regardless of how entry level positions are made available.

      • hoof_in_mouth says:

        This is a good expression of goals, thanks.

    • Conrad Honcho says:

      4, resulting in 2 and 1 and therefore 3 and therefore 5.

    • hoof_in_mouth says:

      I have kids (18,15 and 12 week). I had a happy childhood and revere my parents, but am divorced from my older children’s mother and fairly estranged from them at (what I perceive to be) her behest and I have no illusions that they have such opinions of me or my values. You can want things for and model things for your children but they have their own ideas of what is important. Some kids will take it to heart, some will be indifferent and some will oppose everything you hold dear and that is out of your control. Model the behavior your want, show them what needs to be done and how it is done, punish aberrant behavior in appropriate ways and hope that your child wants to emulate you.

      1) Outcomes for them – Bourgeoisie values of law-abiding, self-supporting, gratification-deferral, value-building
      2) Outcomes for me – they stay somewhat in my life in a positive way
      3) Everything else

    • HeelBearCub says:

      We have two kids, one just out of undergrad, applying to grad school, and another half way through undergrad. We also have another young adult living with us as he tries to get on his feet, and a fourth who was living with us and moved out once they became reliably employed, and we have had numerous others in the circle actively seeking us out for parental type advice.

      Here is the one simple rule that I used to guide me. It’s also what I consistently told the kids about how I viewed my role as a father:

      My job is to help you build the skills needed to live a truly happy life.

      That’s it. That is all. That is enough. But by no means is it simple.

      • Here is the one simple rule that I used to guide me. It’s also what I consistently told the kids about how I viewed my role as a father:

        My job is to help you build the skills needed to live a truly happy life.

        That’s it. That is all. That is enough. But by no means is it simple.

        I agree with most of that, but I think your job also includes helping them live a happy life during the time they are with you. That is, after all, about a quarter of their lifetime.

        Part of my preference for unschooling is based on the idea that it works better, produces happier, better educated and more productive people than the alternative. But part is on my memory of spending a sizable part of my childhood sitting in classes being bored–and I had it relatively easy compared to modern children who also have to spend a lot of time sitting at home being bored doing homework that is of no interest to them and dubious use.

    • aristides says:

      I don’t have kids, but am trying. My list of priorities:

      1-Physical Safety (Probably so obvious it doesn’t need mentioning, but can’t hurt)

      2-Make sure kids have good outcomes for them (Research suggests there is little I can do to help long-term outcomes, but whatever there is, I’ll do)

      3-Give kids good values (Religion in my case, which also helps with meaning of life)

      4-Happy childhood (Probably the one thing on this list I have the most control of)

      5-Good outcome for me (Stay in contact, and I Really want grandchildren, but my best to help with that is my plan to have 3-5 kids, odds are one of them will have children)

      I don’t have concrete knowledge or skills on the list, since I only consider them a means to an end, and don’t have tools to deal with/find meaning in life, because my only method is religion, and if my child rejects that, I am completely unqualified to help them. Hopefully they will meet some rationalists or EAs that can help them find meaning.

  11. Odovacer says:

    What would it take for you to trust a robot nanny with your child? This robo-nanny would take care of you newborn through 10 year old child when you’re busy. This includes feeding, burping, changing, washing, supervising, etc. What sort of criteria would satisfy you that the robot wouldn’t accidentally kill or hurt your child?

    • Hoopyfreud says:

      Not that I have an answer, but interesting related question – would you trust machine learning here? Or would you prefer to stick to proofed-out code?

      • Mr. Doolittle says:

        Oh, I like this question.

        The machine learning would have to be a LOT more advanced than anything we have now for the child not to game the system and train the robot to do counter-intuitive things. Or worse, the robot would learn to do things like giving them candy and cake instead of vegetables because that “soothes” the child. I’d love to see how even an advanced robot dealt with temper tantrums that looked like a child legitimately upset or hurt.

        Of course, with set rules, the rules would have to be unbelievably thought out and inclusive or the child would game those as well. I can imagine the rules for a <1-year-old being programmable, but a 10-year-old would figure out every nuance and trick available in the programming to pretty much do whatever they want. I am quite confident in asserting that such a program would essentially mean that the child is raising themselves with a fancy smartphone available. Maybe not all kids, but certainly quite a few.

        If I had confidence in the programmer/programming to be done well? I would think a learning system would be a necessity. Kids adapt too quickly to involve a system that can't also adapt at all. If I didn't have much faith in the programming? Then a mediocre rules-based system would be preferable to a mediocre learning system. At least you can have a better sense of the limitations of the rules-based system, rather than finding out that your child essentially re-programmed your fancy robot to be their slave.

        • Jefferson says:

          I don’t think that giving a child candy is a symptom of the problems with machine learning so much as a that the wrong things are being optimized for.

          • Mr. Doolittle says:

            Well sure, but that was just a handy example that we happen to see biological parents doing, so hopefully the reader intuitively understands my underlying point.

            Presumably a real robot AI would be coded very strongly to regulate a diet such that each child falls within a certain healthy range of foods. Let’s run with that a little further and more complex. Let’s say that the AI wants each daily intake to fall within a certain range on a variety of vitamins and minerals. The child, through argumentation and experimentation, discovers that the robot accepts a very repetitive diet of one or two foods, plus something the kid really likes (let’s call it cookies). The kid doesn’t have a clue why this is acceptable, but it works. The AI, through experimentation with the child, discovers that the child will eat these “meals” – which do provide the necessary vitamins and minerals – but would not eat greater variety. The AI “learns” that this is the optimum meal and provides this meal all the time. Maybe this really is nutritionally acceptable, but a human parent may very quickly recognize a problem – the child is unwilling to try any variation. As an adult, this child may have significant problems finding appropriate food, especially if visiting others or going out to eat, and therefore struggle with socialization. Humans also seem to have a lot more skepticism about repeated bland diets. Sometimes eating the same thing over and over again has unintended consequences. I read a story about a woman who got cancer from eating some incredibly high amount of microwave popcorn every day. Even hooked up to Google, the AI would not have recognized such a thing, because no one had ever noticed it before (because nobody ate 20 bags of microwave popcorn every day for years before).

            I have very little faith that an AI, learning or not, will be able to avoid such situations.

        • AG says:

          I feel like a simpler set of strictly-adhered to limitations might work better. Kids can get upset about a lot of things and be fine without ever been soothed by the robot. There’s no need for the nanny to maintain the kids’ behavior within a certain band, only that they patch them up when they get hurt and provide necessities like scheduled healthy meals/snacks.

          If the kids insist on hurting themselves in such situations (such as choosing not to eat the provided meal), that’s on me to teach them not to do that, and then on the kids themselves to choose not to do such things. It’s not the robot’s responsibility.

          Better for teaching the kids to be more independent, too.

          • Nancy Lebovitz says:

            If you want the robot to do better than a lot of parents, then the robot should be able to tell if the generally healthy food is actually bad for a particular child.

          • AG says:

            Ah, I misread the OP. I was thinking of nanny more in the “babysit for a day/evening” situation, rather than “I will be busy most of the time so the nanny will spend more time with the kid than I do.”

            To which my answer is “through 10 years old is way too long.” Kids can start having decreased supervision as soon as they start school, pretty much.

            As for the food aspect, a healthy menu can be pre-prepared and pre-programmed in (as a person able to afford a nanny robot can certainly afford food prep staff/robots), and the robot’s duty there would be to monitor for poor health reactions like allergies or nausea. That’s just advanced fitbit stuff.

      • Aging Loser says:

        Elderly Chinese ladies roam the streets of Brooklyn after dark or before dawn, picking up bottles for the 5 cents they get for them by sticking the bottles into machines outside of supermarkets; they push mobile towers of bottle-filled mega-sized garbage-bags. It would be far easier and safer to make mech-suits for elderly Chinese-lady-nannies than to make robotic nannies. I’m quite sure that an elderly Chinese lady in a mech-suit would be far superior, as a nanny, to a robot.

    • Paul Zrimsek says:

      I would insist on one that doesn’t follow Asimov’s 2nd Law.

      ETA: “A robot must obey orders given it by adult human beings except where such orders would conflict with the First Law” is OK too.

      • FLWAB says:

        In Asimov’s novel The Robots of Dawn the people of Dawn are all aristocrats and all the work is done by robots, including raising children. They are forced to deliberately weaken both the Second and First Laws significantly in their nanny bots. The Swcond for obvious reasons and the First because otherwise the robots can’t stand to let the children suffer in any way and they grow up completely spoiled. (normal robots can’t stand up to temper tantrums and screaming: it certainly seems to them that the children are suffering. Human parents sometimes have he same problem, so who can blame them?)

    • jaimeastorga2000 says:

      A service record of several years during which robot nannies have been operated by a large costumer base without a higher rate of death or injury than human nannies. Same thing that would make me trust a driverless car with my life, in other words. Let technophiles and early adopters be guinea pigs if they want.

      • quanta413 says:

        I agree with this criteria. I appreciate the rough spots and risks early adopters are willing to deal with! I just don’t want to be in the position of early adopter in this particular case.

    • Well... says:

      By the time I could afford a robot nanny it would be mature enough technology that it’s less likely to kill my child than a human nanny is. Same for most definitions of “hurt” although I’m curious what the psychological and social/cultural implications of being raised by a robot nanny will be: Will my kid turn out weird as a result? Might my kid end up with strange values that I can’t relate to? Assuming the technology is observably “safe”, as is likely, then these other questions are the real thing I’m concerned about.

    • Mr. Doolittle says:

      Because of the emotional attachments involved, I think that anything less than full human adaptation (meaning Blade Runner levels of ability) would be insane for a “nanny” bot.

      That said, a robot that can change a diaper might be handy with a lower bar for acceptance. Such a thing sounds to me like a safety liability on top of an expensive solution to something that isn’t a problem, so probably not that helpful.

      I wouldn’t let a robot take care of a child without a human nearby, and those early interactions (feeding, changing, burping) are great for both parents and babies to acclimate and build relationships. Speaking from when I was a new father, I didn’t know what to do with a baby, so those activities provided a nice baseline for involvement.

    • Nabil ad Dajjal says:

      It would have to be as good at at child-care than my father and in-laws, if not better. Even then it would need to be at least as strongly motivated as a blood relative.

      I’m not against nannies or robot nannies, because the free alternative available to most people isn’t going to be multi-generational families but neglect. But while it’s a good technology to develop it’s still inferior by its nature to having your retired grandparents watching the kids rather than slowly dying in a nursing home surrounded by strangers.

      • albatross11 says:

        I think when your aging parents are capable of watching your toddler for several hours a day, they’re still quite capable of living on their own, so there’s no reason for them to be in a nursing home. By the time they’re in the “warehouse them till they die off” stage, they are no longer healthy enough (or perhaps mentally together enough) to be left in charge of small children.

    • LesHapablap says:

      Robot nannies seem very dystopian. A race to the bottom in which only the rich will be able to afford to take time off work to raise their own kids.

  12. Le Maistre Chat says:

    Who is Oberon?
    Wikipedia asserts that the name, of French origin, is a variant of the German Alberich (“elf ruler”, from the elements alb- and rich), a character in the story of the Nibelungen. At any rate, the form “Oberon” is first attested in a chanson de geste (fantasy story starring Charlemagne and his barons) called Huon de Bordeaux. He fills the archetypal folk tale function of “supernatural aid” for the title hero, and explains that he’s King of Faerie through his mother, Morgan Le Fay, while his father was Julius Caesar (as you might have guessed, this makes King Arthur, who was taken from Britain to Avalon rather than dying, heir to Faerie).
    Oberon next appears in Shakespeare’s A Midsummer Night’s Dream, which as far as I know is the first appearance of the character Titania, queen of fairies. We find them in Attica during the reign of King Theseus of Athens, which contradicts the chronology of the original story (bear in mind that this is the sort of story where Theseus can snark “St. Valentine’s is passed” even though he’s far future). Titania was subsequently added to the cast of “Matter of France” operas when Oberon appeared, cf. Huon und Amanda or Oberon in German and Holger Danske in Danish. They also appear as a couple in Goethe’s Faust Part I, which confusingly has them celebrating their golden anniversary in the 16th century AD.

  13. Machine Interface says:

    Movies through which you’ve discovered/become quite fond of instances of art music (as in, classical music and adjacent things, western and non-western), be it a piece, a composer, or even a genre.

    2001: A Space Odyssey (1968, Stanley Kubrick): the micropolyphonic music of Ligeti, with the pieces Atmospheres, Requiem and Lux Aeterna heard throughout the film.

    Akira (1988, Katsuhiro Otomo) > discovered Gamelan (the classical music of Indonesia, for short) with this film.

    Amadeus (1984, Miloš Forman) > Mozart’s Requiem, of course.

    Army of Shadows (1969, Jean-Pierre Melville) > Morton Gould’s Spirituals for Strings Choir and Orchestra.

    Barry Lyndon (1975, Stanley Kubrick) > Händel’s Sarabande.

    Excalibur (1981, John Boorman) > several themes by Wagner, but most notably the Prelude of Tristan und Isolde (later also used in Lars von Trier’s Melancholia).

    Fantasia (1940, multiple directors) > almost all the music used is great, but I’ve been particularly enamoraed with Stravinsky’s Rite of Spring since I’ve first seen this.

    A Man Escaped (1956, Robert Bresson) > the Kyrie from Mozart’s Great Mass in C minor.

    The New World (2005, Terrence Malick) > Prelude from Das Rheingold, by Wagner.

    Photographing Fairies (1997, Nick Willing) > Allegretto from Beethoven’s 7th Symphony.

    The Shining (1980, Stanley Kubrick) > Penderecki’s sonorist composition, as well as Bartok’s Music for Strings, Percussion and Celesta.

    Woman in the Dunes (1964, Hiroshi Teshigahara) > the haunting music of Toru Takemitsu.

    • actualitems says:

      The Tree of Life (2011, Terrence Malick) > I discovered composer, Zbigniew Preisner, more specifically his work Requiem for my friend, more specifically “Lacrimosa” from that work

    • Hoopyfreud says:

      Memoirs of a Geisha (2005, Rob Marshall) was my introduction to East Asian-Western “fusion” instrumental music, and it still holds up in that genre.

      The Fountain (2006, Arnofsky) was my introduction to Clint Mansell and Mogwai, and I’ve come to be fond of both.

      Blade Runner (1982, Ridley Scott) was my introduction to the music of Vangelis, whose work I think counts as “art music.”

    • knockknock says:

      Rollerball’s (1975) all-classical score makes particularly good use of Toccata and Fugue, plus plenty of Shostakovich

      And don’t forget Rabbit of Seville, Bugs Bunny’s classic 1950 take on Rossini

    • skef says:

      Koyaanisqatsi (1982, Godfrey Reggio) Generally speaking I more of a fan of Steve Reich than Phillip Glass, but this soundtrack is among the latter’s best works.

      Added: I like good “atmospheric minimalism”, but have a hard time finding it. Cliff Martinez’s soundtracks are among my favorite works of that type. Some of the tracks on the Sex, Lies, and Videotape are really very good. And most of the “Drive” and “The Neon Demon” soundtracks, where he composed for/collaborated with analog synth experts, are really worth listening to on their own merit. (Strangely, “U Smile” slowed down eight times also fits into this genre, although not under your category.)

    • C_B says:

      Sunshine (2007) – John Murphy can’t always decide if he’s writing classical music or post-rock, but I’m there for it.

    • Bugmaster says:

      The Fountain (2016, Darren Aronofsky), featuring excellent tracks by Clint Mansell, such as Death is the Road to Awe. Well, plus any Darren Aronofsky/Clint Mansell/The Kronos Quartet collaboration, really.

    • Kuiperdolin says:

      Silence of the Lambs (Jonathan Demme, 1991) : The Goldberg Variations, by Bach

    • RDNinja says:

      Die Hard and Beethoven’s Ode to Joy? It’s pretty famous besides, but it was notably prominent in the movie.

    • knockknock says:

      Not quite to the original question, but you could be turned on to classical music in general by vintage movie composers such as Korngold, Rozsa and Herrmann. They didn’t reference specific classics so much as become classics on their own merit.

      I got a snootful of that stuff as a movie-crazy teen, so maybe it planted the seed for becoming a classical music fan.

      One more note re movies about music: Not to be missed is the hilariously melodramatic Deception (1946), with Bette Davis and Paul Henreid as orchestra musicians and Claude Rains as their imperious conductor, all locked in a stormy triangle.

      • AG says:

        Very very co-signed on this.

        Erich Korngold, Bernard Herrmann, Mikos Rozsa, John Williams, Franz Waxman, Jerry Goldsmith, Joe Hisaishi, all great stuff. Ironically, I recently had a discovery opposite of the direction OP asked for, where I picked up a CD of movie music, and it included a great video game Star Wars track by by Joel McNeely, which led to me finding that entire soundtrack for Shadows of the Empire game is really good stuff.

        The most common orchestration for the big Zelda Theme steals a lot from Williams’ Superman march.

    • Deiseach says:

      Nosferatu (1979, Werner Hertzog) – the Georgian traditional song Zinskaro/Tsintskaro (there are several variant spellings). Film version is by Vocal Ensemble Gordela and is beautiful, even more gorgeous version by famous Georgian singer Hamlet Gonashvili and ensemble Rustavi.

      (Also used on the track “Hello Earth” from Kate Bush’s 1985 album Hounds of Earth).

      Tous les Matins du Monde (1991, Alain Cordeau director) – the music of Marin Marais.

    • LesHapablap says:

      Pi (1998, Darren Aranofsky) Aphex Twin and what used to be called IDM

      A Clockwork Orange (1971, Kubrick) Various

    • gbdub says:

      Master and Commander had some lovely violin+cello/small string ensemble arrangements. The soundtrack has some good pieces, but there’s also a couple volumes of classical music released as “Musical Evenings With the Captain”, not related to the movie directly but just classical music designed to be listened to while reading the Aubrey/Maturin series.

    • Paul Brinkley says:

      It’s television, not film, but one thing I will always be grateful for is The Smurfs, for popularizing several of Mussorgsky’s Pictures at an Exhibition, and for the third movement of Beethoven’s Moonlight Sonata – the most difficult piece I ever fully learned to play on a piano.

      • knockknock says:

        There was an ad for the 1972 horror flick Asylum that evocatively used a scary little passage from Pictures at an Exhibition. And the self-explanatory Mephisto Waltz (1971)

        But wait — I forgot another music-movie gem — The Beast With Five Fingers (1945) features the severed hand of a murdered pianist that crawls around choking the dead guy’s enemies. It also stops to play Bach’s chaconne for the left hand!

    • j1000000 says:

      I rewatched Ocean’s 11 a few years ago and they play Clair de Lune by Debussy at the end. I had pretty much never listened to any pre-1960s music for pleasure, but I really loved Clair de Lune. From there I kept moving back/expanding little by little, and I like a bit of Bach, Mozart, Beethoven, etc now.

      The other one I can think of is Brahms’ Violin Concerto at the end of There Will Be Blood.

    • jaimeastorga2000 says:

      Excalibur (1981) was the first time I heard Carl Orff’s “O Fortuna”.

      The End of Evangelion (1997) has a memorable scene where Asuka fights the Eva Series to the tune of “Air on the G String”.

    • AG says:

      Heh, Mozart’s Requiem was also used to hilarious effect in The Incredibles. But for Amadeus, the big impact was Serenade for Thirteen Wind Instruments, the piece that bowls Salieri over after being gobsmacked by his buffoonery. The older Salieri describing the piece in the asylum illuminated it so much for me.

      I have a great love for the anime soundtrack composers. They do the best classical music work in the world right now, imho, but there are plenty that put out great non-classical stuff, too.
      Gatchaman Crowds alerted me to Iwasaki Taku, which was then reinforced by his brief stint on Jojo’s.
      Yoko Kanno needs no introduction, of course. There’s obviously her Cowboy Bebop stuff, but my first introductions to EDM was through her Ghost in the Shell stuff.
      Sound of the Sky introduced me to Kubota Mina.
      Sound Euphonium alerted me to the piece Takarajima, and how ridiculously popular it is in the Japan wind ensemble scene.
      Beck: Mongolian Chop Squad was pretty instrumental in shaping my opinions on rock music, largely in making me always pay attention to the bass part.

      Kind of cheating, but the musical film Bandwagon introduced me to some great names. First of all, it’s actually kind of a jukebox musical, for the songs of Dietz and Schwartz (in the way that Singing in the Rain was a jukebox musical for Arthur Freed). The arranger for the film was Conrad Salinger, and there’s a great story about how he sat down with one of the singing stars, Nanette Fabray, and figured out what her best singing pitch was, and then arranged her big showcase number (Louisiana Hayride) to finish on a long belt of that pitch. Plus, his arrangement for Dancing in the Dark just can’t be beat. And though they weren’t as involved with the music side for the film, it also introduced me to Comden and Green, whose work I would love in It’s Always Fair Weather, and who worked on On the Town.

      For video games, so many have gotten into classical music because of Nobuo Uematsu, Yasunori Mitsuda, and Koji Kondo.

  14. johan_larson says:

    “Take me down to the paradise city, where grass is green and the girls are pretty, ” sang the singer. But that lyric fits an awful lot of places. Your mission, should you choose to accept it, is to find the city it fits best.

    • The Nybbler says:

      New York, I’m afraid, is right out on the grass front. Ireland is certainly known for green grass and pretty girls (or is it bonnie lasses?), but I’ve never heard an Irish city described as paradise. There’s an actual Paradise City train station in South Korea, but I’m afraid it fails on the grass front; it’s at an airport, so the girls are just passing though. Places described as “paradise” tend to be beach or mountain. Beaches aren’t known for their grass (except on golf courses). Maachu Picchu fits mountain and has grass, but no girls. Katmandu has that paradise vibe and is inhabited, but not so much on the grass.

      Huh. Hard to get all of “city, paradise, green grass, pretty girls”. At least with substantial cities.

      • baconbits9 says:

        Huh? The answer is San Diego.

      • Aging Loser says:

        Moving south through Brooklyn — there’s Fort Greene Park, Prospect Park (huge), Green Wood Cemetery (huge), Sunset Park, and Owl’s Head Park. The problem lies in coordinating the prettiest girls — 16-to-17-year-old Puerto Ricans, in my opinion — with the grass. This does happen in Sunset Park, to some extent (although more along the walkways in between the grass than on the grass itself) but not so much in the other Parks I’ve listed. Still, on a sunny warm day you’ll see a lot of quite attractive professional-class 25-year-olds in Prospect Park — often draped nearly naked across the grass.

      • johan_larson says:

        I think New York is out of the running. “Paradise” just has to have better weather than New York winters.

        How about Miami?

    • Chalid says:

      To qualify, the city has to get lots of rain, and also be warm (and preferably with a beach) so as to encourage skimpy clothing, and fairly rich, and have lots of open space. There is some tension between “lots of rain” and “skimpy clothing” obviously. My first thought was Honolulu, but it turns out to be semi-arid due to being in the island’s rain shadow. My second thought was Rio but it is also drier than I expected.

      Perhaps Miami Beach? I’ve never been there but it gets plenty of rain, it’s a famous resort town so some people must think it’s paradisiacal, it has lots of beautiful people (or so popular culture tells me), and a quick look at Google Satellite View shows lots of grass.

      • A Definite Beta Guy says:

        Perhaps Miami Beach? I’ve never been there but it gets plenty of rain, it’s a famous resort town so some people must think it’s paradisiacal, it has lots of beautiful people (or so popular culture tells me), and a quick look at Google Satellite View shows lots of grass.

        South Beach is definitely awesome, but it definitely has little grass. There are a couple of teeny tiny parks and some golf courses. Flamingo Park seems more “here is where the kids play soccer on Sundays” and less “rager.”
        I do like South Beach, though!

    • Fossegrimen says:

      Let’s solve for pretty girls first; You want either a place that is a disproportionate source of models or somewhere the models keep ending up.
      So your set of cities will be Stockholm, Milan, LA and a few others
      Then you evaluate the greenness of the grass and how important that is to you and choose accordingly.

    • SamChevre says:

      I’ll argue for Nashville. Enough rain to be green, nice enough weather to see the pretty girls, and a disproportionate number of pretty girls because of the music industry.

    • dodrian says:

      Katy Perry seems to have narrowed the search down to California, insisting that “the grass is really greener”, and that the “gurls” are unforgettable, in the attractiveness sense. I’m not sure which city in California though, and obviously some aren’t as green as claimed. Perhaps we could start with a list of which cities have the least onerous water restrictions?

      • johan_larson says:

        How about Monterey? It’s on the ocean, the topography is quite rugged, and it’s in the northern part of the state so it doesn’t get too hot. Maybe a little short on grass; it’s more about ocean and forest.

        • Chalid says:

          California generally is short on green grass, especially in these drought years.

        • Randy M says:

          Monterrey really doesn’t get too hot. Vacationed there this summer and we were wearing jackets and long pants. In California! Apparently the depth of the bay keeps the weather temperate.

          I don’t recall how pretty the girls were; nothing outstanding, probably.

          • johan_larson says:

            I don’t recall how pretty the girls were; nothing outstanding, probably.

            Maybe they’re better across the bay in Santa Cruz.

      • gbdub says:

        Per the Beach Boys, the cutest girls in the world are “back in the [United] States”, but they are able to find attractive girls throughout the country, such that they “wish they all could be California girls” (i.e. that all the variety of U.S. girls could live in their home state).

        While Katy did kiss a girl, and like it, based on the original prompt we’re going to weight the opinion of heterosexual males a little higher here.

        So bottom line, anywhere with nice lawns in the U.S. will be good. I lean more toward Florida here, or somewhere else warm on the south east coast. Most of CA is too dry.

        • Randy M says:

          That’s a bit of an out of date reference, to be honest.

          • gbdub says:

            The GN’R song this whole thread is based on is 31 years old, classic rock and oldies are fair game.

            “Paradise City” (1987) was released closer to “California Girls” (1965) than to “California Gurls” (2010), and obviously Axl Rose could not have been influenced by the opinion of Katheryn Elizabeth Hudson, who was 3 at the time.

          • Randy M says:

            True, but the desirability of pretty girls is timeless; the location of the cutest girls is subject to change.

          • gbdub says:

            Are you saying America does NOT have the cutest girls? Comment reported for obvious Russian troll.

          • Randy M says:

            Let’s just say I’m willing to review the evidence again.

        • Winter Shaker says:

          If we are being picky, we could narrow our search to northern California where the girls are warm.

          That said, this is all starting to sound like one of those terrible triangles:
          -Paradise
          -Has green grass
          -Has pretty girls
          Pick two.

        • Vorkon says:

          For what it’s worth, when I was little and heard “Drift Away” for the first time, I thought it was saying “give me the Beach Boys and free my soul…” and for the longest time I thought that’s what the lyrics actually were.

          I didn’t really understand why this guy liked the Beach Boys so much, or why this song that he apparently wrote as a tribute to them was in such a different style.

        • acymetric says:

          Apparently I misunderstood this song all this time. I thought the point was that all these different types of girls are great, but California girls are the best so he wishes they were all more like California girls. Not that he wishes all these types of girls resided in California.

          • dodrian says:

            That was my interpretation as well.

          • gbdub says:

            Nope, he says some good quality about girls from every region of the US (except the Southwest, for some reason), then favorably contrasts girls from the States (not state) with all kinds of girls from all around this great big world. At no point does he compliment California girls specifically nor state that any type of US girls are superior to others.

            Supposedly Brian Wilson came up with the song on his first acid trip.

          • Paul Zrimsek says:

            The Southwestern girls may have been lumped in with the ones in California, since the Southwest also has the sunshine and the girls all get so tanned.

    • Hoopyfreud says:

      A quick google seems to suggest, based on various values of “paradise” and “pretty,” Moscow, Singapore, Vienna, or Shenzhen. I’ll also throw a word in for New York; it surely snows in paradise. Though I doubt the paradise summers are so miserable.

    • Telminha says:

      My guess is that the city it’s Rio, Brazil. It has the world’s largest urban forest and The Girl from Ipanema — “she’s tall and tan and young and lovely,” sang the singer. Also, you can stay at a beachfront historic condo building called “Paradise,” in Copacabana beach; not too far from Ipanema. There are several hotels/motels called “Paradise” and a “Paradise” neighborhood. The Wall Street Jornal recently called Rio a “paradise,” but there is trouble in paradise, unfortunately.

    • Dack says:

      Axl Rose claims the chorus refers to the midwest. Since he is from Indiana, I would guess one of the college towns near where he was raised: possibly Lafayette, Bloomington, or South Bend

      • acymetric says:

        Also worth noting that it says “where the girls are pretty” not “where the prettiest girls are” which loosens that requirement almost to a non-factor. Also, pretty sure “paradise” was not meant in the sense of a Sandals resort commercial, although maybe pointing that out is against the spirit of the OP.

        Also, the city is a paradise because the grass is green and the girls are pretty. Paradise is a conclusion about the place, not a descriptor. So the range of candidates is fairly large.

      • Tarpitz says:

        Axl’s information may be long out of date, though. Mary Jane – seemingly a typical example of the Indiana girls in question – moved “down here” (presumably Los Angeles) no later than the summer of 1993, per Tom Petty. Perhaps Axl’s youth was the last time such a place existed outside of song, and we are indeed now stuck with Winter Shaker’s triangle.

  15. Nicholas Weininger says:

    What are some underappreciated mechanisms for separating coaching and evaluation responsibilities? What are some cases in which there is a good prospect for achieving a much greater such separation in the future than we now have? What success cases of separation are most worth learning from and generalizing?

    There are several common professions in which these are typically combined, schoolteacher and corporate manager being probably the most obvious. Arnold Kling has pointed out in the past that separating the two in education could improve outcomes through division of labor and specialization. I also suspect it would be effective at reducing abuses of power: coaching done well typically involves a close bond which, combined with the power relationship created by evaluation, can easily produce a strong temptation to corruption. So where should we look for frontiers of improvement here?

    • LesHapablap says:

      Flight training comes to mind. In New Zealand the training is done by one entity, exams are created and given by another, and oral exams and flight tests done by a third entity. All three entities are monitored by the regulator.

      If evaluation and training were done by the same organization you would end up with a lot more fake pilot licenses in third world countries.

  16. Atlas says:

    Career advice/philosophical reflection question:

    So, I am nearing the end of my undergraduate studies, and thus having to think more seriously about finding a career. One possibility is law. However, in addition to the many other reasons why one might not want to pursue such a career, I have a personal bias (or set of biases) that might be unfounded. I would thus like to see if people think my impression is accurate or not.

    Namely, the study of law qua law seems to me to be the study of relatively uninteresting and unimportant questions. In particular, questions of constitutional law, which is what I’ve been most exposed to in various social studies/government/political science courses, strike me as obviously less interesting and less important than the positive questions of social and natural science and normative questions of philosophy that are pertinent to them.

    For instance, in a 1st Amendment class, admittedly the only class specifically on law I’ve taken so far, we discussed the case Edwards v. Aguillard. And I couldn’t help but think that lawyers really have the least interesting contribution to make in the conversation that the issues of the case raise. Namely: Is the theory of evolution through natural selection more or less valid than the theory of creationism? What is the proper boundary, if any, between the religion of the majority and the state? How does one most effectively convince believers in creationism to instead believe in evolution through natural selection?

    Compared to these questions, I find the question of “Does the First Amendment prohibit this state law?” rather dull. Furthermore, it seems like a necessarily fairly trivial question to me, because, whatever the decision of the Supreme Court, if the people/interests who control the Louisiana state government do not substantively agree with the Court’s ruling, there are hundreds of ways that the state government could in practice sabotage the decision while nominally following the law.

    So, if I really cared about the question of teaching evolution vs. creationism in schools, it seems to me that becoming a lawyer who specializes in the relevant legal questions would be a relatively boring and ineffective way to promote the cause I cared about. I would rather, say, become a biologist whose research produces further evidence for or against the theory of evolution; a journalist/pundit who writes books and articles making the substantive case for my favored position; a psychologist who studies how to effectively change people’s minds on the issue, etc.

    In other words, it seems more interesting and important to me to debate what the law should say than what the law does say, and it seems to me that the former debate can only be moved forward and resolved by people who are not (at least solely) lawyers.

    Additionally, while my experience here may be unrepresentative, as I am mostly thinking of people who are politicians, pundits and/or journalists, I find that I am generally very underwhelmed by the lawyers/law professors whose arguments I read/listen to in media. (Of course, that makes me, maybe unwarrantedly, more optimistic about the possibility of succeeding in the profession.) A counterexample to this general trend would be Glenn Greenwald: Whenever I read Greenwald’s writing or hear him in an interview/debate, I am always impressed by the cogency of his reasoning, his ability to identify and latch on to the central points of contention rather than shuffling around with peripheral semantic trivialities, his command of the relevant evidence and his bulldog-like tenacity in defense of his beliefs. (Some other examples of people I find similarly impressive would be linguist Noam Chomsky and the late economist Milton Friedman—as well as valued SSC commenter David Friedman, who, though a law professor, was not, as he proudly proclaims, formally educated as such.)

    If this seemed to me to be representative of the typical intellectual and personal quality of people with a JD, I would be intimidated by but more interested in the prospect of becoming a lawyer. However, I must say that this is sadly the exception, at least among lawyers who speak and commentate publicly. (I would like to provide direct links, but that angers the eldritch gods that control the spam filter, so hopefully you can find what I’m referring to easily enough with Google.) Some random examples:

    —Famed legal scholar Cass Sunstein, one of the most widely cited and well respected law professors in the country. I listened to Sam Harris’ ~1.5 hour interview of him from last autumn, and, I was underwhelmed by his commentary on various social, political and legal issues. I agreed with some of his contentions, disagreed with others, but I was just surprised, as an outsider to the field, that this guy is considered to be one of the most brilliant legal scholars working today.

    Furthermore, some of his ideas have been dismissed by commentators I greatly respect. Nassim Taleb has harshly criticized the central thesis of the “nudge” book Sunstein co-authored with eminent economist Richard Thaler. Steve Sailer quite amusingly mocked Sunstein’s idea to defeat conspiracy theorists through a secret government conspiracy to send agents to “cognitively infiltrate” conspiracy theorist groups and make up conspiracy theories.

    —Justice Sandra Day O’Connor’s statement in the Grutter ruling that affirmative action would be unnecessary to obtain racially proportionate law school admissions in 25 years from the year of the ruling (2003). This is/was obviously a ludicrous contention to anyone familiar with even the most basic relevant evidence, and I find it frankly astounding that, after presumably careful deliberation of the issues in the case, Supreme Court Justices can endorse it. Of course, given that this deals with the hot button issue of race, Justice O’Connor may simply have been either actively dishonest or deliberately ignorant. In either case, I am not impressed.

    —Relevant to the above example, there was a brouhaha recently when Professor Amy Wax, of the UPenn law school, commented that she had rarely seen black students graduate in the top half of the class. While again the particularly touchy nature of the subject is relevant, I nonetheless was shocked by how transparently illogical and/or obtuse the responses to Wax’s comments by law professors, like that of UPenn’s law school dean, were.

    —Moving away from the issue of race, I can’t help but roll my eyes at former Justice Anthony Kennedy’s oft-quoted statement about the “mystery of life.” (Granted, it may be so often quoted in part because it strikes observers as ridiculous.) David Brooks dunked on it pretty effectively, something I don’t often say.

    —Michael Avenatti, who graduated first in his class from GW law school, lawyer of adult film star Stormy Daniels, was interviewed by Fox News’ Tucker Carlson. I thought that Carlson, who I don’t think has a JD, clearly outmaneuvered Avenatti in the interview. (Avenatti’s tactic of repeatedly saying in outrage “How dare you, sir?” failed to convince.) Furthermore, some of Avenatti’s quoted statements, such as that the alleged Russian hacking of the DNC in 2016 was a comparable threat/provocation to the deployment of tens of thousands of troops on the US border, struck me as so palpably ridiculous that only a person deficient in either intellect or honesty could make them.

    —David Frum, Harvard JD, currently a staff writer at the Atlantic, formerly a speechwriter for President George W. Bush. I don’t like to beat up too much on the neoconservatives, for reasons not worth elaborating at length here, but he did really do a bad job of analyzing the evidence about foreign policy at that time, as in his atrocious co-authored book An End to Evil. He’s an… ok… pundit in my view, but his articles and appearances on Ezra Klein and Sam Harris’ podcasts left me unimpressed. Furthermore, you can see on YouTube Noam Chomsky own him pretty hard when Frum attempts to ask him a critical question, and in a Bloggingheads debate Glenn Greenwald conclusively defeats him.

    —Justice Sotomayor’s “wise Latina” comment. Granted, she made the comment before she was a Justice, but it seems like an obviously empirically questionable and self-serving remark. Her later attempts to explain the comment lead me to suspect that she regrets stating her genuine views so baldly and quotably, but does in fact still believe the substance of the comment.

    This has gone on a lot longer than I intended it to, and there are actually quite a few other points I wanted to raise. However, I want to make it clear that this is not meant to be a definitive argument, but rather an explanation of a personal impression that I am welcome to have qualified or changed based on other people’s comments.

    • idontknow131647093 says:

      Well, first of all, just from a pure employment side of things its not a great time to get into law.

      When it comes to becoming “influential” or the like in some of those areas, its possible to do that with a JD (a lot of the most influential “journalists” have them) you have to go to a top university, otherwise you will not be admitted into the clubs either in DC/Sacramento politically or in Boston/Ann Arbor/Etc in the academic arena.

      As to who you find impressive/unimpressive in interviews, I would posit that has 3 axes: 1) How much you align with the person; 2) How little they care about appeasing a “middle”; and 3) (and probably most important) Whether they are rhetorically skilled. None of those is going to change much for you if you get a JD.

      • Atlas says:

        Well, first of all, just from a pure employment side of things its not a great time to get into law.

        Indeed, which is why I linked to Professor Paul Campos’ interesting book about the state of the legal labor market and the dishonesty of many law schools’ marketing towards their prospective students.

        When it comes to becoming “influential” or the like in some of those areas, its possible to do that with a JD (a lot of the most influential “journalists” have them) you have to go to a top university, otherwise you will not be admitted into the clubs either in DC/Sacramento politically or in Boston/Ann Arbor/Etc in the academic arena.

        That strikes me as very true. Generally, as Scott mentioned in a post a while ago, law seems like a winner take all kind of market.

        As to who you find impressive/unimpressive in interviews, I would posit that has 3 axes: 1) How much you align with the person; 2) How little they care about appeasing a “middle”; and 3) (and probably most important) Whether they are rhetorically skilled. None of those is going to change much for you if you get a JD.

        That seems plausible, but I don’t believe it is correct, at least in this specific case. Regarding the first point, I praised people (Noam Chomsky and Milton Friedman) with widely divergent viewpoints, and criticized people (e.g. Michael Avenatti and David Frum) on very different places in the political spectrum. Certainly, people who I think tend to have convincing arguments I tend to agree with to some degree. But I hope these examples suggest that this is postjudice rather than prejudice towards any particular view. Much the same holds for the second point, as well.

        Regarding the third point, I would disagree. I would not say that e.g. Glenn Greenwald is an elegant rhetorician in the sense that I would say that e.g. Vladimir Nabokov is an elegant writer. I am also critical of commentators like the late Stephen Jay Gould who are gifted writers but disregard important evidence.

        I would say that the ability to make cogent and persuasive arguments—even in an informal or challenging setting—is hyped as one of the major benefits of a JD. What I read as your claim that getting a JD doesn’t actually help one at doing this does sound quite plausible to me, though.

        • idontknow131647093 says:

          I got a JD. It teaches you a very particular set of skills. That set of skills (even if you don’t go to Harvard) is the set of skills that make you the kind of persuasive pontificator (on paper not in person) that will get you a job at The Atlantic or Yale. It also will help you persuade judges in the narrow set of circumstances where there is a novel question and/or they have not done the research themselves (or their clerks) and have no experience (easily less than 5% of legal stuff, probably about 1%). Vocal persuasion is the skill of a litigator, and litigation skills are barely taught or even practiced in law school.

          There are ex-litigators that are famous media personalities like Megan Kelley (who was a fairly effective interviewer when she was early on Fox News, because she didn’t try for faux-neutrality, my point #2), and Greenwald actually kind of emulates (to me) a certain kind of litigator.

          I’m not a litigator or professor. I’m a technical writer in patent law. If I could have made what you make with a JD without it I would, and if resumes didn’t include law schools no one would know the difference in my writing.

    • Chalid says:

      To the extent that your complaint is about law being the “study of relatively uninteresting and unimportant questions,” this really could be said to describe a vast swathe of the economy and isn’t a strike against law in particular.

      In your example, the fields of biology, psychology, and journalism as a whole might have more interesting and important contributions to make to the creationism in schools debate than law does, but I’d bet that the majority of actual working biologists and psychologists and journalists spend most of their time doing stuff that would strike you as boring and unimportant.

      • Randy M says:

        That’s a fair point; cutting edge work in biology is probably studying the variations of one particular enzyme in one particular metabolic pathway by loading small, precise amounts of meticulously prepared solutions into complex machinery and then doing statistics on the printout.

        But then again, sometimes you might be sticking electrodes into monkey brains, so …

        • Nabil ad Dajjal says:

          More or less. To the extent that you’re wrong it’s that biological research is less exciting and rewarding on a day-to-day level than you’ve described.

          One enzyme is definitely accurate, but it’s rare for anything to only be involved in one pathway. Evolution is extremely conservative in a way which is inconvenient for researchers who would prefer to only investigate one pathway at a time. But if you don’t check, reviewer two will want to know how your work relates to the enzymes other known functions.

          I wouldn’t call most of it meticulously prepared, although that might be bias in what I consider sloppy. Look at an unedited western blot sometime: interpreting it is almost as much augury as science. The “representative images” in journals are invariably the cleanest ones and typically cut so that most of the smearing isn’t visible.

          Also, sadly, these days the electrodes are almost exlcusively going in rodent brains. Primate research is extremely expensive due to regulation and the need for security. Even at large, well funded institutions almost all of the neuroscience research is done on rodents or in vitro.

    • Deiseach says:

      Michael Avenatti, who graduated first in his class from GW law school

      I now understand all the jokes about GW law school I read online 🙂

    • Statismagician says:

      As late as the my Junior year of undergraduate studies, I was also planning on studying law, and ended up changing plans due to some similar concerns. Instead I’ve ended up in epidemiology and biostatistics. You might consider any of the public health/policy fields as an answer to the question ‘how do I do something practically useful as well as philosophically meaningful?’

    • Randy M says:

      It seems to me (IANAL) that being a lawyer is like being a programmer for computers that don’t have to follow code. This probably results in more rigorous thinking by some and more sloppy thinking by others, but in the end I think it is a practically minded profession, requiring high verbal skills and memory but not otherwise better prepared for abstract questions than any other.

      And so, unless you are of a bent to enjoy interpreting writing elaborate, overly precise formal language about corner cases, you’d probably want to go into Law only if you think you have a good shot at making the big bucks via a law career (which I expect is about 50% smarts and 50% networking), or you want to do accomplish something only lawyers are likely to be able to accomplish, like be an effective cog in the machinery of justice.

    • BBA says:

      I got a JD in 2010, but due to a confluence of my own ineptitude at navigating law school exams and the job market, the Great Recession, and being fortunate enough to secure a high-paying non-legal job through nepotism, I have never practiced law. (I am currently going through my biennial ritual of sitting through hours of continuing legal education lectures and wondering why I bother with all this hassle to renew a license I never use.)

      I think it’s almost certainly a lousy idea for anyone to go to law school. I’m even more jaded than you are about the intellectual quality of law as a field of study – I think basically all of constitutional law is incoherent sophistry dressed in the language of biblical exegesis. Furthermore, working at a law firm is a hellish experience of long hours doing mind-numbing tedium, but you have to keep your hours up in order to hold onto your remote chance of making partner, which is the only way you’ll ever pay off your student loans. Depression, substance abuse, and suicide are all too common among attorneys in America today. I’m depressed and I have a good job at a company that appreciates me, I can’t even imagine how I’d fare in this environment. Not that other sectors of the profession besides Biglaw are any better. I had to unfollow a classmate who works as a public defender, because the daily horror stories were just too much.

      If I sound bitter, well, I am.

      • Brad says:

        I endorse much of this and BBA’s reply to Scott below.

        The interesting questions bit matters if and only if you are weighing a career in legal academia vs some other kind of academia. Legal academics, and a tiny handful of high level judges, aside everyone else is a hired gun that are all trying to persuade rather than answer questions in the first place.

        As being a journalist I think you are confusing correlation with causation. Some top end journalists have JDs for similar reasons to why some top bankers (and lawyers for that matter) have physics degrees.

      • Nancy Lebovitz says:

        There was a prediction that there would be a large number of people with legal training who didn’t get good jobs in law who’d be working on changing the system.

        I don’t *think* this has happened. Have I missed something?

        • BBA says:

          Well, applications to law school are way down and lots of schools have been lowering their admissions standards or downsizing. A few have closed entirely, some for economic reasons, some because the ABA has finally started enforcing its accreditation standards after years of permissiveness.

          I say good riddance to Infilaw, but there really needs to be more widespread reform that neither the ABA nor the state regulators have any interest in enacting. There’s no need for law to be a three-year postgraduate course; it should be an undergraduate major or a one-year “transfer” course, or hell, just let anyone take the bar exam. Now this will necessarily reduce the prestige of law as a profession and of law school faculties, so it has no chance of happening, but it should.

          • Two other suggestions:

            Two years of required courses, plus an optional third year for those intending to end up as legal academics or specialists in some field requiring substantial additional legal expertise. This was more or less Posner’s suggestion.

            Someone, possibly the ABA, AALS, or USNews and World Report, ranks law school on value added–not the average quality of their graduates (bar passage and the like) but the quality of graduates as a function of quality of entering students.

            Note that the decline in applications not only makes it easier to get into law school, it makes it easier to get financial aid—especially for students with good entering credentials.

    • Scott Alexander says:

      I used to think this way, but I think there’s a level on which law can be deep, interesting, and enjoyable. David Friedman’s work convinced me of this, so maybe reading more of his work would be able to convince you. See also http://slatestarcodex.com/2017/08/28/contra-askell-on-moral-offsets/ , which is the closest I come to laying out my thoughts on this – basically, law is a branch of moral philosophy that discusses how to do ethics given certain constraints (ie that it will be enforced by the courts and police)

      • BBA says:

        That’s a good reason to join a law school faculty, not to attend law school. Those philosophical questions are fine to debate but don’t have a lot of relevance when you’re, say, negotiating a prepackaged bankruptcy.

    • aristides says:

      I do not think you should go to law school.

      I graduated from a T14 law school in the top half of my class, and have never practiced law for similar reasons as what you are outlining, except I was not aware until I was already over $100,000 in debt. I find legal questions fascinating theoretically, but if your goal is to improve the world, the practicing law is among least efficient ways to do it.

      Law school is good at teaching the art of written persuasion, but it won’t teach you if what you should persuade others to do. Only 9 justices have any true power, and they are preselected to just do whatever President and Congress want. There are better, cheaper degree options.

      If you want to answer positive questions of socual and natural science, my advice is to become a subject matter expert in some narrow field. Public Health would be my personal choice, so I will use that as an example. You can get a Master’s of Public Health in one year less for half the cost of a JD. You can apply for the PMF program to work as a GS-9 with fast promotion potential to implement what public health programs you believe are effective. Get a DrPH while working, and you’ll become invaluable to the Agency, get promotions, and can study actual questions of social and natural science. There is a similar path for any specific field the government works in, and if you are not sure what specific question you want to work on, a Master’s of Public Policy is a good generalist degree to start out with.

  17. James says:

    Word pairs that mean the same thing but where one is coded positive and the other is coded negative.

    weakness / vulnerability
    childish / childlike

    What else?

  18. Szemeredi says:

    Suppose that at some point in the future we have artificial wombs and the ability to remove a fetus of any age from a pregnant woman and complete gestation in an artificial womb. What will be the new ‘equilibrium’ in abortion rights, i.e. will there be a right only to terminate the pregnancy, or will there be a right to actually end the life of the fetus?

    • Salem says:

      If the technology were developed in the near future, there would still be a right to end the life of the foetus. The “bodily integrity” argument is not the emotional heart of the pro-life position. Different sorts of arguments would be made in favour of abortion, and the bodily integrity argument would be redeployed – for instance, arguing that it is a violation of bodily integrity to compel a woman to have an operation to remove the foetus and transfer it to an artificial womb, rather than kill it.

      However, on the margin, a few people would be swayed, and so there would be a few additional restrictions on abortion. For example, abortion in the UK would probably only be legal up to 20 or 22 weeks, rather than 24. I am less certain as to what would happen in the US, as there the situation is more in the hands of the courts, and so is unpredictable.

      If the technology were developed in the far future, it is very hard to say, as underlying attitudes on abortion might have changed so much as to make this technology irrelevant.

      • Randy M says:

        So, say there is a fetus in an artificial womb because of the some condition the mother had. I wonder how the right to terminate it would shake out? Unilateral, either party? Requiring both parties consent? Or unilateral ability to give up parental rights and responsibilities, with the child automatically given up for adoption at that point?
        Or still gender coded because of inertia/gender war ammo?
        Probably a CW topic to go any further, though.

      • One question is who is responsible for the resulting baby. Bearing a child is a substantial cost, but surely less than the cost of bringing up a child you don’t want. Current norms permit abortion but do not permit infanticide. I’m not sure of the legal status of abandonment, but I think a lot of people who don’t disapprove of abortion would disapprove of a woman bearing a child and then refusing any responsibility for taking care of it.

        If a woman who doesn’t want to bear her baby is required to transfer it to an artificial womb, does the state then assume the obligation for rearing the baby?

        • Hoopyfreud says:

          I think a lot of people who don’t disapprove of abortion would disapprove of a woman bearing a child and then refusing any responsibility for taking care of it.

          Wait what? I’m not under the impression that the pro-choice faction is against adoption.

          • Not if the mother finds a willing adopter.

            But if she refuses any responsibility for taking care of it, that means she is willing to abandon the child whether or not there is someone willing to take over the responsibility.

        • johan_larson says:

          At least at the moment, there is no shortage of people ready to adopt healthy infants. Presumably the state would be on the hook for the unhealthy ones. I’m unsure whether there would be enough to adopt all the infants that would otherwise have been aborted. But lots of places allow no-fault surrender of infants by mothers who don’t feel ready to care for them, and it doesn’t seem to be a budget-buster. So a policy of letting women transfer fetuses they do not wish to carry to term into artificial wombs with the state taking responsibility for the fetuses thereafter seems workable.

        • SaiNushi says:

          “I’m not sure of the legal status of abandonment”

          They are called “Baby Moses” laws, and they are active in most states. You leave the child at a hospital, police station, or fire station.

          My hope is that if artificial wombs become a thing, then they will be used when the mother wants to abort but the father still wants the child. That when the pregnancy is discovered, the mother and father are asked separately “do you want to care for this child?” If either says no, they are given a form to surrender their rights as a parent, and are released from any responsibility for the child. If the mother says no and the father says yes, then artificial womb. If both the mother and the father say no, then abortion.

    • Walter says:

      Abortion will be banned, and there will be a right to evict baby to one of these new robo-wombs.

    • Deiseach says:

      Well, I don’t know if you’ve heard about this, but at least in online pro-life circles there were stories going around about Obama and voting against born-alive infants acts (that is, when surgical abortions carried out so late in pregnancy resulted in delivery of a premature infant rather than a dead foetus, the pro-choice lawmakers wanted to make it law that these were to be considered human persons). Obama voted “no” on the grounds that this would infringe on the right to abortion.

      The Washington Post columnist came down pretty heavily against the pro-life side on this but it is probably more acceptable to quote here as unbiased (or at least not biased in the same way pro-life sites would be biased):

      The anti-Obama claims from Huckabee and Ohden refer to a series of Illinois bills known as the Born Alive Infant Protection acts, which would have defined the term “born alive infant” as “any member of the species homo sapiens” expelled or extracted from his or her mother that exhibits “a beating heart, pulsation of the umbilical cord, or definite movement of voluntary muscles.”

      The 2001 and 2002 measures included a controversial line that proved to be a sticking point. It said, “A live child born as a result of an abortion shall be fully recognized as a human person and accorded immediate protection under the law.”

      Obama took issue with that part of the bill, saying it could interfere with a woman’s right to an abortion, as established through the Roe v. Wade Supreme Court decision. Here is an excerpt of his remarks from the 2001 floor debate:

      “Number one, whenever we define a previable fetus as a person that is protected by the equal protection clause or the other elements in the Constitution, what we’re really saying is, in fact, that they are persons that are entitled to the kinds of protections that would be provided to a child, a nine-month-old child that was delivered to term. That determination then, essentially, if it was accepted by a court, would forbid abortions to take place. I mean, it would essentially bar abortions, because the equal protection clause does not allow somebody to kill a child, and if this is a child, then this would be an antiabortion statute.”

      Notice that Obama referred to “previable fetuses,” or those that do not have a reasonable chance of survival outside the mother’s body. Obama’s primary concern seems to be that the born-alive act would prohibit aborting a fetus still inside the womb.

      Critics contend that this interpretation is not necessarily true because some previable fetuses survive after delivery from an unsuccessful abortion. They argue that Obama essentially opposed protecting the survivors.

      Illinois lawmakers voted down identical versions of the Born-Alive Infants Protection Act in 2001 and 2002 before a new iteration of the bill came before the Senate’s Health and Human Services Committee, headed by Obama. This new legislation removed the controversial line about recognizing live-born children as humans and giving them immediate protection under the law. It also addressed Obama’s concern about previable fetuses, adding a “neutrality clause” that said the measure would not affect the legal status of fetuses prior to delivery.

      Nonetheless, Obama voted against the new bill, which happened to be an almost exact replica — almost to the word — of a federal Born-Alive Infants Protection Act that passed in 2002 without opposition in either politial party. (Updated: The vote in the House was by voice vote and the vote in the Senate was by unanimous consent.)

      So the point of this quote is that if you transfer the “previable foetus” to an artificial womb and permit it to come to term and be delivered, are you transgressing on the right to abortion? Are you forcing a woman to become a parent against her will (after all, this still results in a live, born baby even if the pregnancy of the woman has been interrupted)? Who is the legal parent of the baby born from the artificial womb? There have been women writing articles about how they want abortion rather than carrying a child to term then giving it up for adoption, because mothers-to-be are afraid they would bond with the child after birth and suffer emotional pain from giving it up and thinking about it being raised separate from them.

      So this would not be a solution to “if you don’t want/can’t have a baby, instead of an abortion have it transferred to an artificial womb”, there would be plenty arguing this was an attack on abortion rights.

      • Walter says:

        I mean, those people would exist, but they’d lose.

        Like, pro choice folks aren’t actually pro death. Most of them just don’t think that women should be forced to go through pregnancy and parenthood against their will, and accept as the sad consequence of this that babies have to be killed.

        IF you make the ‘go through pregnancy’ part into ‘have this safe, brief procedure to put the baby in the iWomb’, and the parenthood part into ‘someone else, somewhere else, who wants a kid has one’, then they will, for the most part, declare victory.

        The picture you are sketching, where killing kids is a terminal value that they would defend even if a simple alternative exist, remind me of the caricature that pro lifers are actually all about oppressing women. There might be a tiny part of each movement which resembles these monsters, but they are only influential because of reasonable people around them. If you give those reasonable people an answer to the dilemma, then the maniacs will lose their cover, and vanish away.

        • Deiseach says:

          the parenthood part into ‘someone else, somewhere else, who wants a kid has one’

          Did you not read the anti-adoption articles? They address and argue against this as too simplistic, as false, and as forcing the pregnant woman to ‘grow a baby for someone else’:

          Nobody WANTS to Have a Baby for Someone Else

          Let me correct that; because some women do, and they are called surrogates. I won’t get into that here and now.

          …As I have said many times before, adoption presents itself as having many answers that face a woman with an unplanned pregnancy. They have carefully honed their marketing message and have it down pat. It’s just full of holes, but too many mothers realize that when it is too late.

          …Hence, if the child is born, then abortion is obviously LONG off the table.

          It also means that adoption is NOT a reproductive choice; it’s a PARENTING choice. A woman has already made the choice to reproduce. She is already a mother. She has had the baby. She has decided NOT to parent her child. THAT is adoption.

          …So for ONE FINAL TIME: Adoption is not the Alternative to Abortion.

          And for why it is wrong and really just horrid for people to expect that those facing an unplanned pregnancy have some weird moral obligation to carry their pregnancy to term to fulfill the needs of those who desire babies? I’ll just refer again to this post written by an adoptive parent who says it much nicer than I could: To Pro-Lifers Who Believe Adoption is Always the Answer

          ADOPTION AND ABORTION ARE NOT THE SAME CHOICE

          OK? SO can we be DONE with this stupid debate? No? OK that’s fine. Now I’m just linking to this post. Sick of this shit…

          Plainly for this person a pregnancy that, even if interrupted and transferred to an artificial womb, results in a living child is making a woman a mother, a parent, against her will. See the objections to your suggestion about ‘someone else, somewhere else, who wants a kid has one’. From the linked article in the excerpt quoted above:

          If a pregnant woman chooses to give birth and relinquish her child for adoption, I honor her noble decision. But it is absolutely wrong to suggest that it is the duty of a pregnant woman who doesn’t want to carry her child to term to have her baby anyway so that she can relinquish it for adoption. There are already enough of God’s children on this earth who need a family. If you want to adopt one or more such children, by all means do. But don’t enjoin a pregnant woman to grow her zygote or embryo into a newborn for you.

          No, I don’t think “killing kids is a terminal value”. But I do think there has been so much emphasis on “this is NOT a person!!!!! it’s a clump of cells, an unwanted zygote!!!!” in order to justify abortion as “this is not at all the same as killing babies” that the idea of the ‘end products of conception’ existing after the woman has made the choice to terminate the pregnancy is something that cannot be reconciled. It is considered to be too painful to force the woman to be a mother who has given up a child because of the women who are haunted by guilt and grief after handing over babies for adoption, the adoptees who feel hurt by being rejected by their birth-mother even if the adoptive families are loving and good, the adoptees who hated their adoptive families because they never fit in and felt the burden of gratitude for ‘saving’ them imposed on them – it really is argued that the most humane solution for everyone is to terminate the unwanted pregnancy while it’s still at the ‘clump of cells’ stage:

          I went into my adoption work as a strong advocate of a woman’s right to choose, and I retire this year as an even stronger one. In fact, I wish abortion had been more readily available to many of the clients I worked with.

          …Needless to say, this is different from the narratives frequently trotted out by the anti-choice movement of women placing their children for adoption and feeling, as one website put it, “good and positive about [their] choice.” No matter what the reason was for placing a child for adoption, all of the women I personally encountered did so with a heavy heart. They expressed enormous sadness and guilt, having exhausted every other path. Many had no one they could turn to for help; the social services available to them were so paltry that raising a child seemed impossible.

          It was very difficult to watch these women go through the adoption process: undergoing nine months of pregnancy, withstanding inquiries from family or acquaintances about their plans for a baby, allowing near-strangers or people they had only come to know in the last few months to love and nurture their child, and then trusting those people to follow through on post-placement contact agreements. Some women were, and are, able to get solace from providing a good home for their child and giving joy to new parents. Even so, though, the process also nearly always involved anxiety and long-term sadness.

          And my clients were not alone. Experts have found that many biological parents who place their children for adoption go through an immense grieving process, one that may last for decades. In one study cited by the Child Welfare Information Gateway, three-quarters of birth mothers still experienced feelings of loss 12 to 20 years after placing their newborns.

          And the people so eager for having a kid can adopt all the foster kids and kids in care who are not cute babies instead:

          As I’ve said to many people, Bryan: If you truly believe that adoption is the answer for children without parents, then by all means, adopt a child already born who wants for a family. There are hundreds of thousands of such children already living on this earth, Bryan, many languishing in foster care throughout this country. They are good kids who desperately need families. Most people won’t adopt them, though, because they’re not infants. They might have “issues” of some sort, you know, being in foster care and all (most don’t, actually). Most prospective adoptive parents want healthy, white newborns. Even those willing to parent a child of color want a newborn.

          You won’t overcome so easily or in a short time the results of decades of arguing an entrenched position that abortion is a right, is moral, is not at all wrong, is exercise of bodily autonomy, is the woman’s decision and choice alone, and that the rights of a potential person do not come anywhere near the rights of a real actual born person, and indeed that real actual born women are threatened in their physical, mental and emotional health, their education, career and lives by assigning any value to the embryo over them.

          • ana53294 says:

            But the thing is, as far as I know, most countries do not allow men to stop paying child support to a child born against their will. No country on Earth forces women to go through an abortion to avoid forcing fathers becoming parents.

            So we already force men to become parents against their will. Women have options to become a mother or not, whereas men have no say in the matter once a child is conceived. And this is because of the high costs on body autonomy and health that a pregnancy imposes.

            Men also have to go through attachment, they also sometimes pay a very high price for an undesirable pregnancy. But we don’t let them have any say in the matter, and rightly so; it’s not their body. But once a foetus is easily separable from a woman’s body, it also stops being her body.

          • Walter says:

            I feel like reading the response to ‘give birth then put kid up for adoption’ as also applying to this hypothetical procedure isn’t super useful, yeah? They’d be super different things.

            Like, the hypothetical we are talking about is something that is pretty exactly equivalent to an abortion, right? You are gonna go in, get a procedure, and come out not pregnant.

            I definitely agree with you that the position wouldn’t shift overnight. It would take a while. But I think attitudes towards it would slide like they have towards gay marriage. From the first time folks go into the clinic and see the two options side by side, abortion is on a clock. I dunno how long after that it is relegated to cases where removal isn’t plausible. Call it 2 generations?

          • John Schilling says:

            Like, the hypothetical we are talking about is something that is pretty exactly equivalent to an abortion, right? You are gonna go in, get a procedure, and come out not pregnant.

            But possibly with a strong emotional attachment to another human being, which you didn’t want, and possibly with a legal obligation to pay several hundred thousand dollars over the next eighteen years, which you didn’t want. The status quo allows women (but as ana52394 notes, not men) to opt out of both of those obligations, and they will object to any new plan that might stick them with either one.

            Possibly the right answer is “tough; you implicitly agreed to this and now you’re stuck with it just like the men”, but given the stakes, understand that the opposition will be substantial.

          • Walter says:

            I’m not super sold on the idea that, going forward in this alternate world, the emotional argument would get made a lot. Like, if that’s a big deal, just have an option on the procedure where they don’t tell you what happens to the kid.

            As far as the economics, yeah, I think that is a big deal,and would be the delaying factor on this solution spreading. It seems like ultimately (in the same nebulous ultimately that polyamory and student loan reliefs exist in), the UBI or whatever ends up providing for the removed kids. Federal Ward program or what have you. Heck, you could fund it by pitching it to sympathetic billionaires as a concrete dollar amount you could pay to finally shut the pro lifers up.

          • 10240 says:

            It would largely depend on whether there would be enough adoptive parents to replace all abortions. (Probably not). If yes, then women could give up their fetus without an obligation to support the child. If not, then support for the right to abortion will remain.

            My impression is that most women who have an abortion do so primarily because of financial reasons or other life circumstances that make in inconvenient to become a parent, rather than to get out of pregnancy or giving birth. Those motives wouldn’t change.

          • Deiseach says:

            Like, the hypothetical we are talking about is something that is pretty exactly equivalent to an abortion, right? You are gonna go in, get a procedure, and come out not pregnant

            Except that with artificial wombs, the child lives. And the birth mothers are faced with the prospect of twenty to thirty years down the line, a knock at the door and a stranger going “Hi, Mom!” which they do not want.

            The movement is towards allowing adopted people more and more information; they fought successfully for the right to have knowledge of who their birth parents were (though in most cases it’s only the mother anyone knows about) and sometimes that does not work out because, as above, some women haven’t told their new families about the baby given away and wanted that part of their life sealed. There’s also the push to do away with anonymous sperm donation and give the names of donors to children conceived by such means, even though that means breaking the original confidentiality.

            Abortion prevents all that – there isn’t going to be the “what if?” hanging over your head for the next twenty years about “I wonder what my kid is like/where are they/are they going to come looking for me?” Some women may indeed be very happy to have the chance to meet the grown-up child, but some women won’t want it at all. And if the artificial wombs are being sold to the public as “this solves the problem of abortion”, then there are definitely going to be women who say “we don’t want this, we want our right to abortion preserved” (and there are going to be the heart-string tugging examples of ‘fatal foetal abnormality why be so cruel as to force a child that is only going to die after birth to be born?’, rape and incest, and the usual suspects trotted out – coathanger back alley abortions once again! – whenever the notion of doing away with abortion is floated).

            The reason I’m arguing all this is because I’ve seen in other countries what happened with abortion laws, and in my own country what is happening right now – where the legislation hasn’t even been written yet but already the protests about “there should be no conscientious objection clauses! and it’s too restrictive!” are happening.

            People who argued for the legalisation of abortion swore it would only ever be used for actual physical danger of death to the mother. Or rape and incest. Certainly not as a kind of back-up birth control, no not at all! Well, how has that worked out? 2014 factsheet from the Guttmacher Institute:

            The reasons patients gave for having an abortion underscored their understanding of the responsibilities of parenthood and family life. The three most common reasons — each cited by three-fourths of patients — were concern for or responsibility to other individuals; the inability to afford raising a child; and the belief that having a baby would interfere with work, school or the ability to care for dependents. Half said they did not want to be a single parent or were having problems with their husband or partner

            So not because of danger of death, rape or incest. It’s poor women who are the majority of those having abortions, and I doubt they’ll be the ones able to afford or access artificial wombs should those be developed.

            You really think artificial wombs are going to change attitudes to abortion so easily? Maybe over a long time, but abortion will always be there in the background as a ‘just in case’ or ‘for really hard cases’. Again, to quote the data from the Guttmacher Institute:

            Approximately 926,200 abortions were performed in 2014, down 12% from 1.06 million in 2011.

            So how many prospective adoptive parents are out there?

            There are no national statistics on how many people are waiting to adopt, but experts estimate it is somewhere between one and two million couples.

            So problem solved, right? More people looking to adopt than women wanting abortions, artificial wombs are the way forward!

            Except.

            While there may be that many saying they want to adopt, in reality that many adoptions don’t go through. And particularly children in foster care often age out of the system without ever being adopted.

            About 135,000 children are adopted in the United States each year. Of non-stepparent adoptions, about 59% are from the child welfare (or foster) system, 26% are from other countries, and 15% are voluntarily relinquished American babies.

            There are 107,918 foster children eligible for and waiting to be adopted. In 2014, 50,644 foster kids were adopted — a number that has stayed roughly consistent for the past five years. The average age of a waiting child is 7.7 years old and 29% of them will spend at least three years in foster care.

            So maybe artifical wombs for transplanted unwanted pregnancies may be the solution – or maybe the children of poor, non-white women (or even poor white women) won’t be desirable products on the adoption market if stated demand does not match up with actual rates of adoption as above.

          • John Schilling says:

            There’s room for a reasonable compromise where abortion is legal through roughly the first trimester, and after that it’s pregnancy or the uterine replicator and then adoption. I expect that’s where most of the developed world wound end up, but that there’s too much bad blood and too much partisan advantage to keeping the fight alive in the US for that to work here.

            OTOH, we have an edge on the adoption front in that Bruce Wayne is filthy rich and if we can up the mortality rate on Robins a few notches, he’ll take in orphans as fast as we can crank them out and make sure they are alternately too busy and too dead to go looking up their birth parents.

          • 10240 says:

            There’s room for a reasonable compromise where abortion is legal through roughly the first trimester

            That already works in most of Europe, without artificial wombs.

        • 10240 says:

          Most of them just don’t think that women should be forced to go through pregnancy and parenthood against their will, and accept as the sad consequence of this that babies have to be killed.

          I’m not sure about that. I’ve never regarded abortion as wrong at all. Mainly because my mother didn’t. Or at least she supported the right to abortion, and she never imprinted on me the idea that it was wrong at all, as long as it was early (first trimester). (We hold most of our moral values because our parents and environment did the same, and only a minority of each generation change their views on a given issue. Realizing that should help both sides understand each other a bit better.) I don’t think anyone regards killing fetuses as a terminal goal, but not having it as a goal doesn’t imply thinking that it’s wrong.

          I thought that most people who support the right to abortion thought the same, but I don’t know. I’d expect that people who consider a fetus a baby and killing it murder in the morally relevant sense would oppose the right to abortion.

    • rlms says:

      It depends on various things, a less obvious one is what happens to all the babies if this replaces abortion (there are several times as many abortions in the UK each year as total children in care).

  19. Jo says:

    Are there examples of what happens during SSC meetups, and what is the difference between SSC meetups and LW meetups in this respect?

    • In the SSC meetups I have been part of, some at my house and some elsewhere, most of what happens is small groups of people getting into conversations. There is usually also food. At ours, that includes our feeding dinner to those still around by dinner time. At the outdoors Berkeley meetup I went to a while ago, some people had brought finger food of various sorts, and many people went out to a restaurant together for dinner.

      I don’t have much experience of Less Wrong meetups, but my impression is that they are more structured than that.

      • LesHapablap says:

        Do people drink alcohol, smoke cigarettes or other drugs? Do they talk about intellectual things, or talk about their childhoods, or is it lots of small talk? Are there socially very awkward people that show up?

        • At ours, nobody smokes either tobacco or marijuana. If anyone is using other illegal drugs it isn’t obvious.

          People have sometimes brought beer, and I think last time we had out what was left of a bottle of mead someone had brought to a dinner party in the past, but there isn’t much drinking.

          I think most of the talk is about intellectual things, broadly defined. Some people appear to be socially awkward.

        • cassander says:

          I host the DC meetups. My answers mirror davidefriedman’s, except we always have alcohol, because I provide it (usually others bring some as well, but I always have some) and that helps with the social awkwardness.

          • Paul Brinkley says:

            One of the first topics I found myself in at our last meetup was whether there was anything one could say at an SSC meetup or discussion group that would be taboo, in the sense of causing everyone to just avoid you from then on, or in rare cases, get you punched.

            My feeling was that there might not be anything, to my surprise. Imagine the sort of thing you might say to troll a group of evangelicals, or SJs, or preppers, or pro-choice activists, or… Now I imagine what might troll a group of rationalists, and it’s just not coming. It’s like everyone comes to those tables with the emotion switch already behind clear plastic. The only way you’re going to get punched is if you commit literal physical assault, or perhaps some version of Ryan Stiles’ fighting words.

            Since rationalists types are commonly discussion geeks, the worst you could do, I think, would be to attack the means of discourse itself. Be irrational by an SSCer’s standards, and stick to it.

        • bean says:

          I’ve been to a couple. No drinking at either (we were at a coffee place) and mostly talk of intellectual things. There were definitely socially awkward people there (the median would have probably been the classroom weirdo) although the more extremely awkward usually say they don’t go because awkward.

    • SamChevre says:

      I can’t comment on the difference between Less Wrong and SSC meetups, because I haven’t been to any Less Wrong meetups.

      The Western Mass SSC meetups have 6-12 people in a public place (coffee shop/bar). No one smokes, a few people have a beer. The conversation is not pre-planned, but a few people generally come with a question or topic to bring up to start discussion.

  20. ana53294 says:

    My University recently bought an Illumina MiSeq machine – and everybody in the university is allowed to use it, as long as they paid for the per-use cost. Then more experienced people told me that it would take a couple of years to get actually useful results out of it.

    The reasons why this happens are non-obvious to me, although I understand how PCR works, and how the Illumina machines work. Illumina machines are also the most used machines for sequencing – so I would hope they would have protocols that are useful.

    I know that you need to adapt your protocols for every genotype – but why would you need to adapt to a new machine if you are using a model organism such as Arabidopsis?

    Is there any other type of frequently used machinery for which you need a couple of years to get useful results? Why does this happen?

    • abystander says:

      We have a miseq and I’m sure it didn’t take a couple of years to get useful results out of it. I would have thought that acceptance testing would include getting a good sequence of a model organism like it did with the Pacbio.

      We are sequencing prokaryotes if it makes a difference.

      Maybe the standards of a useful result is really high? Long repetitive sequences are a problem.

  21. Le Maistre Chat says:

    Roland’s Breach on the Franco-Spanish border is 100 meters high, 40 wide, and I-don’t-know deep. Legend says it was made by the paladin Roland after the Battle of Roncevaux Pass when he smashed his indestructible sword Durandal against the rocks with all his strength, in hopes of keeping it out of Saracen hands.
    How many joules would this feat have released?

  22. Callum G says:

    Hey Scott, have you ever thought of paying someone to help moderate the site?

    • Acedia says:

      Seems unlikely he hasn’t thought of bringing in mods. My guess is he’s concerned about the difficulty of finding someone who’d make the same moderation choices he would. Especially with the storied history of internet communities falling prey to ideological entryism as a result of recruiting extra moderators.

    • Scott Alexander says:

      Do you think it needs more moderation? In the sense of not enough people getting banned, or something else?

      • Callum G says:

        Less that, more that there are probably tedious tasks associated with running a community. Having someone to moderate, triage comments, help run events like meetups or essay submission and the like, could help free up time. I would guess that there are things that you’d like to do with the community but are less keen on doing the heavy admin work on.

      • Evan Þ says:

        I think the site needs more moderation in the sense of less polarized opinions on the Culture War. 🙂

        • Paul Brinkley says:

          If the thing to seek is less polarized CW opinions, the obvious solution is to not express them. 😉

          If the thing to seek is less polarized CW opinions from other commenters, that’s harder. But there are remedies, mainly in tools for responding to them.

      • Callum G says:

        Double posting, but an extra pair of hands could also deal with time-zoning issues, interface with the subreddit, do deep dives on posted links to see if they’re from at least somewhat reputable sources. It’s not that you’re doing a bad job now, but you do seem to be an amazingly busy person.

      • Bugmaster says:

        My gut feeling is that the site is currently suffering from too much moderation; but, of course, I can’t be sure — due to all the moderation.

      • Plumber says:

        @Scott Alexander

        “Do you think it needs more moderation? In the sense of not enough people getting banned, or something else?”

        Whatever it is you’re doing now seems to be working well here.

        The Reddit forum on the other hand is too hard to read on my phone so I haven’t looked at it enough to really judge.

  23. bean says:

    Naval Gazing has been very busy since I last posted.
    The 45th Infantry Museum in Oklahoma City – This is a good museum, focused on infantry and military vehicles, including a huge outdoor vehicle park with lots of neat early Cold War stuff.

    Museum Ships outside the US and Europe – The last of three lists on global museum ships. See if there’s one near you and go visit.

    The Falklands War, Part 8 – The sinking of the General Belgrano. What happened, why the British took the shot, and why the fact that she was outside the TEZ is totally irrelevant.

    My commentary on the recent sinking of the drydock holding the Russian carrier Admiral Kuznetsov, and on the collision between the Norwegian frigate Helge Ingstad and an oil tanker that left the frigate a total loss.

    I’ve started reposting my old series on commercial aviation, starting with an essay on revenue management, or how airlines sell tickets to get the most money out of the most people.

    The conclusion to my history of the USS Missouri, running from the start of the Korean War all the way to her participation in the Gulf War and retirement.

    And lastly, my own open thread.

    • Statismagician says:

      Re: the apparent controversy over the sinking of General Belgrano – what real objection could anybody have to sinking a major enemy warship during an actual declared war? Was it just an oddly specific application of general pacifism? Taking any opportunity to score points against the governing party, never mind how silly?

      • CatCube says:

        That always baffled me, too. Same thing for the Highway of Death in Iraq…like, the whole purpose of your army is to ball up the enemy army while minimizing the cost to your own troops. “You don’t win a war by dying for your country. You win a war by making the other guy die for his.” Sinking an enemy cruiser seems to be a central example of that principle.

        • edmundgennings says:

          The basic impulse is correct if potentially influenced by some less valid considerations. Enemy soldiers are still relatively innocent humans and so there needs to be a sufficient benefit to outweigh the bad that is their death. There are lots of benefits in most war situations but in some weird situations, generally involving a retreating army of a clearly beaten foe who will surrender regardless of casualties, one could imagine there not being sufficient benefit to outweigh the fact that deaths of innocents persons is to be avoided.

          • bean says:

            That might apply to the “Highway of Death”, although Tom Clancy once said it should be more accurately called the “Highway of Abandoned Loot”. Definitely doesn’t apply to the Belgrano.

          • abystander says:

            There are many cases where soldiers were severely defeated, retreated then rallied and fought again, especially if they are in possession of heavy military equipment like tanks.

            There was probably also question about whether there was an unstated goal of removing Saddam Hussein from power, in which case even if the soldiers were no threat to Kuwait, killing the soldiers and worsening the defeat would weaken his position.

      • John Schilling says:

        Overly-literal interpretation of the concept of “limited war”. The British had declared a specific 200-mile “exclusion zone” within which they would sink Argentine or suspected-Argentine as part of their effort to defend against the unprovoked invasion of British territory. Some people took this as a promise that they would not sink anything outside that zone, and therefore this escalation represented an uprovoked war of aggression against Argentina.

      • bean says:

        It’s never made any sense to me, either. I suspect it’s driven primarily by politics, initially a desire to score points against Thatcher and more recently by those with a bizarre and pathological need to oppose “western aggression” wherever and whenever they find it. This is a particularly strange case, as the British clearly communicated that they didn’t consider themselves limited by the TEZ, and the Argentine Junta was the sort of government those kind of people hold marches to protest when they aren’t protesting US or British actions.

      • Salem says:

        You are starting from the premise that Britain was fighting the war, and given that, it was a reasonable action. From this frame, opposition looks absurd.

        Instead, start from the premise that Britain should never have gotten involved, that we should have politely petitioned the UN or something, that the war was just Western imperialism, etc. Most importantly, start from the premise that a British victory would be a bad thing, because it would lock in colonialism, and empower jingoistic attitudes at home. Then the sinking looks like lives lost for no good purpose.

        I am not exaggerating. A few years ago, when they were doing the 30 year documentaries about the Falklands, they were showing the footage of the crowds cheering the victorious returning ships, and they had various leftist celebrities and politicians saying how unhappy it made them.

  24. theodidactus says:

    The 24th Interactive Fiction competition has ended!
    Here is a list of all 70+ games, in rank order:
    https://ifcomp.org/comp/2018

    There are some excellent ones this year. Ones this community would like, I would think, are:
    (dark ones)
    1. Devotionalia
    2. Erstwhile
    3. Bogeyman

    (light ones)
    1. Alias the Magpie
    2. Animalia
    3. The Origin of Madame Time.

    …though there are all sorts of games, from the classic parser-based puzzlefests (alias the magpie, temple of shorgill) to contemplative works (dead man’s fiesta, cannery vale) to “gamebook” style choose-your-own-adventures (Grimnoir, Within a Circle of Water and Sand). There are also a few “social sim” games, including my personal favorite, “the Master of the land”

    I’ve got a game in there too, which placed a humble 31st out of 76. I think you’ll like that one too…though I plan to put out a new version in a few weeks with a better parser, since almost every review was like “good game…but the parser is annoying”. This is probably the most accurate review of my game.

    • Walter says:

      Thanks for bringing this to my attention. Looks very cool!

    • dick says:

      Did you write your own parser? If not, are complaints of the form “the parser is annoying” just a short-hand way of saying that the author didn’t add enough synonyms and stuff like that, as opposed to being a complaint about the off-the-shelf parser used?

      • theodidactus says:

        I created the game using ADRIFT, but heavily modified the basic way the parser interacts with the player. A central gimmick of the game is that, as a secret agent trained for a very specific mission, you are afflicted with psychological programming that severely limits your options…also most off-the-shelf parsers are, i dunno how to put it, light and breezy in tone: “Pick up the gun” is answered with “okay you pick up the gun”…just didn’t fit the tone.

        So to answer your question it was a combination of both. I could have done a better job. It is a very large game, and I probably should have gotten more playtesters and had them play longer (I wrote this while in law school, so I suppose I have something of an excuse). It seems to get generally good reviews in spite of the crappy parser.

        Negative reviews also emanated from a general disregard for conventions around interactive fiction: there’s some arbitrary death, there’s a lot of randomness, I can’t say where the game falls on the cruelty scale without ruining several puzzles, etc

  25. A Definite Beta Guy says:

    American SSCers, what are your Thanksgiving plans? Any special traditions or dishes you are looking forward to?

    We’re heading to our in-laws for the Big Day. Looking forward to seeing everyone: I great along pretty well with all my in-laws, since they are basically just big kids and loves video/board games. The Bears also play the Lions and my Mother-In-Law is a big fan, so that should be a good experience.
    Now, the food…my in-laws panic about boiling vegetables, so I’m looking forward to another year of “meh” cooking.
    On the weird side, they’ve gotten really big into this “Turkey Trot” thing. I think running for fun is dumb, so I’m taking it as a free 3.6 mile walk with Mrs. ADBG. I’d still rather not being do that on Thanksgiving, but I guess I’ll survive.

    • dodrian says:

      My wife and I decided last night to go to her mother’s house several hours away, as it looks like our daughter is finally over her ear infection. My mother-in-law cannot cook and insists it is too late to order a pre-cooked turkey, so I suspect dinner will be whatever I can find still at the grocery store when we arrive Wednesday night. Sweet potato pie (my favorite) will be easy, as will green bean casserole (wife’s favorite, but seriously, yuck). I don’t want to cook a turkey in someone else’s kitchen (plus my wife doesn’t really like it), and will probably try and find a nice piece of fish (hard to do where we live, so always something we look for when traveling).

    • KG says:

      My wife and I are going to some rented house with her parents halfway between our homes, where we will probably mostly relax and take occasional walks and then jointly cook our Thanksgiving dinner, an all-vegetarian meal which includes a Quorn “turkey” roast, mashed potatoes, stuffing, corn and broccoli, and apple pie. My favorite parts are always the roast (made by her dad) and the apple pie (made by my wife).

      This has been our Thanksgiving tradition for most of the time I’ve lived with my wife, and I much prefer it to Thanksgivings I had when I was younger, either going to my aunt’s house to eat with a bunch of relatives I don’t know or staying at home and not doing or eating much. Also the fake turkey is easily better than real turkey, which I always found to be the blandest thing at the meal.

    • My nephew and one or both of his kids often have Thanksgiving dinner with us, but this year his mother (my sister) is actually home instead of at a bridge tournament, so they will be having Thanksgiving with her. My older son, my grandchildren, and my daughter in law are going to southern California to have Thanksgiving with her relatives (and escape the smoke from the large fire north of us).

      My daughter therefor invited several SCA friends who she thinks might not have other commitments, but she isn’t up yet so I don’t know whether she has gotten a reply yet.

      It follows that we will have between four (us) and nine (us plus all invited friends) at the table.

      Thanksgiving is a modern holiday, so one meal at which we don’t plan to do any of our medieval recipes. We will reserve an unfrozen turkey, size depending on how many people we are having. Stuffing, cranberry jelly of both sorts, gravy (my department), my wife’s bread, either our daughter’s caprese salad or my wife’s green beans and bacon, and pies–probably both apple and pumpkin, although if it was up to me it would be just apple and tradition be damned.

      My wife’s family tradition included creamed onions, but since none of us particularly like them they dropped out some years ago.

      This will be our second Thanksgiving dinner of the week. My wife’s church does one the Sunday before Thanksgiving. I was in charge of gravy for that one too, on a somewhat larger scale. As usual, I ate more than I should have. I am not a member of the church, being an unbeliever, but am happy to participate in its social activities, since I enjoy both cooking and chatting with friendly strangers.

    • Brad says:

      My family finally came to its senses and dropped the turkey a few years back. It’s decidedly a bottom tier protein and none of us are good enough cooks to spin dross into gold. We’ll be having beef instead.

      • A Definite Beta Guy says:

        Turkey is definitely weak. Really annoyed that this is the Thanksgiving tradition. I’d rather make a big pot of chicken legs or something.

      • Evan Þ says:

        When I was growing up, my mom used to cook ham for Thanksgiving. It was pretty good!

        (Then she switched to turkey when I was about ten or so. I didn’t mind, since that was about the time she started cooking more interesting food the rest of the year.)

      • Randy M says:

        I’m not thrilled with Turkey itself, but it’s good for once a year, and why not at a time when tradition requires half a dozen other side dishes to fill up on?

        It certainly makes a picturesque dish to carve up, too. I don’t know what main course would do better for visuals other than maybe a spit pig.

        • Salem says:

          What about goose? You still see it a lot in England for Christmas, but I’ve never heard of it at an American Thanksgiving.

          • Nornagest says:

            I’ve had goose for Christmas before, but it’s rare. Never had it for Thanksgiving — turkey is so traditional that even vegetarians will eat ersatz turkey made of tofu.

            I did cook a large duck for Thanksgiving once, when a small guest list combined with vegetarianism meant that it was just me and my dad eating meat, but I’m probably going to American Civic Hell for it.

        • Turkey isn’t great, but there are lots of good things you can do with the leftovers.

          And I do like stuffing and gravy.

      • Nornagest says:

        Turkey’s biggest problem is its dryness. The flavor’s fine if you can manage that, but if you don’t, it’ll be tasteless by the time you finish roasting a large bird.

        The deep frying method works well. Although I like simmering it much longer and at a lower temperature than a lot of the recipes you’ll find call for, for a carnitas-like effect.

        • Our solution to that problem for chicken is a medieval recipe in which you cut the chicken in half, wrap it in sage leaves, bacon, and bread dough, then bake it in the oven like bread. The dough keeps the chicken moist.

          We tried it once with turkey, unsuccessfully, should probably try it again sometime.

      • sunnydestroy says:

        Turkey can definitely be delicious. In my cooking method, I go for crispy, flavorful skin and moist meat.

        That means a little bit more work, but I enjoy cooking so I don’t mind.

        First you start by butchering the turkey to promote evenness of cooking. The problem with roasting it whole is different parts cook at different rates, a fact exacerbated by the size of a turkey. You can try trussing it to fix some of the problem, but I like heading it off beforehand. Butterfly the turkey so you can lay it flat in the pan. That means removing the backbone, tailbone, and wishbone at a minimum. I didn’t do it this year, but I plan to also remove the legs from the body since they cover the meat and form a thick slab that takes longer to cook than the rest.

        Another benefit of butterflying your turkey is all of the skin is exposed and kept dry for maximum caramelization and crisping during the roast. It also cooks faster.

        Liberally salt your turkey all over, then lay it in a pan on top of paper towels. Let it sit for at least 24 hours. This is the dry brine stage, where the salt will draw out the moisture, then the salty liquid will be reabsorbed into the meat. This improves flavor and helps maintain moistness during the roast. Some people do a wet brine, but it’s messier and it interferes with crispy skin formation. You can also add your other seasonings/herbs at this time. I like it pretty simple with some pepper, garlic powder, and a light sprinkling of herbs like thyme, rosemary, and sage. I go light because the roast tends to char herbs, which contributes bitterness. Feel free to leave it uncovered in the fridge for some time–it’ll further dry the skin and promote crisping in the oven.

        When you cook your turkey, you would ideally have a leave-in digital meat thermometer that will tell you exactly how close to done your turkey is. Turn your oven to 400 and put the turkey in. If you have a lot of time, you could put it in at 225 to promote more even cooking, then turn it up to 400 at the end to crisp the skin. When you’re 10 degrees off from 160F, you’ll want to start probing around with an instant read digital thermometer to check doneness. 160 is plenty safe for the meat (30 seconds at this temp = safe pasteurization), though for legs people sometimes prefer a higher temperature, 170-180, due to the textural difference and softness in the leg meat. I wouldn’t go higher than 160-165 is the right time to pull it out for the breast meat though, it’ll just get really dry beyond that. Separating the legs gives you differential cooking options here.

        You should be left with a deep brown, crispy delicious skin and moist, seasoned meat. Butterflying also makes for easier carving of the turkey too.

    • Nick says:

      I’ll be staying with friends for Thanksgiving and eating with their family. I do this every Thanksgiving; it’s nice to get away from town, and I try to stay off the Internet and get some reading done instead. I’m considering Thinking, Fast and Slow this year, which I bought last year and haven’t gotten to yet.

      My being along probably means entertaining the little ones. There will likely be Minecraft and may or may not be board games. If there are, I’m hoping to triumph in Exploding Kittens again.

    • Nancy Lebovitz says:

      I’m going to Chessicon, a science fiction convention in Baltimore– there will be a Thanksgiving buffet.

  26. DragonMilk says:

    Anyone ever use BoardGameArena?

    My favorites there are Carcassone, Puerto Rico, and Seven Wonders. When there is no in person quorum for a game, I find it’s a good substitute Games are much shorter.

    Any other recommendations there? Not as big of a fan of race for galaxy, incan gold, or lost cities.

  27. Salem says:

    The Iraqi Coup d’Etat of 1958 – this is in response to this request.

    Why did Iraq have a coup d’etat in 1958? More importantly, why did it take the form it did – the complete overthrow of the previous regime, never to be restored? Why did it lead to one military dictator after another until 2003? I will try to explain. TLDR: because there was no constitutional mechanism to change the government, because the regime had failed to legitimate itself, and because the state was too powerful.

    Iraq in the 1950s was a fairly modern, liberal, pro-Western country. It wasn’t rich by any means, but it was now producing a lot of oil, and the government was investing that money, both in infrastructure, and in human capital – sending Iraqis to the West to learn modern science/engineering/technology, then bringing them back to apply that knowledge domestically and teach others. The legitimating ideology of the state was that it was unifying and modernising the country. It was also a young country, with a high birth rate – it had just under 6 million people (compared to 33 million today, despite a large diaspora). The natural effect of economic development, and the centrally-led modernisation programme, was to raise the return to living in capital – Baghdad had quadrupled in population to just under 600,000, caused by migration not just from the countryside but the other cities too.

    The country was led by a small, elite group, who had more or less run politics since independence. At the head was Nuri al-Said, who had served in every political role possible, and by 1958 had been Prime Minister on 14 separate occasions (“when you have a small deck, you must shuffle the cards often.”). They had been young men in 1922, they were old men now. Their ranks had thinned over the years, due to death and expulsion, but little new blood had come through – instead it had increased the power of those who remained. Their social and political hinterland were the middle-class, foreign-educated Iraqis of the generation below who tried in vain to succeed them.

    Why did a new generation not succeed to power? Partly because the older generation were egotistical – certainly Nuri al-Said himself believed in his own indispensability. But it was also because the new generation didn’t agree with the ideas of the older – a new generation would have changed not just the personnel but the policies. I don’t want to sound like a Death Eater, but Western universities in the 1940s and 1950s were not too keen on liberalism or the West. Instead of being grateful to the government for their expensive Western educations, they all came back Communists who thought the government was evil. They were of course cured of these delusions by Iraq’s brief flirtation with Communism, but by then it was too late.

    Communism was also popular among the mass of the people, along with the other big movement of the day, Arab Nationalism. Both appeared much more “the future” than the old-fashioned liberalism espoused by the government, which made the government’s claim to be “modernising” Iraq look quite hollow. Arab nationalism had taken over in Egypt, long the biggest, most modern, and most important Arab country, economically and culturally, and Nasser was attempting to spread this ideology to the rest of the Arab world. They had already taken over Syria, and now his radio stations were broadcasting Arab nationalist propaganda into Iraq, and the government didn’t have much of an answer to it.

    What’s more, the mass of the people were virulently anti-British, because of past British meddling in Iraq (they had overthrown a popular government in 1941 and occupied the country until 1947), whereas the government was committed to a pro-British foreign policy, partly as a matter of sympathy (wealthy and educated Iraqis were pro-British), partly raison d’etat (without a British alliance, Iraq would be vulnerable to Egyptian/Syrian invasion – see below), and partly for selfish reasons (the last government to defy the British had been violently deposed). Unlike the communists and the Arab nationalists, the government wasn’t making much of a direct economic appeal to ordinary people. They didn’t try and buy support with handouts, instead preferring to enrich the country in the long term with investment. This was the best policy for Iraq, but not the best route to popularity, especially with a low-information population like Iraq in the 1950s. But the government thought they could just ride out any unrest and they’d be thanked in the long run. They tried to take some populist measures, like being cruel to the Jewish minority, but everyone could see that their heart wasn’t in it, and they could always be outbid by more extreme forces.

    For all the above reasons, as more and more time passed, the regime increasingly lacked popular legitimacy, but it was tolerated rather than hated. Even when they used the army to hold off some pretty nasty Communist violence, this was accepted – Iraqi governments had done far worse in the 1930s. Despite this general unhappiness, Iraq was going pretty well, and there was no urgency to any of this discontent. People assumed things would change in time – for example, it looked for a while like Nuri al-Said would start stepping back after Faisal II attained his majority – and prior to 1956 the general conditions were far too contented for serious revolt to be anyone’s primary concern. However, even by 1956, a lot of the damage had already been done. The regime had been sufficiently delegitimised that civil servants and functionaries were used to just executing orders they didn’t believe in, and no-one was going to try and restore the regime if it fell.

    But in 1956, disaster struck. Britain made a secret treaty with Israel whereby the Israelis would invade Egypt, and then Britain would intervene to “separate the combatants” so that they could seize control of the Suez Canal. They did this knowing full well that it would be disastrous for Britain’s allies in the Middle East, like Iraq. Eden must have known that the existence of the treaty would come out within days, revealing him as a treacherous liar. He probably didn’t know that Britain would be forced into humiliating withdrawal, because his Chancellor of the Exchequer, Macmillan, was talking out of both sides of his face, but that is how it ended up. This was the end of Eden’s political career, of Britain’s status as a major power, and – more importantly for our purposes – of the Hashemite monarchy in Iraq.

    It wasn’t merely that the government’s major foreign ally was revealed to be evil liars. Most Iraqis already thought that. It was that they were revealed to be weak – Egypt had stood up to them and won. Nasser was now at the height of his prestige, and the message to Iraqis was unmistakable. We had now moved from indefinite illegitimacy, to urgent crisis. Something had to give.

    Iraq had a form of democracy, but it was controlled. Despite the government’s huge unpopularity, in the 1958 Parliamentary elections, just 5 opposition candidates (out of 128) were allowed to win seats. On the surface, the government had everything under control – riots were put down, the generals were loyal, they even found time to make a gesture to pan-Arabism by forming the Arab Federation, a surface-level union with Jordan. But beneath this seeming control, was real fragility. With no legitimate outlets for dissent, illegitimate ones festered. Besides, the generals may have been loyal, but who launches coups?

    The coup was mostly arranged by Colonel Abd al-Salam Arif, an Arab nationalist. Abd al-Karim Qasim was brought in pretty late, and it is rumoured that he was merely being informed of the coup but instead decided to take part. However, as a Brigadier, he was the most senior officer involved, and so took leadership. It was a fairly disorganised affair, but they did manage, after some delay, to kill the King, the Crown Prince, the Prime Minister, and various other key figures in the old regime.

    Plenty of key figures survived, many going into exile – had the old regime been sufficiently popular, they could perhaps have tried to come back. But no-one ever showed the least interest in restoring Prince Zeid. Feisal II had been personally popular, and a lot of people were very sad that he was killed, but his claims to legitimacy – descent from the Prophet, his grandfather’s revolt against the Ottomans – seemed quaint. The regime itself was not mourned. Meanwhile, the powerful state apparatus built by the Hashemites was now up for grabs, and both government officials and civilians were used to carrying out and suffering what they considered illegitimate, unpopular laws. So it was difficult to form opposition when the government went off the rails.

    Representative democracy was not a popular ideal at the time, so the new government abolished what democracy did exist. But despite being a dictatorship, they had a major falling out over whether the new regime should be Communist or Arab Nationalist. The Communists managed to kill 100,000 people in a few months, which made it very clear that if you were in any way involved in politics, you needed to either gain control over the state security apparatus, or leave the country. As a result, politics was governed by various coups, hard or soft, from then on.

    • Le Maistre Chat says:

      Thank you for the effortpost.
      I get the impression that Iraqi Kurdistan is full of Marxists, nationalists and religious minorities, while the Arabs come across as all orthodox Muslims, just divided by the Sunni/Shi’a split. What was the Overton Window throughout Iraq like in the period you’re discussing? Obviously there were a lot of Arab Marxists, but what were imams & ulema preaching and how seriously did the person in the street who wasn’t an apostate take them? Did Marxists and other apostates have to live in their own social bubble?

      • Salem says:

        There were plenty of non-Muslim Arab Iraqis in this period, Christians and Jews, who mostly lived in Mosul and Baghdad, not in Kurdistan. Mind you, this is the time when the Jewish population were encouraged to leave.

        What makes you think Communists were considered apostates? I mean, maybe the tiny number of super-hard-core activists might have gone that far, but Iraqi Communism didn’t become a mass-movement by asking people to reject Islam. Communists didn’t have to live in their own social bubble – it’s pretty clear everyone knew, informally, who was a communist, but it wasn’t something you necessarily told the police, seeing as the Iraqi Communist Party was an illegal organisation.

        I think the “Overton Window” is a strange term to use in this context. The government was liberal in terms of social and economic organisation, but this didn’t mean there was political freedom. The government determined the “window” of what it was acceptable to publish in a newspaper or broadcast on TV, and it got wider or narrower not based on popular opinion, but depending on how confident the government was feeling at that point. Without opinion polls or mass popular debates, it’s impossible to say what the Overton Window “would have been.” However, there was a lot of Arab Nationalism (which is a secular, socialist movement), and a lot of Communism, so it’s not exactly suggestive of a Thatcherite consensus. This is before Islamic fundamentalism became a big thing.

        The influence of the imams and Ulema was at a low point. The urbanisation of society had broken ties and reduced their influence, as had the introduction of the state education system. In the cities, lots of people drank, lots of women didn’t wear veils, society was more secular. There also wasn’t the same social distinction between Sunni and Shi’a, except to the extent that the latter came from poorer and more backwards parts of the country. There were Shi’a politicians and Prime Ministers. The major religious authorities, Sunni and Shi’a alike, were co-opted into the state, typically holding Senate seats. They weren’t preaching anything radical.

        • Le Maistre Chat says:

          What makes you think Communists were considered apostates? I mean, maybe the tiny number of super-hard-core activists might have gone that far, but Iraqi Communism didn’t become a mass-movement by asking people to reject Islam.

          Ah, good point. Being a Marxist doesn’t necessarily entail using blatant anti-religion propaganda like the “We rule you, we fool you” chart.
          OK, so Iraqis were increasingly urban and Westernized, plus Sunni and Shi’a religious leaders had been co-opted by the liberal state, so fundamentalism was at its lowest ebb. Makes sense.

          • Conrad Honcho says:

            I was under the impression that “true Islam” and communism had a lot in common. It’s a Muslim’s duty to take care of other Muslims. During the brief existence of the Islamic State, everyone got free food, free medical care, etc.

          • That’s complicated.

            The Koranic tax is supposed to go to a list of purposes, including support for the poor, for students, for travelers, for warriors, … . Some schools of law hold that it must be evenly split among those purposes, others that it need not be.

            The tax can be paid to the government to hand out, or the tax payer can himself give it out, or he can use a private middleman to distribute it, who then gets a small cut for himself. So in principle it’s a decentralized system of redistribution.

        • Le Maistre Chat says:

          In the cities, lots of people drank, lots of women didn’t wear veils, society was more secular.

          I didn’t say so, but I took it for granted that Iraqi women wore Western clothing without a hijab and lots of people drank, because that was normative in numerous other countries from the 1950s to ’79 or so, like Egypt, Iran, Indonesia (where we have anecdotes about Barack Obama’s stepfather going to Friday services and then walking out to have a couple drinks), et al.

    • bean says:

      Very interesting. Thanks for writing that.

  28. sty_silver says:

    Say I have 200 000$. My goal is to protect them from inflation (i.e. invest in something) while also withdrawing about 2000$ per month. The value doesn’t need to increase throughout this time, but there should be close to zero risk, and it should be reasonably easy to manage. I live in Germany.

    What would you recommend I do? (This might be a practical problem rather than just a thought experiment).

    • SamChevre says:

      In the US, you would invest in TIPS. Does Germany have an equivalent?

      • pjs says:

        I don’t think you can use TIPS because of tax (may be wrong, but I think you pay tax every year on the notional inflationary gain, even if the bond hasn’t matured). If you can get them in a tax deferred account (which wouldn’t help the poster anyway, as he needs to draw down the capital over a few years) it might help somewhat, but even so keeping up with (say) 75% of inflation at the end still isn’t necessarily the desired outcome.

        So the question is: what portfolio should I have to have a good chance of a TIPS like return, post-tax? TIPS-like in actual distribution, not just in expectation! (But not TIPS-like in that you don’t really care bout a premium above inflation.) There’s still a risk/return tradeoff, albeit an idiosyncratic one, so there’s not a single answer – but there should be a menu of portfolios one could choose from.

        Supposedly a financial advisor can help you (*) choose a portfolio based on helping you choose a point along a risk/return efficient frontier; where risk = variance(change in value), and return = expected change in value. But you are trying to tell him, that for you risk is variance(f(change in value)), return = E(f(change in value)), where f(x) = x for x less than inflation, otherwise = inflation. Gaining real 0% and gaining real 1000%
        are the same to me: you’ve maintained my spending power as I wished. I’ll probably donate my excess gains in the latter case, since I never wanted them. But you can’t use that possibility to justify increased downside risk.
        Surely the new utility function should change the set of feasible options! (?)

        (*) In my limited experience, financial advisors are often just confused by risk vs return anyway; they want to talk about your needs in terms of growth vs income. Which is as bad; to maintain capital you don’t care about growth (beyond a certain point) at all, and to the extent you are willing to sacrifice upside it’s wrong to think of the tradeoff being a desire for increased “income”.

      • ryan8518 says:

        Germany appears to have come around to the TIPS idea recently and only reluctantly. There are some options, but the liquidity doesn’t perfectly match up to what you want so you’ll have to come up with some sort of synthetic exposure (and paying some price for it, though almost certainly less than the excess return you get for the underlying bond) or accepting a bit of risk of market speculation about interest rates between your sale date and the redemption date. As mentioned above, the American (and to a lesser extent the British) markets are much better developed and easier to get into, though carry both currency risk and the risk that inflation in another country is different than yours (though, that last part applies in country, everything is indexed to the national inflation rate, which is likely to more adequate in slow growing parts of the country and less so in fast growing, though how much that affects an individual tends to be driven by whether you own your own home, and in the US is a poor gauge for seniors given medical inflation rates).

    • Walter says:

      My portfolio is heavy on index funds. I guess the risk there isn’t zero, but it has generally served me well thus far.

    • Hoopyfreud says:

      Hypothetically, if you’re more confident in the inflation rate than the market, you could sell OTM covered calls on your entire portfolio pegged to projected inflation.

      Please note that this is almost certainly extremely stupid. I don’t actually know why it would be, but it seems very, very likely.

    • Thomas Jørgensen says:

      Uhm? European monetary policy is really, really hard-money biased. What on earth makes you think the ECB is going to stop undershooting its inflation target every year in the near future?

      But if that is what you think is going to happen, then by all means invest in something that would profit from expansionary money supply – that is, anyone with a broad customer base, so.. index funds? Or just bet on VW actually executing on their e-mobility gambit.

    • Bamboozle says:

      You’re wanting to have your cake and eat it too. You cannot have return without risk. If you want zero risk or close to, you want a current account. If you want income from that amount you are going to eat capital and be eroded by inflation.

      This is how i would explain it to financially minded clients. You can argue inflation exists (or is allowed to exist) to encourage useful investment of funds instead of wealth just accumulating (and to allow national governments to erode their debts). Your $200,000 sitting around doing nothing does nothing for society at large, so inflation will hopefully spur you to invest it in some way that is hopefully productive.

      Back to your situation, there is relatively little that you can invest and get any kind of return for next to no risk. This is because there is a global over-supply of liquidity chasing fewer and fewer assets. (if this isn’t making sense i can try to couch this in different terms, just let me know) So your options are:

      A current account: in the UK for that amount of cash you’d get maybe 1.8%, so you’d get your income but be eroded by inflation over time, not sure what rates are in germany
      A property: you won’t be able to access your money easily and there are significant costs involved with purchase as well as the risks involved of having all your funds in one “basket” so to speak. but you get your income and inflation protection built in.
      Fixed rate bonds: your money is tied up for how ever long you agree, but you are still unlikely to beat inflation with this method either as rates won’t be that much higher.
      Stocks and shares: you should comfortably, over >10 years on average, beat inflation and generate more than the 1% income you are looking for. You could do a mix of stocks and bonds to try to reduce the risk you take with this option as well. As i don’t know what the ultimate purpose is for these funds or your attitude towards risk in general (other than you want 0% risk which we’ve covered is unreasonable) it’s hard to suggest this approach, but this is in general what the vast majority of people trying to do what you’re doing go for and they just have to live with the risk.

      You could use inflation linked bonds, something like TIPS, but in a rising rate environment you’re likely to see your capital reduce as the income increases, and ultimately if you use TIPS (the biggest index-linked bond vehicle) you’d be tied to the US currency. For a lot of recent history interest rates around the world have been falling and this has lead to things like TIPS performing relatively well as this has lead the capital to increase. With interest rates rising you’d likely see your capital decrease and income component increasing, which goes against your 1% income objective. (In the UK you have INXG, i’m not sure about Germany)

      I could get into different vehicles such as REITs or infrastructure funds but my advice would be probably limited to the UK markets.

      You also have options such as absolute return funds but curiously (/s) these only manage to return capital inflation plus income in bull markets, and often not even then.

      TLDR: either have 0 risk and get slowly killed by inflation, or take some risk and have a much better chance of meeting your goals. This information is also generic advice and shouldn’t be read as advice for your situation personally.

      • Bamboozle says:

        Sorry i just read that you said $2,000 per month not per year, which may have been my brain auto-correcting for reasonableness. $2,000 per month is a 12% income return only, plus c. 3% inflation you requested, would be 15% annual which for no risk is insane.

        • sty_silver says:

          no no you misunderstand. I meant I want to withdraw 2000$ per month of the original sum until there’s nothing left of it, which would take about 8 years. I don’t mean that the 2000$ are returns on investment. The only investment I ‘expected’ in the proposition was something cancelling out inflation. (Apologies that this wasn’t clear.)

          • Bamboozle says:

            Ahh i understand. You could invest it in the market and take an income for the rest of your life instead of drawing down the capital, or buy an annuity and be guaranteed income at a certain level instead of whatever the market provides. (though at the record low rates we are at now this is unlikely to be a good option).

            In response to your question you would most likely be best off in a collection of index-linked bonds. Keeping 6 months payments to yourself in cash and supplementing this with the income provided until either a bond matures or you are forced to sell to make up that $2,000.

          • pjs says:

            > Ahh i understand. You could invest it in the market and take an income for the rest of your life instead of drawing down the capital, or buy an annuity and be guaranteed income at a certain level instead

            I’m not sure this answers OP’s question, but maybe it does. But I’d sincerely like to know your thoughts on a (perhaps?) different question: what if income is irrelevant? I want to maintain post-tax, post-inflation, spending power (and trade off risk against that), but an outcome ending up with 1000000000000x my original investment isn’t that different to me from ending with with 0% (real!, post-tax) return – in either case, my portfolio has “succeeded” in what I wanted. But lose real 5% vs lose 4.9%; now I care.
            Who can suggest a portfolio that seriously takes such wishes into account?

    • EchoChaos says:

      I’m not clear what the question is here, so perhaps clarity is in order.

      Are you burning through $200,000 in principal in ~100 months at $2000 a month and just want a little bit of purely safe bonus money? In that case, you want Treasury Bonds or maybe a respected Money Market. The rate of return will be essentially nil, but you will guarantee the return.

      Or do you want to live on the interest and need to capture $2000 interest monthly off of $200,000. Because in that case there is no totally safe way to do it (and not a ton of moderate risk ways). That’s $24,000 a year, or a 12% return. That’s an excellent return for a stock market investment with decent risk.

      You really can’t have your cake and eat it, though.

    • raj says:

      You want a 12% drawdown with no risk? Laughably out of touch with the reality of investment. Disabuse yourself of the notion that you can have “close to zero risk”.

      A common rule of thumb for safe drawdown rate is 4% of your portfolio, so you could withdraw 760$ a month. You might be able to go for higher (5-6%) if you are willing to tolerate riskier investments.

      Mainly you need to get the money invested so it works for you. You need a simple portfolio plan. If you are extremely risk averse you could dollar-cost-average over a long period of time (say, move $15k/year into the portfolio from savings) so you aren’t exposed to a single large crash.

      • sty_silver says:

        No, as said above, I want to have a return similar to inflation, if possible. By withdraw I meant taking from the original sum. Again apologies that this was unclear.

  29. Randy M says:

    What makes a good life?
    No, wait, let’s narrow it down a bit more. What makes an activity that you aren’t directly being paid for more or less worthwhile? Two examples of this consideration:

    In the book Enchantment the main character, in one depressed monologue, considers the opinion he had held previously that reading was valuable and concludes something to the effect of “We book people tell ourselves that we’re better than others who waste their lives with TV, but that’s self-delusion. We’re wasting our time just as much.” Put aside reading to acquire actionable information; is reading really better for you then TV? Is it the medium that matters, or the content, or the quantity?

    Secondly, I consider the time I spend playing video games almost entirely wasted. Here and there I might pick up some useful modes of thinking, skills or knowledge, but by and large its a vice that I enjoy while do it but thereafter regret, at least a little. But the time that I spent helping to create a video game I remember very fondly and only regret it ending. Part of it is the social aspect; part of it is the permanence. Part of it might be that it is more intellectually demanding. It’s kind of paradoxical, though, since I wouldn’t see it as worthwhile if no one else had enjoyed the final product, and yet I don’t see playing games as terribly positive (not that I’m a killjoy about it, mind; I don’t hold anyone to a stricter standard for time spent than myself, and I don’t hold myself to a productive standard).

    Similarly, how does writing compare to reading? Does quality matter? Positive feedback from others?

    • Conrad Honcho says:

      Huh. I find the more time I spend playing video games, the happier I am.

      • Randy M says:

        At the time, or generally?
        Bear in mind that I would be prone to getting up, starting a game of Civilization, then making dinner.

        • Conrad Honcho says:

          Both. There are so many good games right now I can’t keep up. And it’s my favorite form of entertainment. I don’t know why I’d feel bad about it, so long as I’m doing well at my job and my family is happy and fed. What else should I be doing with my time than something I enjoy?

          • What else should I be doing with my time than something I enjoy?

            Possibly something that you think gives enjoyment to others as well? Something that you think makes the world a better place? Something that you feel is of value in itself, aside from its effects on you, such as making something beautiful, or working out interesting ideas, or … .

            Those are all possible objectives.

          • Randy M says:

            What else should I be doing with my time than something I enjoy?

            Well, that’s the question. There’s lots of criteria that you could use and I’m wondering what’s recommended here. What feels good at the time, what gives you good memories afterwards, what you felt some whim to do beforehand, variety of experience, what makes an improvement in your character, what makes you more marketable, what builds lasting friendships, how clean is your room, what challenges you intellectually, etc.

            I could play Call of Duty all night and enjoy it, and if I did slightly better than before, I’d probably end in a good mood. But the next day I couldn’t tell you the particulars of any match, and the only accomplishment is in entirely artificial points on a leader board.

            I could play a single player ARPG like torchlight or path of exile, enjoy the intellectual challenge of optimizing a character and the artificial accomplishment of advancing character statistics and some of the flashing graphics, but in the end the game play is pretty much identical from start to finish.

            I could play Civilization, maybe learn a fact or two on the level of “Siam used elephants offensively at some point”, enjoy all the optimization challenges, the exploration, immersion, the emergent narrative, and then lose to an ai offensive or get bored due to the late game micro and immediately have a strong inclination to start again. Upon stopping, while I’ll be in a good mood barring a particularly humiliating defeat, ultimately I mostly just wonder where the time went (other than the obvious–“waiting for other players” downtime).

            I feel like any changes to myself are done very inefficiently, drowned out by fun but repetitive gameplay. It’s fun, but it’s a kind of manipulative fun that works largely by tricking me into thinking I’m accomplishing something by incrementing numbers that vanish when a save file is erased.

            Playing a board game with my daughters or a friend requires a lot of mental work compared to a computer game, isn’t quite as enjoyable, but is something I look back on with appreciation all week due to the social aspects.

            Reading a book is a lot mental effort to motivate myself to do these days, less efficient at providing fun, but much more efficient at changing how I think or potentially act.

            Sitting down and writing a story is something that I will look back and appreciate for a long time to come, provided I persevere to a finished work. It’s also a good way to work through complex issues.

            These are all just activities I could do for fun, and evaluating which is preferable is more complicated than simply adding up hedons–not that I usually try calculating; usually free time use is dictated by whims.
            In the end, the answer probably comes down to the old “moderation in all things.”

          • Conrad Honcho says:

            But David I largely I do those sorts of things at work. I work in a very pro-social field, solving interesting problems that provide real benefit to people’s lives. Every day I show up at the office the world is slightly improved. And it pays my bills and supports my family.

            So when it comes to making the world a better place, thanks, but I gave already at the office.

            Also, given my political and social views, I’m pretty sure much of the SSC readership is glad I’m spending my time playing video games instead of “making the world a better place” as Conrad Honcho would define a “better world.”

          • Randy M says:

            Also, given my political and social views, I’m pretty sure much of the SSC readership is glad I’m spending my time playing video games instead of “making the world a better place” as Conrad Honcho would define a “better world.”

            I think next thread it would be interesting to list red/blue/grey? ways of making the world better vs universally acknowledged good deeds. I’d bet there’s more than we realize in the latter category.

            edit: (Response to Conrad below) I know you’re joking, but it might be interesting to look at the different tribes based on their view of good deeds.

          • Conrad Honcho says:

            I was mostly joking about that party, Randy. Mostly.

          • dick says:

            I don’t know why I’d feel bad about it, so long as I’m doing well at my job and my family is happy and fed.

            I also enjoy video games, but just having a stable job and fed family are nowhere near enough to avoid feeling like I ought to be doing something more useful. I think video games provide instant gratification, but very little long-term gratification (in the sense that a week later you’d say, “I’m glad I did that”) like you get from a hobby like woodworking or writing the great American novel or whatever. So, even if my id wants me to choose the video game every time, my superego makes me feel guilty for doing that, so I only end up doing it some of the time.

            But, to answer your question:

            What else should I be doing with my time than something I enjoy?

            For me, the answer 99% of the time is unequivocal: I should be sleeping. The only way I’ve been able to maintain a conventionally-successful life and also play a lot of video games is to sleep ~6 hours a day.

          • A Definite Beta Guy says:

            Video games have a social aspect with other people who have video games. Then it doesn’t feel so empty. Do you remember that part of Skyrim when you’re in Shor’s Realm or whatever and use your Dragon Shouts to clear up the mist and banish Alduin? Freaking Epic. I feel like much less of a loser when OTHER people also experienced this.

            Right now I am watching MNF. FUMBLE RECOVERY! Chiefs just stripped the ball. It’s just a 1 score game, even with all the penalties against them, even with KC having one of the shittiest defenses in the league. This probably means nothing to most people here, but it’ll be the talk of the office tomorrow….whereas Skyrim won’t ever be.

          • Conrad Honcho says:

            Randy, it occurs to me you’re just playing the wrong video games. Try playing Zelda: Breath of the Wild, Hollow Knight, or The Witcher 3 and then say “gee, that was a total waste, I’m sorry I did that.” Doesn’t happen.

            I appreciate games as art and they’re far more meaningful than any TV show or movie I’ve ever watched.

          • Randy M says:

            I think I gave the wrong impression that I always regret playing games. That’s not really the case, but games more readily expand to fill all available time while providing rapidly diminishing returns for the modal gameplay minute.

            I appreciate games as art and they’re far more meaningful than any TV show or movie I’ve ever watched.

            Played Breath of the Wild. My kids love catching horses; I liked climbing towers and exploring. Recently I beat Xenoblade Chronicles X, which I’d recommend if you have the Wi U. Assassin’s Creed Black Sails was incredible, and I’ve probably put literal years into Square Soft games. I think all of those are genuinely artistic.

            But games as art is pretty inefficient compared with other mediums. Whatever Final Fantasy Tactics or Assassin’s creed is saying about the human condition is quickly swamped in the dozens of hours spend leveling up another party combination or mastering murdering three dudes at once, respectively.

            I’m not saying books are necessarily better, though. I think there’s interesting things to be said about the impact of the medium, but in general I’m sure the ranges overlap considerably in terms of value (however defined).

          • Mr. Doolittle says:

            @ dick

            I think video games provide instant gratification, but very little long-term gratification (in the sense that a week later you’d say, “I’m glad I did that”) like you get from a hobby like woodworking or writing the great American novel or whatever.

            I’ve definitely played games like this (typically games on my smartphone just to kill some time), and I would agree about those. I try to avoid games where I feel like that after. That’s significantly reduced the number and type of games that I play, but that’s probably a good thing. I still have very positive memories of individual gaming sessions from the early 90s (original Civ…), including one memorable game where I lost that’s probably my favorite gaming moment of all time.

          • Conrad Honcho says:

            I have XBX on my Wii U and it’s on the list of “games I don’t have time to play because I have so many games.” Also the text font was so small I had a hard time reading it so I said “meh” and played something else. I did play Xenoblade Chronicles 2 on the Switch and enjoyed that immensely, especially the Torna: The Golden Country expansion.

            XC2 also had probably the best soundtrack I’ve ever heard in a video game. If there’s any other XC2 fans on here, check out this guy’s youtube channel where he does guitar covers of XC2 tracks. Really good stuff.

          • Randy M says:

            You switched abbreviations from XBC to XC, made me think you were talking about XCOM 2. I really need to get that some day, but I think I need to improve my PC first. Firaxis’ X-COM was incredible too.

        • Mr. Doolittle says:

          I used to have a lot of those days (and Civilization specifically!), and I found out of couple of things about myself that help a lot when thinking about it.

          1) I needed to set goals for myself about my play time. Play for an hour, or play for four, or however long I planned to play, but no more. Having a strong sense of how long I wanted to play before doing so was very important. This didn’t need to be a specific time, but could be a “until X” (often my wife getting home from wherever).

          2) Realize that I enjoyed the journey more than the goal. I also got this from playing Minecraft and other sandbox games. I used to feel more like I wasted my time when I didn’t hit an actual win condition, until I realized that my fun came from other things (emergent narrative for me, to use your term). I also no longer play any games where I don’t enjoy the actual gameplay. This helped me move on from games that hit the “incremental reward” part of my brain, but that I found I didn’t actually enjoy.

          3) Recognize the non-game goal for playing. I play such games to relax and clear my head. Depending on how stressful my work and other life is, I often find it hard to stop thinking about something, but I need downtime to rest. Games can fill that niche for me. They can also provide interesting puzzles for my brain to work on that have no serious consequences. Too many of the “interesting” puzzles in real life come with buckets of stress.

          4) Recognize the relative importance of playing games compared to other things. My family is more important to me than playing a game, and that’s pretty normal. I try not to choose playing a game when given a choice between family and gaming. This is greatly helped by the fact that both of my sons love playing Civ. We all know to drop the conversation when my wife is around, because it drives her nuts.

          I used to play 8+ hours a day and really had no perspective at times. I would end up with the same feeling of regret that you are describing. By thinking about and taking a more active role in the meta-analysis, I think I’ve found a good balance where I almost always enjoy when I play games, and rarely have any regret about it. I wouldn’t intentionally delete a save for no reason, but I also no longer tie my enjoyment to the save as you describe. This ties in with #2 above, where if I enjoy the gameplay, losing a save isn’t a problem, because I’ll enjoy making a new one!

          I also lowered the difficulty I would normally play at and took a more casual approach to whatever game I was playing, rather than pushing myself to break records or be better than other people.

          • woah77 says:

            This matches my motivations and logic behind gaming. It’s a way for me to work through and experience things that let me unwind my brain. I love conquering the world, solving interesting puzzles, or cutting my way through swathes of mooks, not because I’m escaping real life, but because work and life has filled my head and I need to give my mind time to process things while my front brain relaxes into something that doesn’t have any attached risks or stresses.

          • @Mr. Doolittle:

            Looking at the same question from the other end … I concluded years ago that my situation made it possible for me to spend most of my time, at least more than half of the year (I was teaching one semester on, one off), playing–broadly defined to include WoW, computer games, reading fiction, arguing with people online. I also concluded that I would not enjoy that life.

            So I decided to commit myself to spending two hours a day seven days a week on writing projects, defined broadly enough to include writing fiction, writing nonfiction, doing research for both, reading about how to write better. I’ve kept to that pretty strictly for some years now. I don’t try to do it for the two weeks at Pennsic, do for pretty nearly all other activities.

            I think it works–that I am happier as well as more productive than I would be if I followed my natural instincts and spent almost all my time—now that I’m retired it could be through most of the year—playing.

          • Mr. Doolittle says:

            @ David

            I also concluded that I would not enjoy that life.

            I was very surprised the learn the same thing about myself. There was a period in my teenage and college years that I could spend every free minute playing games. But I found myself feeling oddly off afterwards, especially as I got older. I tried quitting entirely a few times, but that wasn’t a good fit either, as I found myself bored or stressed (or both, ugh…).

            A better balance made me much happier, though, and I spend the remaining time doing things considered more “productive” than gaming.

    • Nabil ad Dajjal says:

      I’ve found that the Stoic distinction between eudaimonia and pleasure applies pretty well to my own life.

      There are a lot of things that I do which are fun in the moment but don’t bring any lasting satisfaction. They’re enjoyable but I don’t look back on, say, eating a really tasty cookie as something that gave my life meaning. They’re not good but, unless it was keeping me from doing something more important, it’s not bad either.

      Then there are other things I do which bring lasting satisfaction regardless of whether or not they’re pleasurable or painful in the moment. As a hard-and-fast rule about what sorts of activities provoke this feeling, they usually have to do with self-improvement or self-overcoming.

      Looking at your example through this lens, I can absolutely see why making the video game felt meaningful even though it wasn’t fun at the time and why playing video games doesn’t feel meaningful even though it’s fun at the time. Helping to make the video game probably tested your skills and your drive, the challenges pushed you to better yourself. Maybe if you played games professionally or at a tournament level the act of playing them might also encourage you to better yourself, but from how you describe it it sounds like nothing more or less than a fun diversion.

    • j1000000 says:

      I think most of it is what Nabil outlines, certainly, and I think that’s the best answer w/r/t meaning. I don’t much follow Jordan Peterson but in passing I’ve heard him talk about how happiness is the wrong goal because it isn’t lasting, and I think there’s something to that. (Though perhaps the wonders of the all-beef diet has led him to change his tune and now he thinks happiness is attainable? Legitimately wondering.)

      But specifically re: books vs tv, I myself would say that things like TV, the Internet, video games, and smartphones have something like addictive qualities in my personal experience, and I find I indulge in them beyond the level where I’d be simply enjoying them. I definitely enjoy them all in small quantities, but I often find I can’t stop using them even when I think to myself “I’d rather be doing something else.” So that’s where a lot of the guilt comes from, for me.

    • Bamboozle says:

      I think what you are playing matters. Strangely i always feel unfulfilled after playing a few hours of rocket league or overwatch, where it is just endless multiplayer battles. Even if playing with friends.

      Where as if i sit down and play an rpg or narrative driven game for 3 hours i feel much better and content after. Maybe it’s the fake hit of dopamine for “achieving” something, but i think it’s that i get much more out of exploring and interacting with a narrative than i do fighting to be “king of the hill”.

      Maybe others are right and the stress of multiplayer isn’t as relaxing as exploring at your own pace. Reflecting on this though, i’ve been very highly ranked in a number of games and there always comes a point when you realise you’ve “made it” and feel empty. I’ve just wasted 300 hours on being xxx rank at some game, my friends have moved on from it, what was the point, vs i completed skyrim and it was interested much like a book or movie is.

    • Bugmaster says:

      Any amount of time spent on any type of entertainment is wasted, almost by definition. If you want to be maximally productive, and you find yourself unable to work without occasionally engaging in entertainment, then you should strive to re-train yourself to desire less (ideally, zero) entertainment.

      Note that I said “if”. If being maximally productive is not your goal, then you can allow yourself to engage in some activities that make you happy. But, at that point, it all comes down to the numbers. If reading a book makes you X units/hour happy, but is also educational; whereas playing a video game makes you 10 X/hour units happy; then I suppose you need to decide how much education is worth to you.

      Realistically, the world would probably be better off if more people followed the first path. In the interests of full disclosure, I am unable to practice what I preach.

      • Machine Interface says:

        “Waste of time” is another one of these notions I feel only makes sense instrumentally. Treated as an absolute, everything is a waste of time, because any second spent doing anything is a second wasted for anything else you could be doing instead.

        If I measure the use of my time in regard to say, how well it’s conductive to my happiness, then entertainment is actually one of the most productive activities I can think of, for me, whereas engaging in costly altruistic activities would be a complete waste of my time relative to that goal, since I derive abolutely 0 emotional positivity from helping random strangers.

        Different, less sociopathic people than me will of course have different preferences and thus different internal scales of what constitutes a waste of their time for a given goal.

        But the blanket statement that all entertainment is a waste of time has little actual meaning — it’s just the manifestation of unexamined personal values of what goal is desirable (and thus what kind of activitivy is conductive to that goal).

        • Randy M says:

          This reminds me of the moral economics post.
          We don’t need to separate activities into blame-worthy and praise-worthy in this realm, either.
          But I think it merits consideration of the various kinds of pay-offs (short term enjoyment, long term appreciation, social benefits, etc.) that come from different optional activities, and also consider whether claimed benefits hold up (is reading The Great Gatsby superior to watching Breaking Bad, or is that merely intellectual snobbery?)

          • Conrad Honcho says:

            Mostly intellectual snobbery, definitely given your example. Breaking Bad is hailed as one of the most compelling stories and performances ever put to screen.

          • Nick says:

            I have never understood the love for The Great Gatsby. I was not impressed with Jay Gatsby, though some of my classmates fawned over him as this tragic character. It’s a good book, I guess, but it doesn’t strike me as the Great American Novel.

          • Randy M says:

            I remember two things from the Great Gatsby:
            1) There was a woman named Daisy who was loved by Gatsby and the narrator.
            2) There was a man who pronounced Oxford as Eggsford because he had excessive nosehair.

            #2 was a question on a quiz in English class that no one got right.

          • acymetric says:

            Same here. It has been a long time since I read it (maybe I should give it another go) but I think the reason I didn’t get into it was that I found all of the characters unlikable, so I didn’t care about any of their problems (which meant I didn’t care about anything in the book, more or less).

          • Nick says:

            1) There was a woman named Daisy who was loved by Gatsby and the narrator.

            No, I’m afraid you only remember one thing from The Great Gatsby. Daisy was Nick’s cousin; the girl he liked was her friend Jordan. 😛

          • Randy M says:

            Ah well. If they wanted High School me to remember it, they should have set it on a space ship.

          • Salem says:

            No, Randy M is right. Daisy is Nick’s cousin, but he has a thing for her nevertheless. He does things like pretend to be deaf so she’ll lean in to talk to him closer. At one point she asks Nick if he’s in love with her – playfully, of course, but it comes immediately after he describes how beautiful she looks in that moment, and how hearing her voice is an exhilarating tonic. I think we’re meant to believe that the answer is at least partly yes.

            Nick goes out with Jordan, but he doesn’t much like her.

            Unlike Gatsby and Tom Buchanan I had no girl whose disembodied face floated along the dark
            cornices and blinding signs and so I drew up the girl beside me, tightening my arms. Her wan, scornful mouth smiled and so I drew her up again, closer, this time to my face.

            He only kisses her because she’s there, and he dumps her with no remorse. And who can blame him? She cheats at golf.

          • Le Maistre Chat says:

            Aw come on guys, that’s all you remember?
            Jay Gatsby was a poor Jewish American who thought he was adopted by an Old Money guy after he saved him from drowning in a yacht accident, but he got no inheritance when the guy died. Jay was in love with Daisy Duck, but he ran away out of shame at not having money, and she married rich Pat Buchanan instead. World War I happened, and after coming home Jay turned to the 1920s Jewish mafia to make his fortune. Eventually he fulfills his dream of buying a house close enough to the Buchanans to spy on Daisy. That’s when her cousin Nick shows up.

          • Nick says:

            Salem, doesn’t Nick refer to himself as “half in love with her” after one of his last conversations with Jordan? He was evidently not very invested in the relationship, but it’s not as though he didn’t care at all.

            I’ll grant you that Nick may have a thing for Daisy.

          • Salem says:

            You are right, I’d forgotten that line. Fair enough, he has some feelings for Jordan too.

            Nick is an underrated character. He has a lot going on.

          • Hoopyfreud says:

            @Salem

            I always found that rather the point. Despite Gatsby’s best efforts to “break through,” he’s no more a man than Nick is, and Nick is no less a man than Gatsby. Basically everyone else in the story is lesser than them, though.

          • Paul Zrimsek says:

            I’ve always had a soft spot for Jay Gatsby because of the business of the green light on the dock, and especially the way “his count of enchanted objects had diminished by one” once it ceased to symbolize his unattainable beloved. Both feelings are so exactly like the sort I’d have in the same situation.

          • sfoil says:

            Gatsby tries and ultimately fails to make it in The City; a generation earlier, his character would have gone West (in fiction, if not in “fact”). He’s a WWI veteran, i.e. a product of America’s unprecedented engagement in European affairs. TGG implicitly portrays America’s transition from a wilderness frontier state to a settled modern nation as an accomplished fact and with zero nostalgia.

            At worst it’s a perfectly good tale in its own right. Other points of interest: Old West/Frontier/Colonial narratives have only gotten more mythical since 1925, Coming Up In The City narratives have not. Postwar Americans probably found this and its Roaring Twenties setting more relevant to their experience than the depressing Great Depression. Differences in life patterns mean the typical reader now probably doesn’t like it as much as the typical reader in, say, the 1950s.

    • raj says:

      My take: within any medium, there are high-value and low-value activities.

      High-value activities challenge you, are extremely novel, or are experientially powerful. Low-value activities tend to be comfort seeking, or sinking time. Playing wow for 30 hours a week is low value. Playing a mentally stimulating indie game with an amazing soundtrack is high value.

      Low-value activities have a place, because willpower is finite. But I’m on the lookout for activities with high value-to-willpower-expended ratios, i.e. those things that let you achieve zen.

  30. Uribe says:

    Say X is positively correlated with Y over a population. Then say you take the top Z values of X over that population. Is X still correlated with Y for those Z values of X? Is the correlation diminished?

    • Randy M says:

      Not necessarily still correlated. For example, if you charted “calories consumed during childhood” versus height, I think you’d get a somewhat linear relationship for lower amounts of calories consumed, but at the higher end it would top off as you reached the limit of environmental impact.

      Calories consumed versus weight, on the other hand, is going to be highly correlated at the top end.

      I think this is true for any trait that is partially hereditary and partially environmental. The environmental component can get you to your hereditary maximum, but adding more will not give large increases beyond it, and may well be detrimental after a certain level.

      See also Iodine & IQ, or sun exposure and mood.

      • Uribe says:

        The example I had in mind comes from something I read about how it makes sense that Tom Brady is the best quarterback in the NFL because he is the best-looking and that looks and athletic ability are positively correlated. But I was skeptical because he is playing in a league that has already been sorted for the top athletes in the world. My intuition is that once you’ve made this sort, the correlation must be at least significantly diminished, but I’m too lazy to actually figure out the math.

        • quanta413 says:

          I too am too lazy to even Fermi estimate it, but I’m pretty sure you’re right. Everyone in the NFL is an athletic outlier by several standard deviations. There’s a huge restriction of range problem that should shrink the correlation. Which I doubt is higher than .5 anyways.

          I’m sure most men in the NFL are physically more attractive than normal too. Looking at Tom Brady, it’s not obvious to me that he’s the most attractive man in the NFL. He’s good looking, but I’m not blown away. And I don’t really want to look through hundreds of NFL players and rank them.

          • gbdub says:

            It’s not even clear that Tom Brady is the most “athletic”. Other quarterbacks (let alone other positions) are stronger, others are definitely faster (go watch his combine 40 yard dash if you want to see a young Tom Brady looking very mortal). Heck, part of the reason he was famously drafted so low was because he ceded playing time at Michigan to a more physically talented QB.

            Brady’s advantages seem to be primarily intelligence, control, rapid decision making, and an excellent coach.

          • Randy M says:

            I would expect quarterbacks to be handsomer than other positions, due to them being some of the better known players on a team, and possibly something of a face for the organization. But I don’t think this is directly related to their athleticism.

          • Salem says:

            One reason we might expect quarterbacks to be handsomer than other positions is because the physical demands are less extreme – they don’t need to carry the huge bulk of offensive linemen, the cartoonish necks of linebackers, etc. They are basically just young, fit, athletic men, which is the general ideal of “handsome,” whereas being hyper-specialised in an unusual physical skill makes people look weird. Quarterbacks also get hit less than most other players, which probably helps their looks.

          • A Definite Beta Guy says:

            I agree with Salem, the QB can be more attractive because he is more generically handsome because he doesn’t have to be as athletic or specialized as other positions. Great QBs are obviously athletic compared to us SSCers, but the stuff that makes great QBs is not related to sprinting super fast or lifting a lot of heavy weights (though you do want some rather amazing arm strength).

            Are there any ugly top QBs, though? I can’t really think of one. The most I can think of are some aging QBs that don’t look as good as they used to, like Drew Brees.

          • gbdub says:

            Colin Kaepernick and Roethlisberger are kinda ugly. Nobody’s really below the (high) floor of attractiveness for “tall, generically in shape dude”, but other than Brady, Aaron Rodgers, and maybe Cam Newton, are there any that stand out as unusually handsome?

          • albatross11 says:

            Tall, in-shape, high-status dude probably does wonders for their success rate.

          • baconbits9 says:

            Peyton Manning is definitely not handsome, nor is his brother. Brees is fairly average looking.

    • quanta413 says:

      The correlation could do pretty much anything. It could get stronger (if the underlying relationship is appropriately U-shaped with low enough noise), it could disappear (imagine a step function dependence of Y on X), it could reverse (imagine a step function plus a sawtooth function dependence of Y on X), it could stay the same (linear relationship with 0 noise). It depends on the underlying data set and on the value of Z. I think the most “typical” scenario given some random data with a positive correlation would be that the correlation would be diminished, but I wouldn’t depend on that at all.

      Correlations without a robust underlying theory don’t tell you very much.

      Any more particular scenario you have in mind?

    • Salem says:

      Maybe, maybe not. Depends on the causal relationship. But there are many possible reasons why the correlation might be diminished or even reversed, eg Berkson’s paradox.

    • Björn says:

      You can take the population of (n, n) for n in {1, 2, … , 1.000.000}. Then the first entry and the second entry is obviously very much correlated. Now if you add {1.000.001, 0} to the population, the overall correlation suffers only a little bit, but if you look at the member of the population with the highest first entry, it is gone.

  31. pontifex says:

    From a New Yorker article about the Silicon Valley TV show.

    During one visit to Google’s headquarters, in Mountain View, about six writers sat in a conference room with Astro Teller, the head of GoogleX, who wore a midi ring and kept his long hair in a ponytail. “Most of our research meetings are fun, but this one was uncomfortable,” Kemper told me….

    “He claimed he hadn’t seen the show, and then he referred many times to specific things that had happened on the show,” Kemper said. “His message was, ‘We don’t do stupid things here. We do things that actually are going to change the world, whether you choose to make fun of that or not.’ ” (Teller could not be reached for comment.)

    Teller ended the meeting by standing up in a huff, but his attempt at a dramatic exit was marred by the fact that he was wearing Rollerblades. He wobbled to the door in silence. “Then there was this awkward moment of him fumbling with his I.D. badge, trying to get the door to open,” Kemper said. “It felt like it lasted an hour. We were all trying not to laugh. Even while it was happening, I knew we were all thinking the same thing: Can we use this?” In the end, the joke was deemed “too hacky to use on the show.”

    • skef says:

      “I think even Amazon had an outage because one of the admins fat fingered a DNS or ACL change at one point.”

      From my perspective a few dozen feet away, that was a tense mess with a sad aftermath.

  32. beleuze says:

    I’m moving next year to the Bay Area from abroad to work for one of the big tech companies.
    I’m in my early twenties. I like philosophy (both kinds), music (both kinds) and experimental videogames. I enjoy SSC very much and I kind of identify with ‘rationalist’ way of thinking without having much confidence that it is really right.

    Any tips on
    a) finding people I might get along with?
    b) living as cheaply as possible in SF/the bay?

    • Showing up at SSC meetups is one way of finding people you might get along with.

    • Rebecca Friedman says:

      Welcome!

      I have to ask – what are the two kinds of music? Playing and listening? Instrumental and vocal? Current and historical? Two things I’m not thinking of?

      My own hobbies are dance and music – I’m only indirectly connected to gamers, at least realspace, but if you want recommendations for dance and music I can cheerfully offer them. (You could try BayCon, but I haven’t had much luck socializing there – I may just be an introvert though.) As my father says, SSC meetups are also an option for meeting people; I think we have a reasonable density of them. As far as living as cheaply as possible – do you cook? If you do, the south bay at least has a lot of nice ethnic grocery stores which tend to have both a wide variety of ingredients and very reasonable prices. My favorite locally is Lion Market, but there’s also a very nice Indian supermarket on El Camino a few blocks east of the San Tomas, I’ve even found shelling peas there. Farmers’ markets are mostly not actually farmers’ markets, and tend to be heavily overpriced; be wary of them. (The closest to real farmers’ markets I’ve found so far are stands on the highway that say STRAWBERRIES; those are for real and perfectly affordable but they don’t appear until you’re well into the central valley, so unlikely to come up much).

      Do you drive, and plan to have a car? If so skip this paragraph. If not, public transportation in the south bay is affordable and as far as I can tell safe (at least where I’ve used it) but slow as all get-out; you can get a card to scan for the bus and it generally works fine, but unless where you’re going is on a fairly straight route from where you are Uber/Lyft may be a better bet. Everything I’ve heard about them has been good. This is pretty much all South Bay, I don’t know San Francisco as well.

      Housing is, as you no doubt know, hideously expensive. The best plan I’ve heard is to find roommates you like; there may be rationalist networks for that, I’m not really connected to that community. I’m afraid beyond that, I can’t really give you any advice everyone else couldn’t give better – good luck!

      That’s most of what I can think of. Sorry for the late post, hope you’re still reading! Um, if you are and have any specific questions, feel free to ask?

  33. Vermillion says:

    Rumble thread, hurray!

    I don’t know who all scans to the bottom of an open thread looking for a play by email game of super heroes battling to the death, but looks like that set includes you. The first edition of this is already over, over the last month contestants picked powers, secretly bid on them and had a royal rumble leaving Subject4056 the undisputed champion. Honestly it was all over quicker than I’d expected but then again in an alternate universe… But rather than retread that path why not have a new match?

    Anyone who’s interested in getting in on this round can sign up here: https://goo.gl/forms/J4sX9Ej7qW3FuKlu1.
    Anyone who’d like to submit a power (non-players included) can do so here: https://goo.gl/forms/OhpCaRIwpcKRAeaF3
    Check out previously submitted powers here: https://docs.google.com/spreadsheets/d/1G0ALnq6W7pilKH9C9GwwSNo0zjuKCuAe1e8TQYAHUlw/edit?usp=sharing
    And read the official rules here: https://kevan.org/rumble.cgi?genre=hero&mode=rules

    Signups will be open for the next week. I’m assuming the contestants from the previous round are looking for a rematch but if you’re not feeling particularly brave then feel free to admit your cowardice below.

  34. johan_larson says:

    What dishes are required for a really archetypal fully orthodox Thanksgiving dinner?

    I think you need
    – roasted turkey
    – gravy
    – cranberry sauce
    – mashed potatoes
    – stuffing
    – pumpkin pie
    – some sort of greens (peas, green beans, and Brussel sprouts are all acceptable)

    Nothing else is required. Other dishes may be added, as long as they don’t overwhelm the central dishes. Nothing is specifically forbidden.

    • Aging Loser says:

      Thanksgiving is a national religious holiday. On Thanksgiving, Americans should either thank God for making America great or shouldn’t make the defiling gesture of eating turkey with relatives on this particular day. And to make a special point of thanking God for personal favors on this particular day, without having first thanked God for making America great, is doubly defiling.

    • SamChevre says:

      If you are from the South, I think candied sweet potatoes in some form and pecan pie are also required; mashed potatoes might be optional.

      • Hoopyfreud says:

        Agreed that pecan is an acceptable alternative to pumpkin.

      • dodrian says:

        Yes – swap the mashed potatoes for sweet potatoes, and pumpkin for pecan.

      • Nabil ad Dajjal says:

        I don’t think that either of those are southern. We make sweet potatoes with marshmallows on top and pecan pie up here in New York too.

      • acymetric says:

        I suspect there has been a lot of homogenization of Thanksgiving foods throughout various regions due to the way Thanksgiving has been popularized and promoted, with various regions adopting each other’s dishes until they were all mostly the same. It would be interesting to look into where some of these foods originated. Did a given food like sweet potatoes with marshmallows on top start out southern and migrate elsewhere, start out in New England and migrate to the south, or were they always fairly universal (I’ll admit it sounds like a pretty southern thing to do, but that is far from conclusive)?

        Well, interesting is maybe the wrong word because it isn’t interesting enough for me to actually do a deep dive on it, but I would be curious to know!

    • A Definite Beta Guy says:

      Cranberry sauce is optional, as far as I’m concerned. We stopped serving it at our Friendsgiving after we had a friend from Master Chef prepare his own version and everyone still disliked it. Let’s kick it out.

      Green Bean casserole is usually featured, but not sure how “green” that really is…

      I’m also of the opinion that stuffing needs to come from inside the bird, but more and more people are absolutely grossed out by that. I’m going to keep doing it until I get food poisoning.

      • Randy M says:

        The Gordon Ramsey Master Chef competition?

      • Nick says:

        Yeah, green bean casserole is pretty common.

        I had a friend whose friends went through the Gordon Ramsay experience (I think it was on Kitchen Nightmares). She was loudly critical of him and said him and the show was completely fake. A shame, since it’s just about the only reality tv show I would watch. 🙁

      • Stuffing is mostly from inside the bird but you always make too much to fit and the excess gets cooked separately–ideally with dripping from the bird poured over it from time to time.

        • A Definite Beta Guy says:

          I think the drippings are essential. I made the extra stuffing on the side this year and it was abhorrent.

          • arlie says:

            I once made veggie stuffing, to increase the amount of the standard Christmas dinner the family veggies would eat. No dripping, obviously. They seemed to find it worth eating. (And for the record, from the same year – margarine is not as good as suet in other traditional Christmas recipes :-()

          • gbdub says:

            Wouldn’t shortening, rather than margarine, be the right veggie alternative to suet?

        • gbdub says:

          The problem with in-the-bird stuffing is that, by the time it’s cooked through, the bird is overdone. Is there a good method for pre-cooking the stuffing such that the meat and stuffing reach optimal doneness simultaneously?