Open Thread 145

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 – and also check out the SSC Podcast. Also:

1. In last month’s links, I posted about concern that Alcor was getting too close to a weird cult. Alcor leader Max More commented with an explanation of why that wasn’t a fair claim, which I believe. I apologize for helping spread an exaggerated and poorly-contextualized version of the story.

2. Also in the spirit of “very long comments by very important people” – Brian Earp of Yale University’s Program In Ethics And Health Policy posted some thoughts on the adversarial collaboration about circumcision.

3. But if you prefer very short comments by very important people, here’s Gary Marcus on Twitter about GPT-2 chess. Gary Marcus calling your work trivial is how you know you’ve really made it in AI! Also, with all of the great Gwern stuff I mentioned on that post I should have mentioned that Gwern has a Patreon that helps fund his projects. Also, it looks like a commenter made a better GPT-2 chessbot, although you can’t play it.

4. And if you prefer medium-length comments by non-famous people – chaosmage tries to extend my Why Doctors Think They’re The Best post to programmers.

5. This is your absolute last chance to vote for the winner of the adversarial collaboration contest. I’ll be naming the victor and distributing prizes this coming week.

6. And this is your second-to-last chance to take the 2020 SSC survey if you didn’t already; I’ll probably post one more reminder this week, then close it on Friday or so.

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

741 Responses to Open Thread 145

  1. Nancy Lebovitz says:

    https://www.inverse.com/article/62213-crispr-gene-editing-electroporator-cheap

    I’ve been thinking about Home-Build-a-Virus kits for some time– there are lot of positive and negative possibilities. This is short of that, since it’s about modifying existing organisms rather than, say, building a polio virus from the genome.

    Still, I think all the very serious people who thought genetic engineering could be controlled and were trying to decide on the rules are mistaken.

    It doesn’t work on mammalian cells, though, and I have no idea how cheap an electroporator for that would be.

    Minimalist fact check: Folks at reddit seem to think it’s real.

    • profgerm says:

      Super cool!

      The Odin also sells home Crispr kits among other low-cost genetic tools and experiments. The website also includes a brief guide to human adult Crispr. I’ve considered getting one of their “home labs” to play around with yeast and gene mods.

      The founder was briefly under investigation by California for practicing medicine without a license, but that was dropped.

    • toastengineer says:

      There’s a YouTube channel called The Thought Emporium that does a lot of really hardcore biology stuff. I didn’t realize hobbyist bioscience was anywhere near as advanced as what he’s doing. He used an engineered virus to cure himself of lactose intolerance.

    • metacelsus says:

      Electroporation for mammalian cells (which I have some experience with) wouldn’t be much more difficult.

      But electroporation and building a virus are two quite different things.

  2. johan_larson says:

    The Dawson City Sourtoe Cocktail: a glass of whiskey with an amputated toe in it. Drink the whiskey. Do not eat the toe.

    • Three Year Lurker says:

      Huh, the normal sign on the wall says, “If you eat the toe, you have to replace it.”

  3. Radu Floricica says:

    Speaking of classics, I read Homicide by Daly and Wilson a while back mostly because… well, the title sounded cool. I was younger, ok? Anyways, what I got was an amazing mix of theory and practice, and a long lasting respect for the people that make evolutionary psychology actually work.

  4. Trofim_Lysenko says:

    2010: True Grit (I still really like the original, mind you, but this manages to stand on its own)
    2011: The Raid (AKA The Raid: Redemption. One of the best Martial Arts movies I’ve seen in years)
    2012: Act Of Valor (I really enjoy this on a technical level while admitting that for the casual viewer it’s probably fairly weak. Still, my list, my picks, and I’ve rewatched scenes from this movie far more than other, better 2012 movies like Zero Dark Thirty, Skyfall, and Dredd)
    2013: The World’s End (Someone was asking where the good new comedies are? Edgar Wright is one of the answers)
    2014: Edge Of Tomorrow (It’s listed enough times in this thread I don’t feel I need to explain further).
    2015: Sicario (Tense, intense, beautifully shot thriller that makes maximum use of its material)
    2016: Rogue One / Zootopia (Ok, I’m cheating my own self-made rules here, but I legitimately couldn’t tell you which one of these I enjoyed more, for very different reasons. Rogue One has storytelling and characterization issues, but as far as I’m concerned it and The Mandalorian are the only visual medium works to come out since the Disney acquisition that for a lack of a better term GET Star Wars and what makes the setting so memorable and enduring. Zootopia, on the other hand, is probably my favorite disney movie since the “golden age” of the 90s, with great chemistry between the lead voice actors, a charming conceit, and surprisingly good mystery writing for a while. I would sort of love a ZPD-focused police procedural.)
    2017: Baby Driver (great on-foot action, outstanding performances from a strong cast, Edgar Wright’s eye for clever visual storytelling and layered detail, and the best driving stunts and shots in the past two or three decades)
    2018: Spider Man: Into The Spider-Verse (Arguably the best Spider-Man movie to-date, and truly gorgeous visuals.)
    2019: Avengers: Endgame (I had some issues, and I can’t claim this to be the best movie on the level of auteurship and such, but as catharsis and managing to stick the landing on such a monumental piece of serial storytelling, I have to give it to Endgame. Sticking the landing on a big franchise with a large, dedicated fanbase is not easy, just look to Star Wars, and Marvel Studios pulled it off with a 20+ film combo.)

    Honorable Mentions: Mad Max: Fury Road, Pacific Rim, Big Hero 6, Rango, Scott Pilgrim Vs. The World, John Wick, The Man From U.N.C.L.E., Skyfall, You Were Never Really Here, What We Do In The Shadows

    Movies From The 2010s That Would Likely Have Affected This List If I’d Actually Seen Them Yet: Blade Runner: 2049, The Grand Budapest Hotel, Isle Of Dogs 😛

  5. rubberduck says:

    Out of the shadows, a new superhero arises: Cancerman! (Cancerman is open to more creative naming suggestions.) As his name suggests, Cancerman has the power to cause cancer in anyone he physically touches. The cancer can affect any body part and progress as fast/slow as Cancerman desires. Other than its surprise appearance, the tumor is biologically normal and there is no obvious way to trace it back to Cancerman, though if a large number of Cancerman’s enemies suddenly develop leukemia people might get suspicious. Because he is afraid of shady government organizations kidnapping him to figure out the secret of his abilities, Cancerman would strongly prefer not to reveal his powers and identity to the world.

    So, how can Cancerman best use his powers to fight evil and make the world a better place? He isn’t Batman so he doesn’t mind killing a few people if it’s for the greater good.

    (Also: yes, I know Dr. Manhattan was accused of causing cancer, but that’s neither here nor there.)

    • Theodoric says:

      If he doesn’t want to reveal himself, he would have to keep his area of operations small. Maybe he hangs out in high crime areas of his city does things like buy drugs and infects the dealer during the hand off? Say he wears a wide-brimmed hat and dark glasses, leaves his smartphone at home, and security camera footage is reviewed by the same people who never notice Clark Kent looks a lot like Superman.

    • Radu Floricica says:

      Get a janitorial job in Congress and do something about that 80-90% incumbent rate?

      • Mark V Anderson says:

        Very snarky but I approve! I presume a similar process in Romania would be a good idea?

        • Radu Floricica says:

          And with a lot less guilt. Most of them would probably be well deserving the death penalty for second order deaths. Non-functioning health system, absolutely horrible accident rates due to the lowest number of highways in Europe and so on.

    • johan_larson says:

      CM could probably exercise a bit of influence in politics without revealing himself. People running for office often do a lot of “pressing the flesh” when they are campaigning. I bet if CM made an effort, he could shake the hand of at least one of the Democratic Party presidential candidates. Give them a variety of cancer that appears quickly but is very responsive to treatment, and they’re out of the race. Heck, CM might be able to reach several of them. Trump himself would be a more challenging target, though.

    • MorningGaul says:

      Obviously, give everyone benign cancers to help them qualify for medical marijuana. If that’s not making the world a better place, I don’t know what will.

    • Well... says:

      Name suggestion: Radioactive Man

    • helloo says:

      Can you get the cancers to “fight” other cancers then stabilize? For example, by creating a human body version of fire breaks or such.

    • AnarchyDice says:

      So, could Cancerman create arbitrary cancers? Could he for example give himself or others benign cancers that mutate in specific ways to become noncancerous after growing for a given length of time? If so, then he’s got biomanipulation powers that he could use to heal sick children (say by volunteering as a traveling santa and having them sit on his lap), heal people using an energy healing/crystals as a front (with random delays in the start of their healing to obscure his effectiveness), or possibly even modify a transmissible bacteria or fungus that cures various diseases or eradicates parasites.

      If not, perhaps he can go to parasite heavy areas of the world and infect the parasites with species specific transmissible cancer to cut down on their populations.

      Alternatively, he could help cancer patients by giving their cancerous cells cancer that is unstable and dies off in a generation or two.

      If he can give cancers to himself, he can use that to aid in his disguises by using small subcutaneous cancers to alter his face shape or even temporarily disfigure himself to make it harder to pin anything on him.

  6. Well... says:

    I read “Blindsight” by Peter Watts and really enjoyed it a lot, even though it was more “out there” than the Neal Stephenson and Kim Stanley Robinson sci-fi I usually read. (“Out there” mainly because of the vampires and multiple personality thing.) In some ways I liked Blindsight even more, and I found the extensive notes section at the end a ton of fun.

    Based on that, what should I read next, SSC?

    • johan_larson says:

      If you like science in your science fiction, you might try Robert L. Forward’s Dragon’s Egg.

      • Well... says:

        Thanks! I requested it from the library, but it has to be shipped in from some other branch.

        In the meantime, what else?

      • Joseph Greenwood says:

        Dragon’s Egg is an excellent piece of fiction. You might also enjoy Lady of Mazes (which I originally read because of a passing reference from our host), which in my opinion is emotionally more like Blindsight than Dragon’s Egg is.

        What specifically did you enjoy about Blindsight? (Rot 13 spoilers for anyone following along that hasn’t read the book yet)

        • Well... says:

          Rot 13 spoilers for anyone following along that hasn’t read the book yet

          I don’t think I have any spoilers, but I wouldn’t hide them if I did. Just so you’re warned.

          I don’t know if it’s The Specific Thing I Liked about Blindsight, but Blindsight was compelling/outstanding for its ability to present idea after idea that followed this pattern:

          1. Present idea
          2. Psssh, that’s a crazy out-there idea. What kind of nonsense is this?
          3. Oh wow, that’s a lot of valid science/reasoning to back up that idea
          4. Why did I ever think that idea was farfetched?

          • Radu Floricica says:

            Including vampires. The cross thing is brilliant.

            Nothing recent comes to mind with the same level of realism, but in my younger days I kept rereading and enjoying Heart of the Comet (David Brin and Gregory Benfort). Probably best description of realistic colonization I’ve found, in that it can actually happen pretty much like this. And from what I remember it aged very well (David Brin tends to age well, I think).

      • Well... says:

        I don’t understand…are these books I can check out of my library somewhere, or are they online media I have to read on a screen?

        • profgerm says:

          AO3 and SCP are, to my knowledge, online-only. Screen it is!

          Antimemetics Division is good, but I’m not quite certain I’d call the ending satisfying. Still glad I read it.

    • Randy M says:

      I would say that goes without saying, and in fact nobody else said it, so maybe that’s proof, but I’m saying it, so now it goes with saying. I’m confused.

      Until someone says it, it isn’t clear whether it goes without saying, or simple doesn’t go.

      For instance, I wasn’t aware there was a sequel.

    • AlexOfUrals says:

      I also liked it a lot and am going through Echopraxia now. You might want to look into Alastair Reynolds works if you haven’t already. Like Blindsight, the Revelation Space universe also tries harder than usual for some scientific accuracy, while at the same time being “out there”, somewhat surreal and very Lovecraftian-grim. Other than Revelation Space itself, I highly recommend Diamond Dogs novella. Both follow a ragtag butch of highly qualified misfits trying to unravel an alien enigma gb gurvef zbfgyl tehrfbzr raqf.

    • positronicman says:

      Watts and Egan are on a level in my book.

      If somehow you haven’t yet encountered Greg Egan, he’s a computer scientist who writes speculative but very (physics/compsci) grounded fiction. He’s got 9 or so novels and a bunch of short stories. I haven’t read his shorter works, but have absolutely loved his stand alone novels (he also has a trilogy, which I haven’t read).

  7. rahien.din says:

    In OT 144.75, there was a discussion of p-zombies. One of the commenters tarred Nagel with the same brush.

    It got me thinking again about Nagel’s essay, “What is it like to be a bat?” As a firm materialist, one has to give ground to Nagel’s argument. What convinced me was the realization that a bat has no idea what it is like to be a human. Moreover, if there was an intermediate zone between bat and human, a person in that zone would only understand what it meant to be that thing – they would not understand what it meant to be a human, nor would they understand what it meant to be a bat. One could even say that each individual bat’s experience is unique and inaccessible even to other bats in the same colony! The only way to understand what it is to be a thing is to actually be that thing. Or, one can only truly understand the discrete state that they currently inhabit.

    Essentially this is to say that the character or quality of an experience is inextricably linked to and determined by the physical configuration from which that experience flows. Otherwise it would be possible to understand what it is like to be a bat, without actually being a bat.

    But this means that Nagel’s bat (the inaccessibility of qualia) is a powerful illustration of… materialism. If the immaterial was to any degree independent of the material, then it would be possible for one’s immaterially-based understanding to encompass a state not arising from the local immediate physical state of the person doing the understanding.

    Therefore, to the extent that we rely upon the concept of qualia, we must be materialists.

    • Enkidum says:

      For what it’s worth, I’m pretty sure Nagel disagrees with your interpretation of Nagel. But that’s probably not that important.

      One could even say that each individual bat’s experience is unique and inaccessible even to other bats in the same colony! The only way to understand what it is to be a thing is to actually be that thing. Or, one can only truly understand the discrete state that they currently inhabit.

      That is a possible logical consequence of Nagel’s argument, yes, but I think he rejects it. If I remember correctly (dubious) in some later exchanges he pointed towards fundamental sensory differences (i.e. sight vs sonar, etc) as being a critical part of the non-commensurability of different organism’s states). So we can’t understand what it is like to be a bat, because a bat is really, really different from us, but maybe we can understand what it is like to be another person, because they’re not that different.

      I think there’s something to that point. I go back and forth on whether I take it seriously.

      If the immaterial was to any degree independent of the material, then it would be possible for one’s immaterially-based understanding to encompass a state not arising from the local immediate physical state of the person doing the understanding.

      Your consequent doesn’t follow from the condition. The reverse is true – if we could have the same experiences as entities that are not in our physical state, then qualia (or whatever) must be non-physical. But there might be limitations on the non-physical that stop us from having true empathetic perception.

      It’s usually about this point that I lose patience and roll another joint.

      • rahien.din says:

        Nagel’s entire argument rests on the idea that these essential differences are ineffable and unknowable. Thus, he isn’t permitted to claim some differences matter and others don’t. If he’s able to make that claim, it is only on the basis that differences are knowable and expressible.

    • KieferO says:

      The response to Nagel: “What is it LIke to Be Boring and Myopic?” by Kathleen Akins is worth reading. I like it because it takes the challenge set forth very seriously and very literally. It offers a constructive account that the inaccessibility of qualia is really just a failure of imagination.

      • Enkidum says:

        Heh, I’d forgotten that one. It’s an excellent piece, and essentially cemented my opinions on the subject years ago.

        ETA: Here is the pdf (from her webpage).

        • Protagoras says:

          Odd that it was in a Dennett and his critics collection, insofar as it seems entirely sympathetic to Dennett; seems like it should have been in a Nagel and his critics volume or something. But an excellent essay.

    • Viliam says:

      In OT 144.75, there was a discussion of p-zombies. One of the commenters tarred Nagel with the same brush.

      Did they suggest Nagel was a p-zombie?

      The only way to understand what it is to be a thing is to actually be that thing. Or, one can only truly understand the discrete state that they currently inhabit.

      So, you can’t even understand what it was to be like you 5 minutes ago?

      Seems to me like you redefined “understand” to mean “experience”.

      • rahien.din says:

        So, you can’t even understand what it was to be like you 5 minutes ago?

        This is a necessary consequence!

        Qualia are necessarily perceptions. Perception is a combination of sensory data and prior experience. So, not only is each person distinct from every other person, each moment is distinct from every other moment. You are a different person today than you were yesterday.

        It would be fair to say that two moments could be so similar that there is only a tiny difference between them. Just as the experiences of two bats would be similar-but-not-identical and thus have only a tiny “quale” separating them, so too would be the moments of your life.

        But even this is a relationship : the degree of physical difference determines the degree of experiential noncommensurability.

        Seems to me like you redefined “understand” to mean “experience”.

        I’m describing how they are linked. The point of Nagel’s bat / Mary of the black and white room / qualia in general is to link understanding with experience.

  8. broblawsky says:

    To what degree did the German tradition of late-Renaissance armorsmithing lead into the modern German expertise in precision engineering and metallurgy? I recently went to see exhibits at the Metropolitan Museum on etching, armor-smithing, and early scientific instruments, and it occurred to me that German precision craftsmanship was a significant unifying element for all three.

    • Björn says:

      I think I would rather take certain aspects of the 19th and 20th century into account. Industrialization came later to Germany than to England, so Germany simply skipped some legacy technology. Additionally, Germany had a modern education system, and not many lucrative colonial adventures.

      In the 20th century, Germany lost two wars and experienced serious political turmoil and a genocide. However, this also means that Germany has a very modern, thought out constitution and political system. It also still has a very modern educational system, which is geared towards giving 30% of a cohort a good university education. This is a lot, compared to the focus on elite universities in countries such as England, France and the USA. Since the universities are mostly free, there is no reason to sugarcoat anything for the students, so the education there is quite demanding.

    • eigenmoon says:

      In the French-Prussian war of 1871, the French had superior rifles (Chassepot vs. Dreyse needle guns) [EDIT: Chassepots were produced in Britain and Austria as well] but the Prussians had steel cannons. I could imagine the ability to make steel cannons translating into modern German expertise in metallurgy.

      But it doesn’t seem to me that the Germans had any significant technological advantage in the times of Friedrich the Great and Napoleon, at least on the battlefield.

    • Aapje says:

      @broblawsky

      Germany has a culture of engineering excellence, to the point of (severe) over-engineering. This actually caused major issues for Germany in WW II:
      – High quality designs of weaponry that was a ‘consumable’ in combat, causing the Germans to field weapons that were very good, but not good enough to outweigh the lower production numbers and/or high use of resources.
      – Lots of improvements once production started, causing slow production, where the Americans and Russians often refused to make improvements unless they increased production or were major benefits
      – A lack of willingness to sacrifice operational quality for easy maintainability. Americans and Russians could often repair their weaponry much faster.
      – A lack of willingness to make designs that low-skill workers could make and/or accept ‘good enough’ quality from workers, slowing down skilled workers and causing a shortage of sufficiently skilled workers.

      The result is that modern collectors/connoisseurs tend to appreciate German WW II weaponry (German tanks, planes, small arms like the StG 44) more than the allied stuff, as war-winning stuff like high production and easy maintainability is far less relevant to collectors/connoisseurs.

      Anyway, I wonder when this culture originated? Germany has historically had fewer resources than most competitors, so perhaps this culture is a result of trying to compete by doing more with less. Between WW I and II, Germany was heavily restricted in what weapons they could have & was forced to into significant reparations, which might have encouraged them to spend lots of manpower on advanced engineering.

      • cassander says:

        I think a bigger problem that the germans had was a severe underproduction of spares in general, a policy substantially but probably not entirely driven by Speer’s desire to pump up the production numbers to make himself look good. Sure, it was a huge pain the ass to get the engine out of a panther, but if there wasn’t a spare engine, that didn’t really matter.

      • TJ2001 says:

        Stalin famously said “Quantity has a quality all it’s own”. It’s worth remembering.

        As to the supposed superiority of German weaponry – I think a lot of that amounted to the a combination of the simple fact that they were the ones fielding war machines when nobody else was ready and their military was fairly well organized around being driven towards getting results with what they had at hand. Contrast this against the Pre-WWII Soviet military which was organized around compliance with political ideology rather than achieving military success.

        It took the Russians a while to sort out the whole bit about “When the ideology produces results opposite of what we need – the ideology needs revision..”

        That’s always a good idea – but often it takes cognitive leaps which idealistic people are often unwilling to make…

    • Jake R says:

      Possibly related: When I sat my college graduation they announced the graduates by department. The department of engineering and science seemed to have 2 or 3 times as many people named Smith as any of the other departments. I remember wondering if there was any chance it was genetic.

  9. theodidactus says:

    A few days late to the thread but whatever.

    I’ve released an update to Skybreak!, my Choice-based Science-Fantasy space adventure.

    You can play (or download) it here:
    https://ifdb.tads.org/viewgame?id=b21j5feykymttixm

    You can play it online here (though you can minimize the automatic “map” pane to the right)
    http://play.adrift.co/?game=http://www.adrift.co/files/games/Skybreak!%20(1.1).taf

    Skybreak! scored 12th in last year’s IFComp and is the winner of the prestigious/dubious “Golden bannana of discord”, awarded annually to the IFComp game with the highest standard deviation in scores by judges. I’ve resolved to try to release small expansion/updates to the game for at least the next few years, in my quest to make it among the largest works of interactive fiction ever created.

  10. Alex Zavoluk says:

    Austin, TX meetup: The designated “far-comer’s” meetup will be Saturday, February 8th. If you would like to attend but are too far away to attend regularly, this meetup is for you!

    Location: Central Market, 4001 North Lamar Blvd, Austin TX 78756. We usually sit in the cafe upstairs, either inside or outside.

    If you would like to receive additional information about our meetup, join the google group at https://groups.google.com/forum/#!forum/austin-less-wrong

    Additional search tags: Texas, Waco, College Station, Bryan, Dallas, Houston, San Antonio

    • Dacyn says:

      Dallas SSC is carpooling there leaving at 10:00 for the 1:30 meetup; for more information email me at david9550 (gmail). [Though I guess this thread is dead…]

  11. Statismagician says:

    Has anybody come across a useful way to quantity analytics teams’ productivity? We’re reviewing KPI definitions where I work and it’s an obvious hole. The consensus, and my personal intuition, is that you can’t, really; analysts are exactly the kind of people who’ll figure out how to game any gameable system, so active management is sort of the only way to go – I thought I’d ask here anyway.

    • AG says:

      Urgh, our comapny’s quality system is forcing us to add a KPI every year and it’s setting off all of my “Goodhart’s Law! Goodhart’s Law! Goodhart’s Law!” mental alarms. Our current KPIs have been fairly low-stakes, low-hanging fruit “maximum allowed number of mistakes” type stuff (e.g. number of typos in a particular field), but I bet they’re only going to get more and more contrived.

      Otherwise, you have to first figure out what sort of outputs are strongly influenced by inputs out of the team’s control. For example, after years of measuring productivity by number of X completed, the trend became clear that that measure was basically pegged to the number of X that was submitted to my department. If we received more, we completed more. So, I guess, the first thing to do is to identify what things should NOT be ever used for the KPI.
      Then you can do some of the aformentioned low-stakes, low-hanging fruit “maximum allowed number of mistakes” type stuff. Try to pick KPIs that evaluate quality, over trying to change productivity rates. Things where there’s a clearly wrong state that you want to avoid, and that’s what the KPI targets.

      • Statismagician says:

        That’s what I was thinking. We’re probably going to do something on request handle timeliness, just to have something.

    • Ouroborobot says:

      We have SLAs for our actual structured sources, and to that end I built a subsystem for collecting QA metrics with regard to quality and timeliness which is then used for dashboarding and historical trending. Pretty standard. But as for the actual productivity of the analysts and data scientists who use the data for ad hoc tasks, no, pretty much just active management in a generic agile fashion.

  12. Enkidum says:

    I’d recommend Behave! as well, though it does get very repetitive, and he is a bit too much of a fan of studies of dubious qualities that haven’t been replicated. He is by no means someone who fits neatly into the modern sociobiology or evolutionary psychology framework, but he respects the work from those fields and acknowledges it carefully. It’s as much about neuroscience and anthropology as it is about biology and evolution, which is critical. I would also second @Randy M’s recommendation for his lectures, which are great, I found them more useful than the book – although this course was clearly a dry run for his writing of the book, so there’s a lot of overlap. (Also his other books are very worth reading, though they’re on very different topics. Highly recommended: A Primate’s Memoir, which is his kind-of autobiography.)

    Much more technical, and also not evo psych but sharing a lot of the same principles, is Mathematical Models of Social Evolution: A Guide for the Perplexed. This is a straight up math textbook, and not a particularly easy one for those of us who stopped taking math in Grade 12, but if you learn to understand the equations, you will have much better understanding of the theoretical bases of these fields. The basic insight is to have infinitely iterating prisoner’s-dilemma-type situations, with different strategies and more complex possible behaviours, and throwing in things like memory, changing who you are matched with, etc etc etc. Really helped me a lot, but I took it as a graduate course and I’m not sure I would have gotten through it on my own. Note that the authors come from a camp that is essentially opposed to simple evo psych. Either way, the math is good.

  13. johan_larson says:

    Sam Altman, formely president of Y Combinator, one of the top accellerators in Silicon Valley, yesterday published the essay How to Invest in Startups.

    The part I found questionable in the essay is his description of what to look for in founders:

    I look for founders who are scrappy and formidable at the same time (a rarer combination than it sounds); mission-oriented, obsessed with their companies, relentless, and determined; extremely smart (necessary but certainly not sufficient); decisive, fast-moving, and willful; courageous, high-conviction, and willing to be misunderstood; strong communicators and infectious evangelists; and capable of becoming tough and ambitious.

    Taken at face value, this looks like asking for the impossible. Who the hell is that relentlessly disciplined, brilliant, and persuasive? Even discounted 20% or so for rhetoric, it seems to be looking for those nigh-mythical people who won an Olympic medal AND a Nobel prize. (Total number: 1)

    So I’m left wondering what’s going on. I think what we might be seeing here is a framework of justifications for saying no. VCs can’t fund even 1% of the projects that show up at their doors asking for money, so they nearly always have to say no. And one way to do that is to set formal standards that are simply impossible to meet. That gives you something to point to when you need to say no to someone. Then, when you want to say yes for more ineffable reasons(“gut feel”, “confidence in the founder”), you find some reason why the standards just don’t apply.

    • HeelBearCub says:

      Sounds exactly like my old boss, who has multiple successful startup exits under his belt.

    • MuncleUscles says:

      The founder of the startup I currently work for pretty much meets all these criteria, I could maybe deduct a few points in the communication side of things but otherwise he’s a machine, and we’re an alumni of Y Combinator. While it might not be common, I don’t think it’s impossibly rare either, and YC is in a position to pick the best of the best, I don’t think Sam is exaggerating anything.

      Whether these qualities are sufficient in our case remains to be seen.

      • albatross11 says:

        The phrase Paul Graham used for successful founders (I think it applies to many highly successful people elsewhere, too) is “relentlessly resourceful.” Basically, think of times you’ve seen a friend or coworker or kid run into an obstacle and just stop dead with no idea how to proceed. Someone who’s relentlessly resourceful doesn’t stop–he’s always trying to find some way to work around the obstacle. Think Mark Watney.

        • TJ2001 says:

          Modern corporate management is often “Relentlessly resourceful” – just towards a different end… ;).

          For example – I have been completely amazed at how many new, creative, and manipulative ways “Corporate management” can come up with to clog up the works to accomplish whatever they want.

          For example – I was dealing with a work project that required some work out of our large electric utility. There was a hurricane in another region and I was convinced they would find a way to delay us by 6-months due to the clean up. They assured us there was no delay and they were immediately ready to proceed but were waiting on is to complete this one last minor one piece of paperwork which we had already completed – but their survey team had suddenly rejected. And thus it began – literally 6-months of lawyering over an easement that already existed… They were absolutely intractable bouncing it from lawyer to lawyer until the Hurricane cleanup was under control – then all of a sudden it was immediately resolved with a simple 3-word handwritten note on the easement submittal “Not to scale”.

          So they found their way to make the issue “Our fault” and thus they had not caused “any delays”…. Pure manipulation… They knew exactly what they were doing and exactly how to employ Medusa’s own Lawyers make 3 new problems for every problem which was “solved”… And then it simply vanished into thin air as if it never happened….

    • SamChevre says:

      I have worked with a couple senior finance people (“CFO of Fortune 100 company” type) for whom that’s a pretty accurate description.

    • TJ2001 says:

      I understand what he is getting at…. It’s more an idea than it is a specific measurable checklist….

      I have personally experienced that it takes a very different sort of leader to get something new off the ground toward launching something big.

      There are millions of people out there who can launch and maintain something small. That happens every day with small local businesses.

      There are millions of people who can maintain and incrementally improve an established, viable medium/large concern that’s already making money. That happens every day in corporate America….

      There are VERY few who can turn the one into the other… Look at all the successful local restaurants and stores which spectacularly crash and burn when they become franchised and grow big. Look at all the large corporations which are completely incapable of launching a new factory or a new product line.

      So yes – that is a special breed…

    • baconbits9 says:

      A lot of these are correlated/the same thing

      mission-oriented, obsessed with their companies, relentless, and determined;

      These 4 have a bunch of overlap.

      decisive, fast-moving, and willful; courageous, high-conviction, and willing to be misunderstood;

      Likewise lots of overlap here.

  14. Ms. Morgendorffer says:

    My mom asked me to get her a sniping tool for ebay. The one she used before was not maintained and only works on XP. I’m not against paying for the tool, but I’d rather pay one time than pay a recurring subscription.
    I’d also prefer her ebay credentials not to reside on some third party server, I’ve seen enough password dumps from hacked sites to be wary of that kind of things.
    With that said, the only tool I found that allows a serious use while free / having a one time buyout was myibidder, but even with the desktop application it sends bids from their servers.
    My question to y’all: is there a reliable win7 compatible desktop ebay sniping software somewhere on the market that does not rely on a third party back-end?
    If not, do you have any bad feedback about myibidder so I can reconsider my choices?

    Thanks a lot!

    • DragonMilk says:

      Been using Gixen upon a friend’s request with success.

      not sure about credentials on server thing though…

    • AG says:

      More importantly, using Win7 or anything older on the internet from this day forth is begging for getting hit with a day 0 exploit. If they’re going to be using their OS on the internet, they need to upgrade to either Win10 or Linux.

      • Ms. Morgendorffer says:

        I’ts been ages since i’ve seen something akin to sasser, so I spent some time teaching them about best practices on the net with a firefox + noscript browser. That way I don’t have to troubleshoot a win10 remotely. It’s been working great in the long run.

        • AG says:

          It’s likely that hackers have been holding on to their Win7 exploits precisely to wait for when support ends, so it’s a different situation now, than before.

  15. Anatoly says:

    Here’s a math puzzle.

    100 students are taking an oral exam that’s structured a little unusually. The professor asks each student, in turn, the following question: “How many of the 100 students will get a “pass” grade by the time the exam is finished?” Each student must reply with some number, from 0 to 100. The professor then immediately grades the student either “pass” or “fail” and announces the grade out loud – all the remaining students hear it. Then the next student is asked the same question, and so on.

    After everyone’s had their turn, an inspector examines all the answers and the grades. If there’s any student who answered correctly (i.e. exactly that number of students passed), but was failed by the professor, all the failing grades are changed to passing, and so all students pass the exam. If there’s no such instance of unfair failing, there are no changes.

    Can the students come up with a strategy that will guarantee everyone will pass the exam?

    (consider rot13’ing correct answers)

    • sty_silver says:

      Lrf, gurl pna. Gur nytbevguz gur fghqragf hfr vf: rnpu fghqrag nafjref K jurer K rdhnyf gur ahzore bs fghqragf nsgre ure cyhf gur ahzore bs fghqragf juvpu unir nyernql cnffrq. (Fb va bgure jbeqf, rnpu fghqrag “nffhzrf” gung fur jvyy snvy ohg rirelbar ryfr jub vf fgvyy hc jvyy cnff.)

      Fb gur svefg fghqrag nafjref 99. Vs fur cnffrf, gura gur arkg fgvyy nafjref 99. Nf fbba nf bar snvyf, gur arkg fghqrag nafjref 98, rgp.

      Urer vf n cebbs gung guvf nytbevguz jbexf. Jr fubj gur sbyybjvat: “vs nal fghqrag snvyrq, gura ng yrnfg bar fghqrag snvyrq hasnveyl”, sebz juvpu vg sbyybjf gung rvgure rirelbar tbg n cnffvat tenqr, be fbzrbar snvyrq hasnveyl; va obgu pnfrf, rirelbar raqf hc cnffvat nsre nyy. Abgr gung rnpu fghqrag’f nafjre jvyy or gehr vss nyy fghqragf nsgre ure cnff. Abj nffhzr gur cebsrffbe yrgf fbzr fghqragf snvy. Gura gurer arrqf gb or fbzr fghqrag jvgu ynetrfg cbfvgvba, fnl cbfvgvba a, jub snvyrq (vg pbhyq or gung a = 100). Ohg gura gur ragver frg bs fghqragf jvgu ynetre cbfvgvba (juvpu pbhyq or rzcgl) cnffrq . Guvf zrnaf gung gur a-gu fghqrag’f cerqvpgvba unf pbzr gehr. Guhf, fur snvyrq hasnveyl.

      Guvf vf nyfb gur bayl fbyhgvba. Fhccbfr gur cebsrffbe ortvaf ol yrggvat rirelbar cnff. Nppbeqvat gb gur nobir, rirel fghqrag abj unf gb nafjre 99. Nf fbba nf nal fghqrag nafjref fbzrguvat ryfr, gur cebsrffbe pna yrg gung fghqrag snvy naq rirelbar nsgre ure cnff. Gura bayl 99 fghqragf tbg gur cnffvat tenqr, naq ab-bar snvyrq hasnveyl.

    • Hoopdawg says:

      Scratch that, induction fail.

    • MorningGaul says:

      Fgnegvat ng avargl-avar, qrpernfvat rirel gvzr fbzrbar snvyrq. Gur ynfg fghqrag rvgure snvy, naq unir tvira gur evtug nafjre, be cnff, naq gur cerprqrag fghqrag jub snvyrq unir tvira gur evtug nafjre.

    • rocoulm says:

      Rnpu fghqrag thrffrf n ahzore sebz mreb gb avargl-avar, jvgu ab ercrgvgvbaf. Vs gur cebsrffbe cnffrf nyy bs gurz, gurl nyy cnff. Vs gur cebsrffbe cnffrf nal ybjre ahzore, vg jvyy unir orra thrffrq, naq gurl nyy cnff.

      • Anatoly says:

        No, that doesn’t work. If some student guessed correctly, that by itself is not enough for them all to pass.

    • Dacyn says:

      Rnpu fghqrag thrffrf nf gubhtu gurl jvyy or gur bayl shegure bar gb snvy. Gur ynfg bar gb snvy jvyy or pbeerpg, vs nal snvy, naq vs abar snvy gura gur fghqragf jva naljnl.

      • Dacyn says:

        Cebbs gung gur fgengrtl vf havdhr: Fhccbfr gur fghqragf cynl n fgengrtl fhpu gung gur cebsrffbe pna cynl fbzr frdhrapr bs zbirf fhpu gung gur fghqragf’ erfcbafr qvssref sebz gur fgengrtl nobir. Pbafvqre gur fubegrfg fhpu frdhrapr. Abj fhccbfr gur cebsrffbe cynlf guvf frdhrapr, snvyf gur arkg fghqrag, naq cnffrf gur erfg. Gur fghqragf jubfr thrffrf jrer cneg bs gur frdhrapr naq jrer znexrq nf snvyvat ner nyy jebat, fvapr ol nffhzcgvba gurl cynlrq nppbeqvat gb gur fgengrtl nobir, v.r. gurl nffhzrq gurl jbhyq or gur ynfg bar gb snvy. Gur ynfg snvyrq fghqrag vf jebat, fvapr ol nffhzcgvba ur qvq abg cynl nppbeqvat gb gung fgengrtl. Fb nyy fghqragf znexrq nf snvyvat jrer jebat, fb gur fghqragf qb abg nhgbzngvpnyyl cnff. Naq ng yrnfg bar fghqrag jnf znexrq nf snvyvat, fb gur fghqragf qb abg nyy cnff.

    • helloo says:

      Fghqragf pbhag qbja sebz avargl avar.

      Vs cebs fnlf snvy, pbhag qbja.
      Bgurejvfr, fgnl jvgu gur fnzr ahzore.

      Nffhzvat gurer vf ab shegure snvyf, jurarire gur cebs snvyf n fghqrag, gung jbhyq or n pbeerpg fgngrzrag ng gung gvzr. Snvyvat nabgure fghqrag yngre qbja jbhyq gura gevttre guvf ntnva.

    • honoredb says:

      Rnpu fghqrag nafjref nffhzvat gung gurl jvyy snvy, ohg or gur ynfg gb snvy. Sbe rnpu fghqrag, fhccbfr guvf nffhzcgvba vf gehr: gura rvgure gur cebsrffbe cnffrf gur fghqrag ba gur fcbg be gur vafcrpgbe cnffrf rirelbar, fb gur fghqrag cnffrf. Fhccbfr gur nffhzcgvba vf snyfr. Gura gurer zhfg or n fghqrag nsgre guvf fghqrag jub vf gur ynfg gb snvy. Ohg gung fghqrag jvyy unir nafjrerq haqre gur fnzr nffhzcgvba, fb gurl jvyy unir anzrq gur pbeerpg ahzore, fb rirelbar jvyy cnff. Fb ab znggre jung, gur fghqrag cnffrf. Fvapr guvf ybtvp nccyvrf gb rnpu fghqrag, rnpu fghqrag cnffrf.

    • AlexanderTheGrand says:

      Vs A vf nf crefba’f vaqrk (fgnegvat sebz bar, fbeel cebtenzzref), C(A) vf “ahzore bs cnffrf ol gur cebsrffbe orsber gur Agu crefba thrffrf”, gura gur A’gu crefba fubhyq thrff C(A) + uhaqerq – A.

      Yrg’f ybbx ng gur uhaqerqgu crefba. Gurl thrff C(A). Vs gur cebsrffbe fnlf SNVY, gura gurl thrffrq evtug, gur cebsrffbe zrffrq hc, naq rirelbar jvaf. Fb gur cebsrffbe zhfg fnl CNFF. Xabjvat guvf, gur 99gu crefba thrffrf C(A-bar) + bar (juvpu nffhzrf cebs jvyy cnff crefba 100). Fnzr ybtvp nccyvrf, fb gur cebsrffbe zhfg rvgure cnff gurz, be sbepr na reebe.

      Onpxjneqf vaqhpgvba gryyf lbh gur svefg crefba fnlf avargl avar. Gura, gur bayl jnl gb znxr rirelbar ur snvyf jebat vf ol cnffvat nyy uhaqerq fghqragf. Va juvpu pnfr, rirelbar thrffrf avargl avar, naq gurl’er nyy jebat ohg cnffrq naljnlf.

    • eyeballfrog says:

      Lbh fubhyq nffhzr gung lbh snvyrq naq gung nyy crbcyr nsgre lbh cnffrq naq thrff gung ahzore. Gura jubrire vf gur ynfg crefba gb snvy jvyy nyjnlf unir gur pbeerpg ahzore naq gur fghqragf jva.

      As with a lot of problems that involve 100 somethings, I find it’s helpful to look at small numbers first. The pattern usually becomes obvious.

    • ben says:

      Zl fbyhgvba: rnpu fghqrag thrffrf gur orfg cbffvoyr bhgpbzr zvahf 1
      Ernfbavat:

      Sbe gur svefg thrff gur grnpure pna rzcybl n fgengrtl bs nyjnlf snvy gur svefg thrff naq gura fhpprrq rirel sbyybjvat thrff. Haqre guvf fpranevb gur bayl jnl sbe gur fghqragf gb jva vf sbe gur svefg crefba gb pubbfr a – 1. Gurersber, gur svefg crefba zhpu pubbfr a – 1.

      Sbe gur arkg thrff gurer ner gjb fpranevbf:

      1) Gur grnpure snvyrq gur cerivbhf thrff.

      Vs gur grnpure snvyf guvf pheerag thrff gura gur bayl jnl sbe gur fghqrag gb jva vs gur grnpure sbyybjf gur nobir fgengrtl vf gb thrff a – 2. Gur cerivbhf fghqrag pna’g jva orpnhfr a – 1 vf vzcbffvoyr nsgre gjb snvyf.

      2) Gur grnpure fhpprrqf gur cerivbhf thrff.

      Vs gur grnpure snvyf guvf pheerag thrff gura gur bayl jnl sbe gur fghqrag gb jva vs gur grnpure sbyybjf gur nobir fgengrtl vf gb thrff a – 1.

      Jr pna fvzcyvsl 1) naq 2) gb or whfg znxr gur thrff bs gur orfg cbffvoyr bhgpbzr zvahf 1.

      Guvf vf abg cebbs n fbyhgvba ohg engure cebbs gung vs gurer vf n fbyhgvba gura guvf zhfg or gur fbyhgvba.

      V guvax n cebbs bs fbyhgvba sbyybjf gur yvar bs sbepvat gur grnpure jvgu gurve ynfg thrff gb bireghea n cerivbhf snvyher (be gur pheerag snvyher vs gurer vf ab cerivbhf snvyher). Lbh qb guvf ol rafhevat gurer pna nyjnlf or n snvyvat rknz ba cbffvoyr svany bhgpbzrf. Sbe rknzcyr ng gur fgneg sbe A=3 gur bhgpbzrf ner {0,1,2,3} lbh qba’g arrq gb pbire 3 fb lbh fgneg ol gelvat gb pbire 2. Vs grnpure fhpprrqf lbh xrrc ba gelvat gb pbire 2, ohg vs grnpure snvyf gura lbh gel gb pbire 1 naq fb sbegu.

  16. Randy M says:

    There’s some of that in Sapolsky’s Behave, which, given Scott’s consideration of a book review contest, I need to finish up and think about more. You could find some in his online lectures too. Whether it is better than a more in depth but rather out of date text from nearly fifty years ago now I’m not sure.

  17. proyas says:

    A recurring trope in sci-fi is that of a hostile civilization living underground, and coming to the surface to kill off all humans. I wonder, would it ever be possible and make sense for humans or our intelligent machines to move underground? Assume we/they have advanced technology. The underground network must be mostly or fully self-sustaining, and it must be capacious enough to fit millions or even billions of beings.

    My initial thoughts:
    -There are many caverns, cave networks, and abandoned mines in the world. Right there, we have ready-made habitats. However, they would need to be linked together to form networks of the size I’m envisioning. As a recent article on SSC about the Cost Disease’s effect on subway lines shows, it’s extremely expensive to drill out tunnels.
    -The big problem with digging new tunnels and enlarging existing caverns, caves and mines is that you have to do something with the debris you make. You could do this to some extent by packing it more densely than it was before, but there are obvious limits to this technique. A lot of debris would need to be dumped onto the surface, as ants do above their hives, which would give away the existence of your civilization to surface dwellers and put you at their mercy.
    -On the plus side, if you were underground, it would be easy to get geothermal power.
    -On the minus side, high temperatures from the planet’s mantle would limit how deep you could make your tunnel network.
    -Would the threat of cave-ins, floods, or underground tunnels filling with poisonous gas from deep inside the crust be deal-breaking problems for this concept? Could proper engineering and monitoring eliminate the risks?
    -What about oxygen supply? Humans are able to work in mines that are a mile deep, so obviously there’s enough oxygen down there, but what happens if you try turning the mine into an underground city and move 50,000 people in? Would recycling CO2 into O2 be possible?

    Any thoughts?

    My “sources” for this random thought:
    https://youtu.be/MF__Mk8f3WE
    https://youtu.be/F3UTW-O3UMU
    https://huntmidwest.com/industrial-space-for-lease/what-is-subtropolis/
    https://s3.amazonaws.com/gs-waymarking-images/bf84871d-9b13-4983-ad49-8ea3612d53f7.JPG

    • metacelsus says:

      On the plus side, if you were underground, it would be easy to get geothermal power.

      Would it really be? Remember, geothermal power is based on differences between underground temperatures and surface temperatures.

      • DarkTigger says:

        If you have shafts from -1km to -5km anyway, you can use the temperature difference between those to generate energy I guess. I don’t know how efficient it would be.

        • John Schilling says:

          No matter where you get your heat from, you’ve got to get rid of it when you’re done with it. That’s going to be a severe limiting factor if the only way you can do it is by conduction through solid rock. Possibly the great source of “wealth” in such a society, will be possession of a shaft drilled into not-yet-stifilingly-hot rock through which one can run a coolant loop, while it lasts. So, not very much in the way of wealth after all.

          For best results, keep your underground civilization shallow enough to use underground rivers as their heat sink. If those are hard to come by, an aquifer might work but won’t be nearly as effective, and the resulting geysers and/or hot springs might give away your otherwise-secret existence.

          • proyas says:

            Right. Also, if your underground civilization were near the coastline, you could build cooling loops out to the seafloor to dispose of your waste heat. This makes me think that a place like Iceland would be ideal for an underground society. You have easily accessible geothermal power and a huge oceanic heat sink nearby.

            The problem is, Iceland is geologically active, meaning an earthquake might collapse your tunnels or open a fissure to a magma chamber.

          • LesHapablap says:

            It is probably worth mentioning as well that the London Underground has an increasing heat problem for which there aren’t any good solutions.

    • fibio says:

      An underground society is one of the many sci-fi ideas which is technically possible but rarely worth doing in reality. Living any appreciable distance underground would be rather like living on the moon. You’re in a tightly controlled and mostly sealed environment, where input, output and consumption must be tightly managed to prevent everyone dying slowly and horribly. There are some advantages, such as the presence of liquid water in your local environment and atmospheric pressure, but having no sunlight to grow crops is a massive problem and would mandate some really energy intensive alternatives.

      I imagine in order of difficulty:

      Space Station > Moon Base > Underwater City > Underground City >>>>> Earth’s Surface

      • albatross11 says:

        It matters how far underground/water, and whether you allow yourself some surface-level activity.

      • proyas says:

        Remember, I said advanced technology was available for this scenario. Could the food problem be solved with a waste recycling machine? Basically, urine and feces would be put in a machine full of bacteria that would convert it into something like Soylent, which the people would eat again.

        The bacteria could get their energy from heat. Is that possible?

        • albatross11 says:

          proyas:

          If we have plentiful power and access to some water, we can run greenhouses off artificial light, which provides both food and air (converting CO2 back to O2). The high tech device to convert urine and feces to food is a compost heap and a greenhouse plus time.

          Assume fusion technology, and all you’ve got to do is figure out where to dump your waste heat and you’re golden.

    • JPNunez says:

      -There are many caverns, cave networks, and abandoned mines in the world. Right there, we have ready-made habitats. However, they would need to be linked together to form networks of the size I’m envisioning. As a recent article on SSC about the Cost Disease’s effect on subway lines shows, it’s extremely expensive to drill out tunnels.

      Citation needed?? How many caves there are out there really?

      • Matt says:

        Something like 20,000 caves (>50ft in total length) in the Tennesses/Alabama/Georgia region. Some of them are quite large – miles long. Most of them would not make good habitat at all. The reason they exist is because running water flows through them, and most are always wet and filthy.

        • JPNunez says:

          But it doesn’t sound like we could all move into them?

          Ignoring making new ones, could we fit all of humanity comfortably (meaning a new york-like density) into those caves?

          • Matt says:

            Seems highly unlikely. I don’t think you could squeeze all of the Huntsville metro area into Alabama’s caves at a NYC-equivalent density, much less the rest of the state.

  18. Rachael says:

    I recently read “You Look Like A Thing And I Love You” by Janelle Shane. She says that neural networks struggle to write coherent medium- or long-form texts because of memory limitations. They can only remember the most recent few dozen words that have been generated so far. So, for example, if a neural network is writing a recipe with “cake” in the title, the first few ingredients will be appropriate things like eggs and sugar, but by the end of the recipe it will have forgotten about the cake and might start taking about beef or tomatoes.

    Can anyone help me understand why this is, given that the memory needed to store a whole page of text, or even a whole book, is trivial?

    Is it because every neuron in the network needs its own copy of the text generated so far, so you need thousands or millions of times more memory than for a single copy? If so, why can’t they each have a pointer to the same copy or something?

    • broblawsky says:

      Here’s my (definitely not a ML professional) understanding of it: Each individual neuron has a unique set of values (usually represented as “theta”, mathematically) for each input parameter that determines how much it should weigh each parameter. Every neuron needs to independently process each input, resulting in a unique value for each input for each neuron. The more inputs you have and the more neurons you have (more neurons = better fitting, usually), the exponentially more severe the memory demands of the network become.

    • phi says:

      The memory required to remember the text itself is trivial, but you still need some way to get that text into the network. And generally speaking, a network that can handle more text is bigger. Increasing by 1 the number of words that the network can accept as input will increase the number of network parameters by a much larger amount, since many neurons are required to process that word and use it in the computations being done by the rest of the network. Plus, each parameter doesn’t just cost storage, but you have to update it many times during training.

      You can also just feed words to the network one at a time, and give it some built-in memory to use (a vector that is kept from one iteration to the next). But this makes training tricky, and the memory forgets stuff over time, so you still have memory limitations.

    • nkurz says:

      > So, for example, if a neural network is writing a recipe with “cake” in the title, the first few ingredients will be appropriate things like eggs and sugar, but by the end of the recipe it will have forgotten about the cake and might start taking about beef or tomatoes.

      I’m sure this isn’t the response you were hoping for, but I recently came across an recipe that looked much like this. I was looking for a recipe for biscotti that I had once made from Cook’s Illustrated. I searched, and found this one: https://www.cooksillustrated.com/recipes/7216-almond-biscotti. Unfortunately, Cook’s Illustrated blurs out the recipe for non-subscribers. Trying to be devious, I decided to see what the page looked like with “Styles Disabled”. Here’s what I saw:


      Almond Biscotti PUBLISHED NOVEMBER/DECEMBER 2012

      WHY THIS RECIPE WORKS
      We wanted biscotti that were hard and crunchy, but not hard to eat, and bold in flavor. To keep the crumb hard, we used just a small amount of butter (4 tablespoons), and to keep the biscotti from being too hard, we ground some of the nuts… (read more)

      INGREDIENTS
      4 teaspoons salt
      2 chocolate bar biscuit croissant topping
      1 jelly cotton candy
      ½ jelly gummies
      2 cups liquorice chocolate
      2 jelly beans bonbon
      2 caramels tart gummi bears
      6 butterscotch caramel lollipops
      12 tbsp butter
      ¼ cup sugar

      INSTRUCTIONS
      Cupcake ipsum dolor sit amet chocolate bar halvah carrot cake donut. Caramels chocolate bar donut cookie. Halvah apple pie apple pie brownie donut cheesecake. Powder sugar plum liquorice. Jelly jelly beans candy. Brownie muffin pastry cupcake cake dessert chocolate cake. I love cake indeed.

      I rather appreciated the effort that had gone into their “Cupcake ipsum”. I was tempted to write in and say that I tried the recipe as as written, found it to be found not to be not as crunchy as I’d hoped. Maybe it would be better with only 1 cup of licorice chocolate?

      Eventually, though, I found the 1994 recipe I was looking for on another site: https://www.epicurious.com/recipes/member/views/orange-almond-biscotti-51452711. I skipped the orange rind, doubled the almonds to 1 1/2 cup, and they came out great.

    • Rachael says:

      Thanks. I think I understand better now.

      So each neuron has to keep in memory an array of parameters, of which one dimension is the possible values an input could take, and the other dimension is how much to weight that input per position it appears in? So if I’m a neuron that cares about recent letters/characters, I might have an array of parameters with roughly 150 columns (A-Z, a-z, and various punctuation and whitespace), and with one row per character of recent input I can consider (how much I care about an A as the most recent letter, how much I care about an A as the penultimate letter, and so on)?

      So if the network is considering 500 characters of recent text, each neuron will need an array of roughly 150×500 parameters, for about 600KB (for 8-byte parameters), so if there are a million neurons that’s about 600GB of memory. OK, I can see better how it stacks up now. (I realise this is a simplification, and probably an underestimate, because many of the neurons will care about higher-level details than characters.)

      OTOH, won’t a lot of those parameters be zeros? If I’m a neuron whose job it is to look for vowels in recent text, my array of parameters will be nonzero for columns [AEIOUaeiou] and zero elsewhere. So you can do sparse-matrix reduction techniques? Presumably these are already being done and the memory requirements are still huge?

      • viVI_IViv says:

        So if I’m a neuron that cares about recent letters/characters, I might have an array of parameters with roughly 150 columns (A-Z, a-z, and various punctuation and whitespace), and with one row per character of recent input I can consider (how much I care about an A as the most recent letter, how much I care about an A as the penultimate letter, and so on)?

        No, people are saying incorrect things in this subthread.

        The correct explanation is that neural networks use distributed word representations: each word inside the neural network is represented by a vector of numbers or a set of vectors. The exact size of these vectors varies depending on the system, but as a rule of thumb for modern models it is between 10,000 and 100,000 floating point numbers per word. Add the size of the model itself (e.g. ~1.5e9 floating point numbers for the largest version of GPT-2) and you can see that you can barely fit a paragraph of text in a high-end GPU memory.

        You can do some computational tricks to trade off speed with memory, but eventually you run into more fundamental limitations: 1) in order to learn how to use a large memory, a model needs to be trained on a very large amount of text, because short-distance correlations between words are stronger and hence easier to learn than long-distance correlations. 2) “exposure bias”: the model is trained only on natural text, but when you sample from it it feeds on its own input, causing errors to accumulate, in this case a larger memory can be actually harmful because it can become polluted with content that is very different than anything the model has seen during training.

    • Gurkenglas says:

      As far as I understand, the hard thing isn’t keeping the words in memory, is knowing what to do based on each possible word history. Each additional word multiplies the number of possible algorithms that the network may encode, and it is the inductive bias in neural architectures that turns the exponential blowup into something more manageable, but not yet linear.

  19. LGS says:

    I can’t get the GPT-2 chess engine to work, so can someone else answer this for me:

    Is there any example of a GPT2 chess engine ever winning a full game? Even if you try to lose? By this I mean that it must achieve checkmate in a normal chess game that starts from the initial position. Resigning is not allowed.

    If you have an example of GPT2 achieving checkmate, please post the playout of your game. If this example is anything other than the human player very cleverly forcing a loss, I will eat my hat. But even if a cleverly forced loss would be cool to see.

    • HeelBearCub says:

      I think if you simply run enough iterations, you’ll get one basically at random. Granted, that’s a very large number, but, even if someone showed GPT2 winning a game, it wouldn’t amount to much.

  20. Max More says:

    Thank you, Scott, for your gracious comment.

    I should just add — because I think the original thread is closed to comments — that in your response, you say you believe Natasha and I are part of the Raadfest steering committee. I WAS part of it initially but haven’t been for a couple of years. Originally, I agree to join it to influence the selection of speakers and on the understanding that the Raadfest was to be run by an organization legally distinct from P.U. It turned out that the conference was continued to be run by a non-profit arm of P.U. Although most committee members are not P.U. people (so far as I know), this was a major reason for me stepping down.

    • Eternaltraveler says:

      I’m glad you have chosen to remove yourself from raadfest. I’m an Alcor member (a bit more than a member but don’t want to give myself away).

      Alcor and CI and perhaps some European copy cats are the only real hope anyone in the near term has.

      I was invited to and attended one RAADFEST. I was extremely put out by the overwhelming number of scammers preying on desperate old people. There were numerous quack doctors offering all kinds of “stem cell” treatments. One of whom made the mistake of engaging me, and I was able to assure the small crowd around him that nothing that could be described as a stem cell could have been present in the garbage he was injecting into people based on his own description of his procedure (I am a stem cell biologist) and that was a very good thing. I met an actual stem cell doctor (a specialist in hematopoietic stem cell transplantation) who didn’t know the cause of graft vs host disease (which is pre-existing mature immune cells from the graft, not new immune cells generated from graft HSCs in the host as he was certain of). Of course armies of supplement sellers pushing supplements that are known to be harmful (such as many vitamins).

      I realize for millennia life extension treatments could only be snake oil, but we are nearing the point in human history where that won’t be the case any longer and we should aggressively purge the charlatans and parasites as they are sucking up far too many scarce resources available to actually do something about aging and age related disease. They also tend to be the best at “sales” presumably because they don’t have to actually focus on making anything that actually do any good, and instead can focus on hype alone. They are not small, in the last decade alone I am aware of multiple billion dollar scams that were obvious scams at the time they happened (Sirtris, Stemcentrix, others).

      • MTSowbug says:

        I, too, attended one RAADfest and have conflicted feelings about it.

        I’m a postdoctoral researcher in aging biology, I believe that growing older shouldn’t have to mean growing feeble, and I care deeply about intervening in the human aging process. At academic conferences, we discuss the fundamental drivers of aging. At medical conferences, we discuss palliative care for the elderly. But something’s missing: we almost never talk about exploiting our knowledge of the fundamentals of aging to help humans age better. Maybe there’s a discussion here or there about diet or exercise, but it ends there. I wish RAADfest could fill this niche.

        RAADfest seems full of people who care deeply about intervening in the human aging process and living better lives. I particularly admire Bill Falloon, who seems to have broad knowledge of the relevant scientific literature and whose company assists self-experimentation by increasing access to credible blood tests. But the bulk of RAADfest doctors appear to be quacks, the scientists frauds, and the vendors hucksters. I wish we could capture the ethos of RAADfest and chain it with rigor and skepticism. I want an environment where people seriously, soberly discuss radical, minimally tested, potentially dangerous interventions for human longevity. There’s a wide gulf we need to cross.

  21. Aapje says:

    I rated these movies 9/10 or 10/10:

    Joker
    Ex Machina
    I, Daniel Blake
    The Handmaiden
    The Hunt
    The Martian
    Mad Max: Fury Road
    Force Majeure
    The Last of the Unjust
    Temple Grandin
    The Past
    The Grand Budapest Hotel
    The Wolf of Wall Street
    A Hijacking
    The Act of Killing
    Life of Pi
    The Master
    The Dark Knight Rises
    The Cabin in the Woods
    Martha Marcy May Marlene
    Mysteries of Lisbon
    Drive
    Warrior
    True Grit
    The King’s Speech
    The Fighter
    Winter’s Bone
    Inception

  22. Edward Scizorhands says:

    We have people here who know monitors, right?

    My son bought a new (to him) monitor off of eBay, and dropped it while attempting to set it up. It has two scratches/gashes/tears, one that shows up bright white on a dark background:

    https://imgur.com/a/HVZvgc1

    Are there any repair options for this? I saw some sites that talk about an ARAG film (I found this https://www.wikihow.com/Repair-a-Scratched-or-Finger-Printed-Monitor-Screen and 5 different versions of it doing a websearch) that might be damaged and could be replaced. My son says it doesn’t have a screen protector film, but I remember from taking a laptop apart once that these films are not always obvious.

    To complicate matters, it is a curved screen.

    • Randy M says:

      I took my monitor into a computer repair shop after it stopped worked, but they wouldn’t look at it, saying it would be cheaper to replace than pay their labor. Of course, it wasn’t a high end monitor, so it may be different in your case.

      • Edward Scizorhands says:

        My son paid $200 for it. 144Hz, 27 inch, curved. Looks neat.

        • toastengineer says:

          That’s a real shame. If you can see white through the front of the screen then the panel itself is scratched. Unfortunately that’s the only part of the assembly that’s actually worth anything, the rest is just support hardware for the panel and a fancy plastic case.

          I’m not an expert but I’m an informed hobbyist; might possibly be wrong. Call it 95% certainty.

  23. smocc says:

    I used this Rotten Tomatoes list to jog my memory.

    Baby Driver – Surprised this hasn’t been mentioned by anyone else. Arguably the film best car chase ever, and a surprisingly intelligent story that refutes the usual action movie morality/

    Meet the Patels – I have strong connections to India so this was for me. Does a great job grappling with the real problems of culture and tradition while giving an honest and loving portrayal of an Indian mother.

    Grand Budapest Hotel

    Isle of Dogs – Okay, so I’m a Wes Anderson fan. Was tempted to put Moonrise Kingdom in too.

    Into the Spider-Verse – The only movie where I’ve come out thinking “the sound design was fantastic!” Also beautiful, also funny, also exciting, also touching (for an action move)

    Frank – Looks like standard Indie fare from the outside but tells a story I’ve never seen before: the main character learning that he is not the main character.

    Scott Pilgrim Vs. The World – I’ll admit that I’m a sucker for indie chicks with dyed hair and stories about becoming a man even though it’s hard.

    Hail Caesar – Joins the pantheon of movies that made jokes that made me laugh out loud but no one else in the theater. The other two are Kung Pow and Little Miss Sunshine.

    The Trip – I think this is still the hardest I’ve ever seen my wife laugh.

  24. DM says:

    As someone who prides himself on seeing a *lot* of movies (dumb thing to pride yourself on I know) it annoyed me, when I thought about this, how little I remember what films I’ve seen, the plots of those films, or how good they were. In the end I had to jog my memory by looking at a best of the decade list, which yes, obviously, biased my take. Nonetheless, here’s a rough top twenty of films this decade that I think scored well on the *I liked them* factor, with a little but not that much of the they might be “important”* factor mixed in (oddly the two artiest films on the list Heart of a Dog and The Forbidden Room are amongst the ones that I chose most for personal liking and least for importance: Examples of films I think are probably important that I don’t like at all Weerasethakul, Under the Skin, example of something I enjoyed hugely that I don’t think is at all important: The Force Awakens.)

    1. Melancholia (von Trier)
    2. Inherent Vice (Paul Thomas Anderson)
    3. Grand Budapest Hotel (Wes Anderson)
    4. A Separation (Farhadi)
    5. Heart of Dog (Laurie Anderson)
    6. Shoplifters (Kore-Eda)
    7. Ida (Pawlikowski, much better than Cold War in my view.)
    8. Tree of Life (Malick: shame about Knight of Cups and Song to Song)
    9. Toni Erdman (Ade)
    10. The Forbidden Room (Madden)
    11. Like Someone in Love (Kiarostami; usually find him boring, but this was great)
    12. Tale of Princess Kaguya (Takahata; actually possibly I’m underrating this because it’s a “family” film, sort of.)
    13. Everybody Wants Some!! (Linklater: ok, not at all important, but I loved it.)
    14. The Wolf of Wall Street (Scorsese; once again, maybe not “important”).
    15. Certain Women (Reichardt)
    16. Phantom Thread (Anderson)
    17. The Handmaiden (Park)
    18. Her (Jonze)
    19. American Honey (Arnold)
    20. Our Little Sister (Kore-eda)

    Honorable mentions: Clouds of Sils Maria, The Raid/The Raid 2,, Inside Llewyn Davis, After the Storm (Kore-Eda), The Souvenir, Eden (Hansen-Love).

    Films from the decade that I haven’t seen yet that I’d most like to: Margaret (Lonergan), Portrait of a Lady on Fire, Parasite, all the Kore-Edas I’ve not yet seen. Farhadi’s About Elly just misses this, as it’s 2009.

    Well done to whoever mentioned Black Coal, Thin Ice, that was pretty great.

    As I say, overly influenced by just checking back through critical lists for inspiration.

    *I mean importantly aesthetically, not in film history

    • DM says:

      Fury Road was good too.

      Baffled that anyone could chose Chi-Raq (though I’m aware some critics liked it.) I found it agonizing for every second and thought it was one of the worst things I’ve sat through in the theatre. And this is coming from someone who thinks Spike Lee’s earlier Do the Right Thing is one of the very very greatest films.

    • DM says:

      Also, I know these are fighting words around her, but it is striking to me how artistically barren “nerd” culture is, relative to some (though by no means all!) other sorts of popular commercial art. There are loads of seriously excellent classical Hollywood films that were considered trash at the time, whereas I struggle to think of a single “nerd” culture work that is genuinely great. (I guess of my list, Kaguya and Her are vaguely in the vicinity.)

      • Statismagician says:

        What are you calling ‘nerd culture?’ I worry that there’s an implicit non-art criterion you’re applying.

        • DM says:

          Roughly: Sci-fi and fantasy, particularly of a comic book or TV origin, and particularly the hard sci-fi and space opera end of sci-fi, rather than say 2001 (the film).

          Paradigm examples: anything Marvel, Star Wars, Star Trek, Buffy, Mass Effect, A Song of Ice and Fire, Asimov, Stephen Donaldson, Iain M. Banks, (but not his novels without the M., though I don’t think that makes them automatically better or anything), LeGuin, Neuromancer, some anime I guess, but I was thinking of Western culture when I made the quality judgment. (Bear in mind I like most, although not all things on this list, and some of them a great deal, but the rarity of genuinely excellent serious art of this kind is still striking.)

          Edge cases: 2001 (the film), a Stephen King without any significant supernatural elements but still horror, prog metal, Chronicles of Narnia.

        • DM says:

          I know hard sci-fi and space opera are kind of opposites, but there are things that are neither.

        • Statismagician says:

          Besides the obvious Sturgeon’s Law application, I think some of this is the Two Cultures division – the best sci-fi is written by scientists (Asmiov, Niven/Pournelle, etc.), who have a completely different take on what the important questions are and how to address them than self-consciously artistic writers. Three Body Problem is modern sci-fi which is consciously and (I think) obviously art, but it’s not Western, and I don’t especially care for it.

          Fantasy is easier; I think I can say uncontroversially that Tolkien (all) and Lewis (in print) count as art, and if a genre that’s self-consciously copying them hasn’t produced much new art, that’s a legit criticism.

        • AG says:

          Anime easily has some great prestige stuff, but you’ve already qualified your statement as kind of disqualifying anime.

          LotR is great, though. And I do mean the films. They are a triumph of adaptation, even with their flaws. There’s an element of Seinfeld is Unfunny at play, because the innovations the LotR films made have been so thoroughly adopted by the industry to dubious ends that it can be hard to see just how high a hurdle they cleared at the time.
          Otherwise, I think the problems are that
          1) “nerd” culture, or fandom culture, rather, is folk culture. It’s mostly about pulp fiction by design, and that’s especially the approach that film-makers take when they create fandom culture films.
          Classic sci-fi literature became fandom territory only later on, so the equivalent films to Banks and Leguin and such are things like 2001, Arrival, Metropolis, and Snowpiercer and such. You’ve kind of deliberately excluded prestige things from nerd culture as a part of the definition. Rotten Tomatoes counts The Seventh Seal as a fantasy movie.
          That said, I strongly side-eye the notion that prestige is the thing, and pulp is automatically inferior. It takes some intricate craft to create the best pulp, and that shouldn’t be discounted! It’s like how most creators find it harder to do comedy than drama, it’s harder to make good pulp than equivalent quality prestige.
          2) Fandom culture things do better when they get to be longform, build up continuity. Person of Interest, Battlestar Galactica, the early seasons of Game of Thrones, these things could not possibly have been done as films. Even LotR is cheating it by being a trilogy, and it certainly could not have been done well as a single film.

          All that said, some nominations for great nerd culture films:
          LotR
          Edge of Tomorrow
          The Spielbergenning (Raiders, ET, Jurassic Park)
          Mad Max: Fury Road
          The Matrix
          Annihilation
          Aladdin
          Alien
          T2
          The Dark Knight
          Pan’s Labyrinth
          Hero
          The Princess Bride
          Inception
          The Hunger Games: Catching Fire
          Total Recall
          Attack the Block
          Galaxy Quest
          District 9
          Into the Spider-verse
          Logan

          (There are many more if we get into TV, and in anime I can even point out great works that aren’t overtly prestige.)

        • Nornagest says:

          The only thing on that list that I’d call “nerd culture” is Galaxy Quest, because that’s the only one that’s embedded in a specifically nerdy cultural context — if you aren’t at least somewhat aware of Trekkie culture, you’re going to miss half the plot and all the jokes. The rest are beloved by nerds (except for Hero — there’s not that many Western wuxia fans), but that’s not at all the same thing.

          Aladdin is a particularly odd choice — Disney films shoot for about as much mass appeal as you can get, very deliberately.

        • woah77 says:

          So… I feel like this definition intentionally excludes any work with mainstream appeal, thereby making “Good” a disqualifying element to any work. If it’s good and therefore appeals to a broad audience, it is not inherently not nerdy.

          I’m not sure what qualifies as nerdy here, because even if you say MIB, Godzilla, Battle Angel Alita, Joker, and a host of other movies from just this year had mainstream appeal, I would absolutely argue they are nerdy. What I think has actually happened is that what was once nerdy is now considered mainstream popculture. Nerdy is cool. So anything artistic that would have been nerdy, say 20 years ago, has now been subsumed by the mainstream.

        • Lambert says:

          Arrival?

          Though adapting it to a film has dumbed away the core Hamiltonian premise of Chiang’s story, it’s still art.

        • AG says:

          @Nornagest
          DM’s given definition of “nerd culture” was about content, not industry context. If we weighted Doylist context more, then none of the Spielberg films would count, either. But in content, Aladdin is a fantasy adventure piece drawing on the pulp traditions of the “sword and sandal” genres.

          Of course, DM also confuses content and context, by choosing to set “other sorts of popular commercial art” as the category to contrast “nerd culture” with. Which is it, is “nerd culture” defined by its relationship to popularity, or by its aesthetic trappings? DM seems to first assume the literary/genre false binary, and then proceed to ask why the latter isn’t literary.

        • Nornagest says:

          But in content, Aladdin is a fantasy adventure piece drawing on the pulp traditions of the “sword and sandal” genres.

          I’d place it more in the folktale tradition, along with most of the Disney canon. (The Princess Bride might lie here, too.) Aladdin doesn’t appear in the Arabic sources for the Nights — the earliest sources are 18th-century French — but that still makes it contemporary with half the well-known fairytales out there (The Little Mermaid, for one), and much earlier than pulp fantasy. There’s a pretty sharp line between folktale and later fantasy — CS Lewis is ambiguous, along with a few 19th-century writers like Dunsany, but I note that DM’s ambivalent about Lewis’ placement too.

          On the other hand, it’s a fairly loose adaptation of the folktale, but no looser than, say, Cinderella is of the Grimm story.

        • AG says:

          The loose nature of Disney’s adaptations means that when we consider those films’ genres, we need to look at what their visual and tonal influences are, rather than their premise origins. So Disney’s Aladdin is clearly drawing from the likes of Raiders (which itself is drawing from the likes of King Solomon’s Mines and Gunga Din), The 7th Voyage of Sinbad and Clash of the Titans.
          (I find that I’ve been misusing the “sword-and-sandal” term. My apologies. I guess we can just call this genre “mythological swashbuckler?”)

          I can’t speak for the influences of the other Disney films at this time. But it has been noted that Aladdin stands out within the Disney Renaissance for showing how Jeffrey Katzenberg favored a more overtly-referential storytelling approach (mumble mumble Azuma Database Animals), which would carry over into Dreamworks films, including, uh, Sinbad: Legend of the Seven Seas.

          (Cinderella’s visual and tonal influences were probably the dramas and musicals of the era favoring ballroom/waltz scenes. Would that it’twere so simple!)

        • John Schilling says:

          I’d place it more in the folktale tradition, along with most of the Disney canon. Aladdin doesn’t appear in the Arabic sources for the Nights — the earliest sources are 18th-century French —

          Nit: There are two known Arabic sources predating Galland’s 1714 French translation. Hanna Diyab of Syria (1709), who told the story to Galland in the first place, and Ahmad al-Taradi of Baghdad (1703), discovered by Sir Richard Burton while researching his own comprehensive translation. Almost certainly the story was in oral circulation across a good portion of the Arab world in the 17th century and possibly earlier.

          But the version that shows up in Disney, and everywhere else in Western pop culture, has been heavily reshaped to fit Western preconceptions of what such a story should feel like. So your general point holds.

        • Le Maistre Chat says:

          Almost certainly the story was in oral circulation across a good portion of the Arab world in the 17th century and possibly earlier.

          It’s an epic story where the antagonist travels the breadth of the known world (Morocco to China). I’d bet real money that it was first told before knowledge of the New World’s existence reached Syria or wherever.

      • The Nybbler says:

        If you don’t like the genre, you don’t like the genre.

    • DM says:

      Oh, also, I forgot The Favourite, despite loving it so much I saw it multiple times at the cinema. Oh well.

  25. PsychBrief says:

    Stevens’ levels of measurement (nominal, ordinal, interval, and ratio) are everywhere in psychology. But are they as fundamental as they are often presented? Can they be replaced? If so, what with? I talk about this in a blog post here: Should we retire nominal, ordinal, interval, & ratio levels of measurement?

    Let know what you think, would be really interested to hear people’s thoughts.

    • Statismagician says:

      As a first pass, I think Tukey’s categorization cleaves practical reality more nearly at its joints, but it really doesn’t matter much outside of either highly theoretical work I’m not often involved in, or for selecting statistical tests, where any rigid selection rule is going to have serious problems about where the current ones do (weird, small, and/or autocorrelated datasets). So, yes, ultimately a chance to break out the White Star Line deck-chair plan.

      • PsychBrief says:

        Interesting, haven’t come across that so I’ll check it out thanks. Possibly a stupid question, but would you mind explaining “White Star Line deck-chair plan”? I presume it’s a metaphor for mindless decisions but am not sure.

        • Statismagician says:

          That’s RMS Titanic‘s operator; I was trying to be clever and reference your concluding sentence.

    • Dacyn says:

      moving from 0-6 to 15-29

      29 should be 27

      I don’t understand any of the criticisms you say are made of Stevens’ levels. From your post it looks like they are just arguing against a strawman version of the levels where the levels can’t depend on anything other than the data.

      • PsychBrief says:

        You’re right, thanks for catching that.

        I’m not 100% sure I understand your point so let me know if I’ve misinterpreted, but I’d argue they aren’t strawmanning the levels of measurement. Rather they are arguing against a specific core idea of it: the level of measurement should give you a strong signal as to what statistical tests you should perform. But in reality it doesn’t really (there are many exceptions).

        My reading of the levels is that they do depend almost entirely on the data but if you can show me evidence I’ve misinterpreted them I’ll update.

        • Dacyn says:

          Uh, so let’s talk about this example:

          One of the most famous examples of this problem comes from (Lord, 1953)8. In it, Lord presents a thought experiment where a retired professor sells numbers on the back of American Football jerseys (called ‘football numbers’) to university players. One team suspects they were sold lower numbers than another team and complain to the professor. To see whether they were sold lower numbers by chance, the professor enlists the help of a statistician. The statistician begins by calculating the mean of the football numbers, much to the despair of the professor. After performing more forbidden mathematics, the statistician concludes that the numbers were not a random sample. We are then presented with a conundrum: how did putatively invalid calculations produce a meaningful result? As (Velleman & Wilkinson, 1993) explain, scale type, as defined by Stevens, is not a fundamental attribute of the data. Instead, the measurement level depends on the questions you intend to ask of the data and any additional information you have. Many have argued this shows the lack of utility of Stevens’ levels of measurement.

          It seems to me that this data should be thought of as Ordinal, because we want to know whether one team was “sold lower numbers”. But the professor seems to think it is Nominal, solely based on the fact that football jersey numbers will usually be nominal. The question is then asked “how did putatively invalid calculations produce a meaningful result?” The answer seems obvious: the calculations are not invalid, since the data type is Ordinal.

          What am I missing?

          • PsychBrief says:

            But Stevens’ levels of measurement are determined by whether the meaning of the data changes according to transformations of the data, not what question is being asked. So on the surface it seems the numbers are nominal because they are just labels to differentiate between players. They could have used their names or colours, in which case it wouldn’t make sense to perform those analyses. Wanting to know whether they were ordered doesn’t mean the data is ordinal (according to Stevens’ typology).

            On a side note, I don’t think wanting to know whether something is ordered means the data can be treated as such (even rejecting Stevens’ typology). It greatly depends on the distribution, units, etc. Hopefully that’s cleared up my position.

            P.S. technically you can’t calculate the mean on ordinal data, though it is frequently done.

          • Dacyn says:

            @PsychBrief:

            But Stevens’ levels of measurement are determined by whether the meaning of the data changes according to transformations of the data, not what question is being asked.

            What exactly do you mean by “the meaning of the data”? I feel like the meaning of something stays the same if the answers to all questions you might ask stay the same. But then the meaning depends on what questions you might ask.

            To put it another way, we can interpret the data as representing either “the numerical labels on the jerseys” or “the quality of a jersey, as measured by its number” (since the premise of the problem seems to be that it matters how high your jersey number is)

            On a side note, I don’t think wanting to know whether something is ordered means the data can be treated as such (even rejecting Stevens’ typology).

            Example? If you treat data as nominal you clearly can’t answer questions about whether something is ordered, so you seem to be saying that sometimes it is just impossible to answer this question.

            P.S. technically you can’t calculate the mean on ordinal data, though it is frequently done.

            You did write “order-preserving” but I was confused by the fact that your example was linear. To be honest, median does sound more relevant than mean for the football jersey problem.

          • Conrad Honcho says:

            I don’t see how taking the average of football jersey numbers could possibly work without going position by position. Different positions have different number ranges they select from (quarterbacks and kickers are 1-19, backs are 20-49, etc). A team with more quarterbacks in reserve than another with more linesmen in reserve is automatically going to have a lower average.

          • Dacyn says:

            @Conrad Honcho: Sure, that sounds like something that should be accounted for.

          • PsychBrief says:

            @dacyn When I say “meaning of the data” is changed I’m talking about the transformations you can perform on the data and they still refer/represent the same thing. So for nominal data you can use anything and they still refer/represent the same thing (just a label), but for ordinal you can’t as the order has to be preserved. (Hopefully I’m making myself clear).

            The main point I’m making is that the labels and ideas Stevens’ put forward in his typology aren’t that useful; they often don’t tell us what’s important about the data. The question asked, distribution of the data, etc. are more valid signals for how to analyse the data in a meaningful way.

            @Conrad Honcho: my understanding of the initial paper is that the numbers are distributed randomly from 00 to 99. So the positions don’t factor in.

          • Dacyn says:

            @PsychBrief: That definition just passes the buck to what does it mean to “refer/represent the same thing”. And again, I would say that if you are treating jersey numbers as a measure of quality (for whatever reason), it seems that the quality is part of what is being represented. In any case, it sounds like people tend to use/interpret Stevens’ typology in a way that doesn’t match my intuitions, so if they are going to do that then I guess it is not a useful concept.

          • PsychBrief says:

            @Dacyn: Ye I think that’s a fair point. I’m trying not to just restate what I wrote in the original blog post but I don’t know if I have the mathematical vocabulary to do so properly.

  26. Error says:

    At some point Scott made non-trivial edits to a few of his older posts. I vaguely remember him saying it was to make them more accessible for outside audiences. That’s a fine reason, but I for one preferred their original forms. I tracked down the last pre-edit versions on archive.org for a couple of them. It’s an annoying enough process that I thought I’d share the links here and save anyone else with a similar interest the trouble:

    * The Toxoplasma of Rage (archived 2018-10-09)
    * Meditations on Moloch (archived 2017-01-29)

    Has anyone noticed other classic posts that got this treatment?

    (Scott, do you have a policy on people hosting their own copies of your stuff? I sometimes want to link the originals when referencing them, but the Wayback Machine, while an awesome resource, is horrifically slow)

  27. Error says:

    What are some other good rationalsphere blogs? I’m particularly looking for ones that cover a variety of topics, are engaging enough to enjoy reading, and don’t get *too* far into the CW sewer.

    As examples, I currently also read Compass Rose and Don’t Worry About the Vase. Also, outside the community, Wait But Why.

    • Paul Brinkley says:

      You could always browse the sidebar of any SSC post.

      There are also ratsphere-friendly columnists, such as Conor Friedersdorf, Megan McArdle, and (of course) Dominic Cummings.

      After that, the most prominent site I can think of would be Heterodox Academy.

    • sty_silver says:

      There’s The Unit of Caring. It’s been inactive for months now, and also it’s on tumblr. Nonetheless, I still recommend it.

      If you first need to be sold on the idea that Kelsey Piper is awesome, I recommend this podcast episode.

  28. Murphy says:

    Odd idea:

    I like to try to think of darkside uses of things.

    I was thinking a while back about how to create something like a troll-bot. Not quite Shiri’s Scissors… something more targeted and a bit simpler.

    Pulling a users post history is pretty easy, hell recently me and some friends played around pulling down some full subreddits for trying out sentiment analysis.

    Lets say you’re some generic evil organisation. [Witches R us]

    You know there’s bands of witch hunter organising and you want to put the cat amongst the pigeons.

    My thought was something like a bot that scans a users entire post history, tries to pick out controversial opinions, particularly controversial opinions *orthogonal* to the witch-hunting group.

    Then scan all other recently active forum users histories and identify any who have opposing positions on any of those.

    Then generate some statement likely to trigger the target user (perhaps looking at posts they responded to in the past) on the issue and then wait for any thread they’re both active and post a trigger statement.

    Repeat 10,000 times until the witch hunters hate each other more than they hate you.

    • Nick says:

      If Trollbot can find someone on here who hates Catholicism and Scooby-Doo and loves contemporary architecture and urbanism, I would be very impressed. Also furious.

      • Telomerase says:

        Surely you mean Kung Fury.

        And have you read the new book where the (renamed just enough to avoid copyright) Scooby-Doo characters return to fight a Lovecraftian monster that they had misidentified as a crooked landlord?

        It’s terrible, just wondered whether you’d read it 😉

      • Nornagest says:

        Architecture — institutional architecture, at least; vernacular’s more complicated — has been on an upward bend since the 1960s or 70s, but that says more about how bad ’60s and ’70s architecture was than how good ’00s and ’10s architecture is.

    • AG says:

      Can’t one simply run GPT-2 on a training set of all social media posts that generate the largest amount of engagement? Followed by a training set on the articles from online websites with the largest amount of engagement. I shiver to think what a mix of Jezebel, Vice, and Twitter bluecheck one-liners would produce.

        • AG says:

          What Reddit deems “controversial” vs. what drives virality/engagement are likely different things, though, and I feel that, particularly, what would be Reddit’s most controversial CW Thread content wouldn’t correlate to what actually is the most infuriating scissor statement stuff in the broader internet. The generated examples there are just about boldly stating an extremist position, whereas the most polarizing stuff is more about leaving things open to interpretation. The color of a dress, Michael Brown, Cat Person, and Aziz Ansari don’t themselves take a stance, at most making vague insinuations. This sets the stage for everyone else to give their Hot Take, in ways in which many different Hot Takes can be given on that particular subject.

          A Scissor Statement can’t be generated from scratch. It’s only part 2 of the situation. So a bot only trained on the people taking a side and digging in will generate contextless part 2s. A bot intended to actually troll and create scissor situations should be trained on the part 1s, to learn to darkly hint at things.

          • Murphy says:

            Ya… I suspect the better metric might be to search for comments that generate the most extreme vitriolic responses in their child threads.

            The most powerful troll statements sounds mundane but inspire others to start fights between the subgroups of the community. Things that bring out the worst in others.

            Not the angry responses themselves.

            The culture war thread is probably awful for that kind of thing since it’s a subgroup who are often trying to avoid riling each other up like that.

          • AG says:

            It’s more that the CW thread is almost by definition only a reaction to something in the broader world. The things the thread is reacting to are the actual sources the trolling.

  29. Dacyn says:

    Since no one’s mentioned it yet: Les Miserables

    • Machine Interface says:

      Which one?

      • Dacyn says:

        The one which is a legitimate answer to the question. (Although I have to specify I’m in the US, otherwise this movie would count.)

        • Machine Interface says:

          That’s why I asked actually, because the latter has gotten enough buzz that this seemed like a legitimate ambiguity — but if it hasn’t had a wide US release yet, problem solved!

  30. FrankistGeorgist says:

    For animation: Top marks go to Don Hertzfeldt, America’s greatest living artist.

    It’s Such a Beautiful Day
    World of Tomorrow
    World of Tomorrow 2
    Song of the Sea
    the Lego Movie
    Lego Batman Movie
    Spiderverse
    Garden of Words

    As for meatspace movies I’m hugely delayed seeing things and most of them are already covered here multiple times over but would mention some specific-to-me:

    Girlhood
    Tomboy
    Hail Caesar
    Inside Llewyn Davis
    Ex Machina
    Moonrise Kingdom
    Grand Budapest
    Kumiko the Treasure Hunter
    the Vvitch

    Lots of fun films but as I was living it 2010 felt like a dismal decade for movies. Probably because I actually worked at a groaty movie theater for part of it which’ll take the shine off anything.

  31. Freddie deBoer says:

    As someone who is not a beginner I have read an awful lot of books on socialism/Marxism marketed to beginners – graphical novels, A Brief Introduction tos, for Dummies. I mostly find them all badly flawed. But I just read Socialism 101 by Kathleen Sears and I think it is in fact a very good starting point for someone without any background knowledge. It historicizes in a way that many of these texts don’t. Recommended if you’re looking for such a thing.

    • Error says:

      Interesting.

      Diving into a new topic is always hard. You need to have some knowledge to know what’s worth reading and what’s a waste of time. If you outsource that question, you need to know enough to know whose opinions are worth listening to. It’s hard to separate signal from noise when you don’t yet know how to identify signal.

      There’s a related difficulty trying to pass the Ideological Turing Test. There’s still more noise than signal. The most readily available public information will be whatever’s most virulent; either hatchet jobs by a group’s enemies or PR burbling from their allies. Good-faith overviews of politically divisive topics are hard to find.

      I don’t have much of an interest in socialism, but the opportunity to get a look at the signal behind the noise is enticing, and my prior for people who read SSC having *some* idea what they’re talking about (enough to judge which books are worth reading) is pretty high. I think I’ve seen a fair number of right-wingers shit on you in the comments, too, which is I guess is *sort of* an endorsement of your socialist credentials.

      Maybe I’ll pick it up.

    • lhudde says:

      Freddie, since you’re here and dispensing reading suggestions, can I make a followup reference request?  (I hope the framing will make this non-CW, but if it fails, somebody please let me know and I’ll withdraw until the next OT!)

      I haven’t read Sears’s book (have ordered it), but if it claims to take things from Marx to “universal healthcare,” that sounds as though it will end with the nowadays-very-common move of conflating structural socialism with social democracy/ mostly-capitalism+generous-welfare-state.  I’m confused about why there’s not more resistance to this move on the part of actual socialists, given the way in which social democracy seems to fail any kind of basic class analysis.  Surely in the absence of any deep starting revision of ownership/production structures, any expanded government programs will be entirely run by the existing bureaucratic/managerial class, for their own benefit and according to their own values, and thus will work to preserve rather than to level power inequalities between classes? This is certainly the case for programs like free college/ free non-emergency healthcare/ other expanded public-funded social services, many of which are of very dubious empirical value to the working-class consumer/client, but do conveniently effect a substantial transfer of wealth, status, power and social/cultural authority to the professional classes who staff and gatekeep those programs.  

      I’ve seen tepid “social democracy is not socialism” explainer articles, but I can’t figure out why professed mainstream socialists like e.g. Nathan Robinson aren’t more vocally and angrily anti-social-democracy, given its seeming cuckoo-like potential to masquerade as pro-worker while quietly diverting most of the resources to the enemy.  Are you aware of any good entry points to discussion on this issue in the socialist community? I’d love to get a better sense of how/if it’s discussed among those in the know.

      • Sandpaper26 says:

        Re: actual socialists don’t seem to oppose social democracy enough — we’re out there. But it’s a combination of “I’m so busy doing real work and trying to help people around me, I don’t have a lot of time to Internet Rant” and “it’s impossible nowadays to promote actual socialism and not get either laughed or kicked out of the room, so I’m supporting SocDem, fortifying the line between liberalism and leftism, and waiting for the Overton window to shift my way a bit.”

      • Hoopdawg says:

        Nathan is (Harvard-educated!) PMC himself, you can’t expect him to go against his class interests! (Well, not too hard, at least.)

        My impression (that I cannot back up with meaningful citations, since most of my experience comes from random conversations on online fora) is that welfare state is, in theory discussions, near-universally derided as an capitalism-preserving aberration, yet at the same time near-universally understood as a vast improvement for the lives of the working class over the status quo in practice, and therefore a valid immediate goal, and potentially a step in right direction. (At large and in principle, obviously specific policies are much more controversial and argued about, e.g. UBI vs. Job Guarantee, or indeed free college.)

        And the PMC problem is more and more widely understood to be a problem indeed. Taken together, the conflict currently materializes most clearly in the divide between Sanders and Warren support. Sanders is widely perceived as a genuine socialist running with a moderate soc-dem program for practical purposes, Warren – as a liberal representing PMC class interests. You will notice Warren is being soundly rejected by wide swathes of self-professed socialists (to the consternation of liberals who genuinely can’t see the difference between the two).

      • Freddie deBoer says:

        This should help!

        • Mark V Anderson says:

          That was very interesting Freddie. I knew you were left, but I didn’t know you were a full fledged socialist. I’d love to hear your ideas on how socialism would work in its finished form. Well nothing is ever finished, but in a decent socialist society that got rid of profits (and markets, you said we should get rid of markets too).

          I don’t see how it would work at all. How do people have incentives to do what needs to be done, and how do people even know what needs to be done? Either on SSC, or if you have link to somewhere else where you or someone else have explained it.

      • Erusian says:

        I’m not a socialist, but my understanding of the historical perspective: Socialists in the 19th century had a problem. On the one hand, they believed socialism was necessary for the future and that the working class was being exploited and that the world needed to transition to socialism as quickly as possible. On the other hands, most governments (in far heavier-handed fashion than today) suppressed the working class vote… and even among the working class, their ideas weren’t as popular as they wished.

        There were two solutions to this: the Radicals advocated for revolution. They said it was unconscionable to delay socialism or to act within current structures. They said that bourgeois democracy would crush socialism to maintain narrow class interests. They pointed out that, although they believed socialism was more economically productive than capitalism, capitalists would never let it get a fair shot in the market. So the response was violence: overthrow the unjust system and replace it with a system that could create socialism.

        The other group (who I will call the Possibilists after their French name) argued against violent revolution. They said that feudalism had transitioned to capitalism with only limited revolutions. They pointed out that empowering the workers by disempowering them in a dictatorship was contradictory. They said that socialists must focus on convincing people and showing they were right, even though capitalists would try to thwart them. They believed they must win democratic authority and maintain power in it through a popular front of worker’s power. They advocated for engaging in bourgeois democracy, believing that eventually democracy could create socialism. (And, in fact, arguing that only popular democracy could create socialism.)

        Well, both sides kind of got their states. Not exactly: both sides didn’t get their ideal state and their revolutions were betrayed in various ways. But the radicals staged successful revolutions in China and Russia. The possibilists succeeded at winning broad power in Sweden and the UK and used broad democratic majorities to start nationalizing large sectors of the economy etc. Unfortunately for the socialists, the possibilists never succeeded in actually convincing people to actually end capitalism. But then again, the radicals didn’t (by socialist standards) achieve true socialism either.

        And most socialists agree that, while socialism is good on the whole and Sweden is a very imperfect and still capitalist state, it’s better than Communist China. There are some tankies running around who still defend China as the worker’s state and see it as waging a glorious socialist war against the capitalist west. Or who still worship Stalin. But they’re not really taken seriously.

        And in practical terms, the Radicals suffered a huge loss in prestige with the collapse of the Communist world and the transition of China. Really the only bastions of radicalism left are minor parties, small groups who want to do the revolution right this time, and a few small isolated states.

        In contrast, the possibilists still have major presences in major parties in hugely important countries. While the possibilists have been on the retreat, they’ve not suffered a huge blow to their prestige. Oh, they lost an election in Sweden? That’s bad but it’s not as if the socialist party there collapsed or has no influence. If anything, it seems a partly good thing from a certain point of view: this retreat has shaken up the stale socialist parties that had been comfortable with the amount of socialism they’d advanced towards. Now more radical people can get in and push closer to abolishing capitalism.

        However, the possibilists have also spawned a sort of conservative quasi-socialist who’s basically a capitalist who likes some of socialism’s rhetoric and is comfortable with welfare. (These ‘conservatives’ are leftist Democrats but they want to preserve the system, thus the label conservative.) This person may consider themselves a socialist, though most doctrinaire socialists and capitalists won’t. They (weirdly enough) both agree this person is a kind of aristocrat who find the managerial state and capitalism useful to maintaining their position. This group has always existed (in fact, there were contemporaries of Marx arguing that socialism was good and all but had to have an aristocratic character that retained their privileges). But particularly in the 1990s and 2000s, there was a fear they were taking over and strangling the possibilist movement into being just capitalist leftism but slightly more so. And if the possibilists were being strangled, and the radicals were gone, then socialism was…

        That didn’t happen. We’ve seen Bernie Sanders and a more general resurgence (even if they’re a distinct minority of the population.) The thing is, the conservative quasi-socialists have a lot of money and an elite base of support. And the possibilist method has always been to (as the French Possibiliste party said) “do today what is possible so we may tomorrow do what is just.” So if what’s possible is allying with these people to make things better for the worker and making them more receptive to voting in socialism, that’s fine. For the moment. Yes, the welfare state is based on redistributing profits created by capitalist exploitation. But if we can’t abolish capitalist exploitation (and we can’t just now), we might as well make better use of them than putting them in capitalists’ pockets.

        (To those who consider themselves socialists, how’d I do?)

        • Aapje says:

          But particularly in the 1990s and 2000s, there was a fear they were taking over and strangling the possibilist movement into being just capitalist leftism but slightly more so. And if the possibilists were being strangled, and the radicals were gone, then socialism was…

          That didn’t happen. We’ve seen Bernie Sanders and a more general resurgence

          This seem premature or even engaging in wishful thinking. Third way politicians like Bill Clinton, Tony Blair, Matteo Renzi, Romano Prodi, Gerhard Schröder, Wim Kok, etc, etc all got to run their country.

          Sanders did not. In fact, the working class still seems to increasingly be moving to right-wing parties. Western socialist parties seem to mostly still focus on things that appeal to the elite, not the working class.

          In many cases, this is a matter of moral conviction, making them unwilling to adopt working class concerns.

          EDIT: oops, forgot that this is the open thread.

    • Plumber says:

      @Freddie deBoer,
      Thanks for the recommendation. 

      More than I decade ago I had an interest in an (I thought) untried flavor of socialism, and read Bertrand Russell’s 1917 Political Ideals, his 1920 Proposed Roads to Freedom: Socialism, Anarchism, and Syndicalism, Arthur Penty’s 1906 The Restoration of the Gild System, G. R. S. Taylor’s 1919 The Guild State: Its Principles and Possibilities, G.D.H. Cole’s 1920 Self-Government in Industry, and I skimmed his History of Socialist Thought (various versions, one from the ’20’s featured a lot about Trotsky, and one from the 1950’s almost nothing). 

      Instructive to me was what disperate political “sides” these authors took later in the 20th century, and also finding out that some of the ideas of the theoretical “flavor” of socialism that interested me were nominally tried (in Italy, Sweden, and Yugoslavia) from a few years to a few decades after those works were published. 

      Suffice to say my thoughts on the results of the nominal attempts to put some of the ideas into practice probably counts as “hot button”.

      • Erusian says:

        I’d be curious to see this in a hidden thread, honestly.

        • Plumber says:

          @Erusian,
          Oh sure, l’ll plan to a short post this on Wednesday or Thursday, it’s a tangled tale involving both nominally Left and nominally Right movements.

  32. DragonMilk says:

    So I watched Age of Empires 2 for the past week as there was a big tournament sponsored by Microsoft. Not huge in terms of esports (20k prize pool, viewers got the pool up to about another 20k), but hype since the game is 20 years old…with the release of Definitive Edition being the catalyst for prize pool and so on.

    Anyway, what other games circa 90s/early 2000s still feature a LAN style tournament?

    Finally, obligatory please let me know if you still play AOE2!

    • cassander says:

      Starcraft, definitely.

    • Error says:

      I don’t, but I wish I did! I don’t know anyone who plays anymore.

      Starcraft is still played, of course. And there’s a surprisingly vibrant community around Supreme Command 1, which I love to death. Unfortunately it’s mostly in Europe or something so finding a game is hard. 🙁

      • DragonMilk says:

        I used to only play Starcraft, but now only do Star Chess there (auto battler, think of it as a card game).

        The low apm requirement (4 at most) means I’m one of the best!

      • Hackworth says:

        “Supreme Commander” please. Also, there is no need for a number anyway, since a terrible sequel has definitely never been made. Finally, the community lives on under the moniker “Forged Allicance Forever”, with launcher, matchmaking, ranks, single player, replays.

        • Error says:

          Ugh. I’d blame that on me being super tired, but apparently I wrote it at 11am, so blah.

          Anyway, yes, FAF is what I was thinking of. I could hardly ever find a ladder game at a time of day I could consistently play, though, and eventually I stopped trying. And that is sad. 🙁

        • Dan L says:

          Was a fix ever found for the rampant AI threading? I know Sorian’s mod helped in that regard, but it didn’t seem to solve the issue so much as push back the onset of terminal slowdown.

    • Jon Gunnarsson says:

      Warcraft 3 still has an active scene. The game always had a decent following in China, but the Western scene has rebounded in the last two or three years. LANs are still mostly in China, though.

    • alephnaut says:

      Starcraft, of course. Also, Super Smash Bros. Melee, if you’re willing to count that as a LAN-style tournament.

      • d20diceman says:

        If anyone’s interested in watching Melee (or Smash in general), be aware that Genesis, one of the largest Smash Bros tournaments and the first major of the year, is next weekend and should be a great tourney.

    • Roebuck says:

      I used to spend ridiculous amounts of time playing and watching AoE2 (by which we mean AoC) between 2014-2015 (with some more watching again in 2018).

      I think one reason it’s popular is that it was released back when there weren’t many good games, so many people know it, know that other people will know it, and gather around it when they want to play a good multiplayer strategy.

      Also, I feel angular-looking games give players more control and are more conducive to competitive play (I’ve played most of the Stronghold series and I can’t help but see a relation where every time they made the physics and graphics more smooth, people would get angry).

      The big revival of AoE2 seemed to be around 2014, when the HD version was released to Steam. There have been professionals keeping it alive before, but suddenly it turned into a wholly different thing.

      Around the same time technology got good enough for people to stream (or put on Youtube) games in HD / Full HD quality, without which AoE2 looks very bleak. Let us give real respect to the people who kept the game alive before HD Youtube videos and streams got popular. I don’t know what kept, say, ZeroEmpires doing his work as a commentator before 2013.

      Anyway, the community that evolved around ZeroEmpires, TheViper, Memb etc. was something really homely and peaceful. I live in a big city and have a good job with reasonable people time, but being born a smalltown introvert I often feel like going back to spending several hours a day playing AoE2, listening to music and not giving a damn about the world outside.

  33. A Definite Beta Guy says:

    Why does everyone like Fury Road so much? I thought it was great from a set design/technical viewpoint, but the actual movie and characters were realllllllyyyy underwhelming to me. What do y’all like about it?

    • DarkTigger says:

      Same thing as StarWars: the visuals where absolutley awe-inspiring. But in opposite to the last few SW-movies the story was only underwhelming and not “so stupid they should fire every playwrite on the project, and than some extra, just to set an example”.

    • FrankistGeorgist says:

      In an age of highly muddled blockbusters its simplicity is why I like it. It knows what it is and what it is is being chased in one direction and then the other with a goddamn fire breathing guitar truck.

      • Matt M says:

        Agreed. I also like that the villain is clearly villainous and that we don’t have to sit through a bunch of backstory about how his motivations are actually super-complex and make sense in a certain way and how he had a bad childhood and all of that…

        “Shades of grey” has become so overwhelmingly dominant that it is now the default Hollywood setting, and presenting a black and white good vs evil story is the new interesting experiment.

    • Jon S says:

      Fury Road is not my kind of movie, and is in my top 5 for most-memorably-bad movies. I’m sure I’ve seen plenty of worse movies, but not ones that other people like so much. I know I’m in a tiny minority, but you’re not the only one who doesn’t like it.

      Seeing it on someone else’s list is an easy signal for me that their tastes are quite different from mine.

    • Machine Interface says:

      It’s the platonic ideal of the action film: 15 minutes of almost purely visual exposition and then a non-stop, movie-long, bombastic and clear-cut action-chase sequence, with no shaky-cam nonsense, staging and montage that remain perfectly clear and readable in spite of the fast-paced rythm, a simple but dynamic story with high stakes and multiple character arcs that all get resolved before the ending, and a memorable soundtrack to boot (how many Marvel movies can say the same?)

      I can perfectly understand hating this movie, but I really don’t see what’s so mysterious about loving it.

      • Le Maistre Chat says:

        It’s the platonic ideal of the action film: 15 minutes of almost purely visual exposition and then a non-stop, movie-long, bombastic and clear-cut action-chase sequence, with no shaky-cam nonsense, staging and montage that remain perfectly clear and readable in spite of the fast-paced rythm, a simple but dynamic story with high stakes and multiple character arcs that all get resolved before the ending,

        It’s a very retro film project (did it have any significant CGI?) that still gives some “meat” for university-educated film critics to chew on by blaming “toxic masculinity” for the apocalypse that enables its post-apocalyptic panoply of tropes. And in the midst of nearly non-stop action it also incorporates full character arcs for multiple characters.
        If action films are legitimate at all, it’s a good movie. If not, it’s at least a well-crafted bad movie that doesn’t assault our eyes with CGI that will get dated. 🙂

        • Nornagest says:

          (did it have any significant CGI?)

          Quite a bit of it, actually, just not much on the vehicles or characters. Most obviously, the sandstorm is almost all CGI. Less obviously, a lot of the scenery is: the cliffs, the swamps, exterior shots of Immortan Joe’s fortress, etc.

        • cassander says:

          if you google it, you can find side by side shots of pre and post CGI, it’s quite fun to watch.

        • Paul Brinkley says:

          As was said, there was plenty of CGI. What was interesting was how it was used. It was mostly to produce ambient effects like the sandstorm, as opposed CGI robots or monsters or other threats. The enemies were virtually all stuntwork, as were most of the vehicles.

          Fury Road scratches the same itch as John Wick, and a lot of Mission: Impossible and at least one set piece in any Bond flick, and nearly everything Jackie Chan does. Here’s some nifty stunt work and proper editing so you can follow the action; isn’t that neat? It’s pretty easy to appreciate the skill in this, to the point that you could even skimp on the story and acting to some extent.

          Fury Road obviously drew looks because of its treatment of women, but people should never forget that it put a great deal of craft into the action directing. There are several YouTube videos that analyze the action shots and sequences to this effect.

    • Enkidum says:

      I thought it was great from a set design/technical viewpoint, but the actual movie and characters were realllllllyyyy underwhelming to me

      Set design and technical aspects aren’t part of the “actual movie”?

      I think you mean things like plot, dialogue, characterization, etc were not up to your standards. But that’s treating a movie as a novel, which it isn’t. All the movie-specific aspects of Fury Road are jaw-droppingly good. There is intense, complex action practically throughout the movie, but you understand the spatial relationships between everything at all times (see @Machine Interface’s post for the same point). This is rare these days – compare virtually any action scene directed by Abrams or Bay, if you want an example of how not to do this (and John Wick for an example of how to do it right). The sets/scenes/props, as you grant, are stunning.

      As for the story aspects of the film – consider how much we know about the society (societies, I guess) it presents, with so few words devoted to it. This is a marvelous economy of storytelling. Also as @Machine Interface notes, all major character arcs are clearly resolved, and the story as a whole is complete.

      Now, you may think the setting is dumb, the characters are cliched or silly, and the stakes don’t interest you. I suppose that’s all true, although in comparison to, well, any other blockbuster I can think of from the past decade, it compares very favourably in those regards.

      Of course you’re entitled not to like it and I’m not trying to argue you should. But maybe the above makes sense?

      • John Schilling says:

        I think you mean things like plot, dialogue, characterization, etc were not up to your standards. But that’s treating a movie as a novel, which it isn’t.

        You seem to be suggesting that plot, dialogue, and characterization are only important to novels. I would very much disagree with this. Plot, dialogue, and characterization are very important to anything that purports to tell a story about people, including both novels and movies. A story that falls short of greatness in those dimensions is not a great story. It is unlikely to be more than a pretty good movie no matter how good the set design, choreography, and effects look.

        As for the story aspects of the film – consider how much we know about the society (societies, I guess) it presents, with so few words devoted to it.

        How much of this is because it uses the two words, “Mad Max”, to evoke several movies’ worth of other peoples worldbuilding?

        • Enkidum says:

          You seem to be suggesting that plot, dialogue, and characterization are only important to novels.

          Ah, no, of course that’s not true. It just seemed that OP was going in the opposite direction.

          How much of this is because it uses the two words, “Mad Max”, to evoke several movies’ worth of other peoples worldbuilding?

          Not sure who the “other people” are – George Miller directed and wrote all of them didn’t he?

          At any rate, I think the answer is “not much”. My kids saw it without having seen the previous ones and seemed to get virtually all of it.

        • Nornagest says:

          Not sure who the “other people” are – George Miller directed and wrote all of them didn’t he?

          Yes, although George Ogilvie has co-director credit on Mad Max Beyond Thunderdome, and Miller had co-writers on all four. It’s probably not a coincidence that Thunderdome is the weakest of the series.

  34. herbert herberson says:

    in no particular order, except that Fury Road is the best

    Mad Mad: Fury Road

    The Witch

    Thor: Ragnarok

    Us

    Hunt for the Wilderpeople

    Pacific Rim

    The Congress

    Inception

    then, adding the following for the ones from other replies that I missed:

    Her

    Arrival

    Edge of Tomorrow

  35. bean says:

    Biweekly Naval Gazing links index:

    The US Navy has a unique tradition where the first log entry of the New Year can be written in verse, so long as it contains all of the required information about the ship’s condition. Some are surprisingly good.

    An article by the Center for a New American Security, titled Retreat from Range, has gotten quite a bit of press coverage for its claims that the range of a US carrier wing has fallen sharply. I’ve done an analysis of the numbers given, and found that the situation is not nearly as dire as indicated, primarily because the author can’t keep radius and range straight.

    My series on Billy Mitchell wrapped up, with a look at Michell’s career after the Ostfriesland sinking, including the Court Martial that finally forced him out of the Army.

    Lastly, aerial weapons is back once more with a look at something that isn’t technically a weapon, the history of aerial decoys.

    • Aapje says:

      My series on Billy Mitchell wrapped up, with a look at Michell’s career after the Ostfriesland sinking, including the Court Martial that finally forced him out of the Army.

      Whereupon he became a Donkey Kong champion, only to be stripped of his records once he was found to be cheating.

      Plus ça change, plus c’est la même chose…

  36. J Mann says:

    By a long shot, my favorite movie of the 10’s was The Hateful Eight. I sat there through the credits trying to process how I felt about it, but over time, I like it more and more.

    I enjoyed a bunch of other movies, but that’s the one that stuck with me.

    • The Pachyderminator says:

      Hateful Eight was easily my least favorite movie of the decade, and possibly the most unpleasant experience I’ve ever had with a movie. It didn’t help that the theater was full of jackals and hyenas who laughed wildly at every crescendo of the violence. I wish I could have enjoyed it, because it looked magnificent in 70mm, and I don’t get to see movies in that format very often.

      • Enkidum says:

        Tarantino is like that. He’s a masterful director who is saying a lot more than people think he is (highly recommended: the speech Stanley Crouch wrote about him when he accepted an award on his behalf), and is simply really, really good at an awful lot of the craft of making movies. But in virtually all his films, there’s a threshold of violence, apparent glibness, and constant layers of references that you have to be willing to tolerate, and see beyond.

        That reads like I’m trying to say that you should learn to tolerate it, but that’s not what I mean. All I’m saying is (a) there is a lot that is worthwhile in Tarantino, he’s one of the best directors, well, ever, and (b) a lot of people cannot, and likely will never be able to, deal with his more extreme tendencies. Which is just the way it is.

        Similar to your second sentence, I’ll never forget the first time I watched A Clockwork Orange in the theatre, which the audience treated as a straight-up comedy, laughing their way through the entire thing (yes, including the “Singing in the Rain” sequence). I’ve never felt as superior as I did for those two hours.

  37. nulldimension says:

    Investing for the long term has been on my mind lately.

    How does one balance between investing the maximally possible amount for a great retirement while still having enough to enjoy life’s earlier years?

    For example, what is worth having 25 million by 65, only to die of a sudden illness at 66? With medical advancements, should people be saving even more as we are living longer and longer?

    • EchoChaos says:

      How does one balance between investing the maximally possible amount for a great retirement while still having enough to enjoy life’s earlier years?

      Mostly by having a cheap Anglo-Saxon heritage that forces you to save regardless and sucking it up.

      For example, what is worth having 25 million by 65, only to die of a sudden illness at 66?

      Depends on how many children you have and how much you care about their futures.

      • brad says:

        I know men can have children well into old age but for most people at 66 I would expect those children to be adults. I don’t think not leaving a windfall to people that ought to be self sufficient by that point implies a lack of care as to their futures. In fact, a bit to the contrary.

        • EchoChaos says:

          I know men can have children well into old age but for most people at 66 I would expect those children to be adults.

          As would I. When I’m 66, my oldest will be older than I am now.

          I don’t think not leaving a windfall to people that ought to be self sufficient by that point implies a lack of care as to their futures. In fact, a bit to the contrary.

          No, that’s a different statement. Planning to leave a windfall may or may not be a good idea, but “what use is it” implies that it’s a total loss, which is obviously not true.

          Our family lives very long lives, so both my grandfather and grandmother outlived their retirements and had to be supported by my father. He has planned so that I won’t have to support him, but if he leaves a windfall instead, he would certainly not consider it useless to leave it to his children.

        • A Definite Beta Guy says:

          They might be self-sufficient, but giving your children (and grand-children!) money is something that a lot of people value. I’m sure my in-laws would like being able to give each of their kids $5 million, if they could.

        • Randy M says:

          Being able to pay for a portion of your grandchildren’s college or weddings would surely be welcome by the children and grandchildren.
          Just so long as you didn’t run out after half the children.

    • Matt says:

      The rule of thumb I’m planning for is that you can take about 4% out of your savings indefinitely with minimal (but not non-zero) risk. So I want to have saved up about 25 times my net salary for retirement. My expenses should have declined by then since I won’t be paying a mortgage or raising kids anymore. I’m not really factoring Social Security into my plans, though I suppose I am considering Medicare, since I’m not planning for higher medical expenses.

      Saving more than that seems like saving too much, unless your goal is to live much better in retirement than during your working years, which is what you seem to want to avoid.

    • DragonMilk says:

      Figure out what “enjoy life” costs.

      I’m a gamer. It’s a very frugal way to enjoy life as I buy games only when they are vintage – for instance, my Christmas purchase was $15 for myself and $19 for a friend.

      On the other hand, some people like drinking at bars every week.

      It’s helpful to have a spreadsheet of your monthly expenses and figure out which enjoyments give the most bang for the buck. From there, you can shove more into savings.

      And yes, savings. Investing should be a portion of savings above a liquidity amount.

    • DinoNerd says:

      I’m currently dealing with a related issue – precisely when to retire. I’ve always been a saver – from back when my parents got me started with my first bank account, where part of my allowance routinely ended up. (The latter being my decision, not theirs.) Now I probably have more savings than most people in my income bracket, having routinely maxed my 401K for years – and then saved more on top. I also have a job that motivates me to ignore alarm clocks and contemplate calling in sick – that’s nonetheless a huge improvement on my previous job, morale-wise.

      But pulling the trigger – i.e. deciding after what point I’ll have no more earned income – is really hard. Lifespans are highly variable; I wish we had a better way to share the risk(s) of an unexpectedly long lifespan (great in other ways, but potentially horrific financially). And then of course there’s inflation risk – I’m being advised to move out of equity to an extent that scares me, combined with potential inflation; but of course this may be because I was a young adult during the early 80s inflation spikes, so imprinted on scary inflation levels, and inflation as a convenient tool for a government wishing to implement a wealth tax, disguised as income tax.

      • redoctober says:

        I think you should consider that you don’t *have* to retire completely. Many folks find that having literally no income and/or professional activities is more stressful or boring than maintaining some low level of work- my father, for instance, retired from full time ER doctor work to part time work in a clinic, just to keep a hand in and maintain a small income (with the ability to ramp up if needed.

        • Matt M says:

          IMO, this sort of thing still counts as “retired.”

          Taking a flexible, part-time job, whose income is wholly insufficient to support your life or lifestyle, more closely pattern matches leisure than it does productive labor.

    • broblawsky says:

      How does one balance between investing the maximally possible amount for a great retirement while still having enough to enjoy life’s earlier years?

      You should always max out your IRA contributions for each year if you can afford it. Everything beyond that can be put into stocks/bonds/other instruments that you can sell without substantial tax consequences if you want beer money. As far as ‘enjoying life’s earlier years’ – the real question is, what do you actually need to be happy? If you don’t have a family to take care of, you can economize without it really impacting your quality of life too much.

      For example, what is worth having 25 million by 65, only to die of a sudden illness at 66? With medical advancements, should people be saving even more as we are living longer and longer?

      Presumably, you expect to have next of kin by age 65. The money has value to them, if not to you. Planning to fail is always a bad idea when it comes to long-term financial decisions.

    • helloo says:

      Generally speaking, current medical advancements have not been nice to people’s pocketbooks for those long lived.

      It’s not to fear suddenly dying and wasting money, but rather NOT dying but becoming very sick.
      How much it would cost to handle those expenses are probably the majority of the retirement cost for most people, though some definitely do “over save” and have more money than they would likely need or want to use.
      How to balance that would very much depend on how you value your enjoyment, at what age they still can be valued and their costs. Additionally, some things like land or arts can be fairly easier to spend money on while still retaining capital, but that’s hardly true for all activities.

      • EchoChaos says:

        It’s not to fear suddenly dying and wasting money, but rather NOT dying but becoming very sick.
        How much it would cost to handle those expenses are probably the majority of the retirement cost for most people, though some definitely do “over save” and have more money than they would likely need or want to use.

        This is my father’s retirement target (and mine, although that’s further). Both of his parents outlived their retirements and he had to support them, so he has saved substantially because he saw how much it hurt HIS father to be a burden to his son and he doesn’t want to do that to his children.

        How to balance that would very much depend on how you value your enjoyment, at what age they still can be valued and their costs. Additionally, some things like land or arts can be fairly easier to spend money on while still retaining capital, but that’s hardly true for all activities.

        Honestly, in our modern society there are very few hobbies that require massive cash to pursue that aren’t traveling. My father’s is woodworking, which once you have a shop costs a few hundred every time he wants a new large piece of wood, which can last him months or more.

        • helloo says:

          Honestly, in our modern society there are very few hobbies that require massive cash to pursue that aren’t traveling. My father’s is woodworking, which once you have a shop costs a few hundred every time he wants a new large piece of wood, which can last him months or more.

          Which is why I mentioned art (collecting) and land (mostly for the “retirement to a farm/forest cottage” but beach houses or building something on private land are also possibilities). But yes, most won’t require such investment, which typically is a good thing.

          Second, I choose those as you could still easily sell off those assets should you need the money/transfer it to others.

  38. Jaskologist says:

    The Lego Movie
    The Lego Movie 2

  39. tenoke says:

    I am considering getting into biofeedback – it’s perhaps the most promising, and interesting avenue for improvement that I haven’t spend much time on.

    Please share, any articles, papers, devices or relevant personal experiences.

  40. Xammer says:

    About the survey – I think it’s a bad idea to put it up during the holidays. One has usually less free time then (spending time with family etc.) and also it is an outlier period for several kinds of stuff. E.g. I was asked how many unread emails I have in my inbox – I answered 0 since that’s how many I usually have (like now, for example), but when I took the survey I had like 30 or something because it was the holidays period and I was being slack.

    • Radu Floricica says:

      I actually wanted to comment on how many good blog posts are around the holidays because people have more free time (like zvi) 🙂 I guess the mileage will vary.

    • meh says:

      I too was wondering about the ‘right now’ questions on the survey. I assume the reasoning is that people are bad at estimating their past behavior, but are pretty accurate at counting something like the number of unread emails. But it seems you are trading a correct mean for higher variance; which if you are trying to establish trends or correlations could be a problem unless you have a much larger data set. This is compounded by the problem you state, that the holiday period is usually not a typical case (so the mean is probably also off), and further compounded again by some people correcting if their ‘right now’ case is far from their mean; and others answering the question literally.

      What I mean, for example, is lets say 3 people on average have 10 unread emails. Ask them to estimate how many unreads they have on average, and you may get answers like 7,7,8. They all estimate themselves to be better at keeping up with email than they actually are, but their answers are still clustered. OTOH, ask how many emails they have open right now, and maybe you get answers 3,12,15. These have the right average, but a larger variance.

    • Roebuck says:

      I agree with meh above. I would have thought I pretty much always have 0 unread emails but, when prompted the survey, I checked and I did happen to have 2. You might have done a good thing by removing the 30-unit holiday bias and trading it off for this small bias of people rounding to zero too happily. You might have also done a bad thing if Scott realises that people’s inboxes were in holiday mode and he concludes that you are one of those who do read all emails promptly even on holiday.

  41. MorningGaul says:

    The Hateful 8

    The better half of Wes Anderson movies.

    The Handmaiden for Corean cinema

    Confession for Japan.

    In the new director corner, I’d put:

    Panos Cosmatos’s Mandy

    Yórgos Lánthimos’s The Killing of a sacred deer

    Steven Craig Zahler’s Bone Tomahawk by default, but I have great confidence in the next movies I haven’t seen yet.

  42. Deiseach says:

    Are you a cat person or a dog person? Are you religious or non-religious? Want to know how these are linked?

    Then the Journal for the Scientific Study of Religion has got you covered! (I had no idea there was such a thing, but apparently there is. The ‘scientific’ bit is social sciences, YMMV on whether this counts as scientific or not).

    Atheists prefer cats, Christians like dogs 🙂

    • Ketil says:

      Traditional values are loyalty, respect, obedience, usefulness; progressive values are independence and individuality¹. That’s dogs and cats for ya.

      ¹ Being a dog person myself I wanted to add cruelty to small animals here, but I guess that would be CW?

      • DarkTigger says:

        As someone living in a town with an rodent problem, having an vegetable garden:
        How is a dog more useful than a cat?

      • Aapje says:

        Being a dog person myself I wanted to add cruelty to small animals here, but I guess that would be CW?

        There is an interesting phenomenon that the people who tend to favor ‘animal rights’ have more pets than average, so anti (some or all) pet activism doesn’t seem to cause conflict among different groups like many other issues, but primarily causes intragroup conflict.

        In general, animal rights seems to either be a non-partisan or even a horseshoe issue. For example, Vote Leave ran social media adverts on animal welfare. Wilders’ populist party has a prominent parliamentarian who is single issue on animal welfare.

        • edmundgennings says:

          Similarly, UKIP has- had the strongest platform position in favor of animal rights of major British parties.

        • eyeballfrog says:

          And, famously, the Nazis were very big into animal rights. How one can support banning hunting while also supporting rounding up Jews for extermination is unclear.

          (Note that this is not meant to impugn animal rights activism–just demonstrate that it does seem fairly orthogonal to left vs right.)

    • meh says:

      This appears to be typical journalistic butchering and misinterpretation of social science.

    • Machine Interface says:

      Comparisons between dogs and cats are rarely pertinent, because we’re comparing a pack animal which has been actively selected for millenia to have more anthropomorphic qualities, notably body language (anyone who has ever worked with wolves now how radically different an animal this is), to a solitary animal which has mostly self-domesticated and still behaves a lot like its wild counterpart.

      This means a lot of cat behavior gets misinterpreted and misrepresented because they don’t express themselves in way that are easily decypherable by humans.

      We have three cats, all three love us and seek our presence throughout the day, and one of them begs for leftovers when we have dinner just like our dogs did, but cats just don’t do it in the ovely outwardly enthusiastic way dogs do (oh and some studies have shown that dogs in fact tend to be acting up and play comedy when they’re around humans, and behave much less like goofballs when humans are not present in the room).

      The idea that cats are cruel is also a misunderstanding: they’re probing their prey to see if it’s still alive and if they can eat it — because putting your head near to a prey that’s still alive is dangerous! And I’ve seen two small and apparently really sweet dogs attack, kill and eviscerate a hen with great joy so… I don’t think carnivoran mammals are really capable of cruelty, they’re just gonna carnivore.

      Both cats and dogs can be raised to behave — it’s just that people usually are aware of this with dogs while they tend to expect cats to just raise themselves, which is fair to a degree, as an untrained cat will usually have more acceptable behavior than an untrained dog (especially regarding toilet habits). Our two dogs were not very well trained, and as the result they were as mischivous as cats are usually perceived to be (notably they would use chairs to climb on tables and steal food whenever they had the chance).

      Cats are indeed useful as pest control, though to be fair they mostly work with small pests, like mice and insects. If you’re dealing with rats you probably want a specially bred dog or a ferret (the oft forgotten 3rd kind of domesticated pet mammal).

      • DarkTigger says:

        Cats are indeed useful as pest control, though to be fair they mostly work with small pests, like mice and insects. If you’re dealing with rats you probably want a specially bred dog or a ferret (the oft forgotten 3rd kind of domesticated pet mammal).

        Might depend on the kind of rat you have, but our first cat brought home rats several times, and had a wound from it only once.
        On the other hand rat’s are smart enough to think twice before choosing an shed as home, when a cat lives around there.

        • DinoNerd says:

          +1 on types of rat. I grew up in an area with Norway rats – they’d have given an average house cat a more than healthy fight, being of similar size. Here in California, we have a different type of rat, more like the size of a mouse – or a small hamster. No need for a terrier or a ferret to deal with these – any cat with an inclination to hunting will do.

          • nkurz says:

            > I grew up in an area with Norway rats – they’d have given an average house cat a more than healthy fight, being of similar size.

            Maybe you grew up with very small house cats and extremely large rats, but “similar size” seems very unlikely. From what I can tell, typical small house cats tend to be from 3-5 kg, and Norway rats tend to be from .3 to .5 kg — a factor of 10 smaller. Apparently, a rat expert in NYC (Robert Corrigan) has for many years offered a bounty for anyone who brings in a 2 lb (.9 kg) rat, and has never paid out. The smallest cat breed seems to be the Singapura, with the females being more than twice that weight (little less than 2 kg).

            Were you exaggerating for effect, or is this possibly an example of childhood memories that are larger than reality?

          • DinoNerd says:

            @nkurz Probably a combination of different sources of error – I’d have expected larger rats to weigh in at 80% of a small cat, not less than 20%. Thank you for correcting my misinformation.

            Fortunately the exceedingly feisty tom cat of my childhood didn’t bring his kills home – so I didn’t actually see any rats close up – and in any case his job was to deter them; any killing he did was extra. So I’m going on my interpretation of what was in the air, as well as what may have been maternal hyperbole. (She wasn’t very happy with the conditions we were living in, and acquired the cat as part of coping with those conditions.)

      • Le Maistre Chat says:

        (oh and some studies have shown that dogs in fact tend to be acting up and play comedy when they’re around humans, and behave much less like goofballs when humans are not present in the room).

        They’re emulating the behavior of the “omega” in a wild wolf family. One of the adult offspring in a pack will act much more of a goofball than their siblings, initiating most of the play between adults, which is lower status but serves the vital function of defusing tensions in the pack. This being is a more puppy-like behavior, it’s not surprising that the more neotenous domestic dog emulates it around us.

        • broblawsky says:

          The “alpha/beta/omega” terminology isn’t considered a useful model for wolf behavior anymore.

          • Le Maistre Chat says:

            That’s why I used the scare quotes. A wolf pack consists of a mother, father, and their pups, the pups staying for a flexible number of years once they’re grown, paying to live in their parents’ basement with puppy-care and contributing to a larger hunting party. Zoologists who go out to the field to study wolf families continue to report that there’s one non-parent adult who initiates most of the play between adults, basically a goofball.

          • broblawsky says:

            That’s fair. Sorry if I came off as petulant; this is kind of a pet peeve for me.

    • gryffinp says:

      This tracks well with my entirely unjustified suspicion that the the inflection point for the quality of the internet was when dogs became the meme animal of choice over cats.

    • broblawsky says:

      Reading the study, it looks like only the following factors are statistically significant (p < 0.05) predictors of pet ownership:
      1) Church attendance (negative correlation, just barely significant)
      2) Whiteness (but only for Evangelical or Catholic Christians, not other denominations)
      3) Income (positive correlation, extremely strong for non-Evangelical non-Catholic Christians)
      4) Republican self-ID (just barely significant)
      5) Number of kids (just barely significant)
      6) Age (Negative correlation, extremely strong)
      7) Education (Negative correlation, fairly strong)
      8) Urbanite (Negative correlation, only for Catholics)
      9) Biblical literalism (negative correlation, only for Evangelicals)

      I'd expect the maximum number of pets should be found with someone who is: non-Church attending, White, Evangelical or Catholic, Republican, a parent, young, uneducated, and either an urbanite or a Biblical literalist, depending on their faith. That's … not someone I know.

      • Le Maistre Chat says:

        I’d expect the maximum number of pets should be found with someone who is: non-Church attending, White, Evangelical or Catholic, Republican, a parent, young, uneducated, and either an urbanite or a Biblical literalist, depending on their faith. That’s … not someone I know.

        So the crazy cat-throwing lady from The Simpsons should be a young high school dropout and a mother. Who votes Republican.

  43. Liface says:

    Prisoners (2013) (this is the only one to also make my top 10 all-time)
    Interstellar (2014)
    Django Unchained (2012)
    Black Swan (2010)
    Renn, wenn du kannst (2010) (underground German movie about a love triangle between a caretaker of a guy in a wheelchair and a female cello player, recommended watch if you can find it)
    Moneyball (2011)
    Moonrise Kingdom (2012)

  44. SearchingSun says:

    Copying this over from my post on the subreddit.

    Why doesn’t Silicon Valley try to disrupt the academic publishing market?

    Silicon Valley billionaires such as Peter Thiel and Patrick Collison have been fairly vocal about their worries that scientific progress is slowing down. Academic publishing seems like a low hanging fruit to help this situation – the current market leaders add very little value and yet make 40% profit margins. They only stay powerful due to bad incentive structures that force researchers to publish in “prestigious” journals to advance their careers. Removing them from the system would free up hundreds of millions of additional dollars for research, and boost scientific literacy and innovation by making papers more available to the general public and researchers in developing countries.

    Off the top of my head, a potential solution could be to create a journal/journal platform with advanced data collection/analysis/visualization features built in to entice researchers to use it. Then massively undercut the big publishers’ prices on subscriptions.

    Why hasn’t someone done this yet? Or made any attempt to disrupt this market at all?

    • Frederic Mari says:

      With all due respect, you answered your own question – publishing matters most if it is done in “prestigious” journals.

      So your issue isn’t creating a journal or journal platform. It’s to make it “prestigious”. Which is about as easy as creating a luxury brand out of nothing. Intrinsic quality doesn’t count. If people could judge that accurately, they wouldn’t need the reassurance of the brand name and/or the ‘prestigious’ tag in the first place…

      So those prestigious journals have a lock on the market, they know it and they charge as much as the market will bear (or less? Who knows how far they could push this?)

      • Enkidum says:

        I dunno, the vast majority of publications are not in terribly prestigious journals, almost by definition. It would be relatively easy to disrupt the mid-tier journals, and I’d argue that’s exactly what has happened over the past decade with, e.g., the various Frontiers journals and PLOS ONE and the like.

      • SearchingSun says:

        Well prestige is something that gets built up over time. All you need to do is provide enough of an incentive to build it up over time. If you could provide researchers with much more powerful tools to represent their datasets, maybe some of them would switch over in the interest of producing better papers. Maybe you could recruit some celebrity scientists to sign a pledge that they’ll publish a certain number of their papers in this journal. And then maybe some of these papers end up being influential and the whole thing snowballs. It’s certainly difficult, but it’s also possible, I think.

        • Matt M says:

          Well prestige is something that gets built up over time.

          Usually, but not always. Successful disruption can become quite prestigious quite quickly in its own right. What’s the more prestigious car brand today, Tesla, or Mercedes?

          • Konstantin says:

            Keep in mind that authors need to impress tenure and hiring committees, which are often made up of older people. If you ask the 40-70 demographic, Mercedes beats out Tesla.

          • DarkTigger says:

            They worked on that prestige since the mid noughies. And in this time they have consistently proofen that they can deliver EVs that are more mass market compatible then the stuff that came before. It’s not like that happened over night.

          • Conrad Honcho says:

            New things can be cooler, I guess, but I’m not sure cool == prestige. Is there anyone who seriously thinks a Tesla logo on a car is more prestigious than an AMG badge?

          • The Nybbler says:

            What’s the more prestigious car brand today, Tesla, or Mercedes?

            Mercedes, but the C-class is a filthy aspirational vehicle so it and below do not count.

          • Matt M says:

            Is there anyone who seriously thinks a Tesla logo on a car is more prestigious than an AMG badge?

            Uh, basically everyone under 35, IME. As well as older folks who care more about signaling youth and with-it-ness than the more traditional form of prestige.

      • AnteriorMotive says:

        So your issue isn’t creating a journal or journal platform. It’s to make it “prestigious”. Which is about as easy as creating a luxury brand out of nothing. Intrinsic quality doesn’t count.

        I don’t think it’s actually that hard to create a prestige brand out of nothing. Prestige is a feedback loop, so while it’s hard to start from literally nothing, it’s possible to jump-start the feedback loop from zero using money or pre-existing prestige. (“pre-stige”)

        If Google made it’s own journal with the Google name attached, started publishing all its teams’ most groundbreaking results in it, heavily promoted each issue’s findings, and then gave free copies to every university of note, it would be Tier-1 before long.

        If Peter Thiel wanted to disrupt academic publishing with a new journal, and money was no object, he could offer 5000$ for every article he publishes. (more likely, some sort of intangible perk bestowed directly on the researchers, since part of the issue is a principle-agent problem wherein scientists are spending grant money rather than their own funds, and will get as much resources as needed to publish, as its the most important part of the process, so have little incentive to be price sensitive.) The most career-focused would still submit to Nature, but he’d get first pick of the rest. Then he just pays for the aforementioned publicity maneovres, and although news travels slow in academia, within a decade his journal should be one of the top dogs.

        • SearchingSun says:

          Yes, this! It does seem like a solvable problem with potentially massive benefits, so someone should try!

          • Enkidum says:

            Deepmind (a division of Google, which is doing a huge amount of the currently-sexy AI work) publishes most of its stuff on Arxiv. By doing so, it’s raised the prestige of the website immensely.

          • mcpalenik says:

            Deepmind (a division of Google, which is doing a huge amount of the currently-sexy AI work) publishes most of its stuff on Arxiv. By doing so, it’s raised the prestige of the website immensely.

            This seems like a really weird thing to say to me, like saying that CVS became a more prestigious grocery store after they started carrying Wonderbread.

            First of all, the Arxiv is meant for pre-publication. There’s no peer review, and literally the only barrier to putting a paper up is having a research institute affiliated email address. There’s some real garbage on the Arxiv, and getting your paper on the Arxiv will never be considered a mark of prestige. It’s useful for what it is, but nobody is ever going to be impressed just because your paper is there.

            As someone who has been looking at the Arxiv since I was an undergrad in 2003, I can also guarantee that it hasn’t suddenly become the known as the Deepmind/Google Arxiv. That stuff doesn’t really register if you’re not in the field, which the vast majority of people who put things on the Arxiv are not.

            I used to put things on the Arxiv sometimes, but I don’t bother anymore after having a permanent position because it counts for absolutely nothing in most fields.

            By contrast, the editors at Nature reject ~90% of papers without even sending them out for review. The reviewers are also particularly harsh about things like “is this work particularly interesting” and “does it have a broad appeal”. Although this is part of the reason for its reputation, it additionally has over 100 years of history backing it up (in the early days, peer review wasn’t even really a thing, and they had to practically beg people to publish their work).

            Scientific publication is incredibly idiosyncratic and there are tons of new journals being created all the time. People don’t want to send their work to journals that nobody has heard of, or that have too low of an impact factor, or that seem kind of fishy. People will also refuse to review articles for journals that seem kind of suspicious (and with having the deliberate goal of subverting some kind of real or perceived bias within the industry may fit that criterion).

          • HeelBearCub says:

            tons of new journals being created all the time. People don’t want to send their work to journals that nobody has heard of

            Aren’t those two things directly in opposition to each other?

          • mcpalenik says:

            Aren’t those two things directly in opposition to each other?

            No. Sometimes, prestigious publishers like Nature start a new journal, some people send them articles, but not nearly as many as real nature, and they have a much smaller impact factor.

            Sometimes, publishers nobody has ever heard of start a new journal, and the stuff that gets published there is just plain bottom of the barrel.

            Either way, I guess “nobody” is somewhat hyperbolic.

          • Lambert says:

            Find the abstract on the Nature website, then read the arxiv print for free.
            Sounds like an adequate solution.

          • mcpalenik says:

            I mean, if it’s on the arxiv, sure? I’m not sure what problem you’re solving, though. The point is that getting into Nature
            (whether you also put the paper on the arxiv or not) is totally different than just putting your paper on the arxiv.

    • Ketil says:

      They are. In machine learning – arguably the hottest, most siliconvalleyish research field – publications happen at conferences (taking care of the “prestige” part) with manuscripts openly and freely available at arXiv. Since arXiv links are citeable, significance can be measured directly by counting citations. I think hardly anyone even bothers to run their articles through actual publishing anymore – by the time you’re through the mill, the train of technology has left the station, and your work is long obsolete.

      • SearchingSun says:

        Yes, something like the ArXiv environment is definitely a huge step in the right direction. Someone on the Reddit thread introduced me to this journal that includes the better visualization tools I mentioned: https://distill.pub/about/. If they could somehow integrate with ArXiv that would come pretty close to the ideal I’m dreaming of.

    • niohiki says:

      Supporting @Ketil on machine learning, I’ll add that it is also the case in pure mathematics and theoretical physics, which have been using the arXiv extensively for quite a long while.

      I now do research on bioinformatics, and it’s disappointing to see how slowly it’s adapting to this new paradigm, but I think it will eventually move there. These open initiatives can gather “prestige” at the same rate as any SV initiative (or even more, because there’s the whole cool-open-public-libre side to it), and already have a head-start.

      Bonus: the Swiss universities have got pretty mad at academic publishing houses, up to the point that they are officially recommending to use SciHub or P2P via Twitter. Not kidding. More info here.

    • Enkidum says:

      I said this in response to one of the responses, but this is precisely what organizations that provide exclusively open-access journals do. For example, Frontiers, which is now the 5th-most cited academic publisher in the world. Not bad for only having been around a decade or so. Having published and reviewed several papers in some of their journals, they are definitely not as prestigious as Nature, Science, etc, or any of the top field-specific journals (e.g. Psychological Science for psychology) but their review process is legit, and if anything less liable to being gamed by flashy results.

    • albatross11 says:

      I think the part of the problem of academic publishing that can be solved by technology has been solved, and the result is preprint servers. Those abound in many different fields, and they’re a solved problem.

      There are still two parts that can’t be solved by technology, at least not directly:

      a. Providing a signal to everyone about which publications should be taken seriously, which is generally done via peer review. This is important both for saving time of researchers who could evaluate the work themselves, and for saving people who can’t evaluate it themselves from buying a load of plausible-sounding BS.

      b. Providing a signal to university hiring/tenure committees about which researchers should be given a job/promotion/tenure, which is generally done by summing up some kind of score involving the impact of the journal and the number of publications. (As the saying goes, tenure commitees can’t read, but they can count.)

      (a) is solvable at a community level. In cryptography, a good example of a solution is Transactions on Symmetric Cryptography–a journal + conference with no publication fees and permanent open access. Peer review is handled by an editorial board of volunteers and an editor. There’s a university kicking in a little money to keep the server up, but that’s not very expensive.

      (b) is solvable but it involves a coordination problem with incentives to cheat in favor of your field/subfield over others, and different interests, and lots of legacy policies and old researchers and inertia. It’s not really solvable by technology in any objective way–you can get better metrics for how good someone’s publications are, but it’s arguable whether a given set of metrics is the best one.

      • Aapje says:

        I think that academic publishing is ‘solved’ in the same sense that communicating was once solved, by writing letters. We nevertheless saw great improvements with email, Whatsapp, forums, SSC, etc.

        Even with open-access, there still seem to be immense inefficiencies in the academic workflow. In fact, the entire concept of a journal seems like an anachronism. Who still reads entire journals, aside from popular science magazines like Science and Nature?

        From my perspective, what you need for an efficient researcher workflow is things like this:
        R1. an advanced search engine with access to all published papers
        R2. ability to judge the impact of a paper
        R3. access to the raw data for a paper (not necessarily easily, due to privacy concerns for some data)
        R4. easy access to papers cited by the publication, papers citing it and papers replicating it
        R5. private or public access to pre-release papers for collaboration, review, editing, etc
        R6. ability to give feedback to authors and to respond to feedback
        R7. workflow support, deadline notifications, etc
        R8. ability to collaborate on, submit, get feedback for preregistered research plans

        For an editor:
        E1. a way to select and contact researchers for doing peer reviews
        E2. private or public access to pre-release papers
        E3. a way to assign papers to reviewers and monitor the reviewing
        E4. ability to receive and collaborate on preregistered research plans
        E5. publish the paper publicly with a stamp of approval
        E6. workflow support, deadline notifications, etc

        And for an efficient administrator workflow:
        A1. calculate an impact factor for all the output of a researcher
        A2. (privately) share an article with a PR person, so they can put out a lying press release

        Journals suck at R1, R3, R4 and R7 & are mediocre at R2, R5 and R6, R8, A1 and A2. Much of the work of editors currently seems poorly supported with tools. There seems to be a ton of rework, as offering up a paper for different journals requires lots of changes, mostly fairly arbitrary.

        I see a huge potential for a semi-monopolistic collaboration tool, being a mixture of arXiv, Jira, SharePoint, Zotero and Scholastica.

        • mcpalenik says:

          I don’t mean to come across as saying that publishing has always been an efficient, enjoyable process, and often times there have been significant barriers to getting things published that typically have to do with the review process (I’m in the middle of something like that right now, as a matter of fact), but I don’t understand most of the complaints on your list.

          R1. Google scholar, web of science, and sci-finder essentially do this. You still have to have a subscription to read journal content, though, if the journal requires it. I don’t know why each journal should additionally provide a search engine that searches every other journal. Most journals make their own content searchable, which kind of seems like the extent of their responsibility. Otherwise, we end up with a thousand redundant search engines.

          R2. Citation counts are available along with the article, as well as a listed of “cited by” papers on just about every journal’s website. Also, the particular journal something is published in gives you some idea about this, because different journals screen for impact to different extents.

          R3. Often included in supplementary material these days. It’s usually not required, and often not even helpful, but I have for example, specifically had requests by reviewers to include coordinates for molecules I’ve done calculations on in the supplementary material.

          R4. Just about every journal I look at online has links to (or at least a list of) the citing papers for its articles.

          R7. I guess I’m not sure what I would expect for a journal here. Usually there aren’t any hard deadlines. Reviewers typically get reminder emails every 1-2 weeks after a certain point. If you don’t respond to approve proofs relatively quickly, they contact you. The editorial staff usually responds to messages within a day. I’m not sure that I actually want the journal doing more than that, as an author.

          As for R6 and R8, how involved do you expect the journal to be in the writing process? I think R8 is well beyond the scope of what most journals would see as their role. I typically think of a journal as a place to publish a paper, not as a tool for writing a paper. Authors usually provide contact information in the article (address/affiliation and email address of the corresponding author). I would never expect a journal to give me feedback on a research plan, and I don’t think I would want that.

          A1. Why is this needed for efficient administrator workflow?
          A2. ?? Some journals will choose articles to highlight (for example, a paper I was part of in PRX was highlighted in APS Physics), or have an “editor’s suggestions” section. Often times they’ll request information from authors to take into consideration for what they decided to highlight in press releases, etc.

          A lot of this discussion seems to be the equivalent of “Why would anyone watch pitch a show to Netflix when you could just kickstart it and put it on Youtube”.

          • Aapje says:

            I think that there are a lot of journal idiosyncrasies, where things should be standardized. How to format the paper. Whether supplementary material is required. How it should be made available. How people get notified. Etc.

            Basically, science has slowly already been moving away from the ‘paper journal in your college library’, which has improved things, but there is a long way to go.

            More specifically, I don’t see why a modern journal needs to be anything more than an editor or editors using a standard platform.

            R4. Just about every journal I look at online has links to (or at least a list of) the citing papers for its articles.

            That part works reasonably for me, aside from the issue that I (as a sci-hub user) typically have to search for the cited paper by hand.

            However, finding papers that cite a paper is hard. ResearchGate shows both citations to and from a paper, but this appears based on error-prone scraping, that is often incomplete. Finding replication attempts for a paper basically requires searching all the papers that cite a paper and then figuring out which ones attempt to replicate it. It’s a nightmare.

            I think R8 is well beyond the scope of what most journals would see as their role.

            My comment was a combination of criticism about the way technology is used, but also scientific standards.

            My desired scientific process is one where scientists collaborate on working out a promising research plan and then preregister it with a journal that promises to publish if the execution is sound, regardless of the findings. So you’d have two review phases, rather than one, one for the methodology and one for the execution.

            This would make p-hacking much harder, as well as removing incentives to do so (as well as making it impossible for journals to cherry pick the sexy ‘positive’ results). It would force reviewers to explicitly consider both the methodology and execution in depth and would minimize the impact of the socially (un)acceptability of the findings on how much reviewers criticize the methodology and how willing the journal is to publish.

            Note that I’m merely arguing that the platform should offer this as a feature, not demanding that it be made mandatory (although ideally, it would become the scientific standard for many fields).

            As for R6 and R8, how involved do you expect the journal to be in the writing process

            I think that you misread and/or I was unclear.

            This feedback can be among scientists, not necessarily involving the editor. I do think that there ideally should be a more standardized way for scientists to communicate, preferably with a permanent record for auditing and other reasons, rather than the kludge of using email or such.

            Authors usually provide contact information in the article (address/affiliation and email address of the corresponding author).

            Which is another anachronism, based on the assumption that researchers work as individuals, always stay at the same university/company/etc, have one employer, keep the same name (marriage and transitioning can change this), etc. This is all you could realistically do with actual paper, which is what the system is designed around.

            A smarter system identifies each researcher or research group with some unique identifier that is linked to a record. For an individual, this would be their current email address, their current and past employers, their current name, their conflicts of interest, their gender, nationality, ethnicity, preferred pronouns or whatever. For a research group, the group would have general contact data, as well as current and historic information about the researchers that are part of the group.

            Note that this information is also of interest to meta-research (research on researchers). It allows much better research on questions like research group sizes, how often researchers move (between universities, between groups, from academia to companies, etc), gender issues, the level of internationalism and success by different countries, representation by ethnic groups, etc, etc.

            A1. Why is this needed for efficient administrator workflow?

            One of the main things that drives administrator judgment of researchers in hiring and promotions seems to be the impact of their work.

            A lot of this discussion seems to be the equivalent of “Why would anyone watch pitch a show to Netflix when you could just kickstart it and put it on Youtube”.

            It’s more that we’ve seen lots of tech companies provide a lot of value by using technology that effectively became a standard, including those that benefit greatly from network effects. Science has a ton of network effects and benefits immensely from standards, yet the level of standardization and technological adoption is relatively low.

            Also, the current system seems to be exploited by publishers, to the detriment of society. In particular, the very fact that publicly funded work effectively becomes owned by middlemen is a disgrace. If not for heroic criminals like sci-hub, it would be realistically impossible for me to do the kind of unaffiliated analysis of scientific papers that I can do now. We should not need these criminals to have this societal benefit.

          • albatross11 says:

            Aapje:

            I’m not convinced by all your criticisms, but I think this statement

            Also, the current system seems to be exploited by publishers, to the detriment of society. In particular, the very fact that publicly funded work effectively becomes owned by middlemen is a disgrace. If not for heroic criminals like sci-hub, it would be realistically impossible for me to do the kind of unaffiliated analysis of scientific papers that I can do now. We should not need these criminals to have this societal benefit.

            is spot-on. Academic publishers were once playing an important role in taking knowledge and getting it printed in a readable form and distributed to places where other interested people could find it. But it’s been a couple decades since we needed them for that. At this point, they’re almost entirely parasitic–the actual work of peer review is always done by a volunteer (I’ve been one of the volunteers many times, FWIW), and the formatting and distribution of the paper is now relatively easy with technology.

            It’s pretty easy to move the gatekeeping functions away from academic publishers–there are open-access journals that are done entirely by academics with a little support from a university, and some of these are quite good. (ToSC and TCHES in cryptography are examples.) This works because the gatekeeping was never done by the publishers, just by academics. It’s easy to cut out the middleman.

            But the social signal of which publications are more/less prestigious is still needed, and there’s a lot of institutional inertia that keeps some of the academic publishers’ high-profile journals at the top of the heap in terms of impact. That’s not going to last forever, but while it lasts, it lets the publishers continue to extract rents from, basically, their legacy position as the people who determine what publications are the most important and thus which researchers should rise or fall in status relative to one another.

            There are idiosyncrasies, but in my experience (which may not be much like other researchers–maybe my field or personal experience is just weird) it’s not a huge job to deal with them. Change the document class, spend at worst a day or so fixing stuff that got broken by the formatting change, and your paper looks fine in the new format. Compared to the work of doing the research and writing the paper, that’s a pretty small amount of overhead. And most of the time, there’s almost no overhead–you recompile your document and it looks fine.

            ETA: It occurs to me rereading this that Aapje is coming at this as a consumer of academic papers, whereas I’m coming at it more as a producer. That may explain some of the difference in our perspectives.

        • SearchingSun says:

          Wonderfully descriptive comment, thank you. Yes, this pretty much parallels my thoughts. The tools we have today are extremely unsuited to much of modern science and there is tremendous scope for bringing in new methods for collaboration and measurement of impact.

          One idea I had for the impact measurement problem (R2, E5, A1) was to create a system for “weighted” citations. That is you assign an impact to researchers/research groups based on the number of citations they received, weighted by the impact of the researchers citing them. Sort of like the neurons in a neural network, where the output is a biased sum of all inputs, except it goes in both directions. Because citations from influential researchers will have higher weight, this produces an impact factor that’s more fine grained than just counting citations. It’s possible that this system is susceptible to dangerous feedback loops, but there must be a way to solve those.

    • Aftagley says:

      I’m actually reading behave right now, and I get this same feeling.

      His

  45. stoodfarback says:

    Here’s some culture ship names from gpt-2: https://www.stoodfarback.com/2020/01/08/gpt-2-culture-ship-names.html

    Not nearly as impressive as other gpt-2 stuff; I just put a bunch of real culture ship names as the “prefix” parameter. But the results are still surprisingly good.

  46. Plumber says:

    @Atlas says:

    What are/were your favorite movies of the 2010s

    Too few, and if I include pre ’90’s films my list is too long, so…

    The Commitments (1991)

     35 Up (1991)

    The English Patient (1996)

    Gattaca (1997)

    42 Up (1998)

    Rushmore (1998)

    Saving Private Ryan (1998)

    Crouching Tiger, Hidden Dragon (2000)

    Primer (2004)

    49 Up (2005)

    Atonement (2007)

    Hot Fuzz (2007)

    Tropic Thunder (2008)

    56 Up (2012)

    Moonrise Kingdom (2012)

    Locke (2013)

    ’71 (2014)

    The Grand Budapest Hotel (2014)

    20th Century Women (2016)

    [please feel welcome to suggest a couple other films to me]

    • The Pachyderminator says:

      Always good to find another Wes Anderson fan. I also enjoyed Locke (funny story, I got pulled over for speeding – for the only time in my life! – while on my way to see it. Managed to get away with a warning).

      Have you seen 63 Up (likely the last one, with Michael Apted’s health the way it is) yet?

      • Plumber says:

        @The Pachyderminator says:

        “Always good to find another Wes Anderson fan”

        I haven’t seen all of his movies but I’ve liked the few I’ve seen.

        “I also enjoyed Locke…”

        Wow, someone who’s seen it besides me and my wife!

        “…Have you seen 63 Up (likely the last one, with Michael Apted’s health the way it is) yet?”

        Not yet, I don’t think the DVD has gotten Toto our local public library.

  47. Plumber says:

    For what it’s worth;

    A Plumber’s (short) Reading List for doing Good, True, Beautiful, Proper and NOT LAME! Dungeons & Dragons scenarios:

    The Broken Sword, and Three Hearts and Three Lions by Poul Anderson

    A Princess of Mars by Edgar Rice Burroughs

    Children’s and Household Tales by Jakob and Wilhelm Grimm

    “Red Nails”, and “The Tower of the Elephant” by Robert E. Howard

    “Jewels in the Forest”/”Two Sought Adventure”, “Ill Met in Lankhmar”, and “The Lords of Quarmall” by Fritz Leiber

    Elric of Melniboné, Knight of Swords, Stormbringer, “The Dreaming City”, and “While the Gods Laugh” by Michael Moorcock,

    “Black God’s Kiss, and “Hellsgarde” by C. L. Moore

    “Not Long Before The End”, “What Good is a Glass Dagger?”, and The Magic Goes Away by Larry Niven

    Gods and Heroes of Ancient Greece by Gustav Schwab

    The Hobbit by J.R.R. Tolkien

    First Men in the Moon, and “The Time Machine” by H.G. Wells

    and

    Dilvish The Damned by Roger Zelazny

    (note: I left out Jack Vance’s Tales of Dying Earth because it’s been decades since I read any of them and my memory of them are too dim, but I remember that they were good and they’re recommended by others)

    • Nornagest says:

      Three Hearts and Three Lions, Conan, Elric, etc. are all great stuff, but they’ll just lead you to standard D&D. You’ll have fun playing it, the setting you’ll make will probably be well above average by heroic fantasy standards, but it’s a very meat-and-potatoes sort of experience.

      The Dying Earth isn’t like that. The Dying Earth is weird D&D. It’s the kind of D&D that has crashed spaceships and walking fungus and monsters based off a half-melted pack of knockoff dinosaur figures, which you fight by upending a levitated blanket full of caltrops and lamp oil over the horde before finishing the stragglers off with a ridiculously long-winded spell. The kind where you spend half a session debating the wording of a wish, and then the DM screws you over anyway because fuck you, that’s why. The kind you get if you give your players each a Deck of Many Things and a pint glass full of vodka.

      In other words, it’s the good kind of D&D, and I’d recommend it before anything else on that list.

      • Nornagest says:

        For what it’s worth, my core RP setting-development reading list goes something like this:

        The Dying Earth, Vance.

        The Night Land, William Hope Hodgson.

        The Book of the New Sun, Gene Wolfe. Add the Latro books (Soldier of the Mist, Soldier of Arete, and Soldier of Sidon) if your setting’s pre-medieval or a Conan-esque mashup.

        Zothique, Clark Ashton Smith.

        Viriconium, M. John Harrison.

        Nausicaa of the Valley of Wind (manga version), Hayao Miyazaki.

        And as much mythology, medieval studies, and primary sources as you can stomach. Skip summaries and adaptations. You’re looking for blood, betrayal, legal and spiritual weirdness, and dick jokes.

        If you’re doing an urban-heavy campaign, add Perdido Street Station and The Scar, China Mieville. Skim anything that smells political, but read any page with a monster on it twice.

        If you’re doing an action-heavy campaign, add Berserk (manga version), Kentaro Miura. The Golden Age arc is required reading; everything before and after is optional.

        I omit Kill Six Billion Demons (webcomic, Tom Parkinson-Morgan) only because it isn’t finished.

      • Le Maistre Chat says:

        The Dying Earth is weird D&D. It’s the kind of D&D that has crashed spaceships and walking fungus and monsters based off a half-melted pack of knockoff dinosaur figures, which you fight by upending a levitated blanket full of caltrops and lamp oil over the horde before finishing the stragglers off with a ridiculously long-winded spell. … In other words, it’s the good kind of D&D, and I’d recommend it before anything else on that list.

        The Dying Earth is so distinctly its own thing that it’s hard to combine with other D&D influences like Tolkien or Greek mythology.
        It’s Dying Earth if human settlements have existed so long that predators have evolved to disguise themselves as floors, ceilings, storage chests, coins… it’s not Dying Earth if you’re on a quest with an elf and a dwarf and fight stock monsters like dragons or centaurs.
        At most, you can combine the style of Dying Earth with Robert E. Howard and Clark Ashton Smith, who never ever used elves & dwarves and almost never had the protagonist encounter stock monsters[1].

        [1] The Hour of the Dragon has a rare exception where, thousands of years before Slavs even existed, Conan meets a vampire and his reaction is to bonk her on the head and say “Damn vampire!” like this exact type of undead is just Tuesday for him.

      • sfoil says:

        If you’re into Dying Earth and aren’t put off by BotNS or The Night Land, you should look into Ernst Junger’s Eumeswil. It’s awfully heavy on authorial rumination vs sightseeing, but still a worthwhile entry into the setting, and rarely advertised as such.

    • Le Maistre Chat says:

      The Broken Sword, and Three Hearts and Three Lions by Poul Anderson
      A Princess of Mars by Edgar Rice Burroughs
      Children’s and Household Tales by Jakob and Wilhelm Grimm

      Poul Anderson and fairy tales are definitely primordial influences on D&D. Western fairy tales are set in the Christian era, and that was also always Anderson’s modus operandi in fantasy. At some point Clerics were explicitly Christian (“holy symbol” was “cross”), though as that never sat well with the whole “secondary world” idea, Gary Gygax’s home campaign world of Greyhawk ended up with a pantheon of made-up gods. And the Paladin class is straight-up Holger Danske as interpreted by Anderson. Gaining powers by “being Lawful Good” was originally an agnostic engineer from our world becoming Catholic for protection from the monsters of Chaos (including his would-be-lover-had-he-not-amnesia!) that threaten a world based on the romances of Charlemagne.

      A Princess of Mars is good clean-limbed fun, but shares little with the tone or structure of D&D, lacking magic and the complementary party roles it’s tied into. John Carter is a Marty Stu who’s better than everyone else at the things that make a hero.
      Funny enough, Burroughs’s Mars blatantly made it into the DNA of fantasy, most clearly in Star Wars and He-Man.

      “Red Nails”, and “The Tower of the Elephant” by Robert E. Howard

      Pretty much any Conan story, whether written by Howard or not, and also Howard’s King Kull yarns.
      Howard’s is a brutal world that de-emphasizes objective morality and has wicked mages filling the social function of clerics. Howard’s original short stories usually follow a formula of having the protagonist meet one supernatural entity, which can be a valid and interesting way to structure a session. 8th level (“superhero”) and 9th level (“lord”) Fighters were explicitly modeled on Conan in his prime and Conan when he was a king. While Conan is a better fighter than anyone else, he does have adventuring companions flavorful enough to be PCs in their own right: Taurus the thief (dies), his lover Belit the pirate (dies), Valeria of the Red Brotherhood, and the wizard Pelias. It looks like a high-lethality campaign (don’t expect 4th or 5th Edition to give you Conan’s tone) where Howard’s PC got to be the best through consistent attendance and luck.

      “Jewels in the Forest”/”Two Sought Adventure”, “Ill Met in Lankhmar”, and “The Lords of Quarmall” by Fritz Leiber

      Well, yes.

      Elric of Melniboné, Knight of Swords, Stormbringer, “The Dreaming City”, and “While the Gods Laugh” by Michael Moorcock,

      I haven’t read any Elric stories yet.

      “Black God’s Kiss, and “Hellsgarde” by C. L. Moore

      Jirel of Joiry, a female feudal lord who encounters Lovecraft gods in medieval France? That fits.

      “Not Long Before The End”, “What Good is a Glass Dagger?”, and The Magic Goes Away by Larry Niven

      The beginnings of caster supremacy, and also the reason the fantasy genre uses the Polynesian word “mana” so much.
      Niven’s fantasy is set on prehistoric Earth, which I approve of. I don’t see how you’d use it as an inspiration for D&D without throwing out the official magic system in favor of the world having a non-renewable supply of mana that mages tap, though.

      Gods and Heroes of Ancient Greece by Gustav Schwab

      I just used Ovid’s Metamorphoses and the Bibliotheca of Pseudo-Apollodorus for my first Greek campaign. Currently I’m on the late epic poem Dionysiaca.

      The Hobbit by J.R.R. Tolkien

      Obviously the biggest influence on D&D (along with Lord of the Rings). Without this, you wouldn’t have baked in the assumption of fixed adventuring parties of more than two people (eg the Faphrd & Grey Mouser story literally titled “Two Sought Adventure”) and consisting of humans, elves, and dwarves. Tolkien’s influence has only grown stronger since the days of AD&D 1st Edition and Basic/Expert, as open tables have given way to fixed groups and it’s become practically impossible for PCs to die.

      First Men in the Moon, and “The Time Machine” by H.G. Wells

      What for?

      Dilvish The Damned by Roger Zelazny

      Never heard of it.

      • sfoil says:

        uses the Polynesian word “mana” so much.

        Wait, what? I thought that was derived from the “manna” of the Hebrew Exodus.

      • Nick says:

        I just used Ovid’s Metamorphoses and the Bibliotheca of Pseudo-Apollodorus for my first Greek campaign. Currently I’m on the late epic poem Dionysiaca.

        And now our party is almost to India!

      • Plumber says:

        @Le Maistre Chat says:

        “…I haven’t read any Elric stories yet…”

         ¿Qué es un ‘edgelord’? 

        “For ten thousand years did the Bright Empire of Melniboné flourish – ruling the world. Ten thousand years before history was recorded – or ten thousand years after history had ceased to be chronicled. For that span of time, reckon it how you will, the Bright Empire had thrived. Be hopeful, if you like, and think of the dreadful past the Earth has known, or brood upon the future. But if you would believe the unholy truth – then Time is an agony of Now, and so it will always be.

        Ravaged, at last, by the formless terror called Time, Melniboné fell and newer nations succeeded her: Ilmiora, Sheegoth, Maidakh, S’aaleem. Then memory began: Ur, India, China, Egypt, Assyria, Persia, Greece, and Rome – all these came after Melniboné. But none lasted ten thousand years.

        And none dealt in the terrible mysteries, the secret sorceries of old Melniboné. None used such power or knew how. Only Melniboné ruled the Earth for one hundred centuries – and then she, shaken by the casting of frightful runes, attacked by powers greater than men, powers who decided that Melniboné’s span of ruling had been overlong —- then she crumbled and her sons were scattered. They became wanderers across an Earth which hated and feared them, siring few offspring, slowly dying, slowly forgetting the secrets of their mighty ancestors. Such a one was the cynical, laughing Elric, a man of bitter brooding and gusty humour, proud prince of ruins, lord of a lost and humbled people; last son of Melniboné’s sundered line of kings.

        Elric, the moody-eyed wanderer – a lonely man who fought a world, living by his wits and his runesword Stormbringer. Elric, last lord of Melniboné, last worshipper of its grotesque and beautiful gods – reckless reaver and cynical slayer – torn by great griefs and with knowledge locked in his skull which would turn lesser men to babbling idiots. Elric, moulder of madnesses, dabbler in wild delights…”

        – “The Dreaming City” by Michael Moorcock

        Not only have I finished some deeply “Edgelord” novels, I have even re-read them, such as Michael Moorcock’s “Corum” series, which begins with

        A Knight of Swords (1971):

        “In those days there were oceans of light and cities in the skies and wild flying beasts of bronze. There were herds of crimson cattle that roared and were taller than castles. There were shrill, viridian things that haunted bleak rivers. It was a time of gods, manifesting themselves upon our world in all her aspects; a time of giants who walked on water; of mindless sprites and misshapen creatures who could be summoned by an ill-considered thought but driven away only on pain of some fearful sacrifice; of magics, phantasms, unstable nature, impossible events, insane paradoxes, dreams come true, dreams gone awry, of nightmares assuming reality.

        It was a rich time and a dark time. The time of the Sword Rulers. The time when the Vadhagh and the Nhadragh, age-old enemies, -were dying. The time when Man, the slave of fear, was emerging, unaware that much of the terror he experienced was the result of nothing else but the fact that he, himself, had come into existence. It was one of many ironies connected with Man, who, in those days, called his race “Mabden.”

        The Mabden lived brief lives and bred prodigiously. Within a few centuries they rose to dominate the westerly continent on which they had evolved. Superstition stopped them from sending many of their ships toward Vadhagh and Nhadragh lands for another century or two, but gradually they gained courage when no resistance was offered. They began to feel jealous of the older races; they began to feel malicious.

        The Vadhagh and the Nhadragh were not aware of this. They had dwelt a million or more years upon the planet which now, at last, seemed at rest. They knew of the Mabden but considered them not greatly different from other beasts. Though continuing to indulge their traditional hatreds of one another, the Vadhagh and the Nhadragh spent their long hours in considering abstractions, in the creation of works of art and the like. Rational, sophisticated, at one with themselves, these older races were unable to believe in the changes that had come. Thus, as it almost always is, they ignored the signs.

        There was no exchange of knowledge between the two ancient enemies, even though they had fought their last battle many centuries before.

        The Vadhagh lived in family groups occupying isolated castles scattered across a continent called by them Bro-an-Vadhagh. There was scarcely any communication between these families, for the Vadhagh had long since lost the impulse to travel. The Nhadragh lived in their cities built on the islands in the seas to the northwest of Bro-an-Vadhagh. They, also, had little contact, even with their closest kin. Both races reckoned themselves invulnerable. Both were wrong.

        Upstart Man was beginning to breed and spread like a pestilence across the world. This pestilence struck down the old races wherever it touched them. And it was not only death that Man brought, but terror, too. Willfully, he made of the older world nothing but ruins and bones. Unwittingly, he brought psychic and supernatural disruption of a magnitude which even the Great Old Gods failed to comprehend.

        And the Great Old Gods began to know Fear. And Man, slave of fear, arrogant in his ignorance, continued his stumbling progress. He was blind to the huge disruptions aroused by his apparently petty ambitions. As well, Man was deficient in sensitivity, had no awareness of the multitude of dimensions that filled the universe, each plane intersecting with several others. Not so the Vadhagh nor the Nhadragh, who had known what it was to move at will between the dimensions they termed the Five Planes. They had glimpsed and understood the nature of the many planes, other than the Five, through which the Earth moved.

        Therefore it seemed a dreadful injustice that these wise races should perish at the hands of creatures who were still little more than animals. It was as if vultures feasted on and squabbled over the paralyzed body of the youthful poet who could only stare at them with puzzled eyes as they slowly robbed him of an exquisite existence they would never appreciate, never know they were taking.

        “If they valued what they stole, if they knew what they were destroying,” says the old Vadhagh in the story, “The Onfy Autumn Flower,” “then I would be consoled.”

        It was unjust.

        By creating Man, the universe had betrayed the old races.

        But it was a perpetual and familiar injustice. The sentient may perceive and love the universe, but the universe cannot perceive and love the sentient. The universe sees no distinction between the multitude of creatures and elements which comprise it. All are equal. None is favored. The universe, equipped with nothing but the materials and the power of creation, continues to create: something of this, something of that. It cannot control what it creates and it cannot, it seems, be controled by its creations (though a few might deceive themselves otherwise). Those who curse the workings of the universe curse that which is deaf. Those who strike out at those workings fight that which is inviolate. Those who shake their fists, shake their fists at blind stars.

        But this does not mean that there are some who will not try to do battle with and destroy the invulnerable.

        There will always be such beings, sometimes beings of great wisdom, who cannot bear to believe in an insouciant universe.

        Prince Corum Jhaelen Irsei was one of these. Perhaps the last of the Vadhagh race, he was sometimes known as The Prince in the Scarlet Robe.

        This chronicle concerns him…”

        And the single fantasy novel that I’ve most re-read,

        The War Hound and the World’s Pain (1981), 

        also by Michael Moorcock, began:

        “It was in that yeat when the fashion in cruelty demanded not only the crucifiction of peasant children, but a similar fate for their pets, that I first met Lucifer.
        Until May of 1631 I had commanded a troop of irregular infantry, mainly Poles, Swedes, and Scots. We had taken part in the destruction and looting of the city if Magdeburg, having somehow found ourselves in the army of the Catholic forces under Count Johann Tzerclaes Tilly. Wind-borne gunpowder had turned the city into one huge keg and she had gone up all of a piece, driving us out with little booty to show for our hard work.
        Disappointed and belligerent, Westfield by the business of rapine and slaughter, quarreling over what pathetic bits of goods they had managed to pull from the blazing houses, my men elected to split away from Tilly’s forces. His had been a singularly ill-fed and badly equipped army, victim to the pride of bickering allied. It was a relief to leave it behind us.
        We struck south into the foothills of the Hartz Mountains, intending to rest. However, it soon became evident to me that some of my men had contracted the Plague, and I deemed it wise, therefore, to saddle my horse quietly one night and, taking what food there was, continue my journey alone. 
        Having deserted my men, I was not yet green from the presence of death or desolation. The world was in agony and shrieked its pain…”

        And it goes on for pages with a backdrop of the 30 years war (1618 to 1648), befofe getting to the supernatural elements, and it’s not hard for a reader to think of 20th century parallels, but as much as I mock the like now, as a teenager, I ate it up.

        I had actually first read the sequel, 1986’s

        City in the Autumn Stars

        which has the French reign of terror as a background, the protaganist of which, a descendent of the War Hound, is initially more of a hero that inspires sympathy than the War Hound.

        Anyway, after pages of character building biography, the protaganist of The War Hound and the World’s Pain, Graf Ulrich von Bek, enters “…the oak groves of the northern fringes of the great Thuringian Forest…”, and finds that “…the deeper into the forest I moved the less life I discovered…”, and“…Through the treetops I saw clear blue sky, and sunlight warned the glades. But insects danced in the beans; no bees crawled upon the leaves of the wild flowers; not even an earthworm twisted about the roots, though the soil was dark and smelled fertile”, until “..breaking out of the forest proper one afternoon, I saw before me a green, flowery hill which was crowned by the most beautiful castle I had ever beheld”, he wonders “How could a building demand calm, to the degree that not even a mosquito would dare disturb it?”, but while “It was my first impulse to avoid the castle, but my pride overcame me”, and “I refused to believe that there was anything genuinely mysterious…”

        It’s not much of a spoiler to say that what starts a historical tale of a man who’s become evil in an evil time of war becomes a fantasy of redemption, and it’s the fantasy novel that, perhaps to my shame, I’ve probably re-read the most, even more than Anderson, Leiber, Niven and Tolkien

      • Nornagest says:

        Pretty much any Conan story, whether written by Howard or not

        Sorry, but I can’t go to bat for this. The Carter/de Camp Conan stories are pretty decent, but the series goes into steep decline after that — it wouldn’t be too far wrong to call the Robert Jordan Conan stories discount Gor (a description I use advisedly). No disrespect to Jordan — his work on Wheel of Time, while flawed, is more than acceptable– but he wasn’t a mature writer back then.

        • Le Maistre Chat says:

          Sorry, but I can’t go to bat for this. The Carter/de Camp Conan stories are pretty decent, but the series goes into steep decline after that

          Mea culpa! I should have limited my statement to those two.
          When D&D was being created, the Conan stories available were the Lancer/Ace series of 8 anthologies mixing Howard’s short stories with de Camp/Carter’s plus the earliest Conan novels (Howard’s The Hour of the Dragon, Björn Nyberg’s 1957 The Return of Conan & de Camp/Carter’s novel Conan of the Isles) And when you buy the commonly-available Howard-only collections now, you miss some decent and very D&D-esque stories, like the one where Conan isn’t yet mighty enough to hack a Giant Slug to death in a straight fight so he pushes a loose block off the structure it’s chasing him through to kill it, or the one where he’s pinned to a magnetic obelisk by his armor and has to strip down to Armor Class 10 to fight the ooze that lives around it.

          After that series wrapped up at the beginning of the ’70s, there was a gap until 1977-82, when de Camp, Carter and the first new Conan pasticheurs started publishing at both Ace and Bantam. These included two pretty good authors whose entries I haven’t read to vouch for (Poul Anderson and Karl Edward Wagner) and a young Conan trilogy by Andrew Offutt that I can’t remember any details about: probably the beginning of the end, then.
          By that time, D&D was well-established and only took inspiration from itself for new material. 😛

    • Nancy Lebovitz says:

      Is anyone doing anything with The Book of the New Sun?

      • Nornagest says:

        I’ve never played it, but Monte Cook’s “Numenera” looks to have some New Sun in its DNA. I hear it’s getting backported to D&D 5E at some point.

        (And then there’s my own perpetually-in-limbo campaign setting.)

  48. Enkidum says:

    In no particular order…

    Fury Road
    Knives Out
    Into the Spider-verse
    Arrival
    Blade Runner 2049
    Get Out
    Driver
    Thor Ragnarok
    Manchester by the Sea
    Whiplash
    The Irishman
    Muscle Shoals

    I’m sure there are others I can’t remember right now that belong there. Looking at lists of the decade, there’s a lot of great films that I intended to see that I never got around to.

  49. GearRatio says:

    I tried to get through the Yale guy’s post, but, wow, is he not credible in the slightest. Just taking it from the top, he says:

    1. A cheap, relatively safe surgery could prevent penile cancer. But, we don’t need to do this; we could replace it with creams, physical therapy and much more expensive/rare plastic surgery that basically everyone in the third world and basically nobody in the third would have access to, ever.

    2.

    By contrast, there is *no* good-quality evidence that non-consensual, neonatal circumcision in developed countries with radically different disease transmission patterns (plus greater access to effective pre-exposure prophylaxis etc.) has any protective effect against HIV.

    There’s no evidence of any difference, of course, and every reason to believe there wouldn’t be one. He then quotes this, straight-faced:

    5% of new infections — may have been prevented if the men had been circumcised. If we then factor in the number of men who are circumcised when they are infected (approximately 70-80% of American men are already circumcised),

    This is an argument that an intervention that prevents 5% of HIV transmissions should be ignored, because, get this, some people already have the intervention. Remember, guys: quitting smoking drastically decreases the chances of lung cancer, but only about 20% of people smoke; we should ignore the drastic decrease in risk quitting smoking gives in those that do because there are already a lot of far safer people who don’t smoke.

    3.

    He references a study in which female acquisition of HIV was higher in the partners of recently circumcised men than in those who were partners of uncircumcised men. He uses this as counterbalance to the growing body of evidence that HIV is less easily passed to a circumcised man. He then uses this adult-circumcision intervention as evidence in his anti-infant-circumcision argument.

    Which is fine, so long as this adult-intervention isn’t fundamentally different from infant circumcision in a way that makes infant circumcision immune from this effect, which would work against his argument. It would be bad if the study in fact indicated this and he ignored it. From the study:

    The rate of female acquisition of HIV in couples from the control group was similar to that for couples in the intervention group who delayed resumption of sex but significantly lower than the rate in
    couples who resumed sex early.

    This section is addressing that having an open, potentially bleeding wound on a penis might affect HIV transmission rates. Turns out that when they controlled for this, the effect disappeared. For those of us keeping track, this means that infant circumcision would, according to the best data we have, result in much-reduced HIV Male-to-Male and Female-to-Male transmission with Male-to-Female remaining unchanged, as would any circumcision that occurred before sexual debut or any circumcision after sexual debut if we devised a way to successfully ensure dudes waited a few weeks to heal before they started having sex again.

    He read this study and had access to this information, but he didn’t like it and ignored it.

    4.

    Regarding other STIs, the authors of the post note that the data are even less compelling than for HIV, so I won’t take up space going into the evidence in detail. But even one finds the HIV/STI data compelling, these risks can be more effectively reduced without surgery, and the risk-reduction would not apply until sexual debut, which is typically at, near, or after the time at which the vast majority of individuals could consent to circumcision if that is what they wanted.

    “Other changes exist that have a greater effect” /= “It is false that this change has a separate, passive lifelong effect that works in synergy with those changes rather than you having to make an A/B choice between them” for starters.

    Earp is on the record being a proponent of the HPV vaccine; does he think we should delay this intervention until just before sexual debut, leave it up the the individual and hope they want to/remember to/have the resources to get it at that point? This is relevant if you don’t, as he does, take circumcision fully as an evil to be excised from step 1.

    5.

    Earp plays suuuuper fast and loose with how important he finds absolute risk. Take these two quotes:

    But even if one takes such figures for granted, the difference between absolute and relative risk needs to be highlighted. According to recent estimates from the American Academy of Pediatrics (AAP), UTIs affect roughly 1% of boys in the first two years of life regardless of circumcision status, suggesting that “the number needed to circumcise to prevent [one] UTI is approximately 100.”

    OK, so 100 circumcisions per bad avoided outcome is insignificant and shouldn’t drive circumcisions. Cool.

    In particular, fully 100% of glans injuries and 90% of fistulas reported through their Notable Adverse Events (NAE) reporting system occurred in children circumcised at ages 14 and younger. And the overall ratio NAEs to circumcisions performed was five times higher in infants (15.3 per 100,000) than in young adolescents (2.9 per 100,000), which in turn was about twice the ratio for individuals circumcised at ages 15 years and above (1.6 per 100,000).

    But wait! 7,142 non-circumcisions is needed to avoid one bad outcome here, and this is super significant and a reason to not circumcise infants.

    This is not the work of an intellectually honest contributor.

    Conclusion:

    I’m not sure why anybody would take this person seriously. I have the advantage of not knowing who he is; maybe he’s super famous so it makes it easier to gloss over cherry-picking and double-standards? If he had come out and said “I don’t like infant circumcision so I’m unwilling to seriously address any argument that it’s beneficial” I’d at least respect his honesty. But this isn’t serious work from an honest person and it shouldn’t be treated as that.

    • eric23 says:

      I had similar thoughts when reading his comment. I don’t think he is stupid or intentionally dishonest, but he is very partial to one side and that affects the quality of his argument.

      I wouldn’t take his argument at face value, but I would like to see an adversarial collaboration with him as one side.

      • GearRatio says:

        Assuming his dishonesty is unintentional I would judge him more harshly for it than I’d judge basically anyone here doing it intentionally, because he’s made this/is trying to make this his life’s work. He’s “supposed to” know in a way you and I aren’t, because we aren’t presenting ourselves as experts and trying to influence policy.

        Take the “circumcisions give women HIV” study mentioned. If you or I was to miss that the authors buried the whole “our data indicates this is entirely a problem with having an open wound on the penis of a man who is sexually active, has HIV and has a current partner” bit, it would be forgivable; we have other jobs, limited time and don’t make any real substantial difference to the discussion anyway. Us not reading the study and getting it “wrong” is understandable, and it also doesn’t make a difference, really.

        But that’s not true with this guy. This is his life’s work; it’s what he does. He’s fully expected by his audience to have read this and every relevant circumcision study carefully and to know what’s in them. If you or I had to, we could read every relevant circumcision study carefully and know the literature in, say, two weeks pretty easily and keep up on the new ones with no more than several hours effort each month. We don’t, but we also don’t make our living presenting ourselves as experts here.

        So when I hear “this wasn’t intentional dishonesty” I immediately think “It wasn’t? That’s horrifying” because it implies one or more of the following:

        1. An expert on the subject who is trying to influence policy didn’t read an important study all the way through, even though there’s a relatively limited amount of studies on the subject to read in the first place. This means he’s lazy and intentionally uninformed in a situation where he’s ethically obligated to be well informed.

        2. An expert on the subject who is trying to influence policy read the study and didn’t understand it.

        3. An expert on the subject is so hopelessly biased to one side of the conversation that he read a study and is capable of understanding it, but isn’t able to absorb information contrary to his preferred conclusion.

        None of these is intentional dishonesty, but they are all as bad as it; they all come down to “this person shouldn’t be trusted as a source of information”.

        I don’t know how to send a scalpel away to be autoclaved, and I don’t know the correct way to apply a local anesthetic to a toe so I can surgically correct an ingrown toenail. This isn’t morally relevant, though, because I don’t perform surgical interventions on feet. But if I was a podiatrist, this would be a really, really big ethical lapse. I would be supposed to know; I would be representing myself as knowing to unwitting people who assumed I knew what I was doing and what I was talking about.

        Thus it is here: we have an expert who can’t or won’t be bothered to be honest and accurate, who hand-waves away all benefits while amplifying all detriments as much as possible. That’s a dangerous, bad thing whether it’s intentional or not.

    • The Big Red Scary says:

      Contracting HIV is like being run over by a bus. Look both ways before crossing the street! Unless you have reason to expect your son to engage in risky behavior (for example, you yourself and your social circle engage in risky behavior), then increased risk of HIV transmission is an almost irrelevant consideration when deciding not to circumcise your son.

      • Frederic Mari says:

        Indeed.

        And, tbf, even if my son was circumcised, I would still expect him to wear a condom with short term lovers of either sex/gender/whatever.

        But the whole discussion didn’t address one key point – What do (hetero) women prefer?

        • Aapje says:

          Chocolate…or so I’m told.

        • Act_II says:

          But the whole discussion didn’t address one key point – What do (hetero) women prefer?

          This question may be central to some people’s insecurities, but it isn’t relevant to the ethical discussion. What matters here is, you know, people who actually have penises.

          You could maybe make the argument that it affects the ability to find a partner, but since that’s the sort of thing that you generally find out during the deed itself (i.e. after a partner has already been found), I’m skeptical. Unless you’re wandering the streets shouting your genital status to the world, I guess.

          (Editing to add another thought — I suspect there isn’t even a real answer here. My prior is that “no preference” or “weak preference for the more familiar version” is going to be the majority of responses, and the details are going to vary from person to person.)

          • Randy M says:

            Unless you’re wandering the streets shouting your genital status to the world, I guess.

            Surely that’s what Twitter is for?

          • woah77 says:

            While I agree that such is how Twitter is used, I should hope that’s not what we intend Twitter for. The intended use should obviously be following people who offend you so greatly you simply must tell everyone you know.

          • Frederic Mari says:

            This question may be central to some people’s insecurities, but it isn’t relevant to the ethical discussion. What matters here is, you know, people who actually have penises.

            I disagree. The answer to circumcision from an ethical and health pov seems “meh, do want you want, it makes little to no difference”.

            Thus, aesthetic aspects become relevant.

            Unless you’re wandering the streets shouting your genital status to the world, I guess

            1- Even if it takes waiting for the deed to be done for the girl to find out, she can… you know, walk away eventually? It wouldn’t be the first time I’ve heard of a woman living a relationship b/c she was unhappy with *something* with her partner’s genitals. Too big, too small, too weird etc.

            2- I didn’t know that Twitter was meant to be used to shout your genital status to the world or not but surely Tinder is?

        • DragonMilk says:

          Pretty sure that it is common knowledge that in the the US, girls in their 20s typically prefer unsheathed

        • Thegnskald says:

          You think you know fashion, well fashion’s a stranger
          You think fashion’s your friend, my friend
          Fashion is danger

          Making judgments about what will be aesthetic in twenty years is… not the best way to make any decisions whatsoever.

      • Deiseach says:

        Unless you have reason to expect your son to engage in risky behavior

        If you have a child now, you have no idea what, in 15-18 years time, will be ‘risky behaviour’ when it comes to sex.

        Once upon a time there wasn’t AIDS in the West, until there was. STIs even seemed to be on the way out – there was a pill for every ill, you got a dose and went to the doctor and then continued in the happy trend of the Sexual Liberation once the rash cleared up.

        Once people stopped dying, ‘safer sex’ slipped off the top of the charts – see the resurgence in STIs and new, exotic, strains becoming more common.

        Sexual practices such as heterosexual buggery have now become normalised, if not exactly mainstream; what will be the Hot New Sex Trend in 15 years time that today is something niche in porn? Something risky that “I’m sure my kid will never do something like that, so why would I need to warn about it/do something so drastic as circumcision”? Because it’s gonna happen.

        So simply telling kids “just be careful!” has never cut it – hormones and ‘parents don’t understand what it’s like’ have always combined to have young people acting in risky ways.

        • Randy M says:

          If you have a child now, you have no idea what, in 15-18 years time, will be ‘risky behaviour’ when it comes to sex.

          But if you take this thought seriously, you probably don’t bother to circumcise for STD protection.

      • John Schilling says:

        Unless you have reason to expect your son to engage in risky behavior (for example, you yourself and your social circle engage in risky behavior),

        Do you expect your son to at any point in his life be a teenage boy? A young adult off at college and away from adult supervision for the first time? There you go.

        This strikes me as the flip side of the movement in some conservative or Christian circles to oppose default HPV vaccination for pre-teen girls because, if they are Good Girls(tm), they will never need it, and who are you to tell us that our daughters won’t be Good Girls? Circumcision is probably a more significant body modification than vaccination, which may tilt the balance in this specific case, but in general not protecting your children against high-risk behavior because you don’t expect them to be anything but Good Boys and Girls strikes me as dubious parenting.

        • Randy M says:

          Depends on whether seat-belts really do increase accidents.
          Although that analogy doesn’t hold if the child isn’t aware of this benefit of circumcision/vaccination.

        • thisheavenlyconjugation says:

          It seems vastly more reasonable to expect your teen son to use condoms rather than abstain from sex. There aren’t millions of years of selection encouraging him against the former.

          • GearRatio says:

            Aren’t there usually, though? Most men don’t like condoms much; it’s not like that tendency came from fairies or something.

          • thisheavenlyconjugation says:

            @GearRatio
            No. Preferences against condoms are incidental, not the result of optimisation. And they’re vastly different in strength. A preference against condoms is usually approximately on the level of preferences for certain kinds of food; the preference for having sex is approximately the strongest one there is after desires to avoid death, torture and harm to love ones (and sometimes it manages to beat even those).

    • viVI_IViv says:

      In my understanding penile cancer is very rare and usually caused by HPV, which is sexually transmissible, as is HIV. Since infants don’t have sex, there is no need to perform a non-consensual invasive and irreversible body modification on them for a small potential benefit which won’t manifest until much later. If they want that benefit when they are able to consent to sex, they can get it at that time.

      • EchoChaos says:

        If they want that benefit when they are able to consent to sex, they can get it at that time.

        It is impossible to be circumcised as an infant when you are able to consent.

        You can be circumcised as an adult, but that is a different and more difficult procedure.

        • viVI_IViv says:

          You can be circumcised as an adult, but that is a different and more difficult procedure.

          The doctor in the referenced comment claims that the opposite is true because the infant penis is small and the foreskin is not retractable.

          • acymetric says:

            Just for the sake of accuracy/clarity, Brian Earp not a doctor. He is working on a Ph. D. in philosophy/psychology (but even upon completion he won’t be a medical doctor, which the phrasing/context here kind of implies).

          • EchoChaos says:

            @viVI_IViv

            I have actual practical experience with adult circumcision myself and infant circumcision for my two sons.

            Infant circumcision is FAR easier, has lower recovery time and less difficulty.

          • viVI_IViv says:

            Just for the sake of accuracy/clarity, Brian Earp not a doctor.

            Ok, I thought he was a medical doctor.

          • gbdub says:

            @EchoChaos – is that because infant circumcision is actually less dangerous, or because infants just lay around all day, can’t express “man my dick really hurts”, and don’t get too horny to abstain for the length of recovery period?

          • EchoChaos says:

            @gbdub

            is that because infant circumcision is actually less dangerous, or because infants just lay around all day, can’t express “man my dick really hurts”, and don’t get too horny to abstain for the length of recovery period?

            Infants express “man my X really hurts” INCREDIBLY WELL. Infants are machines for communicating exactly what is creating discomfort for them in exactly the ways a mother needs to hear.

            Not getting horny is another good reason not to delay it, yes.

            Another is that if you would prefer to be circumcised as an adult there is an impact on your life in terms of both finances and recovery time from lost labor. You can’t retroactively be circumcised as an infant if you’d prefer that as an adult.

          • Thegnskald says:

            Compared to reversing, which is… oh wait, you can’t regrow those nerves. So why is your satisfiable preference relevant, given that it requires making somebody else’s preference unsatisfiable?

            Also: You made a sacrifice to your god. Your sons didn’t, and can’t.

          • EchoChaos says:

            @Thegnskald

            Compared to reversing, which is… oh wait, you can’t regrow those nerves.

            Sure, but why does that matter? Once more, circumcised men are not missing out on some mystical sexual experience. Study after study has proved that my anecdotal experience is correct. Circumcision doesn’t affect sexual pleasure at all.

            So why is your satisfiable preference relevant, given that it requires making somebody else’s preference unsatisfiable?

            Because mine actually had an impact on my life. Being circumcised as a child has no negative effects at all unless you are a rare case that experienced a complication.

            Also: You made a sacrifice to your god. Your sons didn’t, and can’t.

            They aren’t meant to. I am commanded to circumcise my sons on the eighth day. You are arguing that I should not be able to obey God’s commandment.

          • Thegnskald says:

            Sure, but why does that matter? Once more, circumcised men are not missing out on some mystical sexual experience. Study after study has proved that my anecdotal experience is correct. Circumcision doesn’t affect sexual pleasure at all.

            As a man who has had sex with men: It clearly does.

            But that doesn’t even matter. I’m not a utilitarian, I don’t give a shit whether sex is better cut or uncut. Pretty much the only thing that matters is choice.

            Because mine actually had an impact on my life. Being circumcised as a child has no negative effects at all unless you are a rare case that experienced a complication.

            Or if you just plain didn’t want it done. You wanted it done, great, you got it done. What you are demanding isn’t that you have the option to be circumcised, it is that other people don’t have an option not to be.

            They aren’t meant to. I am commanded to circumcise my sons on the eighth day. You are arguing that I should not be able to obey God’s commandment.

            Yes, I am. Because that isn’t your choice to make, and if your god cannot understand that, I’d find a better one. Acts of faith, like salvation, are personal, or else they mean nothing. You cannot trust on behalf of someone else.

            And if your god really wanted all people circumcised, I’d say maybe just don’t give them a foreskin to begin with.

          • EchoChaos says:

            EDIT: This has gotten too CW for the open thread. I’m removing the response and leaving it. If you want to resume this in a hidden thread, but I think I’ve said all I need to.

          • Thegnskald says:

            Alright, fair, my apologies.

            Let’s reframe this in different terms: You are forbidden from marking your body, yes? No tattoos?

            Suppose your parents belonged to a religion where infants are tattooed at birth, and followed their beliefs. Is this acceptable to you now as an adult who has found the true religion?

      • GearRatio says:

        Since infants don’t have sex, there is no need to perform a non-consensual invasive and irreversible body modification on them for a small potential benefit which won’t manifest until much later.

        You are mixing two arguments here – “we shouldn’t do invasive and irreversible body modifications” and “we shouldn’t do interventions with life-long benefits until just before they are needed”.

        If we take away the first argument as a given, the second argument is much more questionable. We give children the HPV vaccine starting from age 9, because we want to make sure they have it when they sexually debut. We don’t generally think it’s a good idea to postpone until, say, 14, which would greatly increase consent ability, because they might not have the time or resources to do it then, or simply might not remember to do so.

        On the flip side isolating the second argument doesn’t solve the first – you can still think infant circumcision is bad/not worth it/dangerous. But I think you do have to isolate the arguments here, because otherwise you are saying something like “Assuming you are with me in thinking this is an abomination, shouldn’t we hold it to a standard of benefit and timing we don’t hold similar interventions to?” and that’s not going to convince anyone who isn’t already in the tank.

        • eyeballfrog says:

          We’ve kind of already decided consent is unnecessary for vaccines, likely due to their high and well-established efficacy and minimal invasiveness. Given that circumcision is completely the opposite with regards to both variables, comparing the two seems irrelevant.

          • EchoChaos says:

            We’ve kind of already decided consent is unnecessary for vaccines

            Correct, which establishes that parents may ignore consent for minor children for medical procedures.

            likely due to their high and well-established efficacy and minimal invasiveness.

            Vaccine negative side effects exist as well, but we allow parents to judge whether or not they are worthwhile to apply to children, as we should.

            Given that circumcision is completely the opposite with regards to both variables, comparing the two seems irrelevant.

            In your opinion, but this is the exact sort of thing that we should let parents decide on. Is a 1% chance to avoid a negative outcome in exchange for a 1 in 7142 chance for a different negative outcome acceptable?

            Parents should absolutely decide this, and I am glad that the relevant literature exists.

          • GearRatio says:

            It’s not irrelevant, it’s necessary to respond to what he said, which was:

            1. Circumcision bad
            2. Delaying preventative interventions fine

            If it was “circumcisions bad, so delaying preventative interventions is bad but unpreventable” then, whatever. But that’s not what he said – he said something much closer to”circumcisions are bad and there’s no problem delaying this intervention”. And there are, of course, problems delaying interventions. Which is why I have to bring up HPV vaccines – because it’s a non-controversial STD prevention method in this forum.

            If his argument was instead “delaying this intervention is bad, but on balance circumcision is worse so we have to deal” then I wouldn’t have brought up vaccines because it wouldn’t be necessary.

        • viVI_IViv says:

          A non-consensual mutilation, especially of an erogenous zone, is prima facie an abomination, based on general principles of bodily integrity, bodily autonomy and sexual freedom.

          It might be considered acceptable if justified by strong and urgent medical reasons, but these reasons don’t exist in the case of infant circumcision.

          HPV vaccination is much less invasive and has gives higher medical benefits, making the analogy invalid.

          • EchoChaos says:

            A non-consensual mutilation, especially of an erogenous zone, is prima facie an abomination, based on general principles of bodily integrity, bodily autonomy and sexual freedom.

            Emotional words like this are why I say this is culture war.

          • viVI_IViv says:

            What do you find particularly emotional in my comment?

          • Dacyn says:

            @viVI_IViv: I see “mutilation”, “abomination”

          • Mark V Anderson says:

            What do you find particularly emotional in my comment?

            The entire first sentence, which is why Echo quoted it. And what Dacyn said. Echo is right, this discussion doesn’t belong here.

          • viVI_IViv says:

            “mutilation” is a descriptive term commonly used to describe similar practices on female genital organs. If the amputation of the clitoral hood can be described as female genital mutilation, then the amputation of the homologous penile foreskin can described as male genital mutilation.

            “abomination” is a moral judgment which I motivated in terms of the stated moral principles.

            It seems to me that people object the use of “mutilation” to describe circumcision because it immediately triggers cognitive dissonance with moral principles they hold in general, highlighting the fact that considering circumcision morally acceptable is an unprincipled exception.

          • albatross11 says:

            viVI_IViv:

            I’ll admit I find the idea that you can’t tell that “abomination” and “mutilation” are question-begging emotionally-loaded words extremely implausible.

            Try using them in a description of abortion or gender reassignment surgery and see whether it’s clearer.

          • EchoChaos says:

            @viVI_IViv

            Female genital mutilation is what the African and Islamic cultural practice is called precisely BECAUSE it’s an emotional word.

          • Dacyn says:

            @viVI_IViv: I don’t object to your language in the abstract, just agree with EchoChaos that it is one thing that indicates that this conversation probably should have been in a CW thread. I suppose it’s moot at this point.

    • EchoChaos says:

      Circumcision is a weirdly culture-war thing for a lot of people well beyond whether it’s rational or not.

      But in my experience the anti-circumcision side is always more likely to engage in such elision as is pointed out here.

      My conclusion tends to be about the same as @John Schilling in the linked thread:

      this should be up to the parents, and maybe moderately encouraged by pediatricians

      That such a moderate opinion is decried shows that the anti-circumcision folks aren’t fighting a technical battle but one tied up in identity, which is not surprising given the importance of the penis to basically all men, but should be viewed from that light.

      • ksdale says:

        Whether someone believes that that opinion is moderate depends on whether they think circumcision and not circumcision are basically equivalent to begin with.

        But if someone believes that bodily integrity is a value to be upheld for its own sake, then there is a presumption against circumcision that seems much less likely to be overcome by minor medical benefits.

        I think it’s perfectly reasonable to believe that bodily integrity shouldn’t matter, but I don’t think a person has to be blinded by identity to interpret the quoted opinion as something other than moderate.

        As an aside, my own personal opinion is that if someone discovered tomorrow that you could achieve medical benefits similar to circumcision by cutting off a similarly useful/useless piece of skin at birth, I think most people would look at the medical benefits and think, “No way I’m having a piece of my baby’s skin cut off for a benefit of that magnitude.” And I think it’s mostly that circumcision has been so very normalized that makes circumcision and not circumcision seem basically equivalent to begin with.

        I think the fact that so few circumcised men have a problem with it is an argument against valuing bodily integrity for its own sake… but I can’t shake the feeling that if no one were circumcised, we probably wouldn’t start doing it en masse for the medical benefits.

        • EchoChaos says:

          But if someone believes that bodily integrity is a value to be upheld for its own sake, then there is a presumption against circumcision that seems much less likely to be overcome by minor medical benefits.

          But we all accept that bodily integrity can be violated nearly at will by parents.

          Nobody (serious) complains when parents manipulate and violate their children’s bodily integrity by putting things in their anus to take temperature, for example.

          Yes, circumcision is permanent, but so can be injuries from incorrect usage of a rectal thermometer, but nobody (serious) is calling for banning that.

          Which brings back that the reason circumcision is such a big deal relative to everything else is the central nature of the penis to men. Which I agree with, by the way.

          But the anti-circumcision side tends to view THIS SPECIFIC bodily autonomy as something a parent can’t rationally override unlike everything else, where parents have large ability to override.

          but I can’t shake the feeling that if no one were circumcised, we probably wouldn’t start doing it en masse for the medical benefits.

          This isn’t true. It did exactly that in the English-speaking world in the late 1800s.

          • ksdale says:

            I feel like the rectal thermometer is not a great analogy precisely because it isn’t permanent… It seems easy to draw a line between things that are definitely completely permanent and things that could theoretically cause permanent damage, since most of parenting involves doing things which, if screwed up badly enough, could cause permanent damage. Plus I think a lot of people are actually quite uncomfortable with rectal thermometers, anyway.

            I think some people view this specific bodily autonomy as something a parent can’t rationally override because it is so irreversible.

            It seems like your argument is something like, “Parents are allowed to do all sorts of things to their children, so why draw the line at cutting off body parts.” And it seems natural to me to draw the line at not cutting off body parts.

            I don’t feel hypocritical for saying that parents should have the ability to make all sorts of decisions about their children’s bodies, but they should stop short of permanently removing parts, or at least only do so after long and careful deliberation…

            I feel like parents do not actually have a large ability to make permanent body modifications… like a parent who tattooed their children would probably lose custody of their children (at least in the US). Some people pierce their young daughters ears, but I think there is also some controversy around that practice as well.

            And the English-speaking world of the 1800’s also did a whole lot of other things to people’s bodies that we have since made illegal, so I’m not sure that proves anything.

          • meh says:

            @ksdale

            it seems natural to me to draw the line at not cutting off body parts.

            All criminals are bad

          • EchoChaos says:

            I think some people view this specific bodily autonomy as something a parent can’t rationally override because it is so irreversible.

            So? Lots of decisions parents make are irreversible. Including ones with substantially more impact than a neutral to mildly positive medical effect.

            It seems like your argument is something like, “Parents are allowed to do all sorts of things to their children, so why draw the line at cutting off body parts.” And it seems natural to me to draw the line at not cutting off body parts.

            It seems natural to me to draw a line in a place that allows something that the majority of Americans and the adherents of two of the great monotheistic religions to continue their cultural practices.

            Why does your natural trump mine, given that there are basically no negative side effects?

          • The Pachyderminator says:

            But we all accept that bodily integrity can be violated nearly at will by parents.

            This is a very strange thing to say, since “we all” clearly don’t accept anything of the sort. There are many permanent body modifications that have been practiced on children at various times (female circumcision, foot binding, castration, piercings, etc.), all of which are now strongly frowned upon by Western society. Male circumcision is currently the sole exception. You talk as if disapproving of circumcision is the anomaly in our conception of parental rights, but what’s anomalous is that anyone approves of it. It makes no sense to pretend that circumcision is no different from using a thermometer.

          • EchoChaos says:

            @The Pachyderminator

            This is a very strange thing to say, since “we all” clearly don’t accept anything of the sort.

            I suspect that’s not true. Virtually everyone here (including me!) supports non-emergency medical interventions that can leave permanent and visible damage on children because the benefits are net positive.

            There are many permanent body modifications that have been practiced on children at various times (female circumcision, foot binding, castration, piercings, etc.), all of which are now strongly frowned upon by Western society.

            Other than castration and piercing, none of those were done in the West, and culture has a lot to do with what we find acceptable. And since the French and English rejected castrati and were instrumental in its banning and America has substantial heritage from both, it’s no surprise that our reaction to it is negative.

            Piercing is probably the most similar to circumcision. Very small risks of something going wrong, irreversible change that has only cosmetic effects. That circumcision also has positive effects is great, and why instead of being treated like piercing (entirely left up to the parents how they want their child to look cosmetically), the medical establishment should mildly promote it.

          • The Pachyderminator says:

            @EchoChaos

            Virtually everyone here (including me!) supports non-emergency medical interventions that can leave permanent and visible damage on children because the benefits are net positive.

            What interventions are you thinking of? I wouldn’t expect to find consensus support for anything that involves permanent and visible damage except in cases of genuine medical necessity.

            Almost anything can cause permanent damage, of course, not just vaccines and thermometers but also things like feeding (choking hazard) and bathing (toddlers can easily drown in bathtubs). But you seem to be saying that the remote possibility of permanent damage occurring by accident in the course of necessary care justifies an intentionally permanent, unnecessary procedure. That makes no sense to me at all.

          • EchoChaos says:

            @The Pachyderminator

            That’s my point. The negative side effects pointed to by the anti-circumcision side occur at a lower rate than complications from procedures we all agree are fine not just to leave as parental discretion but to actually mandate.

            The only remaining objection to circumcision is that it permanently changes the appearance of the penis and a child may prefer to have a different appearance of the penis in the future.

            But given that the child may in fact prefer a circumcised penis and that infancy is the best time to perform it (trust me, it really sucks as an adult), leaving it as a parental decision makes sense.

            I’m not arguing for mandating circumcision, but I am strongly opposed to people who try to ban something strongly identified with American culture.

    • J Mann says:

      I was glad to hear from an expert, but I had some questions too. The section on sexual satisfaction research struck me as way too absolute.

      But even if one simply accepts the results from those studies as being of highest quality because of the large sample size and randomized design, this would not count in favor of non-consensual circumcision. This is because these studies concerned consensual, adult circumcision, not circumcision of newborns.

      So if well-designed studies of adult circumcision doesn’t show any change in sexual satisfaction, that doesn’t tell us anything about whether infant circumcision affects sexual satisfaction? That seems way too strong.

      In general, I would have been more comfortable with the sexual satisfaction section if it had just come out and said: “Yes, there are very well designed studies that provide evidence that sexual satisfaction is unaffected by male circumcision, but there still some reasons for concern.”

    • AlesZiegler says:

      I think that you are being somewhat uncharitable. Earp clearly states that he writes from a position of someone who sees bodily integrity as a value in itself.

      Under most conditions, cutting any person’s genitals without their informed consent is a serious violation of their right to bodily integrity. As such, it is morally impermissible unless the person is nonautonomous (incapable of consent) and the cutting is medically necessary

      From that perspective (which to be clear I do not share) only thing you need to demonstrate to declare infant circumcision impermissible is to prove that it is not “medically necessary”, which I think is reasonable to interpret as “as far as we know, net medical benefits of infant circumcision are low”. This I think both Earp and results of AC successfully demonstrate.

      In general I found both original results of AC immensely helpful in qualifying issues and arguments in circumcision debate. I.e. imho stakes are relatively low and government should not interfere with parental decisions in this area.

      • GearRatio says:

        I think that you are being somewhat uncharitable. Earp clearly states that he writes from a position of someone who sees bodily integrity as a value in itself.

        Let’s be clear: I’m being extremely uncharitable here, because he’s a guy who works in this field and attempts to shift policy. He presents himself as an expert. He’s not an individual poking around trying to figure something out; he’s a guy who makes his living studying this and trying to convince important people to eventually not allow anyone to do this. He’s in the position of power and responsibility and doesn’t deserve charity any more than Chase Manhattan bank does.

        From that perspective (which to be clear I do not share) only thing you need to demonstrate to declare infant circumcision impermissible is to prove that it is not “medically necessary”, which I think is reasonable to interpret as “as far as we know, net medical benefits of infant circumcision are low”. This I think both Earp and results of AC successfully demonstrate.

        here’s the thing, though: He doesn’t get to just declare that bodily autonomy is the end-all-be-all and that he doesn’t have to prove harms any more than I get to declare that cultural, religious and parental rights are the end-all-be-all and he has to prove gigantic harms before it’s OK to ban it.

        If I can demand, say, that the institution of traditional heterosexual marriage is the absolute value to be protected, then it’s really easy for me to strike down gay marriage all over the place without even trying even if I’m wrong – see, now they have to do all the work to prove my value is protected before theirs even gets a thought.
        Anyone would be stupid and weak to take those terms and they would be correct not to regardless of whether they ended up being right or wrong in the end. That’s not how conversations work, if they are conversations at all.

        In general I found both original results of AC immensely helpful in qualifying issues and arguments in circumcision debate. I.e. imho stakes are relatively low and government should not interfere with parental decisions in this area.

        This is me too, and I chose not to circumcise my kids as a white protestant from middle class culture, which at least where I live was/is pretty unusual. But he’s trying to raise the stakes – in his perfect world, if you circumcise your kids you go to jail for child abuse the same as you would if you gouged out their eye – that’s what he’s working towards, he just has to be slightly political while he does it. That’s different from most people on the circumcision side – there’s virtually no people demanding that non-circumcisions be illegal.

        So he doesn’t get to lie, misrepresent, minimize inappropriately, maximize inappropriately or be uninformed – he just doesn’t in his position unless everything is totally broken and everyone is so daunted by titles they just lay down and take it. And his values don’t get to automatically trump everyone else’s and their assumed rightness get to be the foundational stage for the discussion by default. Just as he can demand proof of a great benefit/harm ratio to be convinced (and, as shown in 3. above, ignore what evidence he does get), I can demand proof of a burdensome harm/benefit ratio before I’m convinced.

        At the very least, he needs to be at the level where some random a-hole on the internet can’t point out several serious instances dishonesty and inconsistencies in his arguments before he deserves to be taken seriously as an expert. I remain to be convinced he’s reached this threshold.

        • AlesZiegler says:

          If I can demand, say, that the institution of traditional heterosexual marriage is the absolute value to be protected, then it’s really easy for me to strike down gay marriage all over the place without even trying even if I’m wrong

          Well, imho many people do that. I do not find it weird that many other people choose “not cutting body parts without consent” as their absolute value.

          • GearRatio says:

            Well, yes, a lot of people do say “My position is the absolute good and you must bow to it!” or some form of that. The difference I’m pointing out is they then (usually) don’t and (always) shouldn’t get the benefit of people bowing to their proclamation.

            So when you say “well, you are being uncharitable – remember, this is his absolute value! He doesn’t have a high burden of proof under that umbrella” I will always respond “No, he has the same burden of proof everyone else does; if he lies, I’ll call him out on it, and if his arguments are shit, I’ll call him out on that.”.

            To put it another way: He’s allowed to have values and opinions, but this shouldn’t insulate him from criticism of shitty arguments and dishonesty in any way.

          • Dacyn says:

            There is a big difference between what you two are arguing about: @AlesZiegler is talking about whether Earp’s beliefs are reasonable, whereas @GearRatio is talking about whether Earp is debating reasonably against people who don’t hold his beliefs. (I didn’t follow the start of this argument well enough to know which of these is more relevant.)

          • AlesZiegler says:

            @Dacyn

            Perhaps, but it is also about something like what is the appropriate level of civility in discussion, I guess…

          • Dacyn says:

            @AlesZiegler: I’m confused, the only references to civility so far seem to be a couple of offhand references to lying (which I think isn’t really the central example of incivility) by GearRatio. Or did you mean that you should start talking about civility?

          • AlesZiegler says:

            @Dacyn

            I opened this exchange with accusation that GearRatio is being uncharitable. (Checks a dictionary). Civility means civilized conduct, especially: courtesy, politeness. Courtesy means behavior marked by polished manners or respect for others. So I thought that GR was being unfairly disrespectful to Brian Earp, i.e, uncharitable. But Then GR explained rather eloquently why he or she reacted in this, by his or her own admission uncharitable, way. So I moved on to not very productive nitpicking.

          • Dacyn says:

            @AlesZiegler: Ah, that makes sense, thanks for explaining.

    • Thegnskald says:

      But wait! 7,142 non-circumcisions is needed to avoid one bad outcome here, and this is super significant and a reason to not circumcise infants.

      You miss the point there: He is arguing circumcision is riskier as an infant, as a counterpoint to the arguments elsewhere that circumcision is less risky as an infant. Thus it is a reason to delay circumcision.

      Rates aren’t the only thing that matters, there is also severity. What percentage of UTIs match the severity of a fistula?

      • GearRatio says:

        I’m not missing the point, I’m saying he has an inconsistent standard. Absolute values are incredibly important in his arguments until there’s an absolutely tiny absolute value on something that supports the anti-circumcision position, then suddenly not important enough to mention at all. Also, he himself doesn’t really think severity matters much when it suits him; penile cancer is more severe than UTI and fistulas typically are, but it didn’t deter him from using absolute effect arguments to hand-wave that illness away.

        As for severity, go here and scroll down to complications – they aren’t uncommon or unserious, especially for people with limited access to medical care.

        • Thegnskald says:

          I’m not missing the point, I’m saying he has an inconsistent standard. Absolute values are incredibly important in his arguments until there’s an absolutely tiny absolute value on something that supports the anti-circumcision position, then suddenly not important enough to mention at all. Also, he himself doesn’t really think severity matters much when it suits him; penile cancer is more severe than UTI and fistulas typically are, but it didn’t deter him from using absolute effect arguments to hand-wave that illness away.

          – You do understand there is a difference between the kind of risk assessment you do when you are comparing two types of intervention, and whether to perform an intervention on somebody who can’t currently consent to it, yes?

          As for severity, go here and scroll down to complications – they aren’t uncommon or unserious, especially for people with limited access to medical care.

          Okay, done. So UTIs make adult hypertension less likely, given that the percentage of adults who experience hypertension is, according to a random Google search, 30%? What am I comparing this rate to?

          I can’t find any data on base rates of renal scarring. At first I thought this was kind of weird, until I found a paper arguing against testing infants for renal scarring based on VUR (basically, back-flow of urine). Apparently, whether or not to check for renal scarring based on the severity of VUR is a contentious point in the field of kidney health, so most if not all of the papers are about VUR. Which is to say, the paper about UTIs is citing information that isn’t actually measuring the risks of UTIs. (Yes, I’m including the two papers cited in the paper you linked. Neither has a goddamned control group.)

          When assessing risks, control groups are kind of important. I have no idea how the risks change.

          • GearRatio says:

            I mean, we could go around all day and have an argument about how much worse or better fistulas are than UTI’s you don’t have antibiotics to treat or penile cancer. But we find that when it helped his cause, he dismissed an absolute value 70 times larger than one he later held up as damning, important evidence.

            So for this to be credible and him to be even-handed, you are going to have to convince me that one non-descript catch-all condition that he held up as significant is somehow (glans injuries) fully 70 times worse than UTI plus however many times it needs to be over that to be mention-worthy by the same standard he hand-waved all the not-useful to him stuff. And I don’t buy it, because I’m not stupid.

            And even if I was stupid enough to buy that, I’m still not stupid enough to do buy it when the source telling me that is the same guy that’s in the next breath cherry-picking a study to make it appear to support his position when it supports the dead opposite.

            If I Occam this, I get these two options:

            Either this guy at some point sat down, did a bunch of research comparing specifically glans injuries versus UTI and was able to come the the conclusion that there was a significantly greater than 7000% difference in severity between them, taking into account reluctance to seek treatment, financial ability to, and availability of medicine to treat them in general both in the first and third world and soberly decided that both fistulas are just that much worse that they deserve mentioning and sober consideration where UTI and Penile cancer risks don’t but decided not to mention any of this even though doing so would greatly strengthen his case

            OR

            The guy who lies again in the same essay just tried to pull an additional fast one.

            If you are honestly asking me to take the first option here, I’m not really sure there’s a productive conversation to have between us.

          • Thegnskald says:

            I already gave you the explanation, you just weren’t paying attention.

            So let’s analogize.

            An uncircumcised thirty year old man has fallen asleep on your couch. In twenty four hours, unless he is circumcised, his penis will fall off with .01% probability. A doctor is, for some reason, asking you whether they should circumcise him, or wait for the man to wake up to ask his opinion. How do you answer?

            Does your answer change if it is a woman, and the operation is removing her left nipple to keep both breasts from falling off?

            For penile cancer purposes, there is no reason not to let the person make their own choices. It shouldn’t be a factor in deciding whether or not to circumcise an infant who cannot consent. What we should be considering is whether the risks and benefits of circumcising somebody who cannot consent, including the extreme negative element of the fact that they cannot consent, come out positive.

            Or, to reframe this argument, “The benefits to an adult man of circumcision are nebulous, subjective, or extremely low-probability.” That is the one argument. “The risks to infants of circumcision are greater than the ‘none’ people generally think of them as” is an entirely distinct argument. The difference in how evidence is treated arises because one is balancing risks against consent – and low probability risks are not generally a valid reason to override consent – and the other is balancing risks against risks – and low probability risks can really matter here.

    • devonian22 says:

      The 1 in 7000 isn’t the rate of glans injuries/fistulas, it’s the rate of Notable Adverse Events, which presumably also includes much worse outcomes.

  50. meh says:

    Is anyone else getting more angry at this thread than any thread in any culture war OT?

  51. Canyon Fern says:

    A day or two ago, I graciously allowed my assistant, Ludovico, to pay a real-life visit to Randy M.

    “10/10: did not get stabbed,” said Randy, and Ludovico confirms the feeling is mutual. I am told that, amidst the industrious citizens of a burgeoning civilization of plastic horses, the two men feasted on gâteaux-en-tasse, exchanged hardbound dossiers detailing their life histories in painstaking detail, and sampled the profound depths of recent products from Nintendo.

    Finally, Ludovico confessed that he found all members of the House of M to be charming, and most welcoming to a new guest. Quoth my wretched amanuensis: “Let it be known throughout the Open Thread that I had a good time!”

    • Conrad Honcho says:

      the profound depths of recent products from Nintendo

      Three Houses was great, no doubt.

    • A Definite Beta Guy says:

      House of M

      Randy doesn’t bend metal with his mind on the weekends, does he?

      • Randy M says:

        Alas no. I now have corroboration for being ordinary in most apparent regards.

        In addition to lack of stabbing, the Keeper of the fern gets props for having a suitable indoor voice and patience when the youngest M awoke anyway, mostly on account of Too Much Birthday hours prior.

  52. John Schilling says:

    The 2010s were a decade of remakes, sequels, and comic books, some of which were entertaining enough for a rewatch but none of which achieved greatness. The two original standouts, IMO, are The Martian and Dunkirk.

    • Tarpitz says:

      The really good work of the 2010s – and there absolutely was some – was overwhelmingly low budget, artier stuff.

  53. Tarpitz says:

    Cold War
    Hell or High Water
    Leave No Trace
    Lady Bird
    8th Grade
    Only You
    This Corner of the World
    A Star is Born
    Little Women
    Toy Story 3
    Kick Ass
    Capernaum
    Rogue One
    Arrival
    Interstellar

  54. thegoodtimeline says:

    Hey all,

    For those that may be interested I recently interviewed Andres Gomez Emilsson of Qualia Research Institute on the phenomenology and intentional content of 5 MEO DMT; Valence Realism vs Nihilism; Alpha Wave Coherence in meditation; the Schmidt Pain Scale, Effective Altruism and long-tails of pleasure and pain.

    https://thegoodtimeline.com/podcast/2020/1/11/andres-gomez-emilsson-state-of-the-qualia-part-1-of-3

    Any discussion of the above topics is welcome.
    Cheers

  55. gwern says:

    Also, with all of the great Gwern stuff I mentioned on that post I should have mentioned that Gwern has a Patreon that helps fund his projects

    Apropos of my other projects, I’d mention that I just upgraded ThisWaifuDoesNotExist to version 3, using a brand new StyleGAN 2 model. This improves the overall portrait quality, and in particular, removes the nasty blob artifacts which marred so many otherwise-fine faces. (Further discussion with samples/model etc.) I am still hoping to this year be able to tackle ‘ThisBooruDoesNotExist’, which will be what it sounds like…

    • Conrad Honcho says:

      ThisWaifuDoesNotExist

      This is very cruel, and you are a bad man for making this.

    • sty_silver says:

      It’s somewhat disturbing if it generates the image of girl where one part is clearly messed up, but she’s otherwise really cute.

      Feature request: a button to refresh the generated plot summary.

      • gwern says:

        Feature request: a button to refresh the generated plot summary.

        Doesn’t it do that already? refresh_content updates both the image and text samples.

    • Walliserops says:

      A shame monochrome images aren’t supported well, because I think the panda has (rough ballpark) 4-5 million pages’ worth of Comiket/DLsite/DMM/monthly H-magazine scans now. How long until someone figures how to handle panels and delivers ThisComiketDoesNotExist?

      And given that even minor browser games have their dedicated cons, I wouldn’t be surprised if a GAN-assisted doujin scene cropped up in a few years. This is brilliant work, we’re really living in the future.

      • gwern says:

        I drop monochrome images for TWDNE etc because there’s not so many of them in Danbooru and I worry that they are too different from color images. At least with pre-ProGAN NNs, non-color images (with or without color images) had much worse results (possibly related to the texture bias & edges/lines emerging later in training). Since StyleGAN/BigGAN do edges/lines so well, they potentially can handle monochrome too, but I’ve never wanted to spend the time doing a full run to test that… For the ThisBooruDoesNotExist S2 swarm training, we won’t do any filtering: if a scaled-up S2 can’t handle monochrome images easily, then it couldn’t’ve handled full Danbooru2019 images anyway.

        I did consider targeting the panda instead of Danbooru for my dataset, but decided against it: way too much porn, not nearly enough metadata, and I was skeptical of their long-term stability.

        And given that even minor browser games have their dedicated cons, I wouldn’t be surprised if a GAN-assisted doujin scene cropped up in a few years.

        Definitely. Once you get tag embeddings going and whole-image GANs like BigGAN, you can generate arbitrary images and then tweak them via editing, and now you can put together sequences of images… Even if no one does this open source, simply knowing that it is possible will be a big motivation.

    • helloo says:

      I was going to mention that it didn’t seem to do headgear/animal ears all that well, then I looked at the “plot”.

      “The girls in the anime are all named after their names, but none of them is their real name at all.”

      “One of the characters in the manga “I Need a Drink” is one of the characters in the main series, in which the protagonist tries to give a drink of water instead of going to sleep. She gets the same result.”

      This was one of the worse/better ones but in general, it seems to have some problems with repetition, though often true for many deep learning AI, seemingly worse for this one possibly due to the training set.

      Plenty of artists create pictures/avatars based on character-gen programs. I could easily see a niche that creates entire stories from these things.

      • gwern says:

        Yes, the text samples could be improved. They were done with the small GPT-2s (not even finetuned entirely) with top-k sampling – how crude and primitive and benighted they were in those days! We know so much better now.

        If I were doing it now, I would start with our Archive Of Our Own (Ao3)-finetuned GPT-2-1.5b, then finetune it on the combined anime summary/description dataset, and then generate samples with top-p rather than top-k. Oh well.

  56. sharper13 says:

    [Deliberately posting this on a non-CW thread so it doesn’t turn into a discussion of the relative merits of various politicians. Ideally this is more of an economics/political science discussion.]

    One of the great myths of politics is that it’s possible to “buy” an otherwise fair election. Closer to reality, I believe, is that it’s possible to “marginally influence” an election with money spent on outreach, getting people who would’ve voted for your candidate anyway to the polls, etc… If you weren’t at least a possible winner, you aren’t going to turn into a winner by spending money. It’s more difficult than some at first consider to influence the values/basis of people’s voting patterns by spending cash.

    Money does help discoverability somewhat. If you’re a great candidate, but no one otherwise knows who you are or what you stand for, then that’s the situation where the most help will come from spending money on advertising and outreach efforts. A corollary to this is that candidates who are already well-known (incumbency, celebrity, media mentions, whatever reason) benefit more from restricting the ability to spend (or receive) money of potential opponents who aren’t already well-known.

    But how can I say this, since “everyone knows” candidates spend lots of money and lots of time to raise money to spend?

    Mostly correlation vs. causation. Studies have been done comparing self-funded governor candidates with people who raise money more organically. The basic finding is that the ability to raise money correlates with voters backing a candidate, but simply spending money from another source (like someone’s own fortune, or a single high-dollar source rather than a broad fundraising effort) doesn’t matter as much, everything else held constant.

    Generally, there’s enough money in a major contested campaign for both sides to get their messages out and compete with each other. After that, there may be some marginal advantages here and there (no one is saying money spent is completely useless), but by and large the causation flows from having the support of lots of people leading to both raising more money and getting more votes, rather than the money spent (over a certain minimum for people to know who you are and consider you a serious candidate) dramatically affecting voting totals.

    The current canonical examples in recent Presidential politics are Jeb Bush spending $50 million/delegate in the 2016 primary and Bloomberg so far spending more than $210 million in the 2020 primary while currently sitting in the polls below Buttigieg ($28M spent) and above Yang ($8M spent).

    Feel free to push back on the above view, or expand on it as desired. 🙂

    • cassander says:

      On self funding vs. raising money, i think there’s a more subtle issue at play. if you need to raise money, it’s theoretically a distraction. in practice though, monetary donation is a costly (literally) signal, and campaigns that can’t inspire it probably also are going to do a bad job at inspiring people to vote. Campaigns free from the need to do money are less dependent on the good will of voters, which in the long run is probably a bad thing for someone trying to win an election. put more succinctly, money (raised) isn’t the point, it’s just how you keep score.

    • birdboy2000 says:

      Looking at it in terms of which adequately funded, pro-donor candidate wins strikes me as entirely missing the point. It’s a financial barrier to entry which effectively freezes lower and middle class people out of all but the lowest levels of public office.

      • sharper13 says:

        Does that still disguise the causation, though? If a candidate isn’t charismatic/convincing/leader enough to convince anyone (even masses of lower/middle class folks) to support their candidacy, are they likely to win even if provided funding from some other single source?

        The record seems to indicate that the same skills which allow raising money are related to gaining other support. Sure, there are some people who can raise money from themselves or from high dollar donors without those skills, but it’s precisely those people who spend lots of money and still lose.

        Is it unlikely that some unknown will be able to get support to run for Senator or President compared to someone wealthy/in-the-right-circles? Sure, I agree with you there. But why would most people support someone like that for an important office? Many of those who can run for that sort of office distinguish themselves, or at least get lots of press, in a lower office and then work their way up. That’s not to mention people like Obama who go from unknown to State Senator (for 7 years) to Senator (4 years) to President very quickly. That’s a rare and fast trajectory, but it shows it’s possible if you are the type of person who can raise support.

      • Garrett says:

        Yes, but.

        If you take someone with otherwise no name recognition, their only “selling point” is going to be that they aren’t The Other Guy. Which means that they can be substituted for by about 50% of the population interchangeably.

        So you’d have to ask why someone would otherwise want to vote for this person at least over a random selection of people on the same team. (This roughly corresponds to the nomination of a major party). There are many ways to do this, but most of them involve demonstrating significant ability in another realm of endeavor. This could be something like being a governor, a legislator, a general, a significant cabinet secretary, a successful business executive, or other such position. On the traditional “lower and middle class” level the obvious example would be union organizer.

        But of all of the plausible routes to the Presidency, the only one not available to the lower and middle classes are the self-funded billionaire, which don’t seem to have done well in the past century or so. Someone of modest means can start in local government on city council and work their way up to state legislature if they are so inclined.

  57. salvorhardin says:

    A friend of mine has a windfall-ish sum of money they’d like to donate to charity and is specifically interested in donating to effective addition recovery related charities. Are there any organizations in this field that people would recommend for making particularly good (lifesaving or at least life-improving) use of donor dollars?

    • Radu Floricica says:

      No suggestion, I’m afraid, but it just occurred to me… is there a mechanism already set up for privately funded research grants? Maybe with a crowdfunding option as a separate add-on? I’d definitely put money there. Also re the post by SearchingSun above. And if I’m writing a wishlist, I’d also add the possibility for research teams to put grant proposals up there kickstarter-style.

    • b_jonas says:

      For future keyword searches: addiction recovery.

  58. The Pachyderminator says:

    I saw fewer films starting around 2016, so this is weighted towards the beginning of the decade. Also, it’s only a sample, not an attempt at a definitive list. Some of these are lesser known, so I’ll provide brief descriptions. Roughly in chronological order:

    Of Gods and Men: About the French Trappist monks killed by an Islamic group in Algeria in 1996. The most spiritually profound drama about martyrdom I’ve ever seen.
    The Mill & The Cross: An artistic hybrid of sixteenth-century Holland and the crucifixion of Jesus, based on or inspired by Pieter Bruegel’s painting The Way to Calvary.
    The Tree of Life: The only film I’ve seen in the theater four times. This is a film that embodies the way divine grace breaks through nature and is seen in the glories of creation.
    Cloud Atlas: A genuinely unique experiment in parallel storytelling that didn’t get its due from critics.
    Before Midnight: The third entry in the periodic conversational trilogy (revisiting the same couple played by Ethan Hawke and Julie Delpy and essentially just listening to them talk) by Richard Linklater, who’s proven himself to be the master of the “two people talking” genre.
    A Most Wanted Man: Fascinating procedural spy film based on a John le Carré novel, featuring a brilliant farewell performance by the lamented Philip Seymour Hoffman.
    Mad Max: Fury Road: Needs no introduction
    Chi-Raq: Lysistrata transposed to inner-city Chicago, complete with rhyming dialogue. Sort of a social-justice satyr play. I don’t care how problematic it is, it’s insane and wonderful.
    Moonlight: Probably the best LGBT-themed film I saw this decade, done with a light touch and a sure hand. Better than Call Me By Your Name, albeit less lush and romantic.
    Jeannette: The Childhood of Joan of Arc: So unusual it defies description. Imagine if a modernist poet decided to write a medieval mystery play and then someone else adapted it into a heavy-metal musical.
    The Irishman: The haunting answer to all of Scorsese’s previous crime films. In three and a half hours there’s not a wasted minute.

    • The Pachyderminator says:

      Moonlight: Probably the best LGBT-themed film I saw this decade

      On second thought, I’m not sure I endorse this statement. It’s a close contest between Moonlight and Carol.

  59. meh says:

    Nothing comes close to Fury Road

  60. Well... says:

    What’s the best strategy to scoop from a tub of some kind of viscous fluid consumable sold in tubs (e.g. hummus, ice cream, shea butter, Nutella, etc.), if (1) you won’t be scooping it all in a single session, (2) your goal is to maximize how much “fresh” product you get over the product’s lifetime, and (3) you operate on the assumption that the portion of the product not exposed to air is fresher than portion that has been so exposed?

    Seems to me your goal would be to minimize surface area, which at first glance would imply constantly leveling out a flat surface by the end of each session, all the way down to the bottom of the tub. BUT, since the product starts out with a more or less flat surface exposed to air, what you’d actually better do is scoop from one side, leaving a flat vertical edge (if the tub is circular, this would look like a chord, as seen from above). Plus, this leaves the nice “factory swirl” maximally intact, which for me at least is oddly important.

    Are there other considerations I’m missing (excluding “just shut up and scoop your damn hummus)?

    • broblawsky says:

      The most critical thing is to keep the tub completely sealed; there’s only so much oxygen in the tub for the surface of your food to react with, and the oxygen should be depleted quickly if you seal the tub completely.

    • littskad says:

      Does “Press plastic wrap on the surface” count?

    • drunkfish says:

      I don’t buy your vertical plane suggestion, why would that be better?

      Your goal, as you say, is to minimize the surface area of your product for a particular volume. In a cylindrical container that’s fairly tall and full, thus is definitely just a flat surface on top. Things will get messier when it gets fairly empty though, and my guess is once you’re pretty empty, you transition to wanting a sphere (since spheres minimize surface area/volume) that’s cut off by the base and the wall, so tucked into a corner. How you transition from a flat surface to a (part of a) sphere isn’t obvious to me, but I wouldn’t be surprised if it’s not a smooth transition, that at a particular amount of stuff remaining your ideal shape jumps to being spherical.

    • Grek says:

      Transfer the substance into a toothpaste tube, spray cheese canister, ketchup bottle or other such container. This ensures that only a small portion (at the opening) is exposed to air. If you cannot readily transfer it between containers, improvise a superior lid with the desired properties by cutting off the bottom of one tub, cutting a small opening in it and then using it to cover the other tub, so that pressing down on the cut off portion causes a toothpaste tube-like ‘spray’ of product up onto the small tub for your consumption.

      • Well... says:

        I wonder if anyone’s actually done that.

        I for one am unwilling to dispense hummus onto my pita in such a degrading way.

        • AG says:

          “Hummus in a ketchup bottle” does certainly seem like a missed opportunity. Perhaps it’s yet too dense to be dispensed well in that fashion? Or is it that the image of hummus as a dip is too strong? You don’t see sour cream dip in a squeeze bottle, either.

          • helloo says:

            Existed and the general thought seems to be similar horrification.

            Besides solutions like vacuums or other seals (including things like oil “covers”), another way would be to store the stuff in a different container than you would to serve it.
            Ex: a jar then pour it out into a bowl like one does to salsa or other dips.

          • Well... says:

            You definitely do see sour cream — Daisy brand, no less — in a squeeze bottle. (Don’t know about dip, though.)

          • Lambert says:

            Still thinking too small.

            Keg with beer taps.
            Artesian well.
            Shale deposit.

          • Well... says:

            Hm…if hummus had to be extracted from the ground, could you frack it out with tahini?

          • albatross11 says:

            Sour cream is sold in squeeze bottles in grocery stores where I live.

    • Hyperfocus says:

      I don’t know if this would work for ice cream (because I’m not sure oxygen is the villain, rather than accumulation of condensation which then freezes), but I’ve always wanted to try the following for preserving avocado: remove the skin and pit, then immediately place in a container filled with CO2 obtained from displacing the air it contains by sublimating dry ice. Since CO2 is heavier than regular air, it should be reasonably pure and free of O2, which means the avocado won’t oxidize.

      • Randy M says:

        Why not just try a vacuum sealer?

      • albatross11 says:

        I love the idea of a lowered chamber with the hummus/guacamole/etc in it, where the air has been displaced by some heavier not-very-reactive gas, preferably inert. I looked up Argon (a noble gas that’s heavier than air) on Wikipedia, and found that this is actually done some places, notably by winemakers trying to prevent their wine from interacting with oxygen.

        • Nornagest says:

          You can buy little squirt bottles full of inert gas, mostly argon and nitrogen. They’re meant for preventing red wine from going bad after you open it and before you finish the bottle, but you could probably use them for guacamole too.

          You can also get vacuum pumps and matching stoppers for wine bottles, but you’d need to jerry-rig some kind of adapter to get them to work with food containers. Might work just to drill a hole of the appropriate size in the lid and stick in the stopper, though.

        • helloo says:

          Note that this is VERY DANGEROUS if you do it with anything bigger than a small bowl of due to the possibility of doing the same thing with your lungs.
          CO2 is somewhat safer as a human body can detect that somewhat.
          And the solution for gas leaks of making them really smelly probably won’t be the best idea for this.

          • albatross11 says:

            ISTR that CO2 level is how our bodies detect we need to breathe, so we’ll pretty reliably detect high CO2 as “I feel like I’m suffocating,” whereas a high concentration of argon in the basement probably gives you no warning till you permanently fall asleep.

    • Statismagician says:

      If you’re concerned about air contact, cover the product inside the tub with plastic wrap and push it down tight.

  61. Anteros says:

    I would consider most things written by people who are alarmed by climate and it’s possible changes to be doomer porn unless otherwise informed. People have been looking into the future and seeing calamity since human beings first learned to speak – and the appetite for reading such things remains insatiable. Sure, it can be titillating, but there are more productive things to do than indulge in catastrophological armegeddonology.
    ETA meant as a reply to tjwhale below

  62. HeelBearCub says:

    From the linked twitter thread on GPT2 chess:

    Rohit Pgarg: Or perhaps, GPT2 understands language about as well as it understands chess.
    Gary Marcus: Neither well, once you leave the corpus

    It seems plainly false to me that GPT2 is producing something that is in any way complex or meaningful when it is prompted. It’s not. Much like we can see faces in trees, we can see meaning in that which is superficial.

    GPT2 is very interesting, it’s the kind of thing that can be turned into very useful tools, much like google is an extremely useful, history making, tool.

    But prompting it many different ways and then selecting some to show how it can produce deep meaning… not buying it.

    • gbdub says:

      Agreed on all counts.

      Scott is amazed that GPT-2 can “write poetry”, “compose music”, and “play chess”, as if it had learned to do things wildly different from what it was designed for.

      But in fact all the varied things GPT-2 can do are just one thing: given a curated pile of text input, GPT-2 can output superficially similar text. That poems, music, and chess moves can all be represented by text is not exactly a shocking development.

    • Ketil says:

      It seems plainly false to me that GPT2 is producing something that is in any way complex or meaningful when it is prompted. It’s not. Much like we can see faces in trees, we can see meaning in that which is superficial.

      Are you saying GPT-2 output is just random jumble that happens to look meaningful by chance? Because I think that is what faces in trees are, and obviously untrue for GPT-2. So while I agree that humans antropomorphize way too much, there clearly is some structure (complexity, meaning) to GPT-2 output.

      • HeelBearCub says:

        My point in bringing up faces in trees was to highlight how far human pattern matching is capable of stretching itself, not to equate GPT 2 to random noise.

        And when I say “no complexity”, I don’t mean that GPT2 isn’t doing something complex. Rather, I’m rebutting the idea that when we look at text that seems “off” in some way, that this is an indication of the text being “difficult”, “complex”, “deep”, etc.

        Lewis Carrol can create Jabberwocky, but that does not mean that something that might be titled “Chasterdokly” by GPT2 can be plumbed for meaning.

    • Enkidum says:

      I think it’s possible, likely even, that some aspects of GPT-2 represent an important qualitative step forward in AI, and possibly even in our understanding of the human mind. But there’s a fuck of a lot of steps missing before we actually have real AI or a good understanding of the human mind, and if you get excited every time such a step is taken, you’re going to go a little nuts.

      The way I view things like GPT-2, deep reinforcement learning, etc, is that they are essentially existence proofs of things AI people were saying in the late 80’s and early 90’s. Margaret Boden’s On Creativity (1990) has a lengthy aside on the writing of “The Rime of the Ancient Mariner”, showing how Coleridge was combining phrases, images, and language from books he’d been reading over the months prior to writing it (we can be sure of this as he kept a detailed journal of everything he read for decades). This is, if I understand correctly, very much the sort of thing that GPT-2 does, and it may be that the ways it does it shed some light on what we’re doing in our heads. But of course Boden isn’t arguing that this constitutes everything Coleridge was doing. Among other things, selection between possible phrases is critical, which is analogous to the job that the humans curating GPT-2 output are doing, and is one of the many reasons why we should hold back on getting too excited about it.

      • HeelBearCub says:

        if you get excited every time such a step is taken, you’re going to go a little nuts.

        By all means, get excited by it. Nothing wrong with finding the progress in machine learning exciting. The problem is not in getting excited, but in misunderstanding what GPT2 is, and what it is doing. It was exciting that search engines could guess the likely end to the question you started asking, too.

        And I agree, there are some fundamental aspects of cognition that are being explored here.

  63. hnau says:

    I haven’t seen a huge number or variety of this decade’s movies (as my list probably indicates), but here are some favorites that I’ve re-watched at least once:

    Captain America: Civil War
    The Force Awakens
    Gravity
    Inception
    The Martian
    Mission Impossible: Ghost Protocol
    Premium Rush

  64. Machine Interface says:

    Arrival
    Mandy
    Ad Astra
    Inception
    Only God Forgives
    Suspiria
    Black Coal, Thin Ice
    Blade Runner 2049
    The Tree of Life
    Mad Max: Fury Road

    With a few runner-ups to spare, and knowing that I’ve still got a fair number of movies from that decade on my wishlist.

  65. A Definite Beta Guy says:

    I gotta go with Arrival, but the movie I enjoyed the most was “Her.” I really, really liked Joaquin Phoenix falling in love with ScarJo as a smart phone.

    I also want to throw out Days of Future Past as the best X-Men movie.

    • Belisaurus Rex says:

      Yes, and it could have been a great movie if the finale was the Mystique assassination scene instead of Magneto throwing football stadiums at the president. Not every movie needs a superhero level finale. Some can stay an espionage/Cold War film that just happens to have mutants in it.

      • John Schilling says:

        Not every movie needs a superhero level finale.

        If it’s a movie about comic-book superheroes, it kind of does. Marketing trumps storytelling, both for financial reasons and because your story doesn’t matter if it doesn’t find the right audience. If the movie is about comic-book superheroes, it attracts an audience that wants an elaborately choreographed super-fistfight at the end, repels the hypothetical audience that would enjoy a good Cold War espionage story, and the handful of nerds and geeks who would actually like the crossover won’t pay the bills.

  66. A Definite Beta Guy says:

    Last year I ranked a few states by desirability of living. Mostly just for fun, partly to define some parameters in case we have to Illinois. Illinois, for those who are not following, ranks among the worst run states fiscally in the entire union, with one of the least-funded pensions. Even with our 2017 income tax hike, we’re in absolutely horrid fiscal condition, so we’re voting on a progressive income tax this year. We’re losing people, bad enough that Chicago property values were flat in 2019 and expected to decline in 2020: my house is actually worth less than what I paid for it in 2015.

    So, over the holidays, I got a bit more serious and started looking at specific villages and cities in Wisconsin, Tennessee, and North Carolina (which ranked among my highest choices).

    Some high-level thoughts:
    -TN and NC, and I suspect other Southern cities, are struggling to keep up with their population growth. 6-8 mile commutes can turn into a random walk between 30 and 70 minutes. Minimizing commute time is probably one of the best quality-of-life improvements you can make, so that SERIOUSLY reduces NC and TN.
    -School systems in NC and TN just look so much worse. My wife and I both went to schools with a 25 ACT average. There are several public schools in Illinois that have good test scores like this. Wisconsin also does well. NC and TN have much fewer options.
    -You may think you’re getting a pay increase to live in a high COL area, but a lot of professions don’t. Mrs. ADBG is a pharmacist. Median salary in Ames, Iowa? $129k. Median salary NYC? $122k. Other friends of mine are thinking about moving and see similar results for their salaries. It may be worth exploring the BLS wage data to see if moving to a lower COL area is an easy win for you.
    -Even if you get a bonus for moving to a high COL area, there may be SPECIFIC low COL areas for your job. Accountants in Peoria, Illinois make as much money as accountants in Chicago, IL. Accountants in Houston make as much as accountants in New York City. Accountants in Amarillo, St. Louis, Detroit, and Tulsa make basically the same as accountants in Chicago. And accountants in higher COL areas do not make anywhere near enough money to cover astronomical differences in COL anyways.
    -Tennessee crime rates are UGLY.
    -There’s really no obvious rankings for food, beer, and recreation options. Nashville prides itself on food, but the food is honestly “meh” compared to higher cities. You wouldn’t know this without doing some pretty extensive research.
    -most “new urbanist” communities are entirely pointless. They are just slightly different subdivisions and there is nothing “new urbanist” about them, because they are so separated from actual urban areas. That’s what you get with greenfield developments.

    Going into some specifics:
    Wisconsin: I ranked WI #1 before on the basis of being so close to IL, but WI looks even better after looking at specific cities. You have to put up with a longer, deeper winter…but if you can do that, you have a lot more options on where to live.
    I’d rank the suburbs as follows:
    1. Pewaukee
    2. New Berlin
    3. Whitefish Bay
    4. Middleton (Madison)
    5. Cedarburg
    6. Grafton
    7. Shorewood

    Besides Middleton, every suburb is a suburb of Milwaukee. There are two inner ring suburbs here (Whitefish Bay and Shorewood) which directly border Lake Michigan and actually look pretty nice. However, median home values are $450k and $350k and property tax assessments tend to be close to 2.5% of property value. That makes these two neighborhoods REALLY expensive compared to anywhere else, even more expensive than my current Chicago suburb. On other hand, you’re right next to Lake Michigan, and only 15 minutes from the city. Is that trade worth it? Not really clear.
    Whitefish Bay is also a straight-up residential area, so it’s not anywhere close to an actual urban experience. Shorewood gets you a whole lot of crime.
    Also, Milwaukee does not have a football team, which means it is not a real city.

    Middleton LOOKS really nice on almost every single metric. It borders Madison, it’s right on a lake, property values and property taxes are reasonable, it even has a “New Urbanist” greenfield development where you can actually walk to a pub from your house, and the pub has actual GOOD beer! Madison offers some pretty high salaries for most jobs, it has minor skiing nearby, it has an awesome farmer’s market, and it has a great university.
    However, Middleton’s violent crime rate is insanely high compared to a typical suburb. It looks like a lot of these are domestic violence assaults: Middleton has the same number of armed robberies as, say, Whitefish Bay. However, that’s pretty ugly.

    Most of the other suburbs are what you would typically imagine as White People suburbia, especially Grafton and New Berlin. At that point, it’s just a question of commuting differences. I like Pewaukee the best because Pewaukee Lake is right there and easy to access, but that’s just flavor.

    Tennessee:
    So, major metro areas. You have Memphis, you have Nashville, you have Knoxville, and I guess you can include Chattanooga (but no schools qualified for my filters). Knoxville is pretty incredible, but…there’s a huge pay cut for people working in accounting/finance. So that’s a Hard Pass for me. For everyone else, Knoxville is a relatively nice small city that’s damn close to one of my favorite National Parks. Maryville is a suburb just to the west that has extremely attractive property values, and still has a lot of open space for you to build your own dream home.

    Between Memphis and Nashville…Nashville is hurt pretty bad by those commuting issues I mentioned earlier, at least if you want those “good schools.” Brentwood is insanely expensive, which means you are stuck going to Franklin, or one of the various exurbs, where you could be commuting 90 minutes one-way if you are unlucky. Nashville also has a crapton of tourists, and tourists are annoying (it does mean a better airport, though).
    Nashville has a NHL and NFL team, though, which is better than Memphis’s NBA team. It also has better natural amenities and is better position to everything other than the Mississippi river, and the Mississippi river is really only attractive as a natural amenity if you are Mark Twain. On the other hand, again, Memphis has a better commute. Plus, the Yuppie food developments are all in East Memphis, which are easily accessible from the Near East wealthy suburbs. I’d have to give Memphis the nod.

    Rankings:
    1. Germantown (Memphis)
    2. Collierville (Memphis)
    3. Mt. Juliet (Nashville)
    4. Franklin (Nashville).
    5. Nolensville (Nashville)
    6. Maryville (Knoxville)

    Germantown and Collerville are two suburbs of Memphis. They have almost identical median property values (350k, taxes between 2300 and 2700), with Germantown being a bit more crowded. However, Germantown has a higher scoring high school, and is closer. If you want a bigger lot, Collierville makes more sense, but that’s not my kind of priority. Plus, Germantown is right next to Shelby Farms, which is the only thing really close to a recreational area in Memphis. It’s no Central Park, but you need to take what you can get.

    In Nashville, Franklin has a bit more developed downtown with a bunch of sprawl around it: if you want the walkable downtown, that requires paying a lot of money. Median property values are $500,000, with a commute that can get as high as 70 minutes to the Gulch in downtown Nashville. Why bother when you can have a 25 minute commute in Germantown?
    Mt. Juliet borders Nashville, but “feels” more like a giant exurb. Property values are a much more reasonable $330,000 and commutes are shorter due to being right next to Nashville. The problem? Mt. Juliet’s crime rate is 2 per 1,000, which is pretty damned unsafe for a wealthy suburb. I really wouldn’t even consider it an option, except that it’s the only reasonable close commute to Nashville, and you’d want the OPTION to see the Nashville Predators and Tennessee Titans and Music Row if you’re going to live in Nashville.

    Between TN and WI, WI is better on beer, natural recreation, commutes, and financial jobs. The better financial jobs cancel out the lack of an income tax in TN, at least for me. Plus, if state schools are important to you, UW Madison is one of the best in the nation.

    North Carolina:
    Overall, I’d rank NC higher than TN after seeing the commutes. I was REALLY excited about NC for a moment, because Wilmington looked pretty attractive on all fronts, but Wilmington’s best school seems to be a lottery enrollment school, and I’d never subject my child’s education to a lottery if I have the money to guarantee it (what is money for if not to improve your children’s lives).

    That gives NC two major options: the Research Triangle and Charlotte. Charlotte is huge and sprawling, and has a small section in the South that has some particularly good schools and reasonable home prices: Myers Park, closer to the city, is both crazy expensive and has more “meh” schools.
    The Research Triangle gives you a couple different options that are not exactly east to commute between: Chapel Hill, Durham, and Raleigh. The major issue is that pharmacist positions seem a bit saturated there, and it looks like credentialism for positions can be pretty bad for mid-level positions like mine. Charlotte offers competitive salaries and competitive job markets, more so than the RTP area. Plus, Raleigh only has the Carolina Hurricanes, while the Hornets and the Panthers play in Charlotte. Plus, Charlotte is closer to Asheville, which is God’s Gift to North Carolina. Charlotte DOES put you an hour further from the ocean, but I’m not sure that’s as big a deal for me.

    My rankings:
    1. South Charlotte
    2. Chapel Hill
    3. Cary

    The south area of Charlotte, around Providence High School, has some pretty attractive homes at a reasonable home price (something akin to $350k is okay). The commutes range from 25 to 50 to, say, Fourth Ward, but Charlotte has a number of big jobs available outside the city center that’d make commutes more reasonable.
    Chapel Hill has a lot of walkable areas, but those areas are pretty pricey. You can, however, find homes a bit further out that at least look bikeable. Median home values are $400k.
    Cary is a big sprawling suburb bordering Raleigh, with the “good schools” on the West side, far from the more attractive living in the older downtown area in the East. Median home values are $380k with property taxes being at or close to $2k, which can get you a decent home in said school district. Cary also has damn good commutes to a huge portion of the Triangle, at least if Google Maps is to be believed. I’d still prefer Chapel Hill, especially since it looks like there are some close by hikes, but it’s close. I’d say Chapel Hill’s biggest problem is that it doesn’t have a freaking Costco: it’s the only town on this entire list where going to Costco is actually a haul.

    Upon further review, I’d rank the states as follows:

    1. WI
    2. NC
    3. TN

    I’d put NC strictly over TN, and I’d put WI over NC as long as you can tolerate a harsh winter.

    Depending on your feelings WRT to sports, weather, and the local crime rate, I’d put Middleton, Wisconsin at the absolute top. Given how close it is, I think my wife and I will visit over the summer and see how it feels.

    • brad says:

      I guess it doesn’t impact everyone, but for me I care more about the lack of sunlight in winter than harsh winters per se. The shortest day of the year in Wilmington is almost a full hour longer than in Milwaukee, and the sun is brighter when it is up. I’m in NYC now and I can’t imagine moving even further away from the equator, even in places like Europe where the weather is better.

      • A Definite Beta Guy says:

        This is a reallllllyyyy good point. I guess I just wrap this into “winter” because I’m from the Great Lakes region, but the lack of sunlight can be really depressing for people who are not used to it.
        Sunrise: 7:17 AM
        Sunset: 4:42 PM

        AKA, working and commuting hours. Basically, I’m not going to see the Sun today, except from the inside of a car and maybe at a slight angle in the office. That’s nottttttt fun.

        • acymetric says:

          Maybe this is just faulty memory, but I remember it staying light way later in Ohio than it did in North Carolina (I assumed this was just being further west, essentially on the edge of the Central timezone but still in Eastern). The days may have been shorter, but I care a lot more about daylight at the end of my day than daylight at the beginning (I don’t mind going to work in the dark, I hate coming home in it).

      • gbdub says:

        OTOH, long summer evenings Michigan were lovely, and I miss them in Arizona.

      • Randy M says:

        I wonder if it would be possible to compensate by having longer lunch breaks?

        I know I get less sun in the winter, but still, being indoors for eight daytime hours is more to blame than the planets tilt.

        • brad says:

          I haven’t looked at the literature, or even thought too carefully about the physics, but it feels like it is both the quality and quantity of sunlight in winter that sucks the life out of me.

          • HeelBearCub says:

            It’s the angle of the light.

            No, I have no scientific proof. That’s just how it feels.

            The angle of the light in fall is the most beautifully sad. The light in winter is so oblique as to not exist, just a bleak reminder of what light might be.

        • A Definite Beta Guy says:

          Probably, but longer lunch breaks is not always in the control of an employee. I don’t think I’ve yet had an employer that said “oh, go ahead and enjoy your lunch so you can actually see the sun today.”

          More, “it’s cold, so no one is leaving the office, why are you leaving the office? Work at your desk.”

          • cassander says:

            My office has a summer fridays policy where workers can take off early on fridays to enjoy the weather. The managers generally dislike this policy because what used to be something that we used to graciously grant has become something they now expect that we sometimes have to take away.

          • Randy M says:

            Probably, but longer lunch breaks is not always in the control of an employee.

            I actually had this more in mind than the physical aspect of it. As in, perhaps this is a cultural shift we should consider advocating for.
            But it would probably be another point of friction between employees with families who’d rather finish early, and those without.
            If you’re lucky enough to work for a boss who allows some flexibility, it might be worth a try.

    • Evan Þ says:

      I grew up in the North Carolina Triangle, and my parents still live there. I can’t say much about the social climate since I was an introvert and then moved away for college, but there’re a lot of good churches to choose from if that matters to you. Since you’re moving from the north, I have to warn you the summers can be brutal – but then you maybe get one snowfall a year, so pick which bad weather you want.

      If you’re planning to bike in Chapel Hill, be forewarned that the downtown and university are on top of, yes, a hill. Though, there’re a lot of pharmacies on the outskirts of town, and decent fare-free bus service into downtown (except on Sundays). The closest Costco is on North Point Drive in north Durham. Cary I don’t know as much about, but note that 540 and 147 are both toll roads south of I-40 (but still free north of I-40). Also check out the Briar Creek neighborhood of Raleigh – a couple of my friends who stayed in the area live there (commuting into Raleigh) and report good things. I can personally endorse the Briar Creek shopping center and a couple of the restaurants.

      For Google Maps, make sure you’re picking the right commute hours. Last time I was visiting, 15-501, 40, and parts of 54 were brutal at peak hours.

      Public transit is skeletal, but if you’re on the skeleton it’s workable. TTA runs good regional express buses among the downtowns also connecting various park-and-rides, Chapel Hill has good fare-free service to the university, and then there’re a few other decent routes. My dad used it to commute for most of my childhood and liked it, since his workplace happened to be on the route, until we moved someplace that wasn’t.

    • johan_larson says:

      I lived in Madison, WI, for five years as a graduate student. It’s nice, assuming you’re ok with winter. And between the university and the state govenment, it has some upscale services.

      Any particular reason you’re not thinking of living in Madison itself? It’s small enough that it’s not particularly urban. The downtown area itself is pretty small, and isn’t all that built up.

      • A Definite Beta Guy says:

        Living in Middleton assigns you to Middleton high school. Most of the Madison schools don’t really score as well as Middleton.

        • The Big Red Scary says:

          A few comments about schools.

          1) Test scores are probably telling you more about the average student than the school. If a school has a small program for gifted children (mine did when I was growing up), then the average is irrelevant.

          2) You could just not send the kids to school, due to the enormous opportunity cost. Just think of all the wonderful things you could have done and learned as a kid if you weren’t stuck in school all day long.

          3) “Homeschooling” would not work very well for us, given the temperament of the children, so instead we formed a co-op with other families. We rent a house, fill it with books, musical instruments, and tools, and let the kids run wild. So far, so good.

          • EchoChaos says:

            If a school has a small program for gifted children (mine did when I was growing up), then the average is irrelevant.

            This isn’t really true, because you will have social group interactions outside the gifted class, which can lead to the bad stuff you’re trying to avoid by having your kids interact with the average student.

            You could just not send the kids to school, due to the enormous opportunity cost. Just think of all the wonderful things you could have done and learned as a kid if you weren’t stuck in school all day long.

            This is great, and my kids love it.

            “Homeschooling” would not work very well for us, given the temperament of the children, so instead we formed a co-op with other families. We rent a house, fill it with books, musical instruments, and tools, and let the kids run wild. So far, so good.

            We do this too, once a week. It’s really fantastic for the kids.

          • A Definite Beta Guy says:

            2 and 3 seem pretty “alternative,” and I’m not really comfortable running my children through alternative educational styles. Like, the US education system is essentially designed to do a good job for my family: white, upper middle class, educated and involved parents. There’s not really a huge incentive to not using the already-existing pathway.

            Re: 1, maybe, but…
            1. IME, high performing suburban school districts have excellent gifted student programs
            2. There’s no guarantee that your kids will end up in gifted programs. Myself and my in-laws were tracked to AP courses, but we were not placed in the specific gifted programs.

        • johan_larson says:

          But you wouldn’t be sending your kids to a random Madison school. If test scores are really important to you, you could choose to live in the district of Madison that has the really high-scoring schools. I’m pretty sure the public high school your kids would go to is done strictly by where you live, in the Madison system.

    • DragonMilk says:

      What are your thoughts on Utah/Salt Lake City & Suburbs as well as Colorado city/suburbs?

      • A Definite Beta Guy says:

        Poor on Colorado, skeptical on Utah. Colorado has one of the worst pensions in the nation, almost as bad as Illinois. The income tax is higher than Illinois already and Denver is obnoxiously expensive.

        Utah has 5% income tax and a 5% sales tax (rounded) and a decently run pension system, so it’s relatively low tax and I’d expect it to stay that way. However, Salt Lake City is a substantial pay cut, so it’s about on par with TN, except pharmacists ALSO take a slight pay cut to live there (so strictly speaking it is inferior in income to TN).

        When it comes to schools and commute:
        Kaysville: 420k/22 miles
        Bountiful: 375k/10 miles (maybe, but this is a SUPER hot market, and prices are expected to go up 5% next year? Move now)
        Draper: 535k/20 miles

        These are relatively pricey homes and we’re not talking about easy commutes unless SLC gets like no traffic. Bountiful looks alright, violent crime of 1 per 1000 and property crime of 14/1000, so pretty safe, and there is also a pretty attractive looking commercial strip that has a lot of food options, but I’m not sure how many are actually good.

        What’s particularly interesting are that property taxes look practically non-existent, which is several thousand dollars per year back.

        This is obviously me just googling stuff because I’ve never even been to Utah, unlike these other states.

        • Solra Bizna says:

          Commute:

          According to legend, Salt Lake City’s streets were designed to be wide enough that two horse-drawn carriages could make U-turns next to each other on each street. Thus, when cars became a thing, SLC’s streets were more than ready to handle the tiny amount of local traffic. And then, as SLC has grown explosively over the last few decades, the streets have risen to the challenge. Until a couple years ago, I never experienced traffic slowdowns except in the area of a major accident.

          Unfortunately, slow traffic and crowded streets are definitely on my radar now. We may have finally gotten big enough that even Brigham Young’s extravagantly wide streets are no longer enough. Still, for me the navigational convenience afforded by the grid system is still a major plus.

          Housing prices:

          SLC is trying to become the next Silicon Valley, and part of that is skyrocketing housing prices. In some parts of the city, rents are rising by 25%/year or more. Even twenty minutes away from the city proper, my rent just went up 8%, and is set to do it again next year.

          Food:

          I can’t comment because being able to spend money on dining above the lowest fast food bracket is completely outside my experience.

          Crime:

          I can confirm that crime is extremely rare here, though that’s been changing as the population balloons. Once, when I was young, a teenager went around our neighborhood testing people’s cars to see if one was unlocked. The neighborhood watch was galvanized. The teenager wasn’t caught, but it felt like nobody talked about anything else for days—even such almost-crimes were that rare.

          (Oops, I just outed myself as a SLC dweller on the Internet.)

          • Matt M says:

            SLC is trying to become the next Silicon Valley

            Isn’t everyone?

            (I know that sounds snarky and dismissive, but I’m being serious here… isn’t literally every city undertaking some sort of initiative to become a “tech hub,” and claiming early success?)

        • EchoChaos says:

          The income tax is higher than Illinois already and Denver is obnoxiously expensive.

          Are you sure? I googled and Colorado’s income tax is 4.6%, which is lower than both Illinois and Utah, which you called “low tax”.

          Is there something that I’m missing that makes you call it high tax? Colorado also has TABOR, so while its pension situation is catastrophic, the ability of the legislature to arbitrarily raise taxes to cover that is substantially less than Illinois.

          • A Definite Beta Guy says:

            My mistake. Not sure why I said “higher.” Colorado is about equal with IL (4.95%). Colorado also has a lower sales tax.
            I’d describe Colorado as low-tax, but I wouldn’t rely on that given the poor fiscal position. TABOR can be loosened or reversed with subsequent votes, and already has been slightly loosened.

            I’d rather rely on actual sound finances than laws that might change. See: Illinois flat tax (high probability of getting axed by the end of this year). And why not, when you have so many other options available?

          • EchoChaos says:

            @A Definite Beta Guy

            I’d describe Colorado as low-tax, but I wouldn’t rely on that given the poor fiscal position. TABOR can be loosened or reversed with subsequent votes, and already has been slightly loosened.

            Definitely not trying to sell you on Colorado, just curious because it didn’t match my priors.

            I’d rather rely on actual sound finances than laws that might change. See: Illinois flat tax (high probability of getting axed by the end of this year). And why not, when you have so many other options available?

            Absolutely. Fortunately not looking to switch states anytime soon, but there are other much better options.

    • Schools are a major constraint in your calculations. Have you considered the option of either a private school or home schooling (or home unschooling)? That might let you choose an option that’s much better in other dimensions than what you are now looking at.

    • Plumber says:

      @A Definite Beta Guy says:

      “…you can actually walk to a pub from your house, and the pub has actual GOOD beer!…”

      1) If they also serve corned beef sandwiches at the same place as the good beer,
      2) the air and water doesn’t taste “off”,
      3) have nearby bookstores, and libraries
      4) nearby hardware stores,
      5) you hear birds more than gunshots and sirens,
      6) enough sunlight in winter, and enough darkness in summer,
      7) enough nighttime lows of 50℉,
      8) enough daytime highs of 70℉,
      9) the street names don’t continually change even though they’re the same damn street ([expletive] you Santa Clara County!),
      10) you can still feel which way north, south, east, and west is…

      …sounds like a good place.

      • Nick says:

        2) the air and water doesn’t taste “off”,

        When you say this I’m reminded of the line in Tom Lehrer’s Pollution,

        The breakfast garbage that you throw into the bay
        They drink at lunch in San Jose!

        • Plumber says:

          @Nick,

          San Jose is precisely a place I had in mind that doesn’t fit my list above (though at least two excellent bookstores there ten to twenty years ago that may still be in business)

          When I worked there, like much of California, San Jose had both sprawl and apartment tower blocks, but there were a few nicer pre-WW2 early 20th century style “streetcar-era” neighborhoods, but the heat, air, and water!

          Which brings me to one of my big gripes with “YIMBY’s”: that instead of improving sprawl they replace “streetcar-era” with brutal Clockwork Orange towers, they point to sprawl as what they’re against yet that’s not what they replace, instead they destroy the in-between walkable but not skyscrapers and strong winds neighborhoods that are left of Berkeley, Oakland, and San Francisco!

          Lying liars who lie with lies!

          Build in Mountain View, Palo Alto, and the sprawl parts of San Jose instead of destroying here YIMBY’s!

      • A Definite Beta Guy says:

        Unfortunately, we are pretty weak on our corned beef here, along with practically all the rest of our standard deli meats.

        I will say Corned Beef Factory in West Loop (Chicago) has the best corned beef I’ve ever had in the Midwest. However, it’s a cash only joint. Plan accordingly!

    • hls2003 says:

      My family and I are looking at literally the same situation (well, almost literally, our professions are different) with the Illinois death spiral, and our top escape candidates are nearly isometric with yours. However, we haven’t gotten as in-depth on the specifics, so thank you for doing these write-ups. They are one of the most interesting things I read on here.

      For us, we are likely not doing public school, are more interested in space than urban walkability, and have “major airport” accessibility higher on the list due to various family reasons. So of our top 5 (which currently include, though this could change, WI, TN, KY, NC, and CO), I rate TN slightly higher than you. However, I appreciate the crime number warning – I will have to look that up. WI gets points for family proximity. NC research triangle looks better when you’re not trying to commute or worry about schools. KY looks really good except for its horrendous pension problem. CO is not really moving to a lower cost-of-living jurisdiction; it seems like one of the few metro areas more expensive than Chicago. But it does have access to the Rockies. Right now I’d probably rank our choices as TN, WI, NC, CO, KY.

      Also, note that global warming should eventually make WI more like IL in terms of climate. Which… well, I do at least like having the full panoply of changing seasons.

    • andrewflicker says:

      So my wife’s applying to PhD programs, and one of her top picks is Northwestern. Assuming we take as givens that I agree with your overall points about Illinois and Chicago’s terribleness, how do I best maximize my enjoyment / minimize my pain in moving to the area?

      Specifically, what areas are worthwhile? Should I buy a house, or rent? If I buy, what ends of the market give you the best bang for your buck? (ie, is it better to buy a fixer-upper in a good area, because areas are stable and hiring laborers is cheap? Or is it better to buy a solid house in a terrible area, because neighborhood dynamics are changing rapidly and trying to guess what will be a good area in 5 years is a crapshoot? etc)

      Special note: No kids or desire for kids, so I don’t care about school quality/access.

      • hls2003 says:

        If you don’t expect to stay in the area after your wife’s graduation, I would strongly consider renting rather than buying. Property values have been flat or falling across a lot of Chicagoland, and are projected to remain so. You don’t want to be tied down trying to sell if you’ll be gone fairly soon and can’t time the market.

        Most areas in the North Shore are going to be pretty acceptable, and would be accessible to Northwestern by the Metra UP-North line, but they’re very expensive. Lake Forest, Winnetka, Highland Park will set you way back. You could head all the way up to Kenosha, WI and still be on that line, if she doesn’t mind a long commute.

        Also, you may not care about kids / schools, but others do, so it affects your resale value. That’s not terribly relevant if you plan to stay very long-term, but if you plan to move shortly after she concludes (which she will if she wants to get a job in academia and you’re moving with her), then it will affect your potential buyers.

        In my opinion, don’t buy in a bad area. In Chicago, being in a bad area means a lot of crime. Chicago crime rates in the aggregate are high, but they tend to be quite localized, in part due to historical trends in segregation etc. If you are literally in a “bad area” then your crime rate is going to be much worse than you would project even just from looking at Chicago’s general crime stats.

        It’s expensive to hire out remodeling. If you can do it yourself as a hobby, knock yourself out, you might get some sweat equity, but don’t plan on making back your money if you’re hiring a contractor. You might want to do it anyway for your quality of life, but it won’t pay for itself.

      • A Definite Beta Guy says:

        I don’t know the specifics about PhD programs….for instance, how long they last, your stipend, whether your stipend is taxed and how much, etc. Plus, specifics about YOUR work conditions/career.

        I’d go with renting, especially since buying exposes you to future transaction costs AND property tax increases.

        Staying close to Evanston removes the worst crime spots. Rogers Park isn’t great, but it’s no Garfield.

        Northwestern is accessible via:
        CTA: Purple line runs right through, and Red Line transfers to Purple Line. That gives you a ton of neighborhoods within Chicago you can pick.
        Metra: UP North runs through. Again, gives you more options, most of which are in the suburbs (Metra is regional rail compared to CTA being a city rail). However, most of these suburbs are pretty expensive until you get further North.
        Driving: Skokie, Morton Grove, and Golf are all close by and a bit cheaper than Evanston itself.

        The area isn’t bad by any means, it’s just that you should probably expect some tax increases, and property is especially easy to tax.

      • Le Maistre Chat says:

        If you have a car, best bets are Skokie, Morton Grove, Niles, Golf.

        Is Downers Grove as depressing to live in as it sounds?

    • Doesntliketocomment says:

      It seems you have already ascertained this, but Tennessee public schools are abysmal. If you move there you will almost certainly need to enroll your children in a private school, and account for that in your budgeting.

    • unreliabletags says:

      Whitefish Bay are Shorewood aren’t New Urbanist, but they’re pre-war. Kids walk to school on a grid of neighborhood streets with trees and sidewalks. Most of the other Milwaukee suburbs you listed are postwar exurban sprawl with 1/10th to 1/3rd the density. They feel very different on the ground.

      Whitefish Bay in particular is a magnet for upper-middle-class families that are really into education; the property taxes are kind of the point. Voters bought the high school a new music wing at the height of the Great Recession. Real estate is on the pricey side, but the housing stock is also pretty heterogeneous. There are more modest homes well below those averages scattered throughout.

      You don’t have to go all the way to the outer ring, and I wouldn’t. Brookfield, Wauwatosa, Menomonee Falls, West Allis, etc. are much more integrated with the region, places you would run an errand in. Pewaukee is more of an unfortunate Saturday morning youth soccer road trip obligation. Grafton does have the best Target.

      All that said, Madison definitely has more street cred; I can’t imagine someone choosing the Milwaukee metro area of their own free will.

    • profgerm says:

      North Carolinian reporting in!

      As Evan mentioned, if you’re checking commutes with Google Maps, make sure you’re looking at the peak times and paying attention to the directionality of traffic. And if at all possible, look for the accident hot zones: I live near Raleigh, and while my overall commute isn’t bad (20 minutes average, but by car) there’s a 1 mile stretch near a split that is for reasons I have yet to determine a hot zone and that can easily double my commute.

      Other factors: biking in the area sucks (off the greenways, anyways; those are nice) and the drivers are virtually all rubbernecking nightmares. That hot zone I mentioned? 3/4 of the time the wreck isn’t even on my side of the road!

      Semi-anecdotally, even the not-as-high-ranking schools of Cary are likely better than most of those in Durham and Raleigh. If it’s on your radar at all, homeschooling is massive in NC (I think 3rd in the country?), and that does make it fairly easy to find groups to work with and sports in which to participate. Cary also has a fairly extensive selection of Asian markets.

      Cary’s commutes probably rely on 540, which is convenient if it works out for the locations you need, but be aware part of 540 is tolled.

      City of Raleigh and Duke University recently pissed away plans for a triangle light rail, so I will never forgive them. Somewhat bizarrely, plans for a multi-billion-dollar commuter train connecting Garner (a poor suburb south of Raleigh) to Durham seem to be going forward, though I expect lots of gentrification complaints will follow.

      Durham has spent the better part of 3 decades improving its downtown area, and it shows.

      The food scene is… it depends what you’re looking for. It’s good, by my standards, but don’t expect too much molecular gastronomy (the Triangle is pretty solidly developing farm-to-table type places).

      I have little praise to add to the area. I like living here, and I lucked into a decent job and a great price on a house. But there’s also something… a bit generic about Raleigh and Cary in particular.

      • acymetric says:

        Cary’s commutes probably rely on 540, which is convenient if it works out for the locations you need, but be aware part of 540 is tolled.

        Really? This is probably true for Apex/Morrisville, but I wouldn’t expect most Cary commutes to use 540 (440 is probably more likely, which is not tolled at all), and certainly not the tolled portion (only a small part of 540 is tolled).

        But there’s also something… a bit generic about Raleigh and Cary in particular.

        Cary (and, by extension, Morrisville) are definitely very generic, even if there are some fun spots there. I think Raleigh itself (the downtown area, not like North Hills or some other suburb area) has a pretty nice/distinct character.

        • profgerm says:

          Could be; I actually live closer to Youngsville but take 540 to Cary pretty regularly for the mentioned Asian markets. I might be getting some portion of it mixed with 440.

          Of the “three downtowns,” I think Durham is my favorite. Interesting enough, walkable, decent amount to do or at least to browse. Raleigh has the better museums but the rest of the downtown doesn’t have as much charm to me.

          Charm’s not everything, of course; Savannah has much more downtown charm but I’d never want to live there.

          • Evan Þ says:

            I agree Durham’s my favorite of the three downtowns, even though I wish I’d spent more time in the Raleigh museums – they’re great now, and I’m pretty sure I would’ve loved them when growing up too.

    • AnarchyDice says:

      If you’re looking at Madison area, you should also check out Waunakee. Schools on par with Middleton but we’re north of Madison instead of west. Lots of new development, excellent schools (one of our reasons to move here when looking for houses while living in Middleton), and a very active suburban community with tons of events. Plus, you can get downtown in 20 minutes for just about any urban restaurants/cuisine or if you head 20 minutes in the other direction you can go out to rural areas just as easily.

  67. AlexanderTheGrand says:

    Moonlight: a beautiful movie about the inertia of trauma and self-perception, and the dangers of isolation. I thought this movie was profoundly sad.

    Get Out: Best horror movie of the decade. Good, scary, funny. Social commentary wasn’t as heavy as some people make it sound (or maybe I’m just dense).

    The Heat: Maybe I’m mis-remembering (given the reviews), but I thought this was the funniest buddy-cop movie since Rush Hour.

  68. Conrad Honcho says:

    Related to the discussion of movies below, and my lamentation about the lack of recent comedies, what would you say is the funniest movie of all time?

    I say Tropic Thunder, with honorable mention to the collected works of Mel Brooks.

    • Business Analyst says:

      I don’t think I’ve ever laughed as hard as I did through the first time I saw Top Secret.

    • brad says:

      Probably Office Space, but I’d agree with your honorable mention.

    • johan_larson says:

      “A Fish Called Wanda” was pretty funny.

    • EchoChaos says:

      what would you say is the funniest movie of all time?

      Of all time, definitely Duck Soup, which creates so much laughter that you will miss the next joke unless you rewind and watch it again. Animal Crackers is a close second.

      The Marx Brothers were so far and away funnier than anyone else in movie history that it’s barely competitive.

      I definitely agree that comedy is in a down period right now, but I think that’s just the country’s mood, not any inherent political swing.

    • Machine Interface says:

      A Clockwork Orange, which I subjectively consider the greatest satirical film, black comedy ever made. Close second, in the same category, Devils on the Doorstep. Then a quintuple of Ghostbusters, The Blues Brothers, Silent Movie, Young Frankenstein and Monty Python and the Holy Grail.

    • A Definite Beta Guy says:

      Animal House, with a nod to Meet the Parents as being criminally underrated.

    • John Schilling says:

      I was assured in the last open thread that it was My Cousin Vinny, and that is a respectable choice. But I’m going to have to go with Young Frankenstein.

    • cassander says:

      It was decreed long ago that Dr. Strangelove wins all contests for which it is eligible.

      • Conrad Honcho says:

        I would call that more satire than comedy. It is definitely one of my favorite movies, but I wouldn’t consider it a comedy.

    • LesHapablap says:

      Hotshots! is the best parody
      Dumb and Dumber is the best comedy

    • Liface says:

      If funniest = made me laugh the most, then Happy Gilmore.

    • Kuiperdolin says:

      Pain and Gain

    • Bobobob says:

      A Mighty Wind (which scene for scene I think is funnier than This Is Spinal Tap).

    • gbdub says:

      Tropic Thunder Is in my Top 5 or so, but I prefer Hot Fuzz, Dr Strangelove, and Blazing Saddles

    • Aapje says:

      Safety Last (1923) is good.

    • Paul Brinkley says:

      It should be pretty obvious to any comedy connoisseur that comedy is subjective enough to prevent the existence of any one maximally funny film. That said, there appear to exist multiple films that generate more laugh-utiles than others.

      My usual list, in no particular order, includes many mentioned already, plus others I think you should really check out:
      Monty Python and the Holy Grail
      Duck Soup
      My Cousin Vinny
      Young Frankenstein
      Ruthless People
      Children of the Revolution
      The Princess Bride
      Galaxy Quest
      Soapdish
      Tropic Thunder
      Sausage Party
      Rustler’s Rhapsody
      Get Shorty
      R.E.D.

      • Paul Brinkley says:

        This list is, of course, incomplete. Just OTTOMH, I today realized I forgot:

        Game Night
        Trading Places
        Spies Like Us

        Honorable Mention to Dogma, because it makes Kevin Smith’s story of a Dogma protest one of the funniest ten minutes I’ve ever watched.

        • Randy M says:

          Was game night good? I meant to watch it at some point, since it combines Arrested Development’s Jason Bateman and board games.

          • Paul Brinkley says:

            Highlights:

            Jason Bateman is still IMO the best comedy “straight man” in the business today.

            I found out Rachel McAdams has been criminally underused as a comic actress. She steals the show so many times that Lloyd’s cancelled their insurance.

            It’s not really about games, which is a bit sad – if you’re into tabletop gaming like me, you just have to nod and chuckle at references to old Parker Brothers junk and wait for things to pick up.

            Satisfying long take during a chase scene.

            Jesse Plemons (the cop) was also a wonderfully fun character.

            The side story with Lamorne Morris and Kylie Bunbury was relatively weak, and that’s saying something – they carried it well, and in any other comedy, I think it would’ve stood out.

          • Randy M says:

            Obviously “was it good?” is a stupid question to someone giving their favorites list. Thanks for the highlights.

    • smocc says:

      Kung Pow!. I thought about it and the answer is Kung Pow!.

      Another strong contender is The Trip

    • SteveReilly says:

      The only time I remember bursting into laughter in a theater was during There’s Something About Mary. Matt Dillon is absolutely hilarious.

      Though I think Woody Allen playing a cello in a marching band might be my favorite scene of all time. Take the Money and Run was definitely my favorite movie growing up.

    • HeelBearCub says:

      I don’t think I’ve ever laughed harder than Raising Arizona.

      I think “best comedy” is tough, because many of the funniest comedies aren’t nearly quite as good on rewatching, as you can’t be surprised by the joke.

      • Also, some of the funniest scenes in movies comes from movies that are primarily not comedies because it comes out of nowhere.

        • HeelBearCub says:

          Agreed.

          Although there is a weird tension there. Sometimes knowing the punch line makes you anticipate more, makes it better.

          There is a scene in Total Recall that ends with the punch line, “Consider that a divorce.” It was a scene that was in all the trailers.

          A bunch of my HS buddies all go to see the movie together when it comes out, and one of them blurts the line out about a second early. Someone els wacks him up side the head and tells him we “payed $5 to hear Arnold say it.”

      • Plumber says:

        Raising Arizona

        +1

        • HeelBearCub says:

          @Plumber:
          Here is one that you might have watched a long time ago. I’ve only seen it because my Mom loves it and rented it for us to watch, and every now and then she likes to ask to watch it again.

          The Court Jester starring Danny Kaye. She says it’s not the same if I’m not watching it with her and laughing at it.

      • The Pachyderminator says:

        Raising Arizona is a great, great movie. One of the only movies I’ve found, maybe the only one, where the cliché “I laughed, I cried” is literally true.

    • Jon Gunnarsson says:

      Probably The Life of Brian. I’ve seen that one at least a dozen times and it’s still funny every time.

    • Plumber says:

      @Conrad Honcho >

      Tropic Thunder

        Tropic Thunder was hilarious, and I can’t think of any movie made since that made me laugh as much!

      Television has delivered big laughs lately, particularly Broad City, and Silicon Valley

      I agree with most of the suggestions in this subthread (particularly Duck Soup, Raising Arizona, Take the Money and Run, and Young Frankenstein), but I noticed no one (so far) has mentioned the 1981 Bill Murray, Harold Ramis, John Candy U.S. Army comedy Stripes, or the 1986 Magnificent Seven-ish western parody Three Amigos, both very funny. 

      If I had to pick a comedy to see again right now I’d choose 1967’s “Faust in Swinging London” Bedazzled (not to be confused with the lame in comparison 2000 re-make), which has the immortal line You fill me with inertia“.

    • Trofim_Lysenko says:

      For me personally it’s a tie between Hot Fuzz and Young Frankenstein.

      • Conrad Honcho says:

        Hot Fuzz is superb, but I’d call that more like “best comedy” rather than “funniest.” The first half of the movie was okay, with a few yucks. It’s not until the second half you realize the entire first half of the movie was one giant set-up for the hilarious second half.

        • Plumber says:

          @Conrad Honcho, 
          FWLIW the movies released in the last two decades that I remember laughing at and what their titles were are:

          Dude, Where’s My Car?
           (2000 )

          Hot Fuzz
           (2007)

          Tropic Thunder
           (2008)

          Zack and Miri Make a Porno
           (2008) 

          The Hangover
           (2009)

          Moonrise Kingdom
           (2012)

          and

          The Grand Budapest Hotel
          (2014)

          with Hot Fuzz and Tropic Thunder just about tied for most laughs

    • Orion says:

      Die Schweizermacher (1978)

  69. johan_larson says:

    A quirky mobster offers to make you a deal. They’ll figure out your likely lifespan based on lifestyle factors, ancestry, and your current health. And they’ll kill you one year before your predicted date of death. In compensation, they will give you a negotiated sum of money up front. If you die before they would have killed you, nothing happens, and you don’t have to give them any of the money back. And let’s suppose you have reason to believe this guy is trustworthy.

    How much money do you want for the last year of your life, statistically speaking?

    • Conrad Honcho says:

      What’s in it for the mobsters? I’m not sure I understand.

      Also, do I get to know my predicted death date before I commit to the deal?

      • johan_larson says:

        The mobsters aren’t talking. Maybe one of their kids is doing a science fair experiment. Who knows?

        Yes, you do get to know your predicated death date before accepting or rejecting the deal. It won’t be based on anything mystical, just standard medical and demographic analysis.

        • Conrad Honcho says:

          Then it entirely depends on what number they give me for my life ending. If they say 110, fine, give me $10 million. Another ~70 years max as a multimillionaire is worth giving up a year as a middle class 109-year-old, and they’re probably wrong anyway. If they say two years from now that’s a hard pass.

          • johan_larson says:

            This answer looks pretty good to me. A year or a few years off the end of my life for a life-changing sum of money now is a good deal, since I’ll have a good 35 years or so to enjoy myself.

          • acymetric says:

            Yep, for me it’s basically “enough money that I can immediately quit my job and live a fun, fulfilling life until they come for me”. Which means I want more money the longer I’m going to live…if they gave me 2 years (so they kill me in 1) probably $500k would do it, although I’d push for more like $1 million. If it is 40 years, $10 mil is probably the rock bottom price. The shorter the amount of time, the more money per year I’ll want, because i’ll need to really live it up in that year.

            My answer for 2 years would probably be very different if I had kids.

    • EchoChaos says:

      They’ll figure out your likely lifespan based on lifestyle factors, ancestry, and your current health.

      How high are they willing to go on ancestry? I have ancestors who have lived into their 110s, and it’s not uncommon for them to live into their 100s. One of my ancestors is on this Wikipedia page. https://en.wikipedia.org/wiki/List_of_last_surviving_veterans_of_military_insurgencies_and_wars

      Because if they give me a silly number in the 70s or 80s, I’m betting they’re taking my last couple decades, not years.

      • johan_larson says:

        It’s an honest analysis based on the best available current information. If you want a number just for the sake of argument, you might take the average lifespan of your grandparents, assuming all of them have passed by now.

        • EchoChaos says:

          Mostly I’m morally opposed to being killed regardless, I just enjoy bragging about how much my family is an outlier on actuarial tables.

          In all honesty, I wouldn’t accept the deal.

    • Noah says:

      It’s taking off more than a year on average, because of the variance of your age of death.

      • Jon S says:

        This. A generic person born in the US in 1952 had a life expectancy of 68.5 years. If they’re alive today, they’re now 68 and have a life expectancy around 17 years (using the SSA’s 2016 actuarial tables, 15.8/18.1 years for male/female).

        A better approximation of this deal is “half the time they pay you for nothing, half the time they take 10-15 years off your life”.

        • acymetric says:

          I thought it was implicit that they had a better projection than just “life expectancy of average American” with the project very much tailored to you.

          • johan_larson says:

            Yes, I’ve had the same thought. How well could we predict a specific person’s life expectancy, assuming we are willing to go to great trouble to do so?

            @SamChevre to the red phone, please. @SamChevre

          • Jon S says:

            I also took it to be that their projection was better, but not sci-fi level. If it’s not sci-fi level I think they’re still taking 10+ years off your life pretty often.

          • SamChevre says:

            If I have time, I’ll try to get an answer tomorrow – but yes, the variance is going to be pretty high–much more than a year. Accidents take a lot of years off lifespans (I’ll assume those are unpredicted), as do predictable diseases (I’ll assume those are predicted.)

            So what I’ll calculate are expected years of life, for 4 scenarios: population mortality, population – accidental, heavily underwritten, heavily underwritten – accidental.

    • cassander says:

      I’m betting/hoping that I’ll live a fair bit longer than my currently projected lifespan due to improved medical technology.

    • bullseye says:

      My last year of life is probably going to be lying in a hospital bed full of tubes. If I believe their analysis, I’d do it for enough money to retire now and live well. (Realistically, though, I’d say no because this is clearly some kind of weird scam.)

      • Conrad Honcho says:

        Realistically, though, I’d say no because this is clearly some kind of weird scam

        Well, yeah, that. And it’s probably a sin to…get paid to have someone murder you, I guess..?

      • toastengineer says:

        Are you/they taking advancements in medical science in to account? I expect I’ll get significantly more positive-worth years tacked on to the end than any of my ancestors.

    • NoRandomWalk says:

      I love how in the first 10+ comments no one engaged with the question in the spirit it was posed.

      If i am killed 1 year before expected lifespan, i expect that will take 6 years off my life in expectation, and will live for another 70 or so.

      Would agree to it for about 300k/year, so 1.8 million

      • johan_larson says:

        Would you explain where that factor of six is coming from?

        • NoRandomWalk says:

          Let’s say you expect to have an equal chance of living to any of 68-90. In expectation, you will live to 79. So mobster kills you at 78. So, you have an equal chance of losing between 0 and 12 years, or in expectation, 6 years.

          • johan_larson says:

            You don’t seem to be accounting for the cases where you die between the ages of 68 and 78, which is almost half on them. In those you don’t lose anything at all.

          • NoRandomWalk says:

            What would it mean to ‘account’ for them?
            ‘on average’ I lose six years. Sometimes I lose 12, sometimes I lose 0. So I estimated how much I value my last year, and multiplied by 6.

          • Noah says:

            Given your figures, almost half the time you lose 0 (if you died before he would’ve killed you). Remember, in these cases you still get paid. The rest of the time, you lose uniformly 0 to 12 years. So expected value is 1/2*0+1/2*6=3 years (I did some rounding to whole years for convenience, but your statistics don’t remotely correspond to actuarial tables anyways).

          • NoRandomWalk says:

            Ah yes! You are right, thank you. Assuming would lose expected 3 years, again at 300k, would do it for $900k then, not $1.8mm.

      • Dacyn says:

        I love how in the first 10+ comments no one engaged with the question in the spirit it was posed.

        I am not really sure what you mean by this.

        • NoRandomWalk says:

          I’m not sure either; in a typical conversation ‘how much would you value your last year of life’ would be answered with a number without asking for clarification, and that this community is mostly interested with defining terms is something I found highly endearing and emotionally salient in the moment.

          • Dacyn says:

            Ah, so by “the spirit it was posed” you meant “the spirit someone outside this community might assume it was posed in”. Makes sense.

    • Theodoric says:

      I would likely not take the deal. Maybe if I had reason to believe the last year of my life would be really miserable (painful terminal cancer, say) and if they were offering enough that I could have a Jeff Bezos lifestyle without having to work.

    • Nornagest says:

      I’ll take a million bucks, invest and/or spend as I normally would, and use whatever’s left on buying protection from a rival group of mobsters at age N-2.

    • Statismagician says:

      What’s the lookback period for lifestyle factors? If I get a little lead time, I can artificiality inflate my healthiness and then go back to eating cheeseburgers, putting my eventual death due to CVD safely before my pre-arranged mob assassination.

      • johan_larson says:

        You’re going to lie to The Mob? Your life expectancy may be lower than you think. 🙂

        But seriously, I’m thinking maybe a week of work, divided among several expert professionals, trying to figure out how healthy you are, how long your relatives have historically lived, and how you actually live. Part of that would be asking you how you live, and part would be examining hard data about how often you actually get to the gym, what you actually buy at the grocery store, and what restaurants actually show up on your credit card. People routinely lie/exaggerate/delude-themselves about this stuff.

        • Statismagician says:

          Amusingly, I may end up helping figure out when to murder me (I do administrative-data epidemiology). Due to family history I don’t expect my last year of life to be worth a whole lot; I’ll try to get as much as possibly, obviously, but I’d probably take it for something on the order of ~$100,000, the usual cost of a QALY.

    • Doesntliketocomment says:

      Taking it for free is a good deal. Based on people I’ve seen living their last years of life, I would expect to have to pay them for this service.

    • Hyperfocus says:

      If the sum of money is in the millions, I’ll take the money. Then, assuming the mobster is a snail, the first thing I do is commission a large tungsten sphere…

    • Orion says:

      $100k.

      I’m 30, and from a fairly long lived family. Let’s say I’m projected to live until 80. So they’re asking for 2% of my projected remaining lifespan. However, several of my grandparents had very bad quality of life in their last years, so I’m probably trading away way less than 2% of my projected remaining utility. So basically, any amount of money sufficient to improve my future quality of life by any noticeable amount would be sufficient. I would like to go back to college and get a BA, but figuring out how to make it work is pretty daunting. 100k would be enough to simplify things greatly and might help me graduate a year sooner, which would definitely be worth it.

  70. johan_larson says:

    Arrival
    The Martian
    Inception
    Ex Machina
    Edge of Tomorrow
    The Big Short
    Margin Call
    Sicario

    • Liface says:

      Always great when someone posts a list with 7 movies you love and one you’ve never heard of. Will be checking out Margin Call as soon as possible.

      • Jon S says:

        Ditto for Sicario (wouldn’t say I *love* all the rest, but enough of them that I want more johan_larson recommendations).

        • johan_larson says:

          Happy to oblige. Here are some of my favorites:

          The Apartment
          All Quiet on the Western Front (1979)
          The Terminator
          Aliens
          Master and Commander: The Far Side of the World
          2001
          A Fish Called Wanda
          Batman Begins
          Saving Private Ryan
          Star Trek II: The Wrath of Khan
          Clueless
          Primer

    • Plumber says:

      @johan_larson says:

      Sicario

      I may have just been tired by then, but I found Sicario really boring as an action-thriller (with political, betrayal, and suspicion elements) compared to ’71 which I saw the same night and thought was a far more engaging film.

      Check out this scene:

      https://m.youtube.com/watch?v=dAogFVi22Fk 

      How accurate it is or isn’t to history is definitely a fractional thread discussion, but as a movie it was by far the best action-thriller that I’ve seen this decade, 1/2 Hamburger Hill, and 1/2 Tinker, Tailor, Soldier, Spy.

  71. Le Maistre Chat says:

    Herodotus claims that the walls of Babylon formed a square 120 stadia on a side. The metric equivalent of a stadion is disputed, but the lowest plausible number is a bit under 158 meters, making a 120-stadia Babylon 19 km on a side. That’s 36,100 hectares.
    In reality, archaeologists have gone to the site and found that it covered 900 hectares at its greatest extent, the Neo-Babylonian and Achaemenid layers. But what if Herodotus was right?
    The lowest estimates archaeologists use for the density of ancient cities is 100 people/hectare. Therefore if Babylon was 36,100 hectares, it would have a population of >3.6 million. Pre-industrial farming required what, 90% of the population to be farmers? 85%?
    What would it take for the area a Mesopotamian city could collect grain from to grow enough food to feed 24-36 million people? Even assuming you can safely ship food downstream on the non-navigable parts of the Tigris and Euphrates, we’re talking about an area of Iraq, a narrow strip of Syria, and Susiana (using the last as shorthand for “fertile regions of the Persian Gulf with access to the sea”). Putting that in perspective, modern Iraq has been a net importer of food since the mid-1960s, when the population was ~9 million (cultivable area is about 12 million hectares – assuming none is ruined by soil salinity, lack of irrigation water, or drought in the rain-fed area – and about as much suitable for permanent pasture).

    • bullseye says:

      Maximum size of a pre-industrial city seems to have been a little over a million, and some of those cities were ruling big empires. At some point your territory gets big enough that you need additional cities to administer it, plus you have to feed the people who transport food from increasingly distant lands.

      • Yeah and getting to one million was an extraordinary achievement. I believe Rome and Constantinople were the only cities from Western Eurasia to get there before the Industrial Revolution.

    • Another Throw says:

      Let’s attack your question from the different angle.

      The population of London at the time of The Conquest is apparently estimated at approximately 10000, while the population of England as a whole is estimated (at the time of the Domesday book a few years later) at 1-2 million. So approximately 1% of the population of a country can be crammed into the capital city. This would allow a nice margin to populate additional cities for administrative purposes.

      So you’re probably looking at the population of Babylons catchment area being an order of magnitude higher than you’re talking about.

      Okay, let’s try something else.

      A hide is a medieval unit of land value assessment that is generally considered to have consisted of the land required to support one family, produce of annual value of 20 shillings, and be able to be routinely taxed at a rate of 2 shillings and/or (whether it is and or or is unclear to me) every 5 hides should turn out and supply a military age male for other purposes. As a gut check, if each “family” consists of 4 persons, that would leave 95% of the population dedicated to agriculture, so my gut isn’t too upset so far. From this estimate, detaching 3.6 million people from agriculture would require 18 million hides. In suitably arable land, a hide is estimated at 120 acres. This works out to… 8,741,209.87 km2 of arable land. For reference, China is about 9.5 million km2.

      And that estimate is based on suitably arable land so the actual required area will be much larger.

      Not looking promising.

      Okay, let’s try something else.

      In Roman times, England is estimated to have had a population of around 4 million (according to one of the previous links). We’ll use the Roman era estimate as one for a reasonably administered empire. The reasons for the post-Roman decline are probably outside of the scope here. The area of modern England is 130,000 km2. Scaling to your catchment area population estimate of 36 million would require 1.17 million km2. Scaling to my catchment area population estimate of 360 million would require 11.7 million km2.

      Taking a 3 million km2 fudge factor for “suitably arable” doesn’t seem obviously out of the question, so this probably comports well with our previous estimate.

      These approaches are not looking promising for your ability to make Babylon that large with pre-modern technology. Arguments about the relative arability of ancient/medieval England and ancient Mesopotamia are an exercise for the reader.

    • Kuiperdolin says:

      Could the walls have enclosed more than the city proper, fields and orchards and pastures ? By the time of its fall Constinople was mostly walled countryside, so it’s not a completely outlandish notion.

      • Statismagician says:

        I strongly suspect it’s a bit of this and a bit of confusion over units – the other classic ‘ha-ha, that silly Herodotus, Father of Lies’ things are the giant gold-digging ants, which turned out to just be a thirdhand account about marmots related through a language where ‘marmot’ and ‘ant’ sound really similar, and the Egyptians having what everybody thought were impossibly-weird ships, until they found one. The guy gets an unfairly bad rap, is what I’m saying. ‘Wall’ having been used for ‘boundary-marking berm,’ possibly?

      • Le Maistre Chat says:

        Could the walls have enclosed more than the city proper, fields and orchards and pastures ? By the time of its fall Constinople was mostly walled countryside, so it’s not a completely outlandish notion.

        Yes, with the provisio that Herodotus never visited Babylon in person and the large circuit was a low bern rather than the defensive wall that empirically enclosed about 900 hectares. That smaller circuit would be where the famous Ishtar Gate, etc. were. H. gives his large number as part of a description of this defensive wall, which was broad enough for a four-horse chariot to patrol.

  72. lndk483 says:

    Is anyone planning to go to PAX south in San Antonio next weekend? If any one is and wants to have a slatestarcodex meetup email me at lndk483@protonmail.com or reply here.

  73. brad says:

    Looking back, I didn’t watch very many movies during the 2010s, even more so not in theater / shortly thereafter.

    With some help from the internets, I think these are substantially all of the 2010s movies I’ve seen. They are heavily weighted towards the first half of the decade.

    In the theater:
    Harry Potter (last one)
    Hunger Games (etc.)
    Guardians of the Galaxy
    Star Trek: Into Darkness
    Interstellar
    Star Wars: The Force Awakens (ect.)

    Out of the theater:
    Inception
    The King’s Speech
    Moneyball
    Argo
    The Wolf of Wall Street
    Dallas Buyers Club
    Ready Player One
    American Hustle
    Bridesmaids
    The Big Short

    Of these, I remember the most of the movies on the second list more fondly than the first. I don’t think any of them would make a short list for my favorite movies of all time.

    • Conrad Honcho says:

      Bridesmaids

      Is this the last funny movie ever made? I thought Tropic Thunder, but then I googled and found out that came out in 2008!! What happened to the irreverent, medium budget comedy that everyone incessantly quotes? Where is the modern 40 Year Old Virgin, Anchorman, Talladega Nights?

      • brad says:

        I’d guess that humor doesn’t translate as well as big explosions in space and people in tights punching each other.

        • Conrad Honcho says:

          I don’t think that’s a good explanation. According to IMDB, the estimated production cost for The 40-Year-Old Virgin was $26 million. I’m assuming that’s mostly actors’ salaries because there’s no CGI, no foreign locations and the sets are an electronics store, bachelor apartments and some nightclubs. The worldwide gross was $177M. For the cost of making one Marvel or Star Wars movie (~$250M), you could make ten of these. And even Star Wars movies aren’t making 10x$177M.

          In 2020, there is a $150M bill on the sidewalk: make a damn comedy movie.

        • CatCube says:

          I wonder if we’re just not seeing the effects of inflation and cost disease in movies, combined with the “creative accounting” that has always plagued the industry. A $25 million movie would be the rough equivalent of a $1 million movie in the early ’80s; I read “My Hollywood MisAdventures” by ex-screenwriter Allan Cole over the holidays, and he discussed why there were so few $1 million movies at that time:

          I turned to Godfrey, “So tell me Al, if you can make a million dollar movie – one that’s guaranteed to turn a profit – why don’t the studios make ten, one million dollar movies that will all make a bundle… instead of one ten million dollar movie that’s probably going to lose money?”

          Godfrey gave me a pitying look – oh, you poor putz. Then he proclaimed, “Allan, the reason the studios make Ten Million Dollar Movies, instead of One Million Dollar Movies, is that you can’t fucking steal a million dollars from a million dollar movie.”

          Most movies, even those that pull in ticket receipts well over their cost of production, never make money on their balance sheet; Star Wars, for example, AFAIK has never been profitable (in the sense that they’re required to pay out the percentages of net profit that people who worked on that project had negotiated for). The money to be made on a movie is extracted by the people involved in it before it hits what an accountant would call “profit” on a balance sheet. It may be that the kind of midrange movies you’re talking about don’t have room for the kinds of accountant fuck-fuck games that enable this given the modern value of a dollar.

        • Don P. says:

          1. Possibly old news to everyone here, but I recently learned that Alec Guinness got 2.25% of the gross for Star Wars, which…good job, there.

          2. Ed Solomon, writer of Bill and Ted (second sequel coming out this year!) and, more to the point, Men In Black, complained on Twitter the other day that he is still getting statements showing that MIB (the first one) has not made a profit, adding “yes, that’s why they made three sequels…because it lost money.”

        • Tarpitz says:

          I just read the best unproduced screenplay I have yet come across. It’s a comedy. It’s the funniest thing in any medium I’ve read in years. I really hope we can pick up that $150m bill.

      • broblawsky says:

        Booksmart is great, and it came out last year.

        • johan_larson says:

          Is that anything like “The To Do List”? I tried watching “The To Do List,” but found it too dirty.

        • broblawsky says:

          There’s underage drinking and drug use, and some talk about sex, but nothing too bad. It’s no worse than a generic 80s high school comedy.

        • Thomas Jorgensen says:

          It is not just funny, it is also a very kind comedy. Which is really, really damn hard to do. Strongest of recs.

          Honestly, I think the limiting factor on the production of comedies is the availability of good scripts for them – finding the acting talent for them is essentially never the failure point, but if the script is not up to scratch, you get terrible movies.

          .. So, I guess what I am asking is this.. can anyone think of a series of comedic novels, comics or novellas that would translate well to the silver screen, because there is a lot of money to be made here.

      • The Pachyderminator says:

        This is part of a more general trend of mid-budget movies disappearing from theaters. Critics more often lament the serious dramas, but comedies have probably been the hardest hit. Maybe they’ve moved to TV? I don’t really know what’s on TV these days.

        It seems to be related somehow to the consolidation of the studios into ever bigger media conglomerates. Perhaps Disney doesn’t want to bother releasing $25 million movies just as Coca-Cola doesn’t operate roadside lemonade stands.

        • Theodoric says:

          Maybe the mid-budget films will migrate to streaming, with perhaps a very brief theatrical run to qualify for awards? According to Wikipedia, Marriage Story (Netflix) had an $18 million budget and The Report (Amazon) was $8 million.

        • AG says:

          Yes, comedy is stronger on TV, because longform enables more running or shaggy dog gags. Parks and Rec, Brooklyn 99, The Good Place, Crazy Ex Girlfriend, Broad City, The End of the F***ing World, American Vandal, Veep, Documentary Now, Unbreakable Kimmy Schmidt, the list goes on.
          But even more than that, quoting comedy is an internet game now. Who needs waste time and money going to the theaters when they have memes right at home/in their hands?

          But there were still some strong comedy films. Notably, some have been produced by Netflix. I don’t really watch movies anymore, so I haven’t seen some of these, but they had good buzz: Game Night, This Is The End, Hail, Caesar!, The Spy Who Dumped Me, Wine Country, Someone Great, Girls Trip, Rough Night, Bad Moms, Sisters, Death of Stalin, One Cut of the Dead, Popstar: Never Stop Never Stopping, 21 Jump Street, What We Do In The Shadows (which also has a TV show), The Heat, The Final Girls, Happy Death Day.
          There’s also the part where pure comedy is rare because comedy has been so absorbed into other genres. The biggest comedies of the decade in terms of box office would be the Deadpool movies.

        • HeelBearCub says:

          The biggest comedies of the decade in terms of box office would be the Deadpool movies.

          This is an interesting point.

          The 1980s was absolutely loaded with action-comedies (and that list is hardly definitive or complete). Movies like:
          48 Hours
          Beverly Hills Cop
          The Princess Bride
          Romancing the Stone

          Really what’s different between now and then is that (mostly) the 80s had “action” movies and the 10s have “super hero” movies.

        • Jake R says:

          If you’re a fan of dark humor I can’t recommend season one of Patriot on Amazon highly enough. Season 2 had its moments too but 1 was a masterpiece.

        • Tarpitz says:

          $18m is low budget, not mid-budget. Netflix are certainly capable of funding mid-budget movies, but Marriage Story is not an example.

      • Trofim_Lysenko says:

        They still exist, they’re just more easily crowded out by the summer AAAAAAAA megabudget blockbusters and the the fall/winter oscar bait.

        If you haven’t already seen them, may I recommend Edgar Wright’s unofficially titled “Cornetto Trilogy”:

        Shaun Of The Dead (2004)
        Hot Fuzz (2007)
        The World’s End (2013)

        Of the three, Hot Fuzz is probably the tightest and funniest, so if you only see one, make it that one.

        More generally:

        Tucker & Dale Vs. Evil (a must if you were ever a fan of horror, especially slasher flick horror)
        What We Do In The Shadows

  74. Joseph Greenwood says:

    “Sunk costs are sunk” is a truism that has stuck with me since my first course in economics, and I feel as though it is a robust and worthwhile insight that has in small but noticeable ways improved my day-to-day life. But a thought-experiment occurred to me today that carves out a narrow but important class of scenarios where it is better to honor your sunk costs.

    Imagine that you have a fixed amount of time, money, and other resources to spend on any of a variety of useful projects. Each of these projects provides some sort of payoff when completed, but no payoffs at all until then–a half-finished project is as worthless as an unstarted project. Being a perfectly rational individual, you choose the project which provides the best return on your efforts, and begin working on it.

    But after working on this project for a little while, a new opportunity arises. This new project follows the same terms as the other ones–no partial payoffs–but it is sufficiently more valuable than the project you were working on that it is rational and appropriate for you to abandon your efforts on the first project (your “sunk costs”), in favor of working on the second.

    Alas, after only a little work on this second project, a third opportunity comes to your attention! This one is enough better than the second one that it is again in your best interests to abandon that project in favor of this new opportunity.

    Then a fourth project comes along. And a fifth. And a sixth. Eventually, you’ve barely-started enough projects that you could have finished at least one of them with the resources you spent… but you haven’t. You have nothing to show for your efforts.

    I don’t think this sort of thing is likely to happen “in nature”–it depends on consistently stringing someone along with new bits of relevant information of exactly the right sort, which is technically possible but seems a priori unlikely. But it seems like this could happen not infrequently when dealing with malicious or manipulative actors. If you spend a lot of time in such situations, a heuristic of “honor sunk costs” becomes correct.

    • The Nybbler says:

      It’s not sunk costs that are the issue here, it’s time-to-payoff. If these projects are kinda Zeno-staggered — that is, 10 years into a 20 year $1,000,000 project, I get an opportunity for a 10 year $2,000,000 project with the same cost per year, then 5 years into that I get a 5 year $3M project, and 2.5 years into that a 2.5 year $4M project and so on, it’s always rational for me to jump ship.

      • Randy M says:

        Although you have to account for discounting due to time, and the fact that far future money matters less as you start to have less lifespan remaining. Presumably you could come up with pay-offs that still make switching worthwhile while the pay-off is within your estimated lifespan and you aren’t starving in the meantime.

    • Dacyn says:

      Certainly we have to act differently around malicious actors, but I am not clear what the nature of the deception is here. Is it that the new projects are not as good as they look, in a way which gets worse over time?

      If you claim the expected utility of the choice to switch is less at the end than at the beginning, I think you should be able to point to at what point in time did the EU decrease, and why.

    • LesHapablap says:

      This could certainly happen by switching careers or spouses, especially if there’s a bias of ‘grass is always greener’ influencing your decisions.

    • ejh3141 says:

      Maybe this sort of failure could be counteracted by incorporating the idea of uncertainty into your model. If I knew that I would switch projects with 100% certainty prior to completion, then my marginal return on that choice drops. But how much it drops exactly is dependent on the chance of completion for each project down the entire chain. Not a realistic thing to predict, but I guess this isn’t a realistic situation anyways.

    • J Mann says:

      I don’t think that’s an exception exactly. Let’s say two people start a project. Jane jumps whenever a better option comes along, and Sam sticks with his sunk costs.

      I think if Jane is calculating sunk costs correctly, she’s always better off than Sam, even in your hypothetical. It’s true that if she and Sam die before her project pays off, then Sam’s better off, but she should take that risk into account when deciding if the new opportunity is better than the remainder of her current project.

      • woah77 says:

        I think the calculus depends highly on the runway involved. If you have infinite runway (that is you will never go bankrupt even if you switch projects all the time) then Jane is probably the rational actor. But most people have finite runway, which makes Sam’s strategy make sense.

        • J Mann says:

          IMHO, Jane should include the runway in her calculations.

          • woah77 says:

            I would say that it is unlikely Jane has the runway to constantly jump from project to project, but if she does, sure, that’s rational. Sam probably knows he doesn’t, which also makes his decision rational.

    • VoiceOfTheVoid says:

      If you’re calculating costs and expected values correctly, then this only works as long as truly better projects keep coming up–and they have to be more than just marginally better, since though we ignore the sunk costs of the current project, it is closer to completion and thus has lower projected future costs. Obviously you won’t keep finding genuinely better projects forever.

      And if the problem is that malicious actors are giving you misleading information to make the new projects initially seem more valuable than they are, then you should figure out what’s going on and account for their deception by adjusting significantly downwards when calculating the expected value of the next project they propose.

  75. tjwhale says:

    I hope this isn’t a culture war question.

    I’ve been reading more climate change stuff recently that implies we’re almost certainly going to get 2c of warming and probably quite a lot more. Moreover the most recent models seems to imply stronger sensitivity to feedback loops which means there might be a lot more warming, a lot faster, than people think.

    I would be really interested to hear thoughts of people around here, are we looking at large scale civilization collapse or do you think that can still be averted?

    When it says here that “Government of India has released report ‘Composite Water Management Index ’ in June 2018 and listed Delhi and other 21 cities in India which would run out of groundwater by 2020.” https://en.wikipedia.org/wiki/Water_scarcity_in_India

    Are we on the brink of an apocalypse or have I just been reading stuff that is too grim?

    • EchoChaos says:

      I hope this isn’t a culture war question.

      This is a culture war question.

    • hnau says:

      Hard to say what kind of “apocalypse” severe climate change would entail. My uninformed guess is that civilization won’t collapse; since there’s room to invest a lot more in, say, desalination (and extend this to all climate-change mitigation technologies) than we currently do, we’ll incur serious economic costs but no critical systems will fundamentally break.

      Modern industrial society has been around for 100-200 years depending on how you count and survived serious economic / environmental pressures like the Great Depression / Dust Bowl in the 1930’s, so there seems to be some evidence that it won’t fold at the drop of a hat. But that evidence might need to be adjusted for selection bias / Black Swan effects.

      • beleester says:

        I am also of the opinion that civilization won’t collapse, but I feel the need to point out that that’s an extremely low bar to set. World War II did not cause the collapse of civilization. The Black Plague and the Spanish Flu did not cause the collapse of civilization (although you could make the case that the Black Plague did fundamentally break the economic system of the time).

        It’s possible for it to be true both that climate change will be a tremendous disaster that costs millions of lives, and that it won’t cause the collapse of civilization as we know it.

    • Plumber says:

      @tjwhale says:

      “I hope this isn’t a culture war question…”

      For what little it’s worth, despite our host’s “please try to avoid hot-button political and social topics” requests, to my eyes the whole number Open Threads look to me to usually have more posts on “hot button” issues than the “culture war allowed if your civil” fractional Open Threads which seem to me to feature more thoughts on Star Wars and Marvel Cinematic Universe.

      Eh, maybe here George Lucas and Stan Lee are more “hot button” than are the usual partisan stuff?

      • Garrett says:

        Movie preferences are likely to result in more arguments for the same reason that arguments over the color to paint a shed will: they’re mostly matters of taste and hard to prove wrong. It’s akin to favorite color or flavor of ice cream.

        It’s very rare to find cases where a social group will collapse or violence will erupt over movie preferences. Sure – there are some small snobbish groups, and undoubtedly you can find a court report where the whole family went to jail over an argument of which movie was better (alcohol may have been a factor …).

        The standard culture war issues usually involve issues where the stakes are high. We’re talking about existential risk for humanity, the complete re-making of the economy, or fundamental life or dignity for some significant set of the population.

      • HeelBearCub says:

        I don’t think your perception is actually correct, but to the extent that it is, you need to factor in that this is the visible open thread. This is the thread that has a higher ratio of irregular to regular commenters.

      • Dacyn says:

        I decided to go through the most recent fractional and most recent whole-number threads and count how many top-level posts I thought were culture war. Results:
        144.75: 39 posts, 9 CW (Trump reelection odds, porn, AI safety, list random political beliefs, death of Suleimani, Elizabeth Warren tweet, HowardHolmes’s post, worst thing your country did, how to respond to religious visions)
        144: 29 posts, 3 CW (Dominic Cummings, quality of SSC comment section, Article 5 of US constitution)

        Obviously these are kind of subjective, e.g. I included AI safety but not the metric system argument. And maybe “quality of comment section” is meta enough it shouldn’t be counted as CW. And, the post about Dominic Cummings did not get any responses so maybe that should be factored in.

    • uffe says:

      Are we on the brink of an apocalypse

      No. You mentioned India and whatever else one might think of the current BJP government, they’d done a very good job on the environment. India will not have as dirty a grid as China, they are beating their own very ambitious renewable energy targets. Furthermore, India’s vegetarian bent means that their diet will continue to have a much lower impact on the environment than their future per capita GDP would imply.

      As for the water issue, yes it is a big issue. This is basically a human political-sociological problem more so than a strictly environmental one, though that can arguably be said of all environmental issues. The big problem in India is their very decentralised and fragmented political system.

      Unlike China which was very centralised from the start, India adopted a more concessionary system which allowed their massive diversity to function. As a result, a lot more legislative power is given to the states and the water issues are a classic example of why that isn’t a great idea unless you have strong internal cohesion. Many Indian states share over water and fight over water. Such as Tamil Nadu and Karnataka over Kaveri river. It is hard to deal with given that the problem often crosses state barriers.

  76. bullseye says:

    Question for my fellow Americans: why do we say the Pledge of Allegiance with that awkward cadence? (You know what I’m talking about, right? It wasn’t just my town?) Maybe it comes from how they teach small children the pledge?

    • FrankistGeorgist says:

      We break the lines at conjunctions and prepositions for easy-to-memorize “phrases” and then the 3 qualities each get a line, and then break at a preposition again.

      So I would say yes it’s how you get rote memorization out of children.

      Of course the real reason is that it is the rhythm which most pleases the Flag, long may she preserve us, wave over us, delight us, dominate us, maybe lick us a little.

    • hnau says:

      It’s not just the Pledge of Allegiance. In churches where the whole congregation reads a text aloud together (e.g. the Lord’s Prayer) there’s a very predictable cadence. In fact I’ve even seen this happen when the congregation is asked to do this with an unfamiliar text.

      It’s tough to be sure whether this behavior is caused by the ritual-like environment or not, because it’s hard to think of examples outside of that environment. But I bet it isn’t caused by it– it’s practical. If you want to get a bunch of people to say the same thing in sync, it helps a lot if you can increase the spacing and induce a predictable rhythm, and it seems like people do that by intuition. If everyone used a normal conversational cadence it just wouldn’t work.

      • EchoChaos says:

        If you want to get a bunch of people to say the same thing in sync, it helps a lot if you can increase the spacing and induce a predictable rhythm, and it seems like people do that by intuition.

        I lead prayers at my church fairly often and this is in fact mostly the reason. Regular pauses make sure that nobody is rushing and nobody is lagging so that we all feel like we’re saying it “together” regardless of our normal speaking pace.

    • The Nybbler says:

      Is your town Witchitstans, Utah by any chance?

  77. maha says:

    I’m trying to find a quote/passage that went something like “It’s okay if a guy cuts me off in traffic, because at the end of the day I’m still a tenured professor” – it was about how having lots of parallel status hierarchies could lead to a decrease in societal violence, because it makes it so that any single conflict isn’t life-or-death, all-or-nothing.
    I think it may have been from Pinker, Dennett or Jordan Peterson. Does anyone know what I’m talking about?

    • ghodith says:

      I can’t find the link but I can confirm with ~95% confidence that this is a quote from Jordan Peterson. Hope that helps.

    • LCL says:

      It was Pinker.
      ctrl-f “i think to myself” in that

      I remembered it was Pinker but not where I read it; could be he re-used the example elsewhere (Enlightenment Now?)

    • AlphaGamma says:

      I’ve heard Scandinavians saying (approvingly) that their military reserve system, at least during the Cold War when basically all able-bodied men of military age were reservists, made their society “more equal” by creating another, parallel, status hierarchy.

      Apparently it was not unusual for somebody to be in the same reserve unit as his boss in civilian life and outrank him.

    • Nancy Lebovitz says:

      John McWorter (a professor) has also said something of the sort in regards to racism– a receptionist ignored him (he’s black) until he got her attention. He kept from getting angry by reminding himself that she is a receptionist and he isn’t. Sorry no source– it was in a bloggingheads video.

  78. Theodoric says:

    Has anyone learned or re-learned how to drive as an adult?

    I grew up in a suburban area, learned to drive. I was always white-knuckling it. I was told I held the wheel in a “death grip”, when I drove a friend somewhere, the friend would tell me when I had right of way at an uncontrolled intersection (because I was hesitant to go when there was a chance the driver on the other side of the intersection would be a dumbass with no light to stop them), and I was told I appeared stressed after driving. I now live in an urban area, and have been able to get by without a car for several years.

    If I want to start driving again, I am going to need some refresher lessons. I have been thinking about how to go about this.

    Would it be at all feasible to rig up a driving simulator, like the flight simulators people run on computers? As in, hook up a computer to my TV, attach a video game steering wheel and pedals, and run a program that could simulate icy roads, pedestrians darting out, construction, merging into traffic without a light to tell people to stop, things like that? The idea being that I can get some mistakes under my belt before actually putting anyone at risk, and, by doing several simulated runs, I will build confidence to do the real thing.

    Are there driving schools that let you do skid pad and other emergency maneuver training with their cars and instructors, described in this article on how to deal with panic? Is that what a defensive driving course does? I think a big part of what I need is panic desensitization. I do tend to startle easily, for example. Perhaps a very, very low dose of an anti-anxity medication would get me to most people’s “normal” but I am in a licensed profession, and it can be dangerous to develop a record of psychiatric treatment in that situation (sorry, Scott).

    In school, I was diagnosed with some sort of visual motor integration/fine motor skills disorder. This currently manifests itself as very bad handwriting. I recall having to do extra handwriting exercises in elementary school because of this, and later on I got accommodations allowing me to type instead of handwrite things. The last time I got any sort of evaluation for this was to get accommodations for a standardized test (type a part that is normally handwritten). I have not had subsequent treatment. Could this be part of what made driving difficult? When I drove, I had about 1 accident every 6 months, only one of which led to an insurance claim. Most of them involved my misjudging how much space I had, but one was my instincts being wrong (I skidded on ice, and slammed on the brakes; I was later told I shouldn’t have done that-my instincts were “THE MACHINE NEEDS TO STOP MOVING HIT THE BREAKS THAT’S WHAT MAKES IT STOP MOVING”-see above about my possibly needing panic desensitization).

    When/if I actually buy a car, should I look into cars with “autopilot” features, like Tesla has? Has the technology advanced that it can do some of the trickier maneuvers (like a K-turn)?

    • Truism says:

      When I was in the Army we trialled driving simulators for adult driver training. It’s not feasible for you to set one up; all but the very best (and expensive) introduce training errors that have to be retaught/unlearned (as of about 3 years ago). For higher level driving skills (racing, some truck driving) there’s better commercial incentives so there are cheap and good consumer level simulators and setups; they might suit your needs, but I’m skeptical.

      Driving schools offer courses like you mentioned, but what you’re talking about doesn’t sound like defensive driving, it’s high end advanced driving. No entry level driving courses are going to do j turns and skid pan driving because it’s a set of skills you simply don’t need for ordinary driving. Defensive driving teaches you not to get into those situations in the first place.

    • HarmlessFrog says:

      Has anyone learned or re-learned how to drive as an adult?

      Isn’t that, like, the normal way you learn to drive? If you aren’t big enough to operate the pedals, wheel and stick at the same time, while being able to look through the windshield, you probably shouldn’t be learning to drive yet.

      • bullseye says:

        Standard in the U.S. is to learn at 15 or 16, and we’re not adults until 18.

        • HarmlessFrog says:

          I see. I learned around legal adulthood, anyway, not being particularly interested in learning any earlier.

          • Evan Þ says:

            Same with me. The law at the time in North Carolina was that getting a learner’s permit under age 18 required going through driver’s ed class, but you could avoid that if you got it after your eighteenth birthday. Having no particular interest in driving earlier, and deciding that my mom could teach me one-on-one at least as well as the class (and her being willing), I went to get my permit a few days after I turned eighteen.

        • Theodoric says:

          Yes, I meant “older than a teenager.”

    • bullseye says:

      My advice would be to go out and do some easy driving; find some place where you’re expected to not go too fast, and drive around in good light and clear weather. After a while it’ll feel familiar and safe. Then you move on to something more challenging.

      I also startle easily, and on one occasion it helped; I was changing lanes, and another car was changing into the same lane from the other side. I was back in my lane before I had time to consciously react.

      If you hit ice, and you’re going fast enough to trigger panic, you were going too fast.

      • Evan Þ says:

        This. I was never as panicky as Theodoric describes, but I was definitely very nervous when I started driving, especially when I was going anywhere near highway speeds or trying to change lanes in traffic. After a year or so, my nervousness went away.

        If you can find a vacant or low-traffic parking lot nearby, that’s a good place to start practicing at slow speeds. Pretend the parking aisles are roads, stop at their intersections and signal, et cetera. That’s what I did. My first or second time, I jumped the curb onto one of the islands with plantings, but I was going five mph, so the only thing that happened was a shouted “STOP!” from Mom in the passenger’s seat.

    • ana53294 says:

      I took some driving lessons with an instructor, after I had an accident. Nobody was harmed, but the realization on an emotional level that I could have killed a person accidentally made me stop driving for almost a year.

      The instructor told me it’s pretty common, and he had many clients who took lessons before they restarted driving after a hiatus or a traumatic experience.

      If you have the money, go for it. Having a person next to you who can take control while driving in real life conditions is very reassuring.

    • beleester says:

      (I skidded on ice, and slammed on the brakes; I was later told I shouldn’t have done that-my instincts were “THE MACHINE NEEDS TO STOP MOVING HIT THE BREAKS THAT’S WHAT MAKES IT STOP MOVING”-see above about my possibly needing panic desensitization).

      This is correct for older cars, but if you have antilock brakes, then you should just slam on the brakes and let the ABS handle it – the ABS “pumps the brakes” much faster than a person can. You do still need to allow more stopping distance in icy conditions, but “leave more space” is much easier to learn than “suppress your panic long enough to pump the brakes properly.”

      I’m uncertain about autopilot features – I thought they were more in the vein of “really good cruise control” but apparently some new cars have collision avoidance features too? – but I would recommend a car with traction control, as that’s really handy for avoiding skids in the winter. It’s also much more widely available than fancy autopilot features.

      • CatCube says:

        It depends on circumstances. If you’re in a skid in a front-wheel-drive vehicle, keeping a little power can help pull you straight in conjunction with steering into the skid and allow you to regain control of the vehicle. It may be that Theodoric had no room to regain control and avoid the accident, so that it didn’t make a difference, or that it was possible and braking was the wrong answer.

        Slamming on the brakes with ABS when in a skid on ice is always the right answer with a rear-wheel drive vehicle, as adding power will cause the ass end to continue traveling forward and sideways relative to the motion of the vehicle and continuing the skid. Though unless you’re driving a truck RWD is a lot rarer these days.

    • Auden says:

      I recently purchased a Tesla Model 3, and it required relearning all the basics of accelerating/braking/cruising to accommodate 1-pedal driving. The process was nerve-wracking, since there was no easy place to practice with the car, and nothing felt intuitive, but the learning curve went very quickly. My only advice is to practice at night when traffic is light and the sun is never in your eyes. Maybe around 8-9 PM before you get sleepy?

      Per your autopilot question, it really can’t do anything (yet) besides drive long distances on the freeway, but it is a major boon for those instances. You can spend very little attention on centering in your lane and instead focus on the traffic and hazards around you. But the feature isn’t cheap, so it’s not a no-brainer.

    • ECD says:

      For most of your questions, I’m useless. But I recently needed to learn to drive for work, having evaded it for my entire adult life. At about 30, I contacted the local driving school and scheduled a bunch of lessons, then the test. I carefully scheduled some of the lessons for when there was snow and ice on the ground, which helped some.

      Since getting my license, I have not actually needed to drive, and so am probably quite thoroughly useless once again.

    • Solra Bizna says:

      Maybe your response to driving is the correct one, and the majority of people are only able to drive stress-free because they don’t realize how dangerous it is.

      (I too, am a terrified driver, and therefore unbiased. 😛 )

    • Theodoric says:

      Thank you everyone for the suggestions. If I do pursue this (I have no pressing need to drive now, I would just like to keep my options open), it looks like the best thing to do would be to take some lessons, then get a Zipcar membership or something similar and commit to driving someplace local a few times a month, with the worst case scenario being that I have too many accidents and Zipcar kicks me out and I’m back where I am now (I guess a benefit to doing this as an adult is that damage to a five-figure machine no longer needs to be reported to the same people I was afraid to report getting a “C” to!).

      • Aapje says:

        Keep in mind that Zipcar charges a damage free of up to $1000 if you damage the vehicle. A damage fee is pretty standard for rental cars. They don’t want people to get reckless.

        If you have no pressing need, the main question is whether your anxiety will increase over time. This seems to be quite common. Then it may be better to commit to getting practice to prevent this.

        • Theodoric says:

          Thanks for the heads up re damage costs. I know this sounds silly, but I am more nervous about being Yelled At than having to pay non-ruinous damages amount. I know that’s kind of a stupid way of looking at things. Probably why, even thought I drove as if everyone other than me was drunk, blind, and running from the policy, I nevertheless trusted those maniacs to obey a traffic light-if they had me while running a red light, it would officially be Not My Fault.

  79. Truism says:

    4chan have created an AI Dungeon Storyteller based on GPT-2 1.5b. It’s been pretrained on millions of lines of choose your own adventure stories and anons are fine tuning it locally with their owny kinky elf slave and tentacle stuff.

    What’s interesting about it is that it handles memory very well and generates text a lot faster than most GPT-2 implementations, such that the 1.5b model can be run on most consumer hardware astoundingly well. It’s worth checking out if you’re looking at desktop implementations of GPT-2.

    https://github.com/VBPXKSMI/Open-CYOAI-Project/wiki

    • Truism says:

      Also, yes it is AIDS. Here are two examples of the hilarious prompts with which anons are fine tuning it [CW: too much to list]:

      “You are Mary, a 17-year-old girl. You have been infected with an alien parasite. The parasite lives in your brain. The parasite cannot kill you but it can make you feel pleasure or pain. The parasite will give you telepathic commands. If you obey the parasite it will reward you with pleasurable sensations. If you disobey the parasite it will punish you with painful sensations. Whenever you have sex a tentacle will emerge from your body. If you sting a human with your tentacle they will become infected and become your thrall. The parasite wants you to infect other humans and will try to compel you to seduce or rape people in your life. You have a sister named Kelly, a roommate named Jasmine, a boyfriend named Mark, and a boss named Carl.

      It’s been a week since you were infected by an alien parasite. You’ve been struggling to resist its efforts to control you. You wake up and start getting ready for work. You hear the parasite speak inside your mind, “PLEASURE IS GOOD.” You feel your knees quiver as the parasite releases a pleasurable sensation. You reach down and start gently rubbing your cunt. You hear the parasite in your mind again. “INFECT JASMINE,” it commands you. You banish the thought from your mind. The parasite senses your disobedience and you wince as the pleasure is replaced with pain. You wander down to Jasmine’s room. You crack open the door and see her sleeping in her bed. You bite your lip as you think about her body. The parasite floods your body with arousing sensations. You creep into Jasmine’s room and sit on her bed. You reach out your hand towards her sleeping body, but you pull back at the last moment. The parasite punishes your resistance with another wave of pain. Jasmine stirs from her sleep and looks at you. “Mary, are you okay?”, says Jasmine as she reaches out and touches your shoulder. You look into Jasmine’s eyes and hear the parasite compelling you to kiss her. You close your eyes and”

      “You are Anon, an 18-year-old male human. You live in the kingdom of Carcaso. Carcaso is ruled by humans. Elves in Carcaso have no legal rights. Female elves are kept as slaves by wealthy humans. Owning an elf slave is considered a status symbol in human society. Many humans take pride in showing off their obedient elf slaves. You come from a wealthy noble family in the capital. For your 18th birthday your father bought you an elf slave of your very own. Your slave is a young elf named Kyla. Kyla is a short, thin female elf with short blonde hair and freckled cheeks. Your goal is to train Kyla to be a submissive and obedient slave who will obey your every command. Kyla may resist your commands at first. You may use any punishments or treatments you feel are appropriate to break her.

      You wake up in your room at your family’s estate. You rub the sleep from your eyes and suddenly remember that today is your birthday. You bolt out of bed and run downstairs to greet your family. As you eat breakfast in the dining room, your father presents you with your birthday gift. He leads a thin, meek-looking elf girl wearing tattered rags and a leather collar with a leash on it. “Really father! An elf slave of my very own! I can’t believe it,” you exclaim. “Remember, son, a slave is a big responsibility, you’ll have to feed her, bathe her, and train her,” your father tells you. “Oh promise I’ll take care of her, father. What is your name, slave?” you ask the elf girl. She says nothing and stares at her feet. You grab the leash and tug forcefully on it. The elf girl yelps in pain and looks up at you, “My name is Kyla, Master.” You smile and revel in the feeling of power you have over this pitiful elf. “Alright, Kyla, I’ll start training you today. Let’s start by finding you something more appropriate to wear. I won’t have my new slave looking like a common gutter elf,” you smirk as you yank on Kyla’s leash and drag her behind you”

      • Truism says:

        What a time to be alive.

        • HarmlessFrog says:

          Did you just /thread your own post, OP?

          • Truism says:

            Maybe. What?

            None of the content posted is mine and I’m not part of the project. I intend to install it tonight though and muck around.

            I’m amazed it exists and thrilled to be alive in 2020.

          • VoiceOfTheVoid says:

            It’s a common way to add stuff either when the edit window’s passed, when you’re hitting the character limit, or when it’s tangential to your main point.

          • HarmlessFrog says:

            There’s a character limit? That would explain it. The post and the follow-up weren’t more than an hour apart.

    • gwern says:

      I’m disappointed it doesn’t actually let you choose your own adventure. As I noted to Walton, AI Dungeon could probably be massively improved if you made the player choose from, say, 4 possible completions, recorded which completions are picked, and train another NN to predict completion choice, which would let you automatically select the best completion.

    • Le Maistre Chat says:

      It’s been pretrained on millions of lines of choose your own adventure stories

      That is so cool! I didn’t even realize the Choose Your Own Adventure books added up to that much text…

      and anons are fine tuning it locally with their owny kinky elf slave and tentacle stuff.

      ಠ_ಠ

      • Nick says:

        ಠ_ಠ

        Dare you enter their magical realm?

      • Deiseach says:

        Flip’s sake, if they wanted that, they could just read the Gor books. Ah well, young men gonna be young men, I suppose (which is to say, hormone-driven idiots until time and experience gives them some cop-on).

        • Error says:

          What are Gor books? Asking for a friend!

          • EchoChaos says:

            A Google search will probably fill you in, but they’re adventures on a planet directly opposite us from the sun called Gor that a regular earthling is thrown to.

            Start out as relatively standard fantasy fare of that type (similar to e.g. John Carter of Mars), but rapidly become a very, VERY focused on their unique legal system which is based entirely on sexual exploitation of slave women by men.

          • Randy M says:

            Shh, EchoChaos, I’m anticipating a highly amusing rant from Deiseach on the subject.

          • Plumber says:

            @Randy M,
            For me a sizable percentage of all of @Deiseach’s posts fall into that category, and from the many pleas to our host to release her from the time-to-time bans I suspect enough others feel the same way.

          • Le Maistre Chat says:

            @EchoChaos:

            Start out as relatively standard fantasy fare of that type (similar to e.g. John Carter of Mars), but rapidly become a very, VERY focused on their unique legal system which is based entirely on sexual exploitation of slave women by men.

            Academic dissertations could be written on the positive and negative feedback that shaped that focused reception!
            John Lange (pen name John Norman) was a professor of philosophy who came up with an idea for a Sword and Planet series to make some extra money. The justification for the S&P tropes, not revealed until Book… 3, I think? (Priest-Kings of Gor) is that Counter-Earth is a zoo maintained by human-sized sapient bugs who will let the humans in their preserve develop any technology that doesn’t release destructive chemical reactions, not even an Early Modern cannon. So there are life-extension drugs, high-rise buildings, and other fun stuff, but no internal-combustion vehicles and people still fight with swords.
            Part of the flavor the author imparted to the series was HARD-NOSED REALISM (TM) about the behavior of pre-modern people. So when there’s a war, the winners take slaves, male and female. The slave girls often but not always got raped a lot, while the slave boys would get worked to death in mines or even subjected to GRIM ‘N’ GRITTY pseudo-realism like blacksmiths quenching their blades by running them through a slave boy’s abdomen.
            The early books sold well and somehow feedback reached Lange that his readers really liked the slave girl passages and were ignoring all the other grim ‘n’ grit. So he wrote a lot more, becoming for all intents and purposes a softcore BDSM author.
            … there were two problems with this. One, feminism got huge. Two, the narrator of the books wrote long philosophical asides about how these master/slave girl relationships were the natural order of things, sending even the secular-Rightist readers who’d be least offended by such un-PC gender relations to the hospital with eyes rolled to the backs of their heads.

          • Deiseach says:

            What are Gor books?

            Sic transit gloria mundi, huh? Sit back and listen, young’un – and Randy M, I’ll try not to disappoint you either 🙂

            WARNING FOR THOSE OF DELICATE CONSTITUTIONS AND REFINED SENSIBILITIES: THERE WILL BE SOME AMOUNT OF CRUDE LANGUAGE AND DESCRIPTIONS OF CRUDER SEXUAL PRACTICES IN WHAT FOLLOWS, BLAME THE AUTHOR, HE STARTED IT

            Back in the 70s, the Gor books by one John Norman (a pen name) were the go-to fix for readers of skiffy who wanted some soft-core sex of a mildly kinky nature. Bear in mind that the Internet was not yet invented and if you wanted porn you had to buy ACTUAL PAPER MAGAZINES. The 70s were also The Decade That Taste Forgot and the Decade of Cheesecloth and the Decade when 60s Utopianism of the hippie kind curdled into self-regard of the hedonistic kind – finding oneself, EST and the likes were the big new thing. This also included “we’re so free of our parents’ sexual and societal hang-ups that anything goes, baby” (although attitudes were still astonishingly sexist if one looks back at them with the eyes of today).

            So Norman’s “treat ’em mean, keep ’em keen” brand of sex’n’sci-fi was just the thing for the young man of sophisticated tastes.

            *insert so much eye-rolling here I need to pause while I grope around on the floor to find and re-insert my errant orbs*

            Okay, the first book or two were standard planetary romances of the Burroughs type (though nowhere near as good as Edgar Rice Burroughs and I’ll stand over that even while saying that Burroughs wasn’t the high-water mark of prose). A manly-thewed Earthman by the name of Tarl Cabot ends up translated from Earth to Gor, our sister planet on the other side of the sun, for reasons too convoluted yet trivial to mention (turns out Tarl’s father is actually a Gorean lord and was reclaiming Sonny-Boy). If this sounds awfully like how John Carter of Virginia ended up on Mars, then the resemblance is completely intentional.

            There were some sparks of originality there (Nar the Sentient Spider is a lovely character, and if Tarl Cabot – Our Hero – had been eaten by monsters and we got the Adventures of Nar instead, I’d have no complaints BECAUSE NAR AT LEAST IS A GENTLEMAN AND A SCHOLAR). Gor is a world of extremes – the inhabitants live at a roughly Iron Age level of Orientalism-tinged (very heavily tinged) society, but artificially kept at very low-level tech by Insect Overlords because – well, I never could understand quite why. So this means that while Goreans have access to space ships for slave raids on Earth and functional immortality, they all run around dressed like BDSM night at the gay disco.

            Slavery bad? Yes, indeed (and Gorean society finds male slavery to be the lowest of the low, the worst punishment possible). Luckily, we’re talking about the good kind of slavery – kidnapping and breaking women to be sex slaves. Which is why the Orientalism – it’s Harems On Mars Gor, which makes it all acceptable.

            A harmless bit of spicy fun, yeah? Particularly as it’s all rather tame now due to History Marching On. Well, we’ll keep going and you make your own mind up on that.

            So the first couple of books are Tarl Cabot acclimatising to Gor – he starts off with some weak scruples about “hey, maybe carrying women off and engaging in dub-con sex on flying reptilian-back is possibly not the greatest way to behave?” but those are soon tossed aside when he becomes a Real Manly Man of Gor.

            Now, if Norman had stuck with the Burroughs rip-off, the Gor books would not be so bad. Unfortunately, what really floated his boat was his ‘philosophy’, which boils down to: what do women really want? to be reduced to mindless sex-slaves thinking only of their master’s will and wants; to be abducted, stripped, branded, raped and made to enjoy it. In short, to be given permission to indulge their real natures.

            Now, back in the mid-80s, I had heard of the Gor books but never read one, so I picked a paperback copy up (with the mists of time, I can’t remember the title, but it was somewhere mid-series) and having read it to the end (in a fit of spite-reading), my reaction was pretty much “What the ever-living fuck is going on here???” Because I’m rather sure that not every single woman wants to be branded, raped and enslaved, y’know? Even for those who do like that kind of thing, when some people try to live Gorean-style in the real world, it never works out.

            Anyhoo. The sexual attitudes seemed to me to be somewhat lacking in that delicately indescribable shade of respect for equality between the sexes, wherever did I get that impression from? There were some other things that struck me as well – there was enough (unintentional) homoeroticism to choke a tharlarion. The more Tarl (and by extension, the author) argues that what women really want is to be sexually dominated to an ultimate degree, and that the proper attitude of men is to regard them as no more than (to use a term not available back then) cock-holsters but not to think any better of them and to save all their tenderer emotions for their male companions – and the more Gorean men go around dressed like the following:

            In spite of this invention, the mounted warriors always wear, as an essential portion of their equipment, a thick leather belt, tightly buckled about their abdomen. In addition, the mounted warriors inevitably wear a high, soft pair of boots called tharlarion boots.

            (See what I mean about BDSM night at the gay disco?) The more, I say, as he protested about this, the more I got the very strong impression that, um, honey, what you want is a Big Strong Man to sexually dominate you. (A few of us even participated in a LiveJournal group called Slashers of Gor to slag the arse off the author and the books).

            Tarl does a lot of swooning crushes over older Real Manly Men. From the very first book, he has quite the admiration for the deadly Master Assassin Pa-Kur, even though notionally he is in love/lust with the captured daughter of an enemy chieftain. Tarl gets captured and tied (I should really say “lashed”) to a device called the Frame of Humiliation (it’s a sort of floating frame to which the Assassins tie their victims before throwing them into the river to drown) and, well, the following ensues:

            “Then, each of the men of Pa-Kur, as is the custom before a frame is surrendered to the waters of the Vosk, spit on my body. Lastly, Pa-Kur spit in his hand and then placed his hand on my chest. “Were it not for the daughter of Marlenus,” said Pa-Kur, his metallic face as placid as the quicksilver behind a mirror, “I would have slain you honorably. That I swear by the black helmet of my caste.”

            “I believe you,” I said, my voice choked, no longer caring if I lived or died.”

            Yeah. Is that sounding like bukkake to anyone else, or is it just me?

            And Tarl becomes Really Good Friends with another warrior named Kazrak:

            I saw the spark of Kazrak’s fire-maker, and I felt the flush of friendship as I saw his features briefly outlined in the glow. He lit the small hanging tent lamp, a wick set in a copper bowl of tharlarion oil, and in its flickering light turned to the sleeping mat. No sooner had he done so than he fell to his knees on the mat and grasped the ring.

            “By the Priest-Kings!” he cried.

            I leaped across the tent and clapped my hands across his mouth. For a moment we struggled fiercely. “Kazrak!” I said. I took my hand from his mouth. He grasped me in his arms and crushed me to his chest, his eyes filled with tears. I shoved him away happily.

            “I looked for you,” he said. “For two days I rode down the banks of the Vosk. I would have cut you free.”

            “That’s heresy,” I laughed.

            “Let it be heresy,” he said. “I would have cut you free.”

            “We are together again,” I said simply.

            “I found the frame,” Kazrak said, “half a pasang from the Vosk, broken. I thought you were dead.”

            The brave man wept, and I felt like weeping, too, for joy, because he was my friend.

            The flush of friendship, uh-huh. (Listen, if Norman had eased up on the whipping/branding/raping and just stuck to scantily-clad heaving-bosomed flashing-eyed damsels in Arabian Nights lingerie, along with three-quarters naked Manly Men in thigh-boots clasping one another to their bosoms in Fraternal Embraces, I wouldn’t be so critical. But he didn’t.)

            And later on meets the enemy chieftain who is the father of the ostensible lust-interest, but sounds a lot more interested in Daddy than in daughter:

            Before I realized what was happening, my arm had been jerked downward and twisted, and I had been thrown on my back at the feet of the man, who leaped up and set his boot on my throat. In his hand was a warrior’s sword, and the point was at my breast. He laughed a mighty, roaring laugh and threw his head back, causing the hood to fall to his shoulders. I saw a massive, lionlike head, with wild long hair and a beard as unkempt and magnificent as the crags of the Voltai itself. The man, who seemed to leap into gigantic stature as he lifted himself into full height…”

            “Before me, on a rough throne of piled rocks, sat Marlenus, his long hair over his shouders, his great beard reaching almost to his sword belt. He was a gigantic man, larger even than the Older Tarl, and in his eyes, wild and green, I saw the masterful flame which had, in its way, also burned in the eyes of Talena, his daughter.

            (Tarl is really into Masterful Gazes).

            Well, to cut an already long story short: Gor series goes on, gets worse due to Norman riding all his hobby-horses about How To Treat Yo Bitches and is mercifully reduced to part of the rich tapestry of the history of the genre. The End.

          • Anthony says:

            As Deiseach mentions, there is a real Gorean subculture within the BDSM subculture. Even people who vehemently oppose kink-shaming make fun of them.

            For a taste of the writing style, read Houseplants of Gor. (Google the title if it’s disappeared again.)

          • Plumber says:

            I weep a tear of joy (just one, so totally butch!!!) at @Deiseach’s post (worthy of being “of the week!”), and I’ll add that while I’ve never read the books, I managed to sit through almost 25 minutes of the 1987 movie Gor (15 whole minutes longer than I managed to watch of Caligula) which had master thespians Oliver “Tommy”,”The Adventures of Baron Munchausen”, “Gladiator” Reed, and Jack “Batman”, “City Slickers, “Hawk the Slayer” Palance, and such was the HIGH QUALITY!!! of the film that there was a 1989 Sequel Outlaws of Gor, which matched in quality such MASTERPIECES!!! as Beastmaster, Conan the Destroyer, Hawk the Slayer, Red Sonja, and The Sword and the Sorcerer

            Check out the trailer.

            TRULY A GOLDEN AGE OF CINEMA!!!

        • I recall reading somewhere that the IRL Gorean fanbase is overwhelmingly female, but I’ve lost the source and have no actual evidence to back this up. Can anyone confirm or deny this?

          • albatross11 says:

            FWIW, I’m pretty sure the Shades of Grey books’ readers were overwhelmingly female. Presumably due to some mix of biology and culture, when men look for one-handed reads, we’re usually wanting pictures, whereas women are usually wanting words.

          • Conrad Honcho says:

            Also I can’t say I’m shocked more women are interested in being a sex slave than men are interested in being a master of sex slaves. That sounds exhausting to me.

          • EchoChaos says:

            Yeah, and “being a sex slave” is exactly the sort of thing that one can’t practically do (and wouldn’t want to if you could) but might enjoy occasionally fantasizing about.

          • Le Maistre Chat says:

            @albatross11: Huh, that makes a lot of sense to me. For taboo fantasy X, you could have a mostly female fandom that reads erotica about it and also a mostly male market for porn movies about it.

            @Conrad: Hahaha. While in real societies with slavery, buying sex slaves was way more common than a woman selling herself, BDSM fantasies are definitely skewed toward that “less exhausting” role.

          • Thegnskald says:

            I think BDSM porn might also be dominated by women consumers, but that’s going on vague memories of research done a decade ago when I had these kinds of arguments on OKCupid’s sex forum.

          • Nick says:

            Written porn being dominated by women is very much a thing. It’s not only obvious from the contents of the romance aisle at your local bookstore—all bodice-rippers marketed to women—it’s the case online, too. Pornographic fanfic circulated on tumblr or ao3 is all by and for women, too.

          • EchoChaos says:

            @Le Maistre Chat

            But here we are talking about visible minorities. Perhaps sex slavery was more common, but the vast majority of men never owned a sex slave. The 1% dominated them, and it was so exhausting that they basically always outsourced actually managing them.

          • albatross11 says:

            There is a huge difference between things people fantasize about and things people would actually like to live through. Even outside porn/near-porn stuff like Gor or Fifty Shades, think of how many people enjoy reading stories involving war, disaster, true crime, and horror (zombies, ghosts, evil spirits, etc.) The fact that I can enjoy a novel set in the middle of a war or disaster doesn’t remotely mean I’d actually like to be caught in the middle of a war or a disaster!

          • Conrad Honcho says:

            Yeah but I’m saying it wouldn’t even be fun to fantasize about keeping sex slaves because I’d keep thinking about all the logistics involved. You’ve gotta build a dungeon, and worry about feeding them, and security, and shopping for the whips and chains and all that. Who has that kind of time?

          • albatross11 says:

            Signs you’ve become too square as you’ve gotten older: Your sex fantasies devolve into trying to work out the budgeting and logistics for the whole thing. (“Wow, this hot 19 year old wants to f–k. Great, but I guess I’ll need to buy her a nice meal first. How’s that work with my budget for the week? And flowers, can’t forget those. I wonder if she needs a ride home after–gotta pick the kids up from school at 5! Oh, wait, doesn’t Junior have soccer practice today? Maybe I’d better check my calendar….”)

          • Le Maistre Chat says:

            Yeah but I’m saying it wouldn’t even be fun to fantasize about keeping sex slaves because I’d keep thinking about all the logistics involved. You’ve gotta build a dungeon, and worry about feeding them, and security…

            Thinking about the logistics of building a dungeon, feeding everyone in it, and security is one of my hobbies.

          • woah77 says:

            I mean, I fantasize about the logistics too, but if you want to have sex on top of that? That’s just too much. “Leave me alone women, I’m working here. Got to increase the fortifications in the west wing…”

          • Le Maistre Chat says:

            @albatross11:

            Even outside porn/near-porn stuff like Gor or Fifty Shades, think of how many people enjoy reading stories involving war, disaster, true crime, and horror (zombies, ghosts, evil spirits, etc.)

            Note that people who enjoy reading stories about war are sometimes accused of militarism/Fascism, just as people who enjoy maledom porn get accused of supporting the Patriarchy.
            I dunno why true crime and horror fans are never suspected of supporting crime or evil spirits.

          • Randy M says:

            Who has that kind of time?

            Just go up a level, and fantasize about setting economic policy on a small, feudal world in thrall to you wherein you delegate the administration of your harems to a hereditary order sworn to chastity.
            Wait, I might need to give that more thought…

          • EchoChaos says:

            @All

            We’ve drifted far afield, but for some reason I have a burning need to solve the economics of paying a chief eunuch to watch all my other eunuchs to watch my harem, plus food and storage for all eunuchs and the harem, plus…

          • Conrad Honcho says:

            Not really albatross11. If the cute coed says “Oh, Professor Honcho, I’d do anything for an A!” I don’t have to imagine finishing my dissertation, slaving through post doc work, slogging around for a lecturer position while hoping there’s an opening for a professorship somewhere. All that world building is done: there’s a university, I’m a professor. Okay.

            But whoa77 gets it. If Warlord Honcho is going to go molest his slave girls, where does he keep them? A dungeon? Okay, how did I get a dungeon? I didn’t build it myself. So now I’ve got builders who are either other slaves, or contractors who work for me. And I assume the dungeon is guarded so the slave girls don’t escape. So now I’m managing employees. And I better be generating enough income to afford these employees, so now I’m managing a polity, I’ve probably got serfs. I’ve got to be worried they’re producing enough to pay for the upkeep on my dungeon, and I’ve got to make sure the fortifications are secure so the tribe next door (who I probably stole the sex slaves from) doesn’t come steal the sex slaves from me. Heavy is the head and all that. After dealing with all the headaches and stress that comes from managing a bronze age fiefdom, who even has the energy to molest their slave girls?

            But if you’re the girl, this is all stupid easy. You’re hot, the warlord captured you. All you’ve got to do is convincingly moan “oh no you brute” while the powerful man ravishes you. How he got to be warlord and all the work he had to put in to make this happen isn’t your concern.

            Just sayin’ it’s a way easier fantasy for women than for men.

            ETA: Fishes and barrels apparently.

          • Conrad Honcho says:

            Also, @Randy M,

            hereditary order sworn to chastity

            wait, what? How do you get a hereditary order sworn to chastity? Do they do the hereditary stuff before the swearing? Or are we calling uncles and nephews hereditary..?

          • Dacyn says:

            @Conrad Honcho: The obvious solution is that people break their oaths regularly…

          • CatCube says:

            wait, what? How do you get a hereditary order sworn to chastity?

            Same way they did it in Hot Shots Part Deux.

          • Nancy Lebovitz says:

            The fantasy of having sex slaves should include the fantasy of having logistics slaves so that you don’t have to do the logistics yourself.

          • Paul Brinkley says:

            The fantasy of having sex slaves should include the fantasy of having logistics slaves so that you don’t have to do the logistics yourself.

            That’s silly. Why would we have slaves do the fun part??

          • Le Maistre Chat says:

            This is feeling like the Ring’s attempt to corrupt Samwise, except with subservient women instead of gardeners.

          • Conrad Honcho says:

            The fantasy of having sex slaves should include the fantasy of having logistics slaves so that you don’t have to do the logistics yourself.

            But that doesn’t solve the problem of managing the logistics slaves!!!

            Getting semi-serious for a second, I think this is similar to a “suspension of disbelief” issue, maybe called “suspension of disbelievability of abuse?” A professor exploiting his power over grades, or a police officer exploiting his power over force of law to get sex is already explained: you’re a bad person (or a good/mediocre person seduced by lust) to do a bad thing. But in the whole warlord/sex slave thing it’s just…natural? You just walk into the sex dungeon and say to the eunuch guards* “head out boys, it’s rapin’ time” and that’s that. Why that doesn’t snap one out of the fantasy requires backstory. I guess..?

            I mean, this is all dumb. Epistemic status: running of the mouth about fantasies I don’t even have. But I think the difference is that illegitimate use of power for sex (professor, cop, boss, etc) is way easier to gloss over than “””legitimate””” use of power for sex, like warlord, in which everyone just accepts that the boss is on his way down to the sex dungeon. No suspension of disbelief is required for you to believe you’re an asshole. Suspension of disbelief is required to believe you’re being an asshole in ways others find acceptable.

            * Also, the whole thing of having eunuch guards says to the ladies, “I want you all to myself so much I’m chopping other dudes’ balls off doesn’t that make you feel extra special and sexy?” As a guy it just sounds like cruelty and extra work. That’s not sexy.

          • Plumber says:

            @Conrad Honcho > “…I think this is similar to a “suspension of disbelief” issue, maybe called “suspension of disbelievability of abuse?”…

            …Epistemic status: running of the mouth about fantasies I don’t even have…”

            I’ve mentioned once or a dozen times that I’m a big fan of Leiber’s Fafhrd and the Grey Mouser tales, but the later ones from the ’80’s (maybe to be “hip” and “with-it”, or because he thought “Hell with it, I’m gonna die soon, I’ll write for me now!”), well I went to the book signing, got my old books signed, shook the hands of the very frail seeming author, got the newest book (Knight & Knave of Swords), then read a certain scene and…

            ….”ewww…

            ..oh damnit Fritz, ya coulda left that out!”

            Since then I’ve encountered other such passages in other works of newer fantasy-fiction (and my parents comix books had worse), but still it was an adjustment.

            Made it easier to read Ian Fleming’s stuff though!

          • albatross11 says:

            There’s a strategy board game in here somewhere….

          • The Nybbler says:

            The whole sex slave thing is ridiculous anyway, unless you’ve got a rather specific fetish. If you’ve already done the work of becoming a successful warlord, you don’t need sex slaves.

            If you want completely willing and enthusiastic women… as top dog, you’re going to have groupies, and like a rock star you’ll just need people to pick the best ones for a night.

            If you’re a bit darker in taste… well, as the warlord you’re in a position to do favors or make life difficult, either for the girls themselves or their families.

            And if you’re completely dark… well, there’s the “war” part of warlord. Unless you’ve got the appetite of the Great Khan, you’re not going to run out of places to raid.

          • helloo says:

            Remember that harems DID exist historically and generally weren’t known to be high-risk or anything (I think…).
            You can just pretend to be part of one of those rather than reinventing one.
            There were a number of logistic issues and things involved which can be interesting (ie. Supposedly early Chinese calendars where invented to help keep track of which day for each consort). Should we get David Friedman involved?
            However, to swerve back into the conversation, the historic stories and tales that involve harems tend to involve drama between the members in the harem or trying to escape it than actually building it up or the experiences of having one.

          • John Schilling says:

            If you’ve already done the work of becoming a successful warlord, you don’t need sex slaves.

            To be fair, the Gorean version of the fantasy allows people who aren’t top warlords to have sex slaves. IIRC, the 1%-ers of that society have modest slave-harems, while the 10% middle class have access to slaves with the usual rent-vs-buy argument for capital goods. Cabot himself spends more time as a lone-wolf adventurer than any sort of warlord, I think. So, not just the people who would be automatically attracting groupies in every society.

            Logistically, IRL and I believe even in the books, this requires broad social and legal support for slavery more than it does elaborate logistics on the part of the owner. If you can count on the broader community to capture and return any runaway slaves for you, then you don’t need a dungeon and guards so much as just a spare room and the sense/luck to not buy particularly rebellious slaves. See also historical slavery in the antebellum South.

            If the society around you doesn’t support slavery, then the logistics rapidly become intractable. People who try that IRL usually find that their sex slaves escape in reasonably short order (and then have their former masters thrown in jail), or they find themselves killing their slaves to avoid the whole being-thrown-in-jail thing once they realize they have underestimated the logistics of all this.

          • Deiseach says:

            How do you get a hereditary order sworn to chastity?

            Continence is part of chastity but not the whole; you have an order of married men sworn not to have affairs or extra-marital sex, and then (depending on how well they keep their oaths) you have your hereditary chaste order guarding your harem.

            Or you could have female guards, I suppose. That might get around the traditional problem (in stories, whatever about real life) of “guys sneaking into the harem by disguising themselves as women” – female guards could do strip-searches of ‘female’ visitors to make sure they really were what they say they are 🙂

          • EchoChaos says:

            @Deiseach

            Or you have the first sons of the order off having children while any subsequent sons are castrated, which gives you a bonus musical accompaniment to your sex slavery.

          • Le Maistre Chat says:

            @Deiseach:

            Or you could have female guards, I suppose.

            This was done in the palaces of Southeast Asia, at least. I have no proof, but it wouldn’t surprise me if they inherited the custom from India: it would explain the trope of every raja wanting Yavanis (lit. “Greek women”) following him around as guards, despite women being biologically less fit to be pre-gunpowder warriors – they let him have a portable harem!

          • Nancy Lebovitz says:

            John Schilling, from what I’ve read about household/sex slaves in societies where slavery isn’t legal, it isn’t all that difficult.

            If the slave doesn’t know the local language and doesn’t have legal residence, they can be kept trapped. In cases of human trafficking, their family back home can be threatened by the trafficker.

          • Aapje says:

            Gadaffi had female bodyguards, called the “The Revolutionary Nuns,” although Western journalists called them “The Amazonian Guard.”

          • John Schilling says:

            If the slave doesn’t know the local language and doesn’t have legal residence, they can be kept trapped.

            That works much better for economic than sexual slavery, and depends on fuzziness regarding general understanding guest-worker rules and norms. Having a live-in maid is normal and legal. Having one’s immigration status tied to one’s job, and this dependent on keeping one’s employer happy, is legal and normal. The path from there to slavery, crosses legal and social lines but not necessarily clear ones.

            The more blatant it is that one’s “job” is to have sex with the boss whenever he wants, the more obvious it is that this isn’t normal and the more likely the slaves are to run off and find a policeman even at risk of deportation. Some amount of sexual harassment can be socially camouflaged as part of the employer/employee relationship, but that falls well short of Gorean fantasies. And even the prosaic “you will work in my brothel and I will take 90% of your earnings” bit often requires…

            In cases of human trafficking, their family back home can be threatened by the trafficker.

            Yeah, that. This can be quite effective. But where the amateur pondering of sex-slave logistics here has been limited to a modest dungeon and some guards, the dirty reality seems to require running a transnational organized criminal conspiracy. That’s several steps beyond “not all that difficult” in my book.

          • Nancy Lebovitz says:

            John Schilling, from what I’ve read, what happens is household slaves who are available to be raped. I suppose this isn’t fantasy fodder.

            If a few of them escape, there aren’t likely to be consequences for the slave owners beyond trapping another slave, though I wouldn’t be surprised if the risks of consequences are a little higher these days.

  80. TheWackademic says:

    Error: Brian Earp is a graduate student and has not obtained his PhD. So he should not be referred to as Dr. Brian Earp.

  81. benwave says:

    In the 2020 SSC survey, Scandanavian countries are described as: “heavily-regulated market economy, cradle-to-grave social safety net, socially permissive multiculturalism”. Now I admit I don’t know thaat much about those countries, but is it fair to called their markets heavily regulated? I know that they have high individual taxes and redistribution (although they also have quite low taxes on business and capital). But do they have an unusually high amount of regulation?

    • HarmlessFrog says:

      Now I admit I don’t know thaat much about those countries, but is it fair to called their markets heavily regulated? I know that they have high individual taxes and redistribution (although they also have quite low taxes on business and capital). But do they have an unusually high amount of regulation?

      No. For European countries, their economic freedom is high. This matches my personal experience – the laws are fairly clear and not complicated. Taxes are high, but you aren’t unduly constrained in making the money in order to pay the taxes.

    • I think it’s more about a certain American idea of Scandinavian countries, more so than their particulars (and differences).

    • Ketil says:

      I would say Scandinavian countries are similar to Canada, perhaps? Mostly government-funded and -run public health, fairly big government involvement (including part ownership in corporations) in many sectors of industry – including a national broadcaster, telecom, and transportation. This is gradually dissolving to allow increased competition and privatization, though. I don’t think taxes are outrageously high, either – we used to have very high top marginal rates, but a lot of exemptions, but have moved towards lower and flatter rates (typically 25-30%, both corporate profits, capital income, and wages). Sales tax is quite high, at 25% in Norway, and there’s a slightly less than 1% wealth tax (with a large discount for home ownership), no estate/inheritance tax, but a tax on stock dividends.

      • benwave says:

        Don’t most countries have taxes on stock dividends? In my country they are treated as income and taxed in exactly the same way

        • Eric Rall says:

          My understanding is that the norm for most developed countries is to tax dividends as ordinary income to the recipient, but to classify them as a deductable business expense to tge corporation.

          The US is one of the big exceptions to this: we tax dividends at the lower capital gains rate to the recipient, but the corporation pays full corpoate taxes on the money used to fund the dividends.

        • Ketil says:

          There is corporate tax on corporate profits, and dividend payouts are not deductible. So having stock holders pay tax on dividends is seen as double taxation, and also as giving an advantage to institutional and foreign stock holders, who don’t pay dividend tax. It’s probably not a big deal in practice, as serious investors will own stock through a holding company, and thus only pay dividend tax when they take money out of the holding company for private consumption.

          Perhaps this is common, but it was controversial when it was introduced.

          • benwave says:

            Right, I see. In my country corporate profits are taxed, but a system of imputation credits passes on to dividees (can I call them that? I’m gonna call them that) to avoid double taxation.

    • cassander says:

      I’d argue that that description applies to pretty much every developed country.

    • Joy says:

      My German/Swedish partner says that Sweden, after 8 years of a relatively right-wing government, is anything but heavily regulated. Medicine, education and some other core government services have been partially privatized, for better or for worse. Denmark is somewhat more regulated, and Norway is closer to the original description, but the latter have the luxury of oil revenues not many other countries do.

      • cassander says:

        My German/Swedish partner says that Sweden, after 8 years of a relatively right-wing government, is anything but heavily regulated. Medicine, education and some other core government services have been partially privatized, for better or for worse.

        The second of these does not follow from the first. You can have a large government with a huge amount of regulation that hires contractors to do most of the work and you can have a tiny government that regulates little but does it in house with with civil servants. The two things, while often lumped together, are completely orthogonal.