Open Thread 113.25

This is the twice-weekly hidden open thread. Post about anything you want, ask random questions, whatever. You can also talk at the SSC subreddit or the SSC Discord server.

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

1,449 Responses to Open Thread 113.25

  1. Brad says:

    We have a three day rule about discussing any tragedy in any even vaguely culture war context. Not that your comment is very culture war, but it could easily go that way.

    • Plumber says:

      Thanks, I was unaware of that rule, I’ll delete the post.

    • CatCube says:

      Yeah, Scott’s deleted several threads about this already, though both of them were shit-stirring, rather than honestly seeking answers like this one.

  2. Deiseach says:

    Amusing kerfuffle on Romanian Tumblr (English-language version): seemingly the new Romanian Minister of Research and Innovation is/was a dean in a university (that’s not the kerfuffle); he doesn’t believe in evolution because it’s not interesting, he thinks humans came from the future and a parallel universe (hey, I’m only going on what they’re saying) and he writes/wrote Harry Potter fanfic to explore this (because of course he would) on his Facebook page which has now all been scrubbed and memoryholed. Sample translation here 🙂

  3. Nancy Lebovitz says:

    Speaking of hoax threats, remember those bomb threats against Jewish Community Centers that turned out to originate from an Israeli teenager? And maybe he had a brain tumor? For all I know, he really does have a brain tumor.

    I want you to guess what was going on before you click the link. This explanation never would have occurred to me.

    The link

    • albatross11 says:

      Wow. *That* was not at all on the list of things I expected to see!

    • Deiseach says:

      Pffft. Kids these days! In my day, hoax bomb threats to get the school day disrupted were done for free! (As for framing someone else, that was what the IRA was for – phone in and say this bomb was left by the Ra) 🙂

    • j1000000 says:

      Oh, so it may have been a hoax hoax and the threats were really requested by a malicious actor? Perhaps if this CNN article turns out to be wrong it will be a hoax hoax hoax? Buffalo buffalo buffalo Buffalo buffalo.

      I’d say everything just keeps getting weirder but I probably just wasn’t paying enough attention before.

  4. Deiseach says:

    I don’t know what to think about this: potato gin? Isn’t that, you know, vodka? 😀

    (Honestly, recently it’s like everyone and his dog is doing craft beers, craft ciders, and new gins).

    • cassander says:

      I was under the impression that to be legally considered gin, a liquor had to be distilled from a mash that was at least 50 percent rain clouds and pine cones.

    • A Definite Beta Guy says:

      Nah, from my understanding, gin is a drink redistilled to include juniper (plus other herbal flavors if you like).

      Vodka specifically is neutral tasting spirit. You can distill vodka from a potato base(or any base really), then redistill it with junipers to turn it into gin.

      I am sure someone will correct me if I am wrong.

    • dodrian says:

      As gin is an abomination I don’t frankly care what it’s made of. The only thing that’s worse than gin is a gin & tonic: hey, let’s take this awful-tasting medicine and mix it with the most awful-tasting spirit. That’ll make it more palatable!

    • Nornagest says:

      I think gin’s defined more by the botanicals you add (juniper, mainly) than by what it’s distilled from. And on the other side of things, you can get wheat or rye vodkas pretty easily; I’m told grain vodka is more traditional than potato vodka in some countries, although I forget which.

  5. Nancy Lebovitz says:

    https://www.facebook.com/brad.hicks.982/posts/2378846485478034

    Quoted in full because I know an ssc reader who doesn’t have facebook account. Are they the only one?

    ******

    So the first round of forensics are in on the malfunctioning Texas voting machines. There’s pretty convincing evidence that it’s incompetent design, not fraud: some voters for both parties are having their votes distributed at random rather than to the candidates they thought they voted for. Due to a really crappy design, you can follow the instructions on the screen to the letter, watch it register your vote the way you meant it, and then have the machine totally record it wrong. (I’ll link articles and discussion threads in the comments.)

    This comes right on the heels of an extraordinary article on election digital security by Vox. As a retired computer scientist, I have to say that this is the first article I’ve ever read, about the subject of computer science, by a non-scientist reporter, that didn’t have anything cringe-worthy in it. It details all of the things that have gone wrong or certainly will go wrong, sooner or later, with the way that we run our elections in America, and how much attention we’re putting in the wrong places. I can’t recommend it highly enough. (I’ll also link it in the comments.)

    But this Texas fiasco is a perfect, timely example of something I’ve been thinking about ever since reading the Vox article, something we’re going to have to learn and accept, even though nobody’s going to like it:

    No voting system is ever going to be 100% error proof. Not even old fashioned paper ones. It’s going to take heroic effort to make one that’s 99.9% accurate; getting to better than 99.5% accurate is going to be a financial and technical challenge. Which means …

    If the margin of the vote is less than 0.5% (or even more than that, no better than we’re doing now), it will never be possible to say who actually won the election. If the margin of the vote is less than that, we’re going to have to accept that the office was handed out more or less at random.

    Which, in this case, means that if Beto O’Rourke wins the Texas senate seat and the Democrats would only control the Senate with his vote? or, the other way around, if the Republicans hold the Senate by a single vote and that has to include Ted Cruz’s “win” in that seat? And no fraud can be eliminated, and the result survives a recount, and it’s really close? Then we’re just going to have to accept that who gets to chair the Senate committees next year, and whether or not Vice President Pence gets to cast deciding votes, was decided at random, because we’ll never be able to know which side benefited more from the voting machine errors.

    It’s not like it’s never happened before! The negotiations that brought Rutherford B. Hayes the presidency were pretty dubious, enough that some journalists spent his career calling him “Your Fraudulency,” but there was no more definitive way to decide who won the 1876 election, so we had to let him surrender to the Confederate revanchists in 1877. Richard Nixon felt that he had pretty good reasons to believe that the Illinois electoral votes were fraudulently assigned to JFK in 1960 via back-office ballot-box stuffing. It took decades to prove, even tentatively, that he was probably wrong. We certainly weren’t going to wait that long! So he accepted the results with as much grace as he could, ran again in ’68, and won.

    I think it’s pretty clear that both of George W. Bush’s presidential elections were marred by fraud — fraudulent vote counting in Florida 2000, and fraudulent turning away of voters in Ohio in 2004. But neither one could be proved in time. It took the Miami Herald two years to prove the real result of the Florida presidential election of 2000, and we’re still arguing about the polling place problems in Ohio 2004. So, for lack of any better way to decide who won, nobody complained when the Republican majority on the Supreme Court broke every prior election law ruling and every subsequent election law ruling to award the presidency to their party. Or, more to the point, we complained, but we accepted it as settled.

    If the election is close enough to steal, if ANY election is close enough to have been stolen or awarded by mistake, we’ll never ever be able to prove whether it was stolen, mistaken, or not. That’s probably something we’re going to have to live with. Just like our ancestors lived with it before electronic voting was even a thing.

    • John Schilling says:

      Also: God doesn’t actually care whether Prince Charles succeeds Elizabeth II as the King of England, and wouldn’t even if the post still held real power.

      We’re seeing the difference between democracy as a practical method of selecting leaders, and democracy as a legitimizing myth. In practice, if one candidate gets 50.5% of the vote and another 49.5%, either one is probably good enough as a leader (or, alternately, neither is but you’re stuck). In practice, if your party’s candidate only gets 50.5% popular support against an opponent you’re not willing to admit will be good enough for the next term, then you should be ashamed for not having run a better candidate, understand that luck is going to play a potentially decisive role in translating popular support to electoral victory even if every vote is perfectly counted, and accept that all you’ve earned at 50.5% is a bigger slot on the roulette wheel that’s about to be spun.

      But if you need democracy as a legitimizing myth, if people are going to go around sending pipe bombs through the mail unless they are convinced their government is anchored in free and fair elections, then you need to either count the votes perfectly or get everybody to agree not to look too closely at the flaws in vote-counting. And counting the votes perfectly is only slightly more plausible than arranging for a deity to anoint your next king.

      • arlie says:

        I think we’re actually already seeing the breakdown of democracy as a legitimating myth, particularly in the USA. I’m not 100% certain – what civics I learned was in Canada, and many aspects of the US system are very different, in ways that are fairly subtle, but nonetheless important – and more often than not completely unstated. In particular, the US system is more adversarial, probably from its initial design. Thus various things which would look like – and be – irretrievable system breakdown in Canada, may well be business as usual in the US.

        Three small data points, in chronological order as I observed them:
        – Attempts to impose “democracy” on other countries by force. Imposing “capitalism” by force can make sense. So can ousting dictators who have little popular support, and letting local people then rearrange their government to their satisfaction. But forcing democracy? When the voters don’t want it. Huh?
        – Media productions that portray a monarchy, mostly from the view of the royals, with no politics in sight – and call it a “democracy” – as if the name was a magic label giving it legitimacy and approval.
        – What appear to me to be increasing numbers of irreconcilable defeated voters – people who basically don’t concede that their opponents will do an adequate job in power, or even that they were legitimately elected. Birthers and people who went from voting against Trump to immediately demanding his impeachment. Nutters with guns and bombs.

        In the 1960s, there was a lot of conflict, and probably a lot of people who didn’t accept/respect either political party. But those that weren’t rioting, or “turned on … dropped out” seem to have accepted the legitimacy of the other major party.

        This looks to me more like earlier periods that culminated in the US civil war than like anything recent. Not like immediately before, but perhaps the presidency of Andrew Jackson.

        OTOH, I’m not American, and didn’t follow US news at all closely until relatively recently. Maybe all the current controversy really is just business as usual.

        • dodrian says:

          Could you explain what you’re referring to in your second point about the media presenting a monarchy as a democracy?

          The current political animosity doesn’t feel like business as usual, but honestly I don’t think it’s too different from how it’s been in the past. A few months ago Matt Yglesias of Vox wrote an article pointing out that swing voters still matter in elections. In terms of distrust of politics, back just before Trump’s election Bill Maher apologized for some of the rhetoric he had used to decry Bush and Romney, but insisted that this time with Trump he really meant it. There was definitely a lot of polarization under Bush.

          What has appeared to change is how much social media can amplify this. It does look like people are increasingly angry – but it’s unclear how much of this is just the 1% of active social media users being loud, with 99% staying silent (I think this was discussed on the subreddit recently?). I don’t think that people are losing trust in the system en masse just yet.

          • Nornagest says:

            Could you explain what you’re referring to in your second point about the media presenting a monarchy as a democracy?

            Star Wars prequels, maybe? That’s the closest thing I can think of, but I wouldn’t read too much into those tea leaves.

          • arlie says:

            Yep, I think it was one of the Star Wars movies.

            It seemed like a prize case of “US media expects viewers not to know what ‘democracy’ means” – just to think of it as something vaguely positive.

          • Lillian says:

            The Queen of Naboo is explicitly established to be elected for a four year term, and may not serve more than two terms. That’s not a monarchy presented as a democracy, it’s a democracy pretending to be a monarchy.

            If we’re going to take that as an indicator of what US media expects of its viewers. Then the reasonable conclusion is that they expect viewers do know what democracy means, and will not accept a non-democratic government as the good guys.

            Personally, i take it more as an indicator that George Lucas himself sees democratic governments as default good guys, but thought it would be cool to have a country country that only elected young women as President and called them Queens.

        • cassander says:

          – Attempts to impose “democracy” on other countries by force. Imposing “capitalism” by force can make sense. So can ousting dictators who have little popular support, and letting local people then rearrange their government to their satisfaction. But forcing democracy? When the voters don’t want it. Huh?

          Where in recent years has the US forced democracy against the will of the voters?

          • arlie says:

            It seems like Gulf War II would be the most recent prominent example.

          • cassander says:

            @Arlie

            Your contention is that the Iraqis preferred Saddam?

          • Conrad Honcho says:

            I think the Iraqis preferred theocratic government. They kept trying to put Sharia laws in their constitution and the US envoys kept having to nix it (epistemic status: thing I remember hearing somewhere). They may not have liked Saddam, but they didn’t like (or understand) democracy, either.

          • arlie says:

            @cassander – what Conrad Honcho said

          • cassander says:

            @Conrad Honcho & Arlie

            It’s been most of a decade since the US lost the ability to veto Iraqi constitutional changes. They haven’t established a theocracy. They’re obviously not a particularly liberal society, but they’re no worse than what you’d expect for a democratic arab country.

        • Lillian says:

          In particular, the US system is more adversarial, probably from its initial design. Thus various things which would look like – and be – irretrievable system breakdown in Canada, may well be business as usual in the US.

          This is, ironically, because the US system was designed to be non-adversarial. The government the Framers of the Constitution were envisioning wasn’t even supposed to have political parties. Originally the Vice-Precident was supposed to be whoever got the second most votes in a Presidential election. This didn’t work and was amended less than two decades later, but it illustrates the cooperative mindset. As such the system has few mechanisms to handle or channel actual adversarial competition between parties, with the result that said competition winds up being much more antagonistic.

          • dodrian says:

            The US system was designed to be adversarial in the sense that power is divided into different roles. Each role should be fighting to protect its own power in a way that is intended to ensure no one person can get too powerful.

            Contrast with Canada where the executive Prime Minister is also the head of the legislature, and the cabinet is also mostly drawn from the legislature (in the US if a congressperson is appointed to a cabinet post they are expected to step down from their congregational seat).

            In a parliamentary style democracy the party that holds the legislature wields considerable power, and the parties tend to be more tightly controlled by their leadership. In the US, even when the same party holds the Presidency and both houses (Barack Obama’s first 2 years, Trump’s too), because they were elected in different way, on different terms, it’s much harder for them to enact legislation.

          • Lillian says:

            Yes in that sense it is designed to be more adversarial. Ironically one of the most fundamental failures bedevilling the United States is one that the Framers could not have envisioned: Congress has been systematically and voluntarily divesting itself of its powers, either by explicitly assigning them to the Presidency, or by simply refusing to legislate on contentious issues, leaving them instead to be resolved by the courts. The end result is that whichever party doesn’t hold the Presidency feels disenfranchised , while the Courts have become increasingly politicized.

          • arlie says:

            The thing I noticed first was (some) Americans having no concept of a loyal opposition.

            I saw this in a small volunteer organization, where the whole idea of disagreeing over means – but not ends – appeared inconceivable. It was basically a case of “be loyal to The Leader, or get out”. Now that could be merely one corrupt organization – except the basic concepts didn’t seem to exist in any of the members’ minds, not even as things to be given lip service while contracdicting with all your actions.

            Of course I realize that the phrase “Her Majesty’s Loyal Opposition” is a British/Commonwealth soundbite, and so wouldn’t be recognized by many graduates of American “social studies”. But even the concept didn’t seem to ring a bell.

            Other things I later saw in real world US politics merely seemed to add more detail to that first obervation.

          • Nornagest says:

            As I understand it, the “loyal” in “loyal opposition” points to the Crown, not to the party in government? I don’t see how that would make sense in an American presidential context, where the head of state and the head of government are the same guy, or in the context of a small volunteer organization, where it’s not clear what being head of state would even mean.

            Americans have at least historically been willing to impute patriotism to (at least some) people across the aisle, although that does seem to be getting rarer.

          • arlie says:

            @Nornagest

            Loyal to the country, or to the mission of the organization, without being loyal to/partisans of the person currently in charge.

          • Mark V Anderson says:

            The thing I noticed first was (some) Americans having no concept of a loyal opposition.

            I think Americans very much believe in the concept of loyal opposition. How many coups of elected officials have you seen in the US? It is true that the losing side always predicts doom for the country when then other side is elected. But I’ve never seen anyone trying to reverse an election because it was so terrible. The rhetoric each time a Republican president is elected gets worse every time, but I didn’t see any kind of revolution to prevent Trump from taking office. To me that is what a loyal opposition is — when you lose the election you let the other side assume the office and get working on the next election. So far at least the American culture is that the elected winner takes the office, no matter how bad he is. Maybe this will change at some point, but it is still a very strong ethic now.

          • Paul Zrimsek says:

            What’s more, in pretty much every concession speech the loser promises to support the person they’ve spent the last several months telling us will bring the polity to ruin.

        • A1987dM says:

          I agree that these days “democracy” 90% of the times is just an applause light, but how is it incompatible with monarchy? Isn’t there a real meaningful sense in which the Kingdom of Norway is more democratic than the People’s Republic of China?

          • arlie says:

            Not incompatible with a constitutional monarchy.

            But my memory of the movie has none of that explained. Which is why the out-of-the-blue “democracy” claim in a very monarchical presentation seemed so jarring when i watched it.

          • Lillian says:

            Your memory of the movie is faulty. In the Phantom Menace it’s stated by both Queen Amidala and Senator Palpatine that the Queen of Naboo is an elected official, which fits with Governor Sio Bibble’s earlier declaration that Naboo is a democracy. It was probably that first statement that jarred you, since yes up until that point Naboo does appear to be a hereditary monarchy, but as i said it’s later clarified that the monarch is an elected official. The next movie further establishes that the election is for a four year term, and there’s a two term limit. Also that the 14 year old Amidala wasn’t the youngest queen ever elected, which is certainly a rather unique feature of that society.

    • Paul Zrimsek says:

      If the election is close enough to steal, if ANY election is close enough to have been stolen or awarded by mistake, we’ll never ever be able to prove whether it was stolen, mistaken, or not.

      +1, with appended quibble: “we” includes the Miami Herald.

      • Deiseach says:

        I don’t know about American elections, but over here it’s common to have recounts (and more than one in the same centre) because candidates refuse to accept that they’ve missed the quota, or they’re convinced their share of the vote should be bigger.

        When it’s a close election, people will fight for a single vote to go their way. If we didn’t have paper ballots to be recounted, I imagine we’d have a lot more accusations of stolen/cheating/fraudulent results.

        • albatross11 says:

          Basically, election results have an error distribution. If the difference between the winner and loser is small enough, then there’s a reasonable probability that the actual majority of voters wanted the other guy. As John said, this doesn’t cause big problems for democracy as a way of choosing an acceptable leader and giving the public some meaningful feedback.

        • Nornagest says:

          Recounts are uncommon in American elections; with the first-past-the-post system that we have, the only reason to do one is if the margin of victory in a decisive precinct is smaller than the error bars for vote counting. (Or if there’s clear evidence that the tally was corrupted somehow.) That happened for Bush vs. Gore in 2000, but that was basically unprecedented at its scale.

          • Brad says:

            There’s only one major election that applies to. Almost every other election is at-large within a certain district–whether that’s a state, congressional district, city, city council district or what have you. So most of the time when election machines are being used every vote counts. It’s only in Presidential elections that most votes don’t matter.

          • HeelBearCub says:

            Yeah, recounts occur very frequently, simply because there are so many contests. At least a few of them every cycle will be within the legal margin providing the right to ask for a recount. Each state does have different rules on this, but usually it’s on the candidate to ask for a recount. The procedure for this varies from state to state as well, I’m fairly certain.

            States that have of a Board of Elections, rather than running the process via the elected Secretary of State, seem to handle this stuff more professionally, but I can’t really back that observation up.

    • cassander says:

      I think it’s pretty clear that both of George W. Bush’s presidential elections were marred by fraud — fraudulent vote counting in Florida 2000, and fraudulent turning away of voters in Ohio in 2004.

      Rambling on about Bush stealing the ’04 election has always been a sure signal to me that the person writing is a crank. Bush won Ohio by 100k votes, in a state where he had a 53% approval rating, by almost exactly his percentage of the exit polls, and from which bush raised almost twice as much money as kerry. to claim that there was clear fraud is dubious, and it’s an especially rich claim when made a mere paragraph after totally dismissing the claims of kennedy’s fraud in 1960, for which there is considerably more evidence than the nothingburger of ’04.

      The number of elections that are close enough that fraud is possible is pretty small. the number where it’s close enough and of national import is smaller still. this post reeks to me of concern trolling (Not you Nancy, I think it’s a post worth discussing here, I mean that the author was concern trolling). They’re not worried about the legitimacy of democracy, they feel that they’ve gotten (or are about to get) cheated out of some wins for their team and are trying to wrap that feeling up in some higher minded concerns.

      Yes, electronic voting machines are a bad idea and yes, we should have paper ballots, and yes we should have voter ID laws and more rigorous monitoring of voter rolls. But ultimately, the number of elections that are close enough to be within that .5% margin of error is tiny, and the number of national import smaller still. There are far larger threats to democracy than the occasional close election.

      • albatross11 says:

        Several years of participation on this and other fora have demonstrated pretty conclusively to me that Nancy is not a crank. She may be wrong, but her brain works fine.

    • Plumber says:

      @Nancy Lebovitz

      “…I know an ssc reader who doesn’t have facebook account. Are they the only one?…”

      As someone without a Facebook account I thank you!

  6. johan_larson says:

    I keep seeing ads that talk about how “addictive” various online games are and how people playing them are totally “hooked”. It’s strange that advertisers are resorting to these words when trying to attract users. Addition is a bad thing, right? People get hooked on dope of various kinds, and their lives gradually spiral downwards until they end up homeless or in jail. That’s the story of addiction. Yet ad-men are using these words to sell things to people. Did I miss another damn culture shift?

    • Björn says:

      There was an article in the German Vice about a women who made the shitty porn ads you see in the grey area of the internet. She said that it was in fact intended that those ads have spelling mistakes, bad German and are generally stupid. This is because they are targeted at people stupid enough to buy the porn subscription service they offer. I could imagine that the same is true for the video game ads, as they surely are for some kind of browser game or MMO.

    • The Nybbler says:

      I think “addictive” has been used as a positive for video games since at least Pac-Man. The positive implication is that it’s something that you will enjoy indefinitely and be able to fill your time with; it’s aimed at people who are bored, either because they’ve got nothing to do or because the productive things they could fill their time with are tedious.

    • HeelBearCub says:

      I think some of this is euphemism treadmill.

      Compare this to how people self-describe the experience of playing a great immersive game or watching a TV series on Netflix. Words like “binge”, “hooked”, and phrases like “I can’t stop” and “I’m completely addicted” seem common.

      To the extent we are dealing with kind of game that is extremely repetitious, like Candy Crush, etc. people may simply be looking for that experience (which might actually be described as addiction, I think?) It’s the next new high.

      • johan_larson says:

        I had noticed something like that in a different context. “Spoiling” a child used to mean ruining him or her; now it means indulging him or her.

        • The Nybbler says:

          “Spoiling” a child used to mean ruining him or her; now it means indulging him or her.

          I’ve always taken it to mean ruining the child through over-indulgence. A random website (always reliable sources of knowledge!) suggests it had this meaning as far back as the 17th century. A slightly less random website seems to agree.

          • Lillian says:

            In a dating context, spoiling someone is an euphemism for giving them objects of value other than actual money. Such as buying clothes, jewellery, shoes, accessories, fancy dinners, spa getaways, etc. It’s a popular way to have transactional sex while pretending it’s not.

          • johan_larson says:

            I live a couple blocks from a store called “Spoiled Baby”. They sell baby clothes and toys. If spoiling a child retained its original negative meaning, I think they would have called it something else.

            If watering down terms is the new thing, perhaps it is time for an abortion clinic named “Ruined Daughters” and a tough-love tutoring service named “Worthless Sons”.

    • Nick says:

      I don’t know, I think some folks are okay, consciously or not, with describing their habits like they are out of their control. I know I fall victim to this all the time. (Irony fully intended.)

    • beleester says:

      I played on addictinggames.com in middle school. It’s not a recent cultural shift.

      It’s not the only case of negative words being a good thing in an advertising context, either. How often have you seen “decadent” or “sinful” chocolate cake on a menu?

      • johan_larson says:

        Oh, right. “If you’re a jaded second-generation trust fund scion, we suggest you try our deliciously decadent chocolate cake. It’s a welcome distraction from getting high and buggering the servants.”

  7. The Nybbler says:

    3-day rule.

  8. Nornagest says:

    Not cool, dude.

  9. Le Maistre Chat says:

    Update on the SSC Old School D&D campaign:

    In the days of King Minos, before he had a great navy, the north-central coast of Crete was plagued by pirates who dropped their oarsmen off to become land bandits. A Follower of the king named Eumedes hired a wandering Cleric of Apollo (Protus), a witch named Kirke, a foreign Fighter (Outis) and a local Satyr (Talaos) to flush them out of their hideouts for him and his javelin-men.
    Between clearing out two hideouts (a coastal cave and a captured villa), they attended a procession honoring Rhea at the Labyrinth of Knossos, where they picked up a female Thief named Theano and Alek the Explorer (Ranger). They all found amber beads and a gold figurine of a lion-goat monster with a snake for a tail, raising the question of where the pirates had stolen loot before landing here.
    These heroic deeds have impressed the authorities of the Labyrinth, and Kirke in particular is seeking employment there, hoping to learn new spells in the archive. Meanwhile, there is a mission for them that requires borrowing a ship, where they might meet new PCs…

    We play Saturdays on Discord starting at 3/6 Pacific/Eastern.

  10. Nancy Lebovitz says:

    https://www.theatlantic.com/family/archive/2018/10/crime-runs-family/573394/

    There are families with very high proportions of criminals, and it seems that the family culture makes a difference.

    • Mark V Anderson says:

      Not good journalism there. They stated several times that “there is no such thing as a crime gene,” even as the article seemed to indicate differently. Based on research that indicates personality is highly inheritable through genes, and almost none is passed down through parental behavior, it seems likely that the criminal behavior appearing in families is biological, not environmental. IF the article talked about adopted kids being as criminal as those not adopted, then they’d have some evidence. As it is, the article is bunk.

      • HeelBearCub says:

        Did you miss the part where they dramatically lowered recidivism simply by incentivizing the ex-cons to not move back to their old location?

        • quanta413 says:

          I think your overall point about some significant fraction of criminal behavior being environmental or environmental x genetic is correct. Pretty sure heritability of criminal behavior isn’t close to 1. Even if you assumed the impossible like heritability of desire to commit criminal behavior was 1, it seems clear you could affect the actual incidence rate of criminal behavior by making committing crime harder. Someone far away from anyone else has a much more limited range of crimes they can commit.

          Minor nitpick though in that moving someone to a new location where the police don’t know who they are may also affect the odds they get caught. I assume whoever ran the study did correct for any obvious confounds like how likely someone is to be caught in different jurisdictions overall. But if the local cops have your entire family down as troublemakers, that’s rather different and hard to control for.

        • Mark V Anderson says:

          Did you miss the part where they dramatically lowered recidivism simply by incentivizing the ex-cons to not move back to their old location?

          Somehow I did miss that part, oops. Okay it wasn’t total bunk.

          But digging a little deeper, that could still be related to personality research indicating that little behavior comes from parental behavior.
          1) What quanta said — it was just easier to hide.
          2) Even more, I suspect the reason might be the environment does make a difference to criminals. just not the parental environment. At least that is what the research seems to indicate — that direct parental influence has little effect, and yet environment is still 50%, which implies it is other things. So removing them from those other things could well be effective, and still consistent with the latest research.

          Certainly, I’m not saying it is a slam dunk that the parents’ behavior hasn’t effected the crime. For one thing, I think that research occurred in middle class environments; perhaps a different result would occur in criminal environments. Also, I don’t know that criminal behavior was one of those things that have been part of the research. My point was that the article didn’t get us any closer to the truth. The piece I missed initially was somewhat useful, but only somewhat.

      • arlie says:

        *sigh*

        They stated several times that “there is no such thing as a crime gene,” even as the article seemed to indicate differently.

        If there were “a crime gene”, it would be flamingly obvious, kind of like Mendel’s famous peas – single gene inheritance, nice Mendelian pattern, etc. etc. And most importantly, a lot of the offspring would not inherit, unless no one ever bred with someone lacking 2 matching copies of the gene, which would be hard to arrange.

        OTOH, if there were a large number of genes each of which had some variants that statistically raise the probability of criminal behaviour, it would be a compex tangle at least as complicated as intelligence, with raging net.arguments to match. There’s certainly room for that, without there being “a crime gene”.

        Damned if I know what the authors meant, but given the overall level of clue defecit disorder in the average person’s understanding of genetics, I’m betting that the journalists were thinking of wrinked and smooth peas, or sickle cell anemia, not e.g. something like intelligence.

        • a reader says:

          If there were “a crime gene”, it would be flamingly obvious, kind of like Mendel’s famous peas – single gene inheritance, nice Mendelian pattern, etc. etc.

          Actually, there is a kind of “crime gene” with “single gene inheritance, nice Mendelian pattern” :

          X-linked borderline mental retardation with prominent behavioral disturbance: phenotype, genetic localization, and evidence for disturbed monoamine metabolism.

          We have identified a large Dutch kindred with a new form of X-linked nondysmorphic mild mental retardation. All affected males in this family show very characteristic abnormal behavior, in particular aggressive and sometimes violent behavior. Other types of impulsive behavior include arson, attempted rape, and exhibitionism. Attempted suicide has been reported in a single case. The locus for this disorder could be assigned to the Xp11-21 interval between DXS7 and DXS77 by linkage analysis using markers spanning the X chromosome. A maximal multipoint lod score of 3.69 was obtained at the monoamine oxidase type A (MAOA) locus in Xp11.23-11.4. Results of 24-h urine analysis in three affected males indicated a marked disturbance of monoamine metabolism. These data are compatible with a primary defect in the structural gene for MAOA and/or monoamine oxidase type B (MAOB). Normal platelet MAOB activity suggests that the unusual behavior pattern in this family may be caused by isolated MAOA deficiency.

          https://www.newscientist.com/article/mg14018970-600-does-the-aggressive-gene-lurk-in-a-dutch-family/

          More disturbingly, the male members of the family were prone to aggressive, impulsive outbursts, during which they would threaten and fight with people. One had raped his sister, and two had repeatedly set fire to houses. Another tried to run over his employer when he was told that his work was substandard.

          ‘Based on the information gleaned by the family, there was a pattern that was quite compatible with a genetic disorder,’ says Han Brunner, a clinical geneticist at the University Hospital. The pattern of inheritance – afflicted males and normal, female carriers – immediately suggested that a genetic defect causing the disorder might reside on the X chromosome, which sons always inherit from their mothers.
          […]

          The gene codes for an enzyme called monoamine oxidase A (or MAOA), which plays a pivotal role in brain chemistry. MAOA breaks down brain chemicals, such as serotonin, noradrenaline and dopamine, which transmit messages from one nerve cell to the next. In the afflicted males, however, a mistake in the coding sequence prevents proper production of MAOA. As a result, abnormally large quantities of these chemicals are found in the blood of the affected males – and, presumably, in the brain, too.

          Fortunately, this gene variant, with inactive MAOA gene, is present only in that one Dutch family. What exist in population at large are variants like 2R and 3R, with reduced MAOA gene activity (if I understand it correctly) that make those individuals more predisposed of crime, especially if they were abused in childhood, but not always inevitably criminal.

  11. Brad says:

    Outside of a professional context, I agree they are usually bad. The best solution I’ve seen is to have questions written down and handed to some sort of moderator to read through and ask the good ones.

    • I have an hour to give a talk. The basic point I am trying to explain takes forty-five minutes. One possibility is to pad the talk with additional arguments, examples, etc.

      But I think it usually works better to spend the last fifteen minutes responding to particular points raised by the audience, the subset of the ideas related to the basic point that people in the audience are particularly interested in, which I don’t know in advance.

      It also reduces the problem of people in the audience thinking that there is an obvious hole in my argument which would have collapsed it if I had been forced to face it.

  12. How fast can a (double action) revolver fire? I’m having trouble searching this up, because the results are all about how fast people can shoot with them in contests, whereas I’m interested in the mechanical limit of this kind of weapon. In theory, if you had some kind of mechanical linkage to the trigger and you pulled it as fast as possible repetitively until it was empty, how fast could you do that until the mechanism screwed up? As fast as a fully automatic weapon?

    • HeelBearCub says:

      The return of the trigger to its catch position, plus the time it takes the hammer to fall, both done by spring action, would likely be the limiting factor? I’m thinking the length of travel of both won’t let you reach full auto speeds, no matter how fast you pull the trigger from catch to release.

      It’s also not the way actual people get the fastest fire rate from the weapon, which seems relevant, regardless of the hypothetical.

    • John Schilling says:

      Ed McGivern used to do his trick-shooting with double-action revolvers because he didn’t feel that automatic pistols(*) were fast enough for him. He repeatedly demonstrated cyclic firing rates in excess of 650 rounds per minute, while maintaining 2.5 milliradian precision.

      What the actual mechanical limit is, I’m not sure has been rigorously measured or studied. The lock time should be in the 2-10 ms range, which would correspond to 6000-30000 rounds per minute, but long before that you’ll start literally bending metal as you try to accelerate the relatively heavy cylinder through sixty degrees of rotation and bring it to a dead stop in milliseconds, using mechanisms designed for much slower operation.

      * Which are actually semiautomatic weapons, but the action cycles about as fast as it would for a fully automatic weapon.

      • HeelBearCub says:

        But did he actually use the double-action of the revolver? Or did he prepare the striker with his other hand?

        Edit: looking at YouTube videos of Jerry Miculek who broke some of McGivern’s records, I’m simply wrong here. He is utilizing the double action.

        And isn’t the travel distance of the striker significantly longer in a revolver than a typical semi-auto?

        • John Schilling says:

          I don’t think the difference is significant between revolvers and hammer-fired semiautomatics. Striker-fired semiautomatics like the Glock may be a bit faster, but unless McGivern were a big fan of the Austro-Hungarian Imperial Cavalry, there weren’t many of those in his day.

          OK, now I want to see someone run an IPSC three-gun match with a Roth-Steyr ’07, a Federov Avtomat, and a Browning Auto-5.

      • 650 rounds per minute… That’s pretty impressive.

  13. Le Maistre Chat says:

    Your mission is to choose a real volcano, dormant or active, as a gameable/cinematic lair. Explain your choice: setting can be contemporary, ancient-supernatural, or whatever.
    Highest points for underground chambers with visible flowing magma.
    High points for a permanent ice cap with a level carved out of it.
    Bonus points for anything that makes economic sense, like a geothermal power plant or exploitation of volcanic soil for farming.

    • Nornagest says:

      Mount Erebus, Ross Island, Antarctica. It’s got a permanent lava lake (one of only a few in the world), nearby glaciers, dramatic fumaroles, and a really cool name. (It’s also very close to Mount Terror, with an even cooler name. They’re both named after the British Navy bomb vessels that supplied the Ross expedition.)

      My second choice is Erta Ale, Ethiopia. No glaciers, but it’s got another very rare lava lake, and it lies in the Afar Depression, which is about as close as the real world gets to Mordor. The nearest inhabitants are hostile rebel factions prone to kidnapping and murder. Tourists are advised to hire mercenaries.

      • FLWAB says:

        Mount Erebus is a fantastic idea. The triple threat: ice, lava lake, and a cool name.

        • Le Maistre Chat says:

          Mount Erebus is a fantastic idea.

          Well I like your Mt. Rainier idea too. Either way, the fumaroles are too awesome to pass up for volcano without glaciers. If you had a system to direct the steam, you’d be a steampunk volcano villain.
          What would the people in Mt. Erebus be doing, though? Building warm facilities from which to launch studies of the Mountains of Madness? And where would the volcano lair’s food come from?

          • Nornagest says:

            If logistics was a major problem, I wouldn’t be building it in a volcano. I could be blackmailing McMurdo Station for supplies, though; it’s only a few miles away.

            What I’d be doing there is a tougher question. Could be a control station for an orbital superweapon in polar orbit: the extreme southern latitude means I’ll have line-of-sight more often than a more temperate location would. Could be anything that benefits from being really remote and inhospitable, like bioweapons research. I could be tracking down the remnants of the rumored Nazi South Polar base (which will, naturally, turn out to have been zombified). Or I could be breeding an army of killer albino penguins (tekeli-li!).

    • FLWAB says:

      Mt. Rainier. At the top, around the remnants of a crater from the last time it erupted, there are multiple steam vents that carve their way through the thick layer of ice. I would build my lair in the summit glacier. According to the National Park Service the summit contains:

      a maze of steam-riddled snow caverns. Climbers have often taken refuge from storms by descending into the caves from entrances along the crater rim. Geologist Eugene Kiver and Martin Mumma mapped over a mile (1.2 km) of passageways and found a network of tunnels at slopes of 30 to 40 degrees. Geologist Paul Kennard used radar to determine the thickness of Summit Crater Glacier and found it to be about 200 feet (60 m) thick.

      My ice labyrinth lair would be hidden from the outside world, and defended by a maze of ice tunnels filled with poison gas traps. From there I would develop a fleet of ice tunnelers who would expand my network among all the other 25 glaciers on Mt. Rainier, allowing multiple escape routes to take after activating a self destruct sequence. It would be a simple matter to drill down into the volcano and build geothermal generators. It would be the perfect lair: and the perfect place to perfect a device that would threaten to bring about an eruption, at which point I could hold Seattle, Tacoma, Puyallup, Orting, and several dozen other communities ransom! They will meet my demands or face the wrath of my devestating Lahar Engine! Bwahahaha!

    • ryan8518 says:

      Mount Barbaro, home of Carney Park, a US navy recreational facility built into the caldera of an extinct volcano, that at the time we were deployed there growing up was under the control of Commander James Bond, and the home of scout troop 007. It served as the home to many thriller based “spy operations” as soon as we could sneak out of adult supervision.

  14. cassander says:

    My trouble with panels is that while I like the idea, they seem to be one of two types, excessively cordial groupings of people in the same industry/discipline/etc. where they fall over each other to agree, or shouting matches between partisan hacks on television. The latter can be dismissed as unreformable, the questions is how to fix the former. I think you can do it with very good selection of participants and moderators, but it’s not easy and it’s rarely done.

  15. hyperboloid says:

    The FBI has made an arrest in the mail bomb case. One Cesar Sayoc Jr, 56, of Plantation Florida has been taken into custody. A self described native American Trump supporter, he apparently drives a van covered in MAGA imagery, and paranoid ranting. So in other words obvious thing is obvious.

    Except for the native American part, that’s just odd.

    Everybody who engaged in a lot of motivated reasoning about false flags should think over their line of reasoning. Wasn’t it always far more likely that somebody sending pipe bombs, functional or otherwise, to prominent liberals
    was going to be an unstable wight winger?

    If the opposite had happened and bombs had showen up at Trump tower, Fox news, and the private residences of the Koch brothers what conclusion would you have drawn?

    • Randy M says:

      Does native american here mean from a tribe like Cherokee, or someone born here? If the former, isn’t that usually capitalized?

      Everybody who engaged in a lot of motivated reasoning about false flags should think over their line of reasoning. Wasn’t it always far more likely that somebody sending pipe bombs, functional or otherwise, to prominent liberals was going to be an unstable wight winger?

      I believe most people here who speculated about odds had already agreed that it was far more likely to be malice, a threat, or craziness than a false flag.
      Somebody giving it a 10 or even 20 or 30 percent odds of “false flag” does not necessarily need to update when it turns out that it was genuine; odds can’t be proven one way or the other by a single instance, but they in fact were betting the right way.

      • Montfort says:

        Unless they already have an extremely large sample size of comparable incidents to generate their base rate, they should update one way or another. How much they should update, though, is up for debate. Updating too much and too little are both problems.

      • AG says:

        That’s why HBC made his snarky thread, though. “> 1% odds of the outgroup being [potshot at outgroup]” is such a squirrelly veneer for making the potshot at the outgroup.
        “Oh, I didn’t actually believe they were doing it, look I only said > 1%!” and then writing paragraphs of how ridiculous the situation is for the ingroup to have done it is not good faith, no matter the numbers.

        They should be forced to turn it around with a statement more like “I am >90% confident that the outgroup is NOT being [bad thing]” to get any sort of benefit of the doubt.

        • acymetric says:

          Right, the problem with that whole comment thread was that a lot of the sub-comments came off more as “this is what I think happened” than “there is an incredibly small chance that this happened” regardless of the 1% threshold presented in the original comment (which I think got lost deeper into the subsequent comments), to the point that some of the commenters were actively arguing against other [more] likely scenarios (which you would do in a discussion of what you thought was true, not one where you were presenting an argument with a 2% likelihood). Some people really were discussing what they thought highly unlikely but possible scenarios, but the majority read like a bunch of people happy to have an outlet for the conspiracy theory based scenario they actually sincerely believed.

          • albatross11 says:

            The problem with that subthread was a bunch of people speculating about something with no data at all, mostly based on their tribal affiliations.

          • Slicer says:

            My priors were:

            “Nobody remotely sane actually working on behalf of his tribe is going to send his enemies political ammunition” (this turned out to be true)
            “Couriers do not deliver obviously suspicious packages” (false)
            “Mailroom employees do not open obviously suspicious packages” (false)
            “Mailroom employees do not stand around taking pictures of things they think are bombs” (false)

            This is why a hoax seemed much more likely, and I’m still gobsmacked that nobody twigged to “hey maybe we should take a look at the suspicious package checklist at some point”.

          • Matt M says:

            Man, imagine if these things were real and at least one of them actually went off in close proximity to its target.

            How incredibly screwed would the courier industry about to be?

          • baconbits9 says:

            “Couriers do not deliver obviously suspicious packages” (false)
            “Mailroom employees do not open obviously suspicious packages” (false)

            The outsider problem is how often do couriers get asked to deliver weird packages and how often do mail room employees see weird packages. If someone has worked in a mail room for 2 years and opened (random semi large number) 1,000 packages and had 10-20 odd looking ones and had none of them actually dangerous then why would they look at these packages with anything other than detached boredom?

          • Slicer says:

            “How incredibly screwed would the courier industry about to be?”

            Trending tag: #BennyDidNothingWrong (for people who get the joke, New California is out)

          • Slicer says:

            “If someone has worked in a mail room for 2 years and opened (random semi large number) 1,000 packages and had 10-20 odd looking ones”

            While it’s obvious that failure to recognize an unusual situation was a big part of this, I find it doubtful that even 1% of its packages are anywhere near this weird. A couriered package with a Florida return address and stamps on it, to someone who doesn’t even work there?

            I do believe that CNN receives crank letters every day, which probably share many (but not nearly all) of the same characteristics, but that’s only compounding my confusion – if you know that cranks are sending you weird and threatening letters, why would you ever open a package from a super-crank? What if it was full of fentanyl powder or something?

            I know, obvious in hindsight, and you can definitely bet that nobody’s opening such packages anymore (until everybody forgets about this).

          • baconbits9 says:

            I do believe that CNN receives crank letters every day, which probably share many (but not nearly all) of the same characteristics, but that’s only compounding my confusion – if you know that cranks are sending you weird and threatening letters, why would you ever open a package from a super-crank? What if it was full of fentanyl powder or something?

            I don’t know what weird stuff the mail room workers open, but I do believe that in general you get desensitized to stuff quite quickly. Something that looks totally crazy to us might well just look on a far end of “normal”. I believe they probably open up letters from the crazies to make sure they say “you are lizard men following the orders of the Illuminati” and not “I’m waiting for you in the parking lot to do to you what I did to this cat (see enclosed cat)”.

            The reason they don’t worry about fentanyl powder is the fact that they get untold numbers of letters and packages and this type of thing almost never happens (even this thing wasn’t an actual bomb by the current account). Implementing effective protocols that get followed for really rare events is difficult, and generally expensive.

          • HeelBearCub says:

            My wife recently ordered a toilet plunger from Amazon. It looked exceedingly odd when it arrived.

            It’s not exactly clear to me why people think they, as a mail room employee, get to decide that Bob’s tube of ash from Mount Kīlauea doesn’t get opened.

            Hindsight is always 20/20 is a truism for a reason.

          • Slicer says:

            A package from Amazon is from Amazon. It’s got a standard shipping label on it. It’s actually addressed to your wife. Its shape looks weird but the markings on it make total sense. Whatever misgivings that a reasonable person would have about a weird shape would naturally be blown away by the very large piece of evidence that is the shipping label.

            Bob’s tube of ash would presumably be from somebody he knows, or a recognizable company, and would be addressed to Bob, who’d probably be expecting it.

            Now I’m really curious if any amount of suspiciousness would have caused an average worker to stop and think, “hey, something is wrong here, should I talk to somebody before opening this one?” I’d like to go back before this happened and test if something with letters made from clipped-out magazine bits (in the traditional movie ransom note fashion) ostensibly from Amway and addressed to Donald Duck would get couriered, delivered, and opened. Should probably stick an old UPS label on there too.

          • Matt M says:

            I once worked in a mailroom in the US Navy.

            We were given some token training on finding and reporting suspicious packages. In practice though, the job was basically “sort through these 500 pieces of mail every day (450 of which are just junk), and as soon as you’re done, you can either do something more interesting or go back to browsing the internet on your computer”

            In three years I think I found one piece of mail suspicious enough to show to my supervisor. Lord only knows what he did with it. Most likely just threw it away, or handed it to the Admiral and hoped for the best!

          • Slicer says:

            Then, Matt, to truly punctuate this, I have just one final question:

            Would you have reported the particular package couriered to CNN as suspicious?

        • Montfort says:

          I agree with this criticism, and have seen at least one other thread that turned out a similar way for similar reasons (though they are quite rare, afaik). My prior on people being malicious here is low, so I don’t think such things are intentional, but it’s emergent behavior I’d rather avoid if possible.

          In the future, I think if I post such hypotheticals I’ll try to phrase them in a way resembling “Who do you think is responsible for X?” and then give a few broad classes as an example response. The harder version is when you’re considering a fringe theory, as in “Chance lizard people rule Latvia? I put it at 0.05%” – replace lizard people with something contentious, and it looks a lot like someone hiding their power level.

        • Hoopyfreud says:

          I apologize; I sincerely didn’t mean it as a shot, and I can only ascribe my phrasing to inattentiveness. I’ll do my best to follow Montfort’s example.

          • HeelBearCub says:

            Had nothing to do with you. Your phrasing didn’t matter in the least, IMO.

          • Hoopyfreud says:

            I’m not sure I agree; a more neutral phrasing probably wouldn’t have had people overjustifying the improbabilities quite so much. Besides, I made a similar point about one of Honcho’s posts elsewhere, and I feel a bit like I’m in a glass house right now.

      • hyperboloid says:

        I’m pretty sure he meant Native with a capital N, as he apparently owns a business called Native American catering. If you’re going to point out the typographic errors in my posts there are plenty of bigger nits to pick.

        I’m not sure about the tribal question, the name Cesar, not to mention the obvious fact that the man is from south Florida, says Hispanic to me. Perhaps he is just a Latino who strongly identifies with his indigenous ancestry.

        False flag terrorist attacks are incredibly rare, and other than things like the Lavon Affair planned by state intelligence services the only one I can think of are the 2001 anthrax attacks. A lone person sending mail bombs is putting themselves at enormous personal risk, anybody with enough rational foresight to game out the political consequences of a hoax is very likely to be able to find something better to do with their time.

        A ten percent probability seems like it might be a reasonable accounting for uncertainty in cases like this, but if you put it much higher than that then I suspect working from an incredibly strong prior that people on the right are so innately honest and law abiding that there isn’t likely to be one in the whole country wiling to do this.

        • Randy M says:

          I’m pretty sure he meant Native with a capital N, as he apparently owns a business called Native American catering. If you’re going to point out the typographic errors in my posts there are plenty of bigger nits to pick.

          Without the capitalization, I assumed he was a self-professed native who was anglo and taking the identity in an anti-immigrant statement. With the capitalization, it is clear he traces his identity to a pre-Columbian tribe. Not trying to make you look dumb, just clarify.
          (Congratulations, political correctness, you managed to replace an ambiguous label with an equally ambiguous one.)

          A ten percent probability seems like it might be a reasonable accounting for uncertainty in cases like this, but if you put it much higher than that then I suspect working from an incredibly strong prior that people on the right are so innately honest and law abiding that there isn’t likely to be one in the whole country wiling to do this.

          Honest and law abiding ain’t got nothing to do with it. Everyone here instantly knew that this did not in any way help the side that did it. No one was intimidated, and if there had been deaths, it would have done nothing to set back the movement of those targeted. If anything, people putting low odds on it being a Trump supporter or the like were failing in modeling low intelligence people’s thought processes.

        • Sniffnoy says:

          False flag terrorist attacks are incredibly rare, and other than things like the Lavon Affair planned by state intelligence services the only one I can think of are the 2001 anthrax attacks.

          And note that the 2001 anthrax attacks weren’t even really “false flag operations” in the sense discussed here; while the perpetrator attributed them to another group, this doesn’t seem to have been with the intention of turning people against that group. Probably it was just done to throw off the trail.

          • Sniffnoy says:

            Correction: The above comment is probably wrong, see discussion above with Douglas Knight about the motive.

        • A lone person sending mail bombs is putting themselves at enormous personal risk, anybody with enough rational foresight to game out the political consequences of a hoax is very likely to be able to find something better to do with their time.

          It doesn’t take a lot of rational foresight to conclude that a bunch of apparent bombs sent to opponents of Trump will hurt Trump’s party in the election. That’s consistent with not enough sense to realize that he is likely to be caught, at which point the effect reverses.

          It didn’t strike me as the most likely explanation, but not a wildly unlikely one.

          • cassander says:

            It doesn’t take a lot of logic, but it also doesn’t seem to happen very often. The closest I can recall are fake hate/over interpreted crimes.

        • Conrad Honcho says:

          In the suspect’s bizarre, Time Cube-esque rantings on social media he describes himself as “UNCONQUERED SEMINOLE.”

        • toastengineer says:

          but if you put it much higher than that then I suspect working from an incredibly strong prior that people on the right are so innately honest and law abiding that there isn’t likely to be one in the whole country wiling to do this.

          Or a prior that people who are crazy enough to do this sort of thing are very uncommon and usually not great at actually getting crazy stuff done.

    • Matt M says:

      Except for the native American part, that’s just odd.

      Wasn’t it always far more likely that somebody sending pipe bombs, functional or otherwise, to prominent liberals was going to be an unstable wight winger?

      Or Elizabeth Warren looking to eliminate her potential primary rivals?

      • Nornagest says:

        This thread is obnoxious, but you aren’t helping.

      • Scott Alexander says:

        Matt M is banned for six months. If you disagree with this decision, I dare you to tell me which two of “true, kind, necessary” this comment satisfies.

        • Lillian says:

          Can we disagree about the length of the ban? It’s true he has been a dick lately, and that comment doesn’t even hit one of the “true, kind, necessary” criteria, so he definitely deserves a time out. But surely a couple of months ought to be enough for him to get the message, no? This is his first offence, unless you’ve warned him previously and i missed it, and when he’s not being a dick he does add value to the open threads. His commentary on Star Trek Voyager on this very OT, for example.

          • The original Mr. X says:

            Seconded. Even if the post itself wasn’t true, kind, or necessary, I don’t think it’s egregious enough to merit a six-month ban.

          • Hoopyfreud says:

            Thirded; this seems like a duration that would discourage Matt from coming back, not encourage him to moderate himself. I’ve been at least tangent to some of the less-good discussion he’s been involved in recently, and I’d say he’s been a bit obnoxious with comments like this, but not that he’s fundamentally unpleasant. I fear that if he decides to live on Twitter instead of here, that may change.

            If warnings have been previously issued, ignore this post.

          • HeelBearCub says:

            I think a comment I made was eaten, perhaps because I may have named a banned commenter exactly, which I think is verbotten by the system.

            Initially I though the ban was too long and abnormal, especially given that we have seen other comments like this not be banned, but looking back at the register, we have two recent precedents that seem relevant.

            Ancient Geek was banned for a full year for a one line comment directed at Matt. Sailer was banned for two months for redirecting unrelated threads to favored CW topics. The fact that Geek was banned specifically because of Matt also seems relevant.

            We only have one Scott and he is busy. The reign of terror continues. Frankly I just wish Matt had kept a leash on himself, it’s not like he wasn’t aware of what he was doing.

          • Nornagest says:

            Yeah, I’d be happier with two months. I don’t think TAG is a very good parallel; he got his ban, as I recall, for repeatedly attacking other commenters. Matt clearly broke the rules here, but the post was at least directed at a public figure, and we’ve historically cut that a little more slack.

          • AliceToBob says:

            Thoughts:

            A) Scott: that you’re asking people to comment on this ban hints that you’re not completely comfortable with issuing it.

            B) You’ve explicitly been lenient with left-ish commenters, choosing not to ban in cases when you normally would [1]. Presumably, this is to improve the thread quality: even if such an individual oversteps, their comments provide a net good. Perhaps the same leniency is appropriate here, and for the same reasons.

            C) I find HBC’s comment incomplete. There are also instances where left-ish commenters should have been banned and were not. For example, HBC’s short comment of “Fuck off” to Bugmaster [2], for which no ban was issued.

            D) I find BBA’s comment disingenuous. The remark “…just call Warren a c*nt and be done with it” implies a misogyny that Matt has (to my knowledge) not displayed, but which BBA is insinuating nonetheless.

            I’m arguing for no ban at all. I’m also in favor of an apologetic statement from Matt M., as I believe that such things can be useful.

            [1] “You may have noticed some leftists saying things that should have gotten them banned. After some thought, I’ve decided to keep them around anyway with warnings instead (this means you, Brad and Freddie). I will still ban leftists for more serious issues. This doesn’t mean other people will be able to get away with this kind of behavior, so consider yourself warned.” from OT112: OPENTAGON THREAD.

            [2] DIFFERENT WORLDS, Oct 2, 2017.

          • BBA says:

            I meant that to accuse Matt of engaging in content-free insults rather than specifically of misogyny.

          • HeelBearCub says:

            @Nornagest:
            I don’t think his comment was directed at Warren. I believe he indicated several times in the past that he likes to provoke other (liberal) commenters, as sport. It wasn’t a sarcastic reply that was on topic.

            But like I said, it struck me as over the top, but maybe we are all supposed to take something out of that.

          • skef says:

            So if six months is so long as to have too discouraging an effect, and two months is more likely to have the appropriate effect, what is the effect of two months plus twelve people begging for two months because after all Matt M is just a delightful misunderstood scamp?

          • AliceToBob says:

            Using data on the ban list, the durations are roughly [1]:

            indefinitely = 61
            order of days = 3
            order of weeks = 10
            1 to 3 months = 25
            1 year = 1

            So, unless you’re AncientGeek, prior bans tend to send one of two clear messages: (a) go away and never come back, or (b) go away, reform, and then you’re welcome back.

            It makes sense to have these two types, and I’m not convinced that longer, intermediate bans have value. What degree of reform do you expect for a 6-month-to-1-year ban that you won’t obtain via a 3-month ban?

            [1] There are a handful of cases where the ban length is not explicitly listed, and I didn’t count the ban duration.

          • AliceToBob says:

            So if six months is so long as to have too discouraging an effect, and two months is more likely to have the appropriate effect, what is the effect of two months plus twelve people begging for two months because after all Matt M is just a delightful misunderstood scamp?

            It’s hard to read this as anything other than gloating over the banishment of a ideological opponent. Your attack on Conrad Honcho [1] was more personal and vitriolic than most comments I’ve read on SSC, and for that, you received a ban of 1 month.

            [1] OPEN THREAD 104.25

        • meh says:

          I think you should consider making the ban longer, six months seems too short

        • Deiseach says:

          While it may not have been true, kind or necessary, neither was it meant as anything more than a joke. A bad joke, but a joke. If we’re going to apply that standard to comedy, there will be a drastic reduction in the number of comedians down to about three, one of whom would be the late Tommy Cooper.

          Six months is a lot, can’t you reduce it down a bit and grant him clemency?

        • albatross11 says:

          A month or two seems like enough to send an effective “stop being a dick” message; six months seems like “go away and never come back.” I think “stop being a dick” is a better message here than “go away and never come back.”

          Alternatively, have you considered banning people only from the CW-allowed thread and CW-allowed topics, or is that too hard to keep track of?

        • Plumber says:

          The comment seemed like kind of a joke and sometimes levity is neccesary

          • toastengineer says:

            I like to add “true, kind, necessary, and funny” to the “pick two” list.

          • Guy in TN says:

            Terrible idea.

            “x group was just attacked? They probably did it to themselves, being the scum that they are…

            …haha jk guys! Funny joke!”

          • I don’t think the “true, kind, or necessary” criterion works very well. A large fraction of our posts are neither true nor false, since they are not statements of fact. Most are not kind, although only a minority are unkind. Almost nothing we say is necessary.

            Consider the string of Russian jokes we had a while back. Almost certainly none were true. I don’t remember any that were kind–and one might view anything poking fun at a society as at least mildly unkind. And telling jokes isn’t necessary. Zero out of three.

            They were, however, funny.

        • Guy in TN says:

          Counterpoint: Extremely toxic “jokes” that are essentially just smears against the outgroup (and in response to the outgroup receiving an assassination attempt, of all things!) is how you turn into /pol/.

          • Plumber says:

            You make a strong point.

            To me the idea that Warren would do something like that is so ridiculous that it could only be a joke, but I suppose they are madmen who believe the ridiculous i.e. the D.C. pizza place “scandal”.

          • I think the criterion makes more sense if we read “kind” as “not actively hostile.”

        • BBA says:

          I’m no fan of Matt’s (to put it lightly), but I didn’t think it was that much worse than his usual. I still think it’d be more honest and less of a waste of anyone’s time to just call Warren a c*nt and be done with it.

          • Speaking for myself, I have no idea whether Warren is a bad person, which your epithet implies. As best I can tell she did one mildly wicked thing once (claiming to be a minority when she wasn’t) and one mildly dishonest thing repeatedly (claiming a closer connection to Amerind culture than she really had). She also holds political views I disagree with.

            That’s consistent with her being, on net, a better than average person.

        • Protagoras says:

          I support the reign of terror. Bans are good, and cluttering up comment threads arguing about them is bad. Apologies for contributing to the latter myself.

        • bean says:

          Matt has been more toxic than usual lately, and I don’t disagree that a ban is probably warranted. But as many others have pointed out, offenses of his kind have traditionally drawn a month or two, not six.

    • Deiseach says:

      So this was done by Florida Man? Not much of a surprise there.

      I’m going to guess wildly that someone with a name like “Cesar” is of Hispanic in some degree ancestry, so the “native American” part I take to mean “born here in the good old USA unlike them illegals!”

      (Quick Googling says the surname “Sayoc” is from the Philippines, so yeah I’m going to go with “Hispanic ancestry”).

      • Matt M says:

        The van, which was parked in a Plantation, Florida, neighborhood, appeared to be covered in stickers and photos of President Donald Trump and Vice President Mike Pence, in addition to other stickers about politics and soccer.

        Probably an Evertonian.

        • Deiseach says:

          That made me laugh, Matt M 🙂

          And further news reveals that Mr Sayoc is indeed claiming to be Native American (and not just native American), as in “Seminole”, so do we have the DNA test results on that yet?

          • Matt M says:

            Taking bets now, is he over or under 1/1024 native?

          • BBA says:

            You can snicker all you want about Warren (or just call her the c-word, I won’t judge) but the Seminoles are a mixed-race tribe that postdates European colonization, having substantial African descent as well as Native American. DNA won’t tell you much about someone’s connection to their community. (A more extreme example is the entirely synthetic Lumbee, who have lived as an “Indian tribe” for centuries despite negligible Native ancestry.)

            That said, the Seminole Tribe has stated that Sayoc is not and has never been an enrolled member, so that should be the end of it.

          • Matt M says:

            You can snicker all you want about Warren

            Thanks, I believe I will!

            That said, the Seminole Tribe has stated that Sayoc is not and has never been an enrolled member, so that should be the end of it.

            It should, but it wasn’t the last time we dealt with this sort of thing, so let’s have some fun, first!

          • Deiseach says:

            You can snicker all you want about Warren

            The coincidence is irresistible, especially as it’s the one funny element in this unfortunate mess of someone who seems as if he should definitely have been receiving professional help (and maybe some involuntary committal) instead being free to send fake? so badly made they might as well have been fake? (it’s hard to tell if he genuinely wanted to blow people up but was so staggeringly incompetent he failed at this) bombs through the post/courier services.

            Lizzie got to bang on about how proud she was of her native heritage via family tradition and any questioning of “yeah, exactly how much if any native heritage, blonde white woman?” was presented by her supporters as nothing more than a mean-spirited rabid hater, so let’s let this guy have his NATIVE PRIDE as well!

    • cassander says:

      Wasn’t it always far more likely that somebody sending pipe bombs, functional or otherwise, to prominent liberals was going to be an unstable wight winger?

      Yep. Did anyone around here actually say otherwise?

      • hyperboloid says:

        Conrad Honcho had it at a 50 percent chance of being a deranged liberal, I’d like him to explain his reasoning.

        • alexkidd says:

          How do you explain when your 50 percent predictions are wrong??? That happens half the time.

        • Matt M says:

          I’m not Conrad, but my reasoning was based primarily on numerous publicly visible threats in recent years that did, in fact, turn out to be left-wingers threatening themselves in order to garner sympathy for their cause, combined with the fact that the bombs seemed to be shoddily made almost to the point of being intentionally non-operational.

        • Thegnskald says:

          I an going to guess that Conrad still places pretty high odds on that possibility, albeit probably not as high as this morning, and is waiting to see if another shoe drops.

          And, honestly, I can sort of see why someone might put such high odds. The set of “Matters of concern” and the set of “Matters of concern that the media focuses on” don’t have nearly as much overlap as one might expect. Indeed, if something makes the national news, it lowers my own odds that the most obvious political angle is correct, because it takes a certain degree of salaciousness to achieve that, and ordinary crazy is boring. You need something exceptional, and crazy and dysfunctional are both boring and not conducive to exceptionality.

          Which is to say – there is a difference, to take an example from upthread, between the probability that graffiti that says “Jews are bad” was written by a Nazi, and the probability that graffiti that says “Jews are super terrible bad in these ten specific and super stereotype-of-a-Nazi ways, signed genuine Nazi” was created by a Nazi.

          This case feels a little bit more like the latter than the former. Not a lot, but enough to cause people to update in erroneous directions.

          ETA:

          TLDR: Things that make the national news tend to be weird and unusual. Weird and unusual implies less reliable base rates. Conrad’s estimate probably reflects an unconscious reaction to this, and indeed may generally be accurate, with regard to the things that make national news. Unfortunately, this implies Conrad’s predictions of more common occurrences, and indeed his perception of base rates, may be substantially skewed.

        • cassander says:

          I’d say that if it were a deranged liberal, which was unlikely, it’s at least an order of magnitude more likely that it was someone who thought the people in question were selling out the cause by being too right wing than that it was a false flag operation.

          • hyperboloid says:

            A valid point, and it would be a potential motive if the Clintons, or for that matter John Brennan, were the only targets. But Maxine Waters? I don’t see a left wing crazy going after a women of color who once blamed the CIA for the crack epidemic.

          • Matt M says:

            Yeah, the list of victims really reads like a whos who of “prominent Trump critics” as defined by Fox News.

            “Far left person mad that the party isn’t left enough” is way way way down on my list of possibilities, well below “legit right wing guy.”

          • cassander says:

            @Matt M

            “Far left person mad that the party isn’t left enough” is way way way down on my list of possibilities, well below “legit right wing guy.”

            I agree completely.

          • I’d say that if it were a deranged liberal, it’s at least an order of magnitude more likely that it was someone who thought the people in question were selling out the cause by being too right wing than that it was a false flag operation.

            It occurred to me that if this was a thriller/detective story plot instead of the real world, one of the bombs would have been real, aimed at someone who is both a Trump enemy and hated by the person who sent the bombs. That simultaneously conceals the real motive, which might be non-political, and makes the false flag more believable by having one bomb actually blow up.

            Has any author used a version of this for a murder mystery? You arrange for apparent attacks on a bunch of people who all share a characteristic that implies an obvious motive for the attacks. The one attack that succeeds is against the one of those people who, for some unrelated reason, you really want to kill–perhaps you are his heir, or rival in business or politics, or something.

            The closest I can think of is a Father Brown story where an officer gets a lot of his people killed somewhere in order to hide the body that he killed, but that isn’t very close.

            [Added later as an edit]

            My wife points at Agatha Christie’s The ABC Murders, where the third victim is the one the murderer wants to kill, the first two designed to imply a pattern of alphabetic murders by a nut. Not quite as elegant as my version, but in the same ballpark.

          • Hoopyfreud says:

            @David

            Similarly, Borges’ Death and the Compass.

          • Chalid says:

            @DavidFriedman

            It was speculated that the Beltway snipers did what you said; that the older one wanted to kill his ex-wife, and believed that police attention would not focus on him if her death seemed to be a random part of a larger killing spree.

          • Matt M says:

            Undergoing random spree killings in public for several weeks seems like a poor plan to achieve the goal of less police attention on you…

          • John Schilling says:

            The Lee Child novel “One Shot”, and later Tom Cruise movie “Jack Reacher”, had a single targeted assassination concealed as one of five deaths in a sniper’s spree killing. Neither the real assassination nor the spree killing were presented as political, though.

            The mastermind behind the assassination also framed a plausible third party for the killings, which seems almost necessary for this to work. Nutjob-style murders, political or otherwise, rarely go unsolved. If you give people one of the exceptional mass murders that isn’t quickly solved, that will attract more investigative resources and a greater willingness to consider oddball theories than the much simpler “make it look like a robbery gone wrong”.

          • Montfort says:

            @Matt M
            I wonder if a tradeoff between generalized suspicion/attention “there’s a crime going on, who did it” and specific evidence “John’s been acting weird, what’s going on?”/”We know it was John, can we prove it?” would make an interesting mechanic for games where the player commits a lot of crimes.

            But yeah, I think he underestimated just how intense the investigation would be.

          • Matt M says:

            Not quite the same, but there’s also the book/movie Shooter, in which a plausible threat is made against the US President, but the real assassination target ends up being a visiting African dignitary he was meeting (much deeper and overly complex conspiracy follows)

          • Nancy Lebovitz says:

            Some decades ago, I read a story, possibly by Robert Bloch, which included “Where do you hide a tree? In a forest”, and a murder which (from very faint memory) was concealed among what looked like random poisonings.

          • “Where do you hide a tree–in a forest” sounds like the Father Brown story I mentioned, but the deaths were not by poisoning.

          • Nancy Lebovitz says:

            Even fainter memory: “Where do you hide a murder? In a war.”

        • Conrad Honcho says:

          I posted my reasoning when I made my 50% prediction. It was because there have been several high-profile “hate crimes” that the media flogged heavily and that turned out to be hoaxes, like the black church burned before the election with “VOTE TRUMP” scrawled on it. Turns out it was committed by a black member of the church.

          This “bombing” effort looked so comically bizarre, with the misspelled names and the catalog of boogeymen and the TV prop looking “bomb” that it looked like someone impersonating a caricature of a “Trump supporting rural retard.”

          But it turns out that comic figure actually did exist. I saw somebody on twitter describe his van as “what Steve Bannon would turn into if he were a transformer.” Truth is stranger than fiction.

          • Chalid says:

            Not trying to be a dick, but did you make any effort to find a denominator to divide your examples by?

          • Conrad Honcho says:

            I did try googling “high profile threats against democrats” but that’s all dominated by stories about the bomb threats. Off the top of my head I couldn’t think of any highly-publicized threats / hate crimes in the Trump era that turned out to be real except for that one early in the election where some Trump supporters beat up some guy in New York. But that’s the simple kind of hate crime and not the “massive media attention” type of “hate crime” that I think is more likely to be a hoax.

            When I saw this, I thought of the church fire, I thought of the Jewish center bomb threats and the Jewish cemetery defacement that were done by an Israeli teen and an anti-Trump muslim respectively but were immediately blamed on Trump and Trump supporters by the media.

            This pattern matches. The media being right about this one is broken clock territory.

            I did say there was a 20% chance it was a deranged Trumper. I’m batting about as well on that as Nate Silver did on the 2016 presidential election.

            ETA: I am NOT saying “hate crimes are likely to be hoaxes.” I am saying highly publicized anonymous attention grabbing hate crimes are likely to be hoaxes. Another example is the “social experiment” one where a black activist posted “whites only” signs on campus bathrooms or water fountains.

          • Matt M says:

            I did say there was a 20% chance it was a deranged Trumper. I’m batting about as well on that as Nate Silver did on the 2016 presidential election.

            Heh, I was just thinking about how, after the election, plenty of people were right here delivering hot takes of “Just because a 95% probability event didn’t happen, doesn’t necessarily mean the estimation was somehow flawed, we’re just living in the 5% scenario!”

            If Nate Silver can be off by that much and still considered respectable, surely Conrad can get by with losing a coin flip!

          • albatross11 says:

            Matt:

            I think Nate Silver was predicting about a 1/3 chance of a Trump victory. Sam Wang was the guy making a 99% sort of prediction–he promised to eat a bug on TV if Trump won, and he *did*. (Skin, or at least stomach, in the game!)

          • lvlln says:

            Heh, I was just thinking about how, after the election, plenty of people were right here delivering hot takes of “Just because a 95% probability event didn’t happen, doesn’t necessarily mean the estimation was somehow flawed, we’re just living in the 5% scenario!”

            If Nate Silver can be off by that much and still considered respectable, surely Conrad can get by with losing a coin flip!

            IIRC, 538 – Nate Silver’s outlet – gave 30% chance of Trump winning in 2016. It was places like NYTimes that gave 5% or lower chance.

            I guess 20% isn’t that much lower than 30%, though I’d still say that being lower by 33% is actually fairly significant.

          • Conrad Honcho says:

            meh, those are not the type of crimes I’m talking about. Those are all things that actually harmed people or attempted to harm people. I’m talking about the “sending a message” type of “hate crime” that is designed to be as salacious as possible and garner media attention while maintaining anonymity and not hurting anyone.

            Church fire hoax.

            Jewish Community Centers bomb hoaxes.

            Jewish Cemetery desecration hoaxes.

            University at Buffalo “whites only” bathroom hoaxes.

            Trump supporters attack muslim girl and rip off her hijab hoax.

            You’ve got a bunch of “make america white again” graffiti with swastikas and stuff where we never find the perpetrator, but where are the anonymous, salacious “sending a message” hate crimes where we find the perp and sure enough, yup an actual nazi?

            Given all of this, is it that unreasonable to think this comically bizarre crime was not yet another hoax?

          • dndnrsn says:

            The impression I get is that violent crimes where people get seriously hurt – beatings, stabbings, murders – are almost exclusively the real deal. Property damage could go either way. Someone on a university campus gets a threatening note? That’s what’s to be most skeptical of.

            However, “false flag” is the wrong categorization for most hate crime hoaxes. It’s usually someone looking for sympathy, or someone trying to avoid consequences for something else and they try to throw up a smokescreen.

          • Nornagest says:

            where are the anonymous, salacious “sending a message” hate crimes where we find the perp and sure enough, yup an actual nazi?

            It’s at least conceivable that we don’t see those because it’s newsworthy when swastika graffiti is found on a synagogue, and it’s newsworthy if said swastika graffiti turns out to have been written by a dude named Avi Goldstein, but it’s not so newsworthy if it turns out to have been written by a dude named Bubba Wallace with an ugly haircut and a garage full of tiki torches. I mean, that’s what everyone, or at least everyone in news, assumes anyway; it’s a dog-bites-man story.

          • Conrad Honcho says:

            Oh, and there’s a mechanism for this. We know that

            1) extremists are likely to have poor models of their political opponents.

            2) Left-wingers have worse models of right-wingers than right-wingers have of left-wingers.

            3) Anyone who engages in either hate crimes or hate crime hoaxes is an extremist.

            Now I know lots of Trump supporters and don’t know any who like nazis or swastikas* or frightening foreigners in their neighborhoods. But left-wing extremists are very likely to think that we’re all really nazis, and so when modeling us would think “oh man, you know what those nazis would do? They’d send nazi hate messages with frequently backwards swastikas!” So that’s what they do when they make the hoaxes. They incorrectly model right-wingers as hating blacks and loving nazis so that’s what the hoaxes look like.

            * internet edgelords not withstanding

          • dndnrsn says:

            I don’t think I’m super-confident in the ability of actual Nazis to do the swastika the right way around, honestly.

          • Hoopyfreud says:

            @Conrad

            Now I know lots of Trump supporters and don’t know any who like nazis or swastikas* or frightening foreigners in their neighborhoods.

            What a coincidence, I don’t know any leftists who spray paint swastikas to make it look like trumpists did it…

            When the leftist extremists fake the actions of rightist extremists, they look to the internet edgelords because they’re visible; it’s probably not actually that unreasonable to look at /pol/ going “gas the bikes, war race now” and assume that that’s what right-wing extremists think, just like it’s not that unreasonable for right-wingers to look at Tumblr and see “castrate all men for the glory of Trotsky” and assume that that’s what left-wing extremists are like.

            My point is that your point 2 is probably not actually that relevant, especially given that there have been several times that I’ve seen you extrapolate the views of idiots on the internet onto the larger left wing. It’s true that typical left-wingers are worse at modeling typical right-wingers, but I don’t necessarily think that’s true of violent extremists modeling violent extremists.

          • Conrad Honcho says:

            But still, you get the idea, right? High profile. Salacious. Media-flogged. No one getting hurt or in any real danger.

            Given the many other hoaxes that match that pattern, and the lack of proven non-hoaxes that match that pattern, was I unreasonable to consider this a hoax?

            ETA: And more importantly, was the media reasonable to immediately declare it legitimate and Trump’s fault with no error bars?

          • Hoopyfreud says:

            @Conrad

            Eh. I think 50% is probably much too high, but I made my OP at a time when I knew very little about the capability of the bombs. I think my presumed P(false flag) never got higher than like… 10%, maybe? It would have gone higher, but the details that made me suspect the bombs were fake (as in, made by a democrat partisan to aid the party) also made me suspect that the person responsible was mind-bogglingly stupid. I have a lot of trust in the sincerity of the stupid, and this whole thing was baldly stupid to the core.

            Also, the false flags you’re talking about are malicious, but they’re basically petty and all designed to produce outrage; I think all the actual acts of political violence I’ve seen recently (Rand Paul getting shot, that guy who shot up a pizza parlor) have been sincere, and I put fake bombing (which I’d call borderline terrorism – it’s certainly a direct threat) closer to violence than outrage-bait.

            As for the media, they definitely had some people who said unreasonably inflammatory things, but avoiding that shit is why I read (not watch) my news.

          • Conrad Honcho says:

            Eh. I think 50% is probably much too high

            But why? What pattern are you matching to?

            If right-wing nuts want to do terror, they tend to do terror. If the bombs had been real I’d be right with you.

            To my knowledge, there does not exist a pattern of right-wing nuts faking terror attacks.

            There does exist a pattern of left-wing nuts faking right-wing (fake) terror attacks.

            What did you pattern match this event to? Who else has done anything like this? To me this looks like a black swan. The one time a right-wing nut faked a right-wing terror attack.

          • Hoopyfreud says:

            @Conrad

            I think, and have always thought, that this is a very stupid “real” right-wing terror attack, not a real “fake” right-wing terror attack. And I think most of the examples you bring up are not “fake” terror attacks so much as… something like “rabbling?” They’re meant to look like they’re to make people feel hated, not threatened, IMO.

          • meh says:

            meh, those are not the type of crimes I’m talking about.

            @Conrad,
            So I don’t think this reinforces the probability you assigned though.
            Your mental model contains crimes that are only left-wing hoaxes. You can defend your prediction percentage against this model, but you need to consider you are using the wrong model.

            If Nate Silver got 10 elections wrong in a row where he predicted an 80% winner, he would look silly claiming ‘But my model predicted 80% each time, and considered these things… therefore it is a correct prediction!’

            I think one area (but not the only) that can be considered is the scope of the security threat of the hoax. If I was personally tasked w/ a hoax, I would consider how likely a large scale follow up would be, and my chance of being caught, and the repercussions if caught. Graffiti or false report is on one level. Sending bombs (even if designed not to go off) through the mail is another level.

          • albatross11 says:

            I think this is an almost impossible topic to think straight about, without digging into actual hard data. Media bias plus availability heuristic plus confirmation bias means it’s hard to even have confidence about any assertions about the differences between right and left wing terrorists.

          • Conrad Honcho says:

            @meh

            But then it still just looks like “bigger version of the same hoax.”

            Remember from the beginning, no matter who did this it was definitely someone deranged. No sane, normal right-winger does this. We know that because there’s 60+ million people who voted for Trump and all but one of them didn’t do this. There’s 60+ million people who voted for Hillary and none of them did this.

            Right-wing nutjobs don’t threaten to bomb abortion clinics. They bomb abortion clinics.

            Left-wing nutjobs don’t beat minorities, pretending to be right-wing nutjobs. They do, however, “threaten” minorities, pretending to be right-wing nutjobs.

          • Matt M says:

            Speaking for myself, I’ll say that the one thing about this that should have caused me to re-consider my priors were the fact that there were multiple victims.

            Most of the fake hate crimes are ones where there is one victim, and its the alleged victim who turns out to be the perpetrator. They send the fake Nazi letter to themselves, not to some other random black student. So long as we can rule out Barack Obama, Joe Biden, and Robert Deniro as likely suspects, this one has less in common with the typical fake hate crime than I originally mapped it to.

          • Chalid says:

            What pattern are you matching to

            it matches *incompetent* attack, of which there are very many. You probably encountered the occasional forgettable news story about some sad-sack loser who wanted to do something like assassinate Obama but who was under FBI surveillance the whole time and was arrested the moment he started to put his plan into motion. Or there was a guy a few towns over from me who drove a truck into a Planned Parenthood earlier this year, hurting nobody; I doubt anyone outside the region ever heard of it.

            This guy just happened to hit on a way for his incompetent attack to make national news.

            Right-wing nutjobs don’t threaten to bomb abortion clinics. They bomb abortion clinics.

            getting kind of far afield, but this is just availability bias. Threats aren’t newsworthy but that doesn’t mean they don’t occur.

          • The Nybbler says:

            The fake JCC bomb threats had multiple victims (and multiple perpetrators with different motives)

          • meh says:

            Remember from the beginning, no matter who did this it was definitely someone deranged. No sane, normal right-winger does this. We know that because there’s 60+ million people who voted for Trump and all but one of them didn’t do this. There’s 60+ million people who voted for Hillary and none of them did this.

            Right-wing nutjobs don’t threaten to bomb abortion clinics. They bomb abortion clinics.

            Left-wing nutjobs don’t beat minorities, pretending to be right-wing nutjobs. They do, however, “threaten” minorities, pretending to be right-wing nutjobs.

            I think I’m missing the relevance and conclusion for b oth parts here…

            1. I agree it was high probability they were deranged, but I don’t follow the argument here. There are 120 million people who didn’t hit a home run in the world series today. How does that make the ones who did deranged?

            2. you’re saying right-wing nut jobs bomb people, and left-wing nut jobs threaten people? I don’t think that is true; and also this was a bomb…? You’ve imposed a classification on the event so that it only matches to left wing hoaxes, which looks to be incorrect.

          • fortaleza84 says:

            For what it may be worth, I agree with Honcho’s reasoning; what I found persuasive was the New York Times report that the bombs appeared somewhat fake.

            I wouldn’t have put the odds of a hoax at 50%, but I think 20-30% was pretty reasonable.

          • Slicer says:

            Actually, having mulled over this for a while, I want to say something about the words “fake” and “hoax” here.

            We say “fake” and “hoax” because the one we saw is fake, really fake, laughably fake.

            But I have a sneaking suspicion, and I’m not sure what percentage to give it, that the perpetrator did not intend to send fakes. I suspect that he actually believed that these devices would explode and that he did wire an alarmless clock and a silver-painted PVC pipe (possibly full of black powder) together in the belief that it would blow up when someone opened the package. Because it’s a bomb, and that’s how you make bombs, right? By wiring things together the way you see them on TV?

            In other words, I suspect that the perpetrator is a cargo cult mail bomber. (The world’s very first?)

          • Nancy Lebovitz says:

            https://www.cnn.com/2017/08/08/politics/court-filing-unsealed-michael-kadar/index.html

            It’s hard to tell what the motives behind that bunch of JCC bomb threats were, but it was being run as a business. Some it it may have been kids wanting to get out of class.

            And there I was, thinking it was an Israeli with poor judgement trying to scare American Jews into moving to Israel. Silly me.

          • Nancy Lebovitz says:

            Hoopyfreud, Rand Paul wasn’t shot.

            He was assaulted and injured by a neighbor.

            He was present at a shooting attack on politicians.

            https://thehill.com/news/337721-rand-paul-shooting-would-have-been-much-worse-if-capitol-police-werent

      • engleberg says:

        Registered D party member and MAGA supporter- a twofer!

    • alexkidd says:

      Youre still motivated in your reasoning. Wait for the facts to come out… what are chances he acted alone??? Did someone pay him off to do this? If so, who?

    • Le Maistre Chat says:

      Unstable wight winger? Does he fly around clumsily and dive at people he hates to drain a level?

      • hyperboloid says:

        Another one of my patented typos, I have a medical condition that makes it difficult for me to type sometimes. I’ve mentioned it here before.

        • I assumed it was deliberate, a Bugs Bunny cartoon reference.

        • HeelBearCub says:

          I just can’t spell. I’ve always found it sort of cheap to attack people for typos, so I feel your pain.

          Although, I never mind funny little bits about my malapropisms, which I think this was intended to be?

    • idontknow131647093 says:

      In that thread my comments were limited to criticism of the “Russia sent the fake bombs” conspiracy theory floated by Chuck Todd and other prominent MSNBC media person and other “Russia Truthers”.

      I suppose I am now vindicated? I don’t know. Pointing out that one stupid conspiracy theory is beyond stupid doesn’t mean my theory of the case was correct. This guy lives much further from DC than I would have expected, for instance.

    • skef says:

      Everybody who engaged in a lot of motivated reasoning about false flags should think over their line of reasoning.

      The phenomenon is more general than this. What tends to happen in these discussions of probability estimates is that a higher estimate serves as a magical thinking contaminant on the side of the actor for that estimate. “Because the evidence shows a 20% probability that side X did this, rather than 1% as you argued, side X is worse than you’re admitting!” This kind of thinking (or really rhetoric) doesn’t follow and mostly functions as a boo light.

      I would say that the primary lesson here is that much of the discussion focused on tracing back implications of reasonable motivations for doing this, and that only applies to the extent the agent responsible is sensitive to reasons.

  16. Le Maistre Chat says:

    Tabletop RPG discussion, CW edition: so I just heard that there’s a new edition of Vampire: The Masquerade, and it includes trigger warnings and orange and red circles players are supposed to hold up to make the Storyteller change their story whenever it makes them uncomfortable? But a Google search for this also turns up accusations that this edition panders to the Alt-Right? What the heck?

    • Machine Interface says:

      Wouldn’t be the first time White Wolf is in the middle of topical controversies. I remember back in 1997, the critical backlash over “Charnel Houses of Europe: the Shoah”, a supplement for Wraith that detailed how the Holocaust happened in the WoD.

      There was also the infamous “Black Dog” label, which they used to publish WoD material for an 18+ audience, notably the classic “Montreal By Night” and its occasionally pornographic illustrations.

      • Gazeboist says:

        Speaking of 18+ erotic RPG suppliments, anyone else remember that time Book of Erotic Fantasy inexplicably contained one of the most flavorful, evocative, and distinctly nonsexual high level spells in 3.x D&D?

        (It was a resurrection-class effect with a duration as a top level bard spell.)

    • woah77 says:

      I haven’t been paying close attention, but the accusation that it panders to the Alt-Right strikes me as truly and thoroughly unlikely. Onyx Path Publishing is far more in line with the Social Justice crowd than the Alt-Right and the only “pandering” to the Alt-Right that might exist is an updated characterization of certain clans, like Gangrel and Brujah.

    • Nabil ad Dajjal says:

      Is this just the God Machine rules update or an entire new edition?

      Having a mechanic for letting the Storyteller know that you’ve been triggered is silly and redundant when you’re sitting right across the table, but since Vampire has a huge LARP community they may well have asked for it.

      On a practical level, anyone trying to write GM advice should include suggestions for what to do when players get uncomfortable. It’s really not always easy to anticipate: for example, one of the players in my D&D game has a fear of whales and giant squid so I’ve had to think about how I want to handle ocean encounters with her. Sensitivity is important but that means actually listening to your players and not just banning a laundry list of politically inconvenient topics.

      • woah77 says:

        Couldn’t be the God Machine, because that is Vampire the Requiem and not Vampire the Masquerade. Otherwise, I think you’re entirely right about you need to be sensitive to your players, although I think that Onyx Path Publishing is overly heavy handed in their suggested approaches (from what I’ve seen in the new edition of Scion).

      • Lillian says:

        Is this just the God Machine rules update or an entire new edition?

        About three years ago CCP Games sold its White Wolf IPs to Paradox Interactive. Except for Exalted, Scion, and Trinity, which are now owned by Onyx Path. Paradox constituted a new White Wolf Publishing to handle them. To reduce confusion between the old World of Darkness and the new World of Darkness, they had the nWoD renamed to Chronicles of Darkness.

        Onyx Path continues to write and publish all CofD material, as well as the 20th Anniverasy Editions of oWoD. nuWW for its part has been writing, and recently published, a 5th Edition of Vampire the Masquerade. Presumably to be followed by new editions of other classic White Wolf properties.

        It is nuWW’s Masquerade 5th Edition that has been attracting controversy. In a post below i try to explain why.

        Having a mechanic for letting the Storyteller know that you’ve been triggered is silly and redundant when you’re sitting right across the table, but since Vampire has a huge LARP community they may well have asked for it.

        The mechanic in question was invented for convention games. It makes sense in that context, since the people playing don’t know each other, and a lot of nerdy types have difficulty speaking up and asserting boundaries. Efforts to shoehorn it into games between normal groups is, however, pretty much just virtue signalling.

    • Nornagest says:

      Haven’t heard of it, but I’m out of the loop on Storyteller system games. This wouldn’t be the first, or even the fifth, time that White Wolf or its successors tried to shoehorn in some ham-fisted social commentary, though.

      There have even been entire gamelines about it. The difference, of course, is that playing Angry Furry Captain Planet is fun.

      • Le Maistre Chat says:

        I think the Golden Rule of World of Darkness was always “Don’t be pretentious; insteAd use the rules and fluff as a toolbox to have fun.”
        Violent Green furries are fun. Mage leveraged postmodernist dogma to do something fun. (I was never attracted to any of the other lines.)

        • Hoopyfreud says:

          Wait, you’re supposed to play White Wolf games unpretentiously???

          Everything I thought I knew is wrong and makes less sense.

          • Nornagest says:

            Depends on the game. It takes some work to make Vampire or Mage unpretentious, but Exalted is pure weeaboo power fantasy and totally unapologetic about it.

            It’s kind of charming, really. Over the course of your average campaign, you’re actually expected to power up to the point where you can kick the (local equivalents of the) Devil in the nuts so hard he explodes.

          • Le Maistre Chat says:

            Someone once recommended Exalted to me as doing mythical heroes better than D&D. I bounced off the “exalt” mechanic/fluff as too specifically a weeaboo power fantasy. Mythic heroes do not work that way.

          • Lillian says:

            Mythic heroes usually have divine lineages. The difference in Exalted is that rather of screwing your mom, an Olympian instead chose you as their champion. Either way, the power of a god makes you mightier than other men. Seems pretty mythic and heroic to me.

    • Unsaintly says:

      Here’s the basis for the accusations of alt-right pandering, in no particular order. Apologies if I get minor details wrong, this has been a long standing problem with VtM5e and some of it happened over a year ago

      One of the developers answered an interview comment saying that [Far-right people, such as nationalists, nazis and the alt-right] (I can’t remember the exact term used, but they were pretty clear that it was in reference to the recent right wing movements in Europe and USA) are part of their target demographic, and they want to make a game these people can enjoy

      One of the playtest materials includes 1,4,8,8 in an example die roll

      A LARP organized by the new owners contained very strong anit-consent language, stating that people who planned to say no shouldn’t bother showing up

      Vampire: the Masquerade media (I don’t remember if it was explicitly associated with 5e, but it definitely happened around the time 5e was in development) was written by a person known for harassing trans people, and included a villain based on one of his victims

      A playtest document released very near the anniversary of the pulse nightclub shootings was set during/after a gay nightclub massacre, and included rules and language that felt like it was reveling in the suffering to an extent beyond normal “playing as monsters” stuff

      The updated setting contains a lot of traditionally anti-semitic imagery and terms, and reframes the position of vampires in a way that reflects certain alt-right conspiracy theories in a way that’s hard to explain.

      The weakness of the Brujah clan was “Triggered” and they included an alt-right neo-nazi as one of the example concepts. The Brujah have often been portrayed as heroic underdogs in V:tM, and listing a neo-nazi as one of their examples gives the appearance of supporting the narrative that the alt-right are heroic underdogs as well.

      The trigger warnings and circles thing appears to be a later addition that is a response to the criticism they received over their previous issues. In response to the comment below regarding Onyx Path, Vampire: the Masquerade 5th edition is not being developed by Onyx Path. It is being developed by White Wolf under the direction of Martin Ericsson. Onyx Path is more Social-Justice friendly, while recent White Wolf publications have been more explicitly anti Social-Justice apart from the nod to trigger warnings. Additionally, later press statements by White Wolf have explicitly condemned far-right and nazi groups. Not everyone believes them.

    • Lillian says:

      Okay so let me explain. First off, the original White Wolf Publishing has been dead dead for years now. CCP Games, the owner of its properties, had been hiring a third party called Onyx Path to handle the publishing of the materials. Then Paradox bought the White Wolf IP’s (except Exalted, Scion, and Trinity, which are now owned by Onyx Path), and constituted a new White Wolf to handle them. The guy they put in charge on nuWW is a Swede by the name of Martin Ericsson. He’s an oldtime White Wolf player and superfan, and is sometimes called “Dracula” because he was the character model for the full page illustration of Dracula in Vampire the Masquerade 20th Anniversary Edition.

      Now Dracula’s goal is to bring back the glory days of 90s era White Wolf. He was frankly never going to succeed in this, but nonetheless that’s what he was aiming for. He is himself very much still of a very 90s sort of mindset, and as such was totally unprepared to handle the new cultural zeitgeist. The thing is, the original White Wolf was both very leftwing and very transgressive. A lot of their themes and writing pushed on the boundaries of taste and acceptability. While in many ways they hit a lot of left-wing shibboleths about inclusiveness in that they tried to have women, minorities, homosexuals, and transsexuals among their characters, they were also never particularly concerned with political correctness or propriety. Their games have always dealt with a lot of themes like sex, rape, bestiality, abuse, torture, slavery, wanton violence, and other such dark and edgy things. Nor did they always handle these things in a sensitive fashion, much of the time it would outright glory in the transgressive nature of it all.

      Basically back in the 90s you could totally be a leftwing edgelord, and that’s fundamentally what Dracula is, a leftwinger who is also a total edgelord. His problem is that things have changed, the modern left wing is extremely concerned with propriety. In fact the whole transgressive edgelord schtick seems to have been taken up by the alt-right. They are the ones pushing on the boundaries of taste and acceptability. So this leaves Dracula in an unhappy position that when he tries to be edgy and cool 90s style by saying, “I don’t care about propriety!” the modern leftists interpret that as “I’m an alt-right sympathizer!” He is not at all an alt-right sympathizer, if one actually reads the material nuWW has put out, it’s evident that it is still very much on board with the progressive agenda. To my understanding, as far as the actual alt-right is concerned Martin Ericcson is a cuck. But when it comes to signalling, tone matters much more than content, and to the modern social justice crowd, vintage 90s edgelordism looks more like the alt-right than it does like themselves.

      Once the idea set-in that Dracula was secretly pandering to the alt-right, pattern matching set-in, and every single misstep is seen in the worst possible light. Not helping things is that not only is Dracula’s cultural context 20 years out of date, so is his PR handbook. He is so used to criticism about White Wolf’s lack of propriety coming from the right, particularly the religious right, that when the promotion for his new edition of Vampire got similar such criticisms, he reacted with indifference and contempt. This worked fine when the complainants were his outgroup, people who would never buy his products anyway, but it’s not been working out quite so well now that it’s coming from the extreme side of his in-group. The very people he used to be part of.

      The thing with the trigger warnings and the circles were last minutes additions, a desperate attempt to signal, “Hey guys, I’m still on your side.” Needless to say, a lot of people are not convinced. He’s pretty much stuck at this point. He’s too out of touch with the modern left to send the right signals, and since he’s already suspected of being a member of the outgroup so he gets no benefit of the doubt.

      • Le Maistre Chat says:

        Thank you; framing it as IP sale and personal psychology I was unaware of makes a lot of sense.
        White Wolf circa 2002, when I first became aware of it, was obviously Leftist… by way of ’90s Seattle, if that makes sense? Green, espresso-drinking, hip to Derrida, BDSM edgelord post-Marxist New Left, which hilariously is no longer New enough.

      • Deiseach says:

        He is so used to criticism about White Wolf’s lack of propriety coming from the right, particularly the religious right, that when the promotion for his new edition of Vampire got similar such criticisms, he reacted with indifference and contempt. This worked fine when the complainants were his outgroup, people who would never buy his products anyway, but it’s not been working out quite so well now that it’s coming from the extreme side of his in-group. The very people he used to be part of.

        Which, for stodgy old right-wing religious and social conservatives stuck in a mindset preceding the 90s such as myself, is totally hilarious to watch as all the hipsters (not that this was the term en vogue when they were at the height of their flourishing) get eaten by their own side for being stodgy stick-in-the-muds who are probably secretly right-wing. Nothing gets dated as fast as something that was cutting-edge cool in its day.

        For a parochial example where I am particularly indulging in Schadenfreude, there’s the recent case of Graham Linehan (who is an example of 90s leftwing edgelord, or rather typical Dublin southside liberal raised Catholic-now atheist headed off to London to be cool in the media scene there type) a co-creator of amongst other shows Father Ted (which in amongst the comedy was indeed intended to be a barbed attack on Irish Catholicism in particular and Catholicism and religious belief in general). He’s been accustomed to getting praised for the bold brave resistance to the power of the Church and the forces of Irish social conservatism during the 90s and after, but recently he got a visit from the police in England and a warning due to getting on the wrong side of the trans rights movement due to a row on Twitter with a trans person. Progress has marched on, and now he’s insufficiently leftist/progressive to fit in with the new orthodoxy, which has much more power to actively punish him than the dying stages of the old orthodoxy.

        It’s darkly amusing to see the people who pilloried you for certain offences in certain terms (and who got plaudits and applause and praise for same) having the same terms used against them in accusations of the same offences, i.e. being insufficiently zealous for the cause of rainbows and the bending arc of the right side of history 🙂

        It’s like the Liberal Bishop in Lewis’ The Great Divorce, who maintained he fearlessly faced all kinds of danger in publicising his heterodox opinions, but is reminded that his heterodoxy was perfectly in tune with the new orthodoxy of the time. Well, now the heterodoxy is old-fashioned and out of tune, and the purveyors of same are facing real opposition, not the toothless tigers of the old foes.

        “Mine certainly were. They were not only honest but heroic. I asserted them fearlessly. When the doctrine of the Resurrection ceased to commend itself to the critical faculties which God had given me, I openly rejected it. I preached my famous sermon. I defied the whole chapter. I took every risk.”

        “What risk? What was at all likely to come of it except what actually came – popularity, sales for your books, invitations, and finally a bishopric?”

  17. j1000000 says:

    With the recent turbulence in the market, I was inspired to go back and read the Lehman Brothers page on Wikipedia. By all indications, Lehman’s final and longtime CEO Richard Fuld is still extremely rich. Am I economically misguided for thinking that is perverse?

    I’m not even saying he needs to go to prison, but shouldn’t he feel enough shame to give away his fortune?

    • arlie says:

      I’m pretty sure that empathy is strongly selected against in the upper levels of any modern business, not to mention compassion or a sense of shame, and the financial industry is worse than many others. I wouldn’t go so far as saying that every CEO is a high functioning sociopath, but if I wanted to recruit subjects for a study of socipathy without severe impulse control issues, I’d certainly recruit from executive suites 🙁

      • Mr. Doolittle says:

        You need people who can throw the fat man in front of the train to save five. Business leaders need to make hard decisions that cut sometimes tens of thousands of people, and affect the lives of hundreds of thousands. Otherwise the entire business may be in jeopardy and many more people could lose out.

        In selecting for those people, you will inevitably get many others who are willing to not only make hard decisions for the benefit of the overall, but also make decisions for their personal benefit, or the benefit of shareholders/owners, or whatever. It’s exceedingly hard to get people in the first group without getting people in the second group.

        • 10240 says:

          A business wants people who will make decisions to benefit the shareholders (mainly by selecting leaders who want to benefit themselves (i.e. most people) and then trying to align the interests of the leaders with the interests of the shareholders), and that’s perfectly obvious and natural.

          • Mr. Doolittle says:

            Yes, agreed. One common failure mode for “most people” who might be able to make those decisions is to hold off on making hard decisions. That is, decisions that require a significant amount of pain for other people. That’s the layoff situation you responded to.

            I happen to think that this is one of the most common failure modes in huge corporations that seem to slowly sink. GM, Microsoft, Blackberry, a whole list of companies that sat on slowly degrading situations until new leadership came in and made some hard decisions. Those decisions saved the company, at the expense of many individuals who got the boot.

            You need a certain type of person who is able to consistently make those decisions, and not let emotions block an otherwise good/needed decision. (There is an opposing failure mode of someone who slashes early and often for personal gain, but that’s a different conversation).

      • cassander says:

        I would disagree. Understanding the motivations of others is critical to getting them to advance your agenda, and thus getting ahead. Some sociopaths can figure it out without having feelings themselves, but most cant, and they’ll be outmaneuvered by the people who can build those connections.

      • 10240 says:

        I don’t think selection is relevant, I think very few people would voluntarily give away their fortune in his situation because their company went bankrupt. Some people would give away a significant part of their fortune if they were rich, and many wouldn’t because they normally follow their interests, but the latter category wouldn’t even if their company went bankrupt (because, again, they follow their interests).

        More generally, why do you think that empathy is strongly selected against in business leadership?

        • Argos says:

          Strongly agree with the first point. It reminds me of the hypocritical complaints about how rich people use loopholes to evade taxes, usually describing them as morally corrupt, when the reality is that most middle class people use at least some loopholes to bring down their tax burden. “But those rich people have so much more money than me, they don’t need three yachts, so they are able to afford those taxes!” In reality, however, this middle class person is usually also able to pay the full tax bill, and is living in luxury from the perspective of people from third world countries.

          Or in other words: Greed is universal

          However, I do think it is reasonable to assume that heartfelt compassion and the capacity to feel shame is not something that will drive you to the top of an investment firm, given the well known importance of office politics in these kind of organisations.

          • Lambert says:

            I think the issue is that the richer you get, the better the loopholes scale.
            I can’t afford one of those fancy panamanian bank accounts, not a personal tax-avoidance adviser telling me how to exploit the loopholes. But the super-rich can, making it regressive.

          • meh says:

            yes. it is amazing how many mid to lower middle class people will cheer when someone rich goes to prison for white collar crime, yet brag about how they commit insurance fraud.

          • 10240 says:

            @Lambert If you don’t do it because you can’t afford it, that’s irrelevant to a question of morality. Someone who would do it if he could do it, and someone who can do it and does it should be considered morally equivalent. For the purposes of questions like “should the government do more to prevent tax avoidance through loopholes by the rich”, that’s a different question.

    • Matt M says:

      Eh, I dunno. Assuming he did his job as best he could (which I’m sure is what he believes) he shouldn’t have to give back much of anything. It seems unlikely the blame for the financial crisis can fall solely on any one man (or any 100 men, for that matter).

      And even though he’s still extremely rich, he’s presumably less rich than he would have been if Lehman didn’t go bankrupt, which is what really matters as far as incentives go.

      • j1000000 says:

        I guess I find it hard to believe that only walking away with $100,000,000 or so for running a company that, while obviously not solely to blame, is partially to blame for a ruinous financial crisis means the incentives are working.

        • 10240 says:

          Whether incentives are working is mostly equivalent to whether business leaders attempt to avoid their business going bankrupt most of the time as much as possible, and to maximize their profits. I don’t think Lehman Brothers is evidence that this is not the case.

        • Matt M says:

          In order to evaluate this claim, we’d have to estimate how much he would have made had Lehman stayed solvent and rebounded from the crisis.

          Let’s say (total speculation here, I have no idea) it was $200M.

          In that case, the man essentially was penalized $100M for his mismanagement of the company. You don’t consider that proper incentive?

          There’s a reason that most CEOs are mainly compensated via stock options. It’s for this exact purpose.

          • acymetric says:

            It seems like you could easily use that same reasoning to find that he was rewarded $100M for his mismanagement, with that compensation providing incentive for the mismanagement. Had he behaved/run the company in a way that was…more responsible, he might well have received less compensation (because prior to the crash the company likely would have shown lower results and he may have been shown the door). I would argue he was heavily incentivized to run things exactly the way he did. Maybe running Lehman Bros responsibly would only have rewarded him $50M.

            I also think it shows a skewed logic to suggest that not getting the hypothetical $100M is a “penalty”. There is a huge difference between being penalized and not receiving additional compensation.

  18. Statismagician says:

    You are a member of an alien anthropological expedition, secretly studying the peculiarities of [old-school BSG voice] the life-form known as ‘Man’ [/old-school BSG voice]. The time has come to submit your final report. What do you think are particularly odd/noteworthy/interesting characteristics of humans, especially things they themselves don’t realize are unusual?

    • Tatterdemalion says:

      The speed of the changes they’re causing to their planetary ecosystem at the moment. I think it’s likely that most of the time most planets are changing much less rapidly most of the time.

    • helloo says:

      Mostly harmless.

    • The fact that they have somehow missed the marsupial solution, continue to produce offspring far enough developed to survive outside the mother’s body. The triple constraint, female pelvis/infant head/infant development, means that their brains can never get much bigger than they now are. It follows that they are unlikely to ever become smart enough to join the galactic civilization as equal partners.

    • Their striking ability to live in harmony with other species. Areas of dense habitation are shared with mobile species, aerial and terrestrial, ranging in size from within an order of magnitude of their own down, and with immobile species some of which are more than order of magnitude larger than an individual human.

    • cassander says:

      Their uniqueness in alone possessing, of all the creatures of earth, the ability to learn from the mistakes of others. And their striking disinclination to do so…..

    • Paul Zrimsek says:

      Their stubborn adherence to antiquated political and moral ideas, as opposed to adopting the obviously superior ideas of the author in whose story I appear as an incredibly advanced alien.

      • FLWAB says:

        Indeed, but there is hope: among their number are a few brilliant sparks, an ember that may be nurtured into a flame among the stars, made of hopeful young humans who share the authors politics.

      • Nick says:

        There’s a nice subversion in Ursula K Le Guin’s The Dispossessed, where uhznaf ner gur barf jub yrnearq gur yrffba nyernql, gur uneq jnl bs pbhefr.

      • Chevalier Mal Fet says:

        And I was so looking forward to Erikson’s Rejoice, too…

  19. Well... says:

    IME they’re good if the questions dive deeper into the points discussed in the lecture/debate/panel (henceforth, “lecture”), or explore the same topic from angles not covered in the lecture — provided they’re interesting questions, as you implied.

    They are a useless waste of time when the lecture is on topic X but the lecturer/debater/panelist is also known for statements about topic(s) Y, and then the questions are about Y.

  20. johan_larson says:

    So, the Ontario Cannabis Store is struggling to fill orders due to unexpected volume. They say they are working on it, 24/7.

    https://twitter.com/ONCannabisStore/status/1054856978852429824

    If that’s true, some dudes in small-town Ontario are picking, packing and shipping dope at four in the morning.

    • Well... says:

      Hah, I’d love to be the fly on the wall that got to hear this conversation:

      “Soo they’re aboot to legalize pwot fer recreational use.”
      “Dont’cha think we oughta get our order-filling process* ready fer a huge influx of folks tryin’ ta buy pwot from us?”
      “Nah, it probably won’t be too bad, eh.”

      *rhymes with “no mess”

  21. Deiseach says:

    (1) Our presidential election is happening today, with six candidates going for the job including the incumbent, Michael D. Higgins. Voting turnout so far has been very low, primarily because everyone expects Michael D. to get the nod and the selection of alternative candidates are not very appealing; three of them are businessmen whose most notorious accomplishment was being on the Irish version of Dragon’s Den, the two women are a Sinn Féin politician who is not unpopular but not particularly popular either and the other a member of our upper house (so misleadingly in American terms is a Senator, a much less powerful and prestigious position than the American office) who is well-regarded for her mental health work but – seemingly – is also considered too socially conservative.

    So Miggedly for the second term seems the obvious outcome! Everyone (including myself) is fairly happy with him and while there are minor quibbles, no alternative candidate is really inspiring enough. The most interesting thing that happened in our campaign, unlike the American versions, is a genuinely awful ‘attack ad’ put out by one of the would-be presidents attacking Michael D. about the terrible scandal of his – dog grooming bills? Link to it here, I would recommend you view it to see how dreadful and aimless it is!

    (2) We are also having a referendum (that’s how we make constitutional changes over here) on removing the word “blasphemous” from the Constitution in order to strike down the law making blasphemy a criminal offence. This one will probably pass as well, which I find ironic since we’ll remove the religious offence of blasphemy but are fast adding in the secular version of it under the guise of hate speech laws etc.

    (3) Brexit is still stumbling onwards with no resolution in immediate view. Why the Irish border really does matter, from the point of view of the Northern Irish – you know, the people that are as British as Finchley, the ones Boris Johnston seems to think live the same way as if living in London – “there’s no border between Camden and Westminster” – which rather seems to give away the whole point of “the United Kingdom of Great Britain and Northern Ireland” if Northern Ireland and the Republic of Ireland are the same thing as different postal codes in London, thus making the entire island one state. Thanks for your support on a United Ireland, BoJo!

    These aren’t the Paddies down South being troublemakers, these are your fellow subjects of the Queen trying to explain why if the English don’t get their heads out of their arses, there will be trouble.

    (4) And why the Border is a big deal, explained as the late Seamus Heaney would say – phrases make history here. The Good Friday Agreement was something unprecedented, something we never thought we would see. To be so cavalier about it now is appalling. A song explaining why it was so special:

    I was born in Londonderry
    I was born in Derry City too
    Oh what a special child
    To see such things and still to smile
    I know that there was something wrong
    But I kept my head down and carried on

    I grew up in Enniskillen
    I grew up in Inis Ceithleann too
    Oh what a clever boy
    To watch your hometown be destroyed
    I knew that I could not stay long
    So I kept my head down and carried on

    Who cares where national borders lie?
    Who cares whose laws you’re governed by?
    Who cares what name you call a town?
    Who’ll care when you’re six feet beneath the ground?

    From the corner of my eye
    A hint of blue in the black sky
    A ray of hope, a beam of light
    An end to thirty years of night
    The church bells ring, the children sing
    What is this strange and beautiful thing?
    It’s the sunrise
    Can you see the sunrise?
    I can see the sunrise
    It’s the sun rising

    • Lambert says:

      Unfounded accusations of abusing the expenses system? Ha! You’ll never out-duckhouse us Brits.

      And yeah, I’d hate to see the Irish Question re-opened.

  22. Well... says:

    Can anyone link me to some sources (articles, experts talking, etc.) that provide sound, proven guidelines about how kids should comport themselves around strangers in order to be safe?

    The conventional wisdom (e.g. contained in materials sent home from my kids’ school) seems to be to teach your kids not to talk to strangers and I’ve even seen advice to teach your kids to ignore them, but this strikes me as obviously bad advice. I figure the types of people who prey on kids probably look for kids who they think will likely stay quiet and freeze up. Teaching kids to basically do exactly that can’t be the right way to go.

    But to make the case, I’d want some substantial research or experts backing me up. Are there any?

    • Björn says:

      I think in general the chance that children become the victim of a crime perpetrated by a stranger are very low, and have been sinking for at least 30 years. The only reason why people are paranoid about strangers is that very very rarely a shocking crime happens that the media then obsesses about for months.

      A much more substantial threat (and still it’s not that likely) is that a child is sexually abused by a person they know, like a relative, a priest, a teacher or a coach. We know this because there have been sexual abuse scandals that dwarf the victim counts of all American serial killers combined. Just think about the recent abuse scandal in the Catholic church that goes back decades. But don’t think that just because you’re not Catholic, this is not your problem. Abuse happens in all organizations whose members have power and an access to potential victims.

      To protect your children from that, teach them a healthy gut feeling. If a person does not respect their boundaries, this person is not their friend. If they encounter such a person, they should tell their parents, or any other person they trust. See that they get good sex ed so shame can not be used against them. And if they have a bad feeling about a situation, they are right and should get out of it.

      • Well... says:

        Yes, I agree with all that, and hopefully it becomes more widely understood and accepted. But it’s separate from the matter of how we teach our kids to act around strangers, which I think is a legitimate topic in its own right, and that’s what I’m asking about.

        • Aapje says:

          In many ways it should be the same as for people they know: be critical and check with their parents if they are unsure.

          Teaching them that strangers are massive more dangerous than strangers creates excessive mistrust of strangers and excessive trust of people they know.

      • albatross11 says:

        My impression is that stranger danger is massively overhyped. There’s nothing wrong with kids being willing to interact with strangers, as long as they know not to, say, get in a car with them or something. (And that kind of crime is extremely rare.) Teaching kids to be afraid of everyone and treat everyone like a potential predator seems like an amazingly bad way to raise your kids[1].

        [1] I have three kids, and I’ve tried to raise them to not be afraid of strangers, while still recognizing that crime can happen.

        • Matt M says:

          I’ve always thought the key distinction here (for strangers) was who approaches who. If you’re lost in a crowd, approaching a random stranger and asking for help is a fairly safe action, because you are sampling a random population, the overwhelming majority of which are decent people who want to help you and not abduct you or whatever.

          But if you’re in the park minding your own business and a stranger approaches you for something, it’s worth updating your priors and considering that this person wants something from you, and what they want is somewhat likely to not be in your own best interest. I think this is true for children as well as adults. If I’m walking down the street and someone tries to stop me and talk to me, it’s about 10x more likely they are asking for money or wanting to sell me something than it is that they’re asking for directions or something otherwise harmless.

          • arlie says:

            *rofl* I like my neighbourhood better 😉

            At least 50% of the time, what they want is directions, which cost me nothing. Next most common are compliments on the dog I’m walking. Beggars do exist, and folks wanting me to support their politics etc., but mostly in places like the entrances to grocery stores. And the most recent grocery store parking lot encounter I’ve had involved a perhaps overly chatty store employee, who wanted to collect my shopping cart once I finished unloading it into my car.

            Of course I’m a middle aged adult, but I recall similar encounters in childhood – requests for directions, and at least once a disabled person needing assistance with something a non-disabled 10 year old could easily do.

        • acymetric says:

          Yeah, this is true of almost all victim-based crimes. There are a lot of people who are overly afraid of being robbed, mugged, etc. where the likelihood of it happening are incredibly small. Of course, that doesn’t mean you shouldn’t be careful, and avoid making yourself a target, but I know people (friends and family) that are basically consumed with these fears either for themselves and their children, and it doesn’t seem particularly healthy to live that way.

    • brmic says:

      How would that substantial research look like?
      Are you looking for interviews of unknown-kids-raping perpetrators about their strategies? Statistics on victims of stranger-danger and the circumstances of their victimization?

      I’m kind of curious about the matter as well, but OTOH: The important thing to know is that stranger danger is low and getting lower. So, unless the advice is counterproductive, it probably doesn’t matter much either way. Given that, I think some general anti-bullying advice along the lines of don’t look like a victim + some general advice about boundaries + some specific advice like ‘don’t get into a stranger’s car, ever’ probably works almost as well as an optimal strategy and has the advantage of being applicable to a much wider range of cases.

      • Well... says:

        I guess I’m mainly looking for links to trustworthy, possibly-well-known experts giving their views on this matter.

        • brmic says:

          Yes, but how would they come by their views? People confidently asserting this or that are a dime a dozen, but their competence is likely illusionary. Initially, you were looking for ‘sound, proven guidelines’ and so I’m spinning that along, asking how those might come about. And the best I can come up with is identifying ‘predator behaviour’ or ‘victim behaviour’ from, well, actual predators and victims and then checking whether any of these are sufficiently specific that a general rule can be derived from them. (In some ways this is similar to breast cancer screening: For certain women the base rate is so low that even a very good test necessarily produces lots of false alarms.)
          When we’re talking stranger danger, I think our database of ‘victim behaviour’ and ‘predator behaviour’ is necessarily limited, plus we know predators can adapt (e.g. spin the whole no-candy-from-strangers thing first as something they’d never offer, then a private joke, then a shared secret (but we’re no longer strangers, so it doesn’t count) when grooming their next victim) so I doubt there are any rules beyond the obvious that don’t produce lots of false alarms.

  23. johan_larson says:

    It seems to me that colleges’ policy of legacy preference generally hands an advantage to those who are already amply advantaged. I mean, if daddy went to Stanford, your family is probably doing pretty well, and they’re had ample means to give you a good start in life.

    I’m thinking it might be useful for the government to ban legacy preferences because they are detrimental to social mobility (and meritocracy, really.) Is there some second-order negative effect of doing this that I’m missing?

    • SamChevre says:

      Here is my argument FOR legacy preferences (I went to a selective college, neither of my parents graduated from college.)

      One centrally-important thing a good college offers is a network–not just your classmates, but graduates from 20 years ago who can talk to you, parents of classmates, and so on. For that network-building function to work well, people need to have the college as a significant piece of their identity. Legacies are especially-likely to have the college as a significant part of their identity, and so are their parents or other relatives. This provides a significant advantage to the students who need the college to provide them a network that is already built, because the networks they have are not the ones they need. The legacy students already have valuable networks in general; having a significant number of them make the college network much more valuable to the non-legacy students.

      To look at it from another perspective, if your father went to Stanford, and you will go to some school that’s roughly as good as Stanford, you will have your father’s network to draw on; if you go to Stanford, though, your classmates ALSO get your father’s network, while they won’t get as much of it if you go to Columbia.

      In general, though, I’d like to see a heavy tax on endowments that aren’t being spent (targeting a 50-year life for an endowment that is not getting new donations) and then let colleges use whatever admission criteria they think good.

      • HeelBearCub says:

        The idea of network is probably a good one, but I’m not sure legacy gets you that right off. Does being a “Penobscot man” really get you access to any other Penobscot graduate’s network? I suppose for the very rich it may, but it seems more of a holdover from a certain aristocratic notion of higher status. I don’t know that it sustains simply under the notion of legacy admissions.

        • SamChevre says:

          My guess–and it’s a guess–is that “being a Penobscot man” opens doors most effectively with other people who strongly identify as “Penobscot men”–and the the parents of legacies are disproportionately likely to be in that category.

          • johan_larson says:

            What’s this Penobscot you are referring to, Sam? There is a bunch of things named Penobscot, but none of them seem relevant. I was expecting a prep school or something.

          • HeelBearCub says:

            I referred to “Penobscot” as a fictional well-heeled college for the elite in the phrase “Penobscot man”. It’s a stand in for something like “Harvard man”.

            The evocation leans on the trope that the Northeast is/was the home of the elite old money who looked down on the rest of the country.

      • gbdub says:

        I’m not sure how much “the network helps the students” plays into it. What almost certainly does play into it is that multigenerational alumni families are more likely to be donors. And probably would start taking their donations elsewhere if their latest progeny gets rejected.

        • Matt M says:

          Well in a practical sense, sucking up to large donors is absolutely why legacy admissions exist.

          Although in a societal sense, I personally think we’d all be a lot better off if donations to large, already well-endowed universities were reduced by a couple orders of magnitude and shifted to more urgent causes.

          • johan_larson says:

            I suppose there’s a conservative argument to be made that the US elite universities are really very good as places of scholarship, whatever their other flaws. And we should be wary of messing with what is working. Something like 18 of the top 20 universities are in the US.

          • rlms says:

            @johan_larson
            Actually 11 by my count. I’m not sure that the US has particularly many elite universities in proportion to its population (compared to other developed countries).

          • Brad says:

            If you go by that list, the UK (66MM people, 16 of the top 100 including 3 of the top 10), Singapore, and Switzerland punch well above their weight and Germany is a big disappointment (82MM people, 3 of the top 100 all in the 60s).

            The US looks lower top. It would need 78 of the top 100 spots to match the UK’s ratio, but to match Germany’s it would only need 12. It has 31, including 5 of the top 10.

          • johan_larson says:

            There are other rankings. This one has the US at 16/20:

            https://cwur.org/2018-19.php

            This one has the US at 15/20:

            https://www.timeshighereducation.com/world-university-rankings/2019/world-ranking#!/page/0/length/25/sort_by/rank/sort_order/asc/cols/stats

            And this one is 17/20:

            https://thebestschools.org/rankings/best-universities-world-today/

            It seems to me the US is clearly in the lead over the rest of the first world in creating tip-top universities. Going by population, a list like that should have a roughly even mix of US and EU schools, a smattering of Asian institutions (mostly Japanese), and the odd entry from elsewhere. But instead they have mostly US schools, a handful from the EU, and the odd entry from elsewhere.

          • rlms says:

            Having looked into the issue slightly more, it appears that Germany is possibly underestimated by these rankings because a lot of research there is done in separate institutes rather than universities. They might also favour Anglosphere universities in other ways, for instance by overvaluing larger universities because they don’t do per capita adjustment.

    • HeelBearCub says:

      State schools pretty much don’t do legacies (AFAIK).

      The second order negative effect on private schools probably involves funding streams, although the more popular state schools tend to have large endowments anyway.

      Then there is the idea of tradition and culture, of which legacy admissions are certainly a part. Whether that tradition is a net good could be argued.

      • gbdub says:

        Back when U of Michigan had an explicit point system, I’m pretty sure there were bonus points for having a parent or grandparent who had attended U of M. But they are unusually selective and are unusually well-endowed for a state school, so that might be a factor.

    • Aapje says:

      I can see the Ivy League deciding to stop taking federal funding entirely. Harvard has $37 billion, so they can manage.

    • A Definite Beta Guy says:

      I’m thinking it might be useful for the government to ban legacy preferences because they are detrimental to social mobility (and meritocracy, really.) Is there some second-order negative effect of doing this that I’m missing?

      I don’t know how incentives might change. But right now, Harvard has a strong incentive to admit smart kids, because it makes the rich kids look smarter. Don’t let Harvard admit so many rich kids, and it changes the incentive to admit the smart kids.

      • Matt M says:

        I’m not sure how this plays out in the undergrad world, but in business schools, the perceived “quality” of your school correlates almost 100% with the outcomes your students achieve (primarily measured by job placement rates, average starting salary, and longer term variables like “number of alumni who are now CEOs”)

        To the extent that real life outcomes largely favor the rich and well connected (and who is going to deny that they do?) the school therefore has every motivation to attract and admit the rich and well connected. And, as you imply, this will benefit the non-rich (but smart) students as well, as they will be exposed and have the opportunity to network and socialize with the rich and well connected (and might be mistaken for them themselves).

    • Chalid says:

      Admitting the rich, connected kids is good for the smart kids who get in, and vice versa.

      The genius from a poor immigrant family gets access to connections and wealth, and the rich kids get to meet the most talented people around.

    • Tatterdemalion says:

      Absolutely. I’m always struck by how much more hostile attention and opposition affirmative action on racial grounds (which I’m also tentatively against, but seems much less unambiguously inexcusable) generates compared to legacy admissions.

      I suspect that the reason they persist is that they’re a good way of encouraging alumni to donate, but I don’t think that excuses them.

      • Matt M says:

        I think for various cultural reasons, people grudgingly accept that sometimes they will be treated unfairly as opposed to richer people from a higher social class with more connections. We accept that one of the benefits to being a highly connected CEO is that you can pull strings to get your son admitted to Harvard, even if he wouldn’t otherwise qualify. (And, notably, that’s part of the reason we want to become a CEO ourselves, we would fully expect to utilize those benefits personally if only we had them).

        Affirmative action turns this completely around. Suddenly, people are being told that they have to be treated unfairly as opposed to significantly lower-status people. It’s one thing for your son to lose his spot to the heir of an oil tycoon (a tradition of the aristocracy pushing the little guy around that humanity has been dealing with since before the Pyramids). It’s a completely different thing for your son to lose his spot to a poor kid from the ghetto whose only parent is a single mom that drives the city bus. There’s no dignity in that, and it turns every human instinct of power and status completely on its head.

        Some people, of course, see that as a good thing. Some might even venture to declare that’s the entire point of affirmative action. But you can’t expect the victims of it to be happy about it.

        • albatross11 says:

          Also, not infrequently, I think the beneficiaries of the affirmative action Harvard admission are children of wealthy American or foreign black parents. Your example is the best case (in terms of making some moral argument for affirmative action).

          • Matt M says:

            I would also suggest that a part of it is surely recency bias.

            Why are so many people right now outraged that Trump is placing new tariffs on steel, when existing tariffs on sugar are also bad for the exact same economic reasons?

            Because one is something that happened/is happening right now, and another is something that happened before many of us were even born, (and that notably managed to not destroy society or make it prohibitively expensive for Americans to eat sugar.)

            Bias in favor of the powerful has existed literally forever. Bias in favor of the officially designated victim classes (which, by sheer coincidence im sure, happens to include groups that are all politically aligned with each other) is relatively new, was introduced recently by fiat, and can be just as easily undone without disaster (whereas, most attempts to undo the power advantages enjoyed by high status people tend to involve guillotines and mountains of skulls).

      • albatross11 says:

        Tatterdemalion:

        One part of that is probably that we spent several decades hammering home the idea that racial discrimination was wrong, and then decided that actually it’s virtuous when done on behalf of some races, but wicked when done on behalf of others.

        Interestingly, right now, the Harvard AA battle seems to be basically the same principle, but in favor of both whites and blacks vs Asians. As I understand it, the practical implication of getting rid of the de facto discrimination against Asians, while keeping affirmative action for blacks, will be decreasing the number of whites on campus by quite a bit. Eliminating AA would get rid of most of the blacks on campus, and I think would slightly lower the number of whites–all those spots being taken by Asians.

        • johan_larson says:

          As is discussed in a Slate article I referenced in an earlier OT, based on research by the Harvard Office of Institutional Research, the highest representation of Asians in the models they studied was 43%, for the model where only academics are considered. Considering other factors such as legacy, athletics, extracurriculars, and personal interviews all push down the portion of Asians admitted, while increasing the number of whites and modestly increasing the number of blacks. Model 3, including all these factors, has this distribution: Asians 25.99%, African-Americans 2.36%, Whites 50.63%.

          Model 4, which adds demographics into the mix (AA, essentially) shifts the ground dramatically for African-Americans, to the detriment of both Asians and Whites: Asians 17.93%, African-Americans 11.12%, Whites 44.08%.

    • Brad says:

      I’d like to see governments less involved with higher education, not more.

    • rlms says:

      Yes, it’s obviously evil from a non-US perspective.

  24. HeelBearCub says:

    I put the odds that Republican operatives have intentionally coded electronic voting machines to change straight party Democratic ballots to vote for Republican candidates as 1% or less.

    Anybody disagree and think it is more likely?

    • idontknow131647093 says:

      I’d put it less than 1% as well. Its a terribly stupid way to cheat and easy to discover. There is a reason almost all voter fraud we have discovered is through absentee/mail ballots: Its freaking easy to do and hard to detect.

    • gbdub says:

      Two wrongs don’t make a right, man.

      • HeelBearCub says:

        a) a theory was proffered (not by me) of which this is a test.
        b) If this is indeed a second wrong, then the first is wrong. And it would be excellent if people recognized what is wrong about them both.
        c) The specifics here are pretty weird, as this isn’t even a touchscreen system.

        • gbdub says:

          a) Your intention of formatting things precisely this way was pretty clearly to prove some point about how we’re all a bunch of right wingers, a fact which you would then pull out against posters you have a beef with, with an end goal of I don’t know what but “good faith effort to improve dialogue” is almost certainly not it.

          b) I don’t think either is wrong in isolation, but the fact that you believe the first to be wrong makes this straight hypocrisy. Again, not exactly a good faith effort to improve dialogue.

          I’ve been trying to give you the benefit of charity that you really are concerned about the quality of dialogue here, but stuff like this makes it a lot harder because it looks like bad faith ego stroking that is justified because “he started it”

          c) if you honestly are interested in the specifics I’ll post some thoughts in direct reply to your original post.

          • skef says:

            People have been openly policing norms here on the basis of what’s “fun” for a long time. Who are the people acting in good faith you wish to contrast HeelBearCub with?

          • gbdub says:

            I’m not attempting to contrast HeelBearCub to anyone other than my earlier more positive impression of his intentions.

          • HeelBearCub says:

            @gbdub:
            I’m deferring my reply at this time, but I will respond later.

          • gbdub says:

            Honestly at this point I’m spent. This was all originally intended as good faith constructive criticism, but now I’m just frustrated and probably going to be repeating myself so I don’t see a lot of good coming from carrying this on longer than the thread and a half it’s already gone on.

          • Paul Zrimsek says:

            Much as I share your annoyance, gbdub, it’s unfortunate that you could only express it by adopting the usual HBC role: deflecting the object-level discussion into Question The Motive. Propriety debates are inherently bad in much the same way that bravery debates are.

          • Thegnskald says:

            Paul –

            My impression is that this isn’t gbdub saying “This is another case of the left also behaving badly”, but rather a more personal “I thought you were better than this, and now I am rethinking that”.

            Which is to say, this isn’t a metacommentary about how debate is conducted here, so much as a statement of dismay as gbdub updates his view of HBC in a negative direction.

            That is, this conversation is -personal-.

            It is the issue I have with HBC and (certain other people here), which I don’t have with, say, Matt M or Conrad. Matt and Conrad say the same sort of shitty things as HBC and (others), but they say them because they don’t really realize they’ll be perceived as shitty; they are saying what they think, and aren’t deliberately violating norms. And will apologize and/or retract their comments if they are convinced they were being shitty. HBC just insists being shitty is justified by other people being shitty.

          • Paul Zrimsek says:

            I get all that; I just hate the Meta Turn no matter which particular form it takes. (Not least because there’s no way to complain without making yourself part of the problem, so I’ll leave it at that.)

          • Thegnskald says:

            Fair.

            There is a certain… brinkmanship, to it.

            Flip side, even discussing the issues it provokes is part of it. We are all aware of the problem, but can’t discuss it, or identify solutions to it, without making it worse.

            My own advocation is that people start viewing argument as a spectator sport, and have confidence that the spectators understand what is going on; skip over the worst arguments and address the best, not out of some sense of community norm, but out of an understanding that the spectators aren’t convinced of your side by engaging in endless cattiness, and indeed are put off by it.

            This has the trap, of course, of getting more annoyed with your own side behaving badly. Which only works if there is a sense of comraderie about it – an inclination to listen to your own side when they say “Woah, man, that isn’t cool”, which is itself it’s own trap, a kind of metapartisanship/conflict view of debate.

            It is a hard problem.

          • gbdub says:

            Meanwhile y’all are having a meta-meta-commentary about an argument HBC and I had, that I have explicitly bowed out of and HBC seems to have quietly exited. Which seems a little rude itself, no?

    • roxannerockwell says:

      Changing all the straight Democrat ballots would probably be too obvious. A better plan would be to change only a certain percentage, but I’m not sure what the optimal ratio would be.
      Another option is that it is really a glitch, but the republicans conspire not to fix it.

    • gbdub says:

      Not enough information. If it is an ongoing glitch as the officials claim, it ought to be reproducible and probably ought to affect other straight party votes (not just Dem), thus easy to test and investigate.

      The specifics are pretty weird though, and it’s problematic whether it was intentional or not. The photos on the Twitter thread would be trivially easy to fake, but the election officials acknowledge a known glitch so we should probably assume it’s real.

      I think the probability of intentional partisan is still low though, because if you’ve got the access and ability to plant a hack like this, why on Earth would you make it easily detectable by an attentive voter? Seems too risky. Make the changes on the back end somewhere.

      I’ve been involved in enough software and testing to know that this kind of bug, and a failure to test for it, is sadly quite plausible, which is why I advocate for paper ballots.

      What might be meaningfully higher than 1% is the chance that this bug was noticed and somebody found it convenient to ignore it. That at least has plausible deniability.

      • Evan Þ says:

        I’m also involved in software, and I agree the Texas Secretary of State’s explanation sounds very plausible. This is the sort of bug that should be caught in testing, but too often is not. Even so, it’s a very serious bug that should be fixed promptly – and we should go back to paper ballots before one of these bugs happens to swing an election without our knowing.

      • skef says:

        I think the probability of intentional partisan is still low though, because if you’ve got the access and ability to plant a hack like this, why on Earth would you make it easily detectable by an attentive voter? Seems too risky. Make the changes on the back end somewhere.

        If I wanted to manipulate voting software to push results towards one outcome, my target wouldn’t be changing the vote to a different value than the one the selected and was displayed to the user. That kind of fraud has the advantage of being virtually impossible for the voter to detect, but the disadvantage of being hard to disguise, in code inspection and especially in independent testing. And there’s a tradeoff there: if I try to do something different in production than testing, that’s all the more difficult to hide from inspection.

        Better, then, to concentrate on UI features that will change the outcomes and can be blamed on the voter.

        I’m not arguing that’s what happened here. But how does a UI explanation like that of the Sec State account for these results. Why, in any design, if a user selected neither “Republican Ticket” or “Ted Cruz” would it be acceptable to present “Ted Cruz” on the last screen? Or is the idea that because of “screen refreshes” the user actually did select “Ted Cruz”?

        The creepy part is that this sort of problem opens the door for manipulation at the municipality level. Say that the vendor is entirely innocent and this is just a bug. If the behavior is consistent and the local elections folks notice, they might influence the outcomes by how they load the slates.

        • gbdub says:

          I’m not sure at what point the actual names get input into the software before being deployed to the polls, or how the software is set up to handle straight party voting. But I would guess the base code is set up to be agnostic so it’s not that “Cruz” overwrote “O’Rourke” despite “Democrats” being selected, it’s something more like one object with a meaningless numerical designator getting a pointer to the wrong second object with a meaningless numerical designator.

          It probably doesn’t help that the software is expected to handle stuff like randomly reordering the list of candidates and to support picking additional not straight ticket candidates individual after selecting straight ticket (as the Twitter voter did).

          I agree someone discovering the bug elsewhere and maybe manipulating the bug to give a desired result is the more likely path of nefariousness.

          Incidentally I really wish the straight party single checkbox option would be banned. I think it really skews down-ballot races in a negative way.

        • Evan Þ says:

          Or is the idea that because of “screen refreshes” the user actually did select “Ted Cruz”?

          This. According to the Secretary of State’s explanation, what happens is that the machine starts displaying the new screen, but it’s delayed for some reason; and then before it finishes displaying, the user hits “Enter” or uses the selection wheel. The machine registers that as selecting the next candidate, who happens to be Cruz. It’s a bug; it should’ve been caught in testing; it wasn’t.

          • HeelBearCub says:

            It doesn’t seem to match with the evidence that the selection you are most likely to be sure of is the top line contest. The “slow” loading screen has to be really slow, so slow that it already shows you Beto selected, then registers some input as a change, then goes on to the next screen without showing you your new selection.

            Frankly, on a dedicated voting system, that seems slightly implausible.

      • albatross11 says:

        Yeah, if you were going to tamper with a DRE (no paper trail to check against), you’d just tamper with the back-end totals.

        Tampering with a VVPAT or an optical scan system (hand-marked ballots scanned when they go into the ballot box) depends on the auditing procedures. If nobody ever looks at the paper (some systems are badly-enough designed that checking paper vs electronic records is needlessly painful), then you tamper with the electronic totals. The more you’re worried about an audit catching your tampering, the more you need to make this look like an error instead of an attack, but that’s not so hard if you’re willing to settle for a small advantage. (Mess up the template so votes for the other party’s candidate for governor or senate tend to get more undervotes than your candidate.)

      • Brad says:

        I’ve been involved in enough software and testing to know that this kind of bug, and a failure to test for it, is sadly quite plausible, which is why I advocate for paper ballots.

        I don’t understand the driver is for electronic voting and that makes me suspicious. What was the problem that needed solving? The lever machines seem to me to be strictly superior to the electronic machines that are replacing them. Were they just too expensive to manufacture / maintain?

        • John Schilling says:

          I don’t understand the driver is for electronic voting and that makes me suspicious. What was the problem that needed solving?

          Dangling Chad. Seriously. That was a crisis that, depending on the degree of hysteria attendant on certain inconvenient truths, doomed the human race to extinction because we left the counting and interpretation of votes even partially to human beings rather than precise, dispassionate machines. Since it’s maybe still possible to avoid the Apocalypse, it was deemed necessary that we mandate all-electronic vote-counting and that was often easiest to do with all-electronic vote-recording

          And the electronic vote-tabulating systems mostly replaced paper ballots of various sorts, not “lever machines”. Those were I believe only ever used in a minority of states and were mostly retired by the 1990s, before all-electronic systems and with paper ballots (chad optional) as the interim technology. I think only New York went directly from mechanical lever machines to electronic voting machines.

          The mechanical machines were heavy, cumbersome to move into and out of polling places that were used for other purposes most of the year, difficult to maintain, tedious to “reprogram” for each new election whether legitimately or by nefarious hacking, and strictly superior to electronic voting machines in roughly the same way a mechanical calculator is superior to an electronic one.

          • HeelBearCub says:

            Yes, HAVA (the Help America Vote Act) drove the adoption of all electronic voting machines, but not as a replacement of optical scan ballots. Optical scan machines were fine under HAVA.

            It’s imortant to remember that the adoption of punch ballot counters was driven by the increasing need to actually be able to count ballots in a reasonable time frame, especially a problem when you may have 40 or more separate contests in a single election. My recollection is that the manufacturers of these systems stated that their intended use was as a way of establishing whether a hand count was necessary. They weren’t certified to be accurate to determine the result of a close election (but hand counting them is obviously also an issue).

            In the state I worked in, hand recounts of optical scanned ballots could be (I assume still can be) requested by a candidate if the totals fell within a defined margin, that I think was 0.5%. Non-document electronic machines are not hand re-counted, but I think their internal logs are re-processed.

            Lever machines have many of the issues that other non-document systems have, but aren’t subject to easily replicable tampering in the same way software based solutions do.

          • John Schilling says:

            Yes, HAVA (the Help America Vote Act) drove the adoption of all electronic voting machines, but not as a replacement of optical scan ballots. Optical scan machines were fine under HAVA.

            Right, optical scan ballots and all-electronic voting machines were both competing technologies to replace (mostly) punched-card ballots.

            They both have their issues, but the one most commonly attributed to the all-electronic system – the one where an adversary can send agents into the field to individually hack each voting machine in the field – is IMO the least significant.

          • Brad says:

            @HBC
            The lever machines produce a pinched ballot that can be counted. It isn’t totally reliant on an internal counter. And I don’t think counting speed is a serious concern. People may want to know within three hours of the polls closing but they have zero need to know that quickly. Inauguration is at least weeks away and often months.

          • HeelBearCub says:

            @Brad:

            The lever machines produce a pinched ballot that can be counted.

            I was unaware of that, thank you.

            As to speed of counting, I think you are discounting the problems that come with a vacuum of knowledge. Absent relatively quick ballot counting, exit polls will be seen as the de facto results. This leads to some pretty severe consequences if ballot counts don’t line up with the exit polls. The confidence of the electorate in the validity of the results will fall, and quite a bit more than they have already, I believe.

            I’d love a public hand count, like they are able to do, IIRC, in the U.K., but I just don’t think it’s possible in our system with the population totals we are dealing with.

          • dodrian says:

            @HBC

            I don’t think the population is the issue – in the UK the tallies are made locally by volunteers and party representatives then the result called in – that process can scale just fine. The problem is the number of positions being voted for.

            In the UK I think I recall one election where I had two ballot papers – one for my MP and another for my local counselor. In the US, even if we limit elected offices to ones important enough that they shouldn’t be appointed by an executive (as mentioned upthread there are a lot of those!), every two years will at least require the election of a Representative, plus one or more of [Senator, President], plus State-level one or more of [Senator, Representative, Governor]. That’s 3-6 important races, and before we get to ballot propositions and local elections.

            If confidence in voting starts to wain, it might be a good idea to do a hand count or statistical count of the most important race (Senator or President), which could serve to affirm that the less-important races were probably not interfered with.

          • HeelBearCub says:

            @dodrian:

            I mentioned the UK system in a different comment. That’s why I specified “in our system”, as a means of distinguishing from “one vote, Party only” voting systems.

          • Brad says:

            [lever machines are] strictly superior to electronic voting machines in roughly the same way a mechanical calculator is superior to an electronic one.

            The dramatic benefit of an electronic calculator over a mechanical one isn’t the size or weight it’s that it enables entirely new workflows like, say, division not to mention graphing arbitrary functions and doing symbolic algebra. Whereas no one wants or needs entirely new workflows from voting machines, just the same exact things we have always wanted from them.

        • albatross11 says:

          A major driver for electronic voting machines is accessibility. With old-style voting machines, a blind voter had to ask someone to help him vote; with a modern touchscreen system plus a headphone jack, he can vote by himself. That matters a lot to a fair number of blind voters.

          • HeelBearCub says:

            As I mentioned, having alternate voting methods for specific special needs is a valid reason to have these machines available. It does not seem to me to be a good reason to make the vast majority of votes follow this path…

          • albatross11 says:

            Yeah, I think this is pretty common–you have a polling place with one accessible voting machine plus PCOS for everyone else. I know some people have build ballot-marking devices, too–you get the touch-screen interface (with a headphone jack, language selection, screen magnification, etc.), and then the computer fills in your optical-scan ballot.

    • skef says:

      David Graeber has a satisfying section in the introductions to one of his books (more than one, I think*) comparing voting machines to ATMs. All sorts of brands and form-factors of ATMs and varieties of ATM software — when was the last time you personally witnessed or even heard of one spitting out anything other than the selected amount? And part of that design involves dealing with individual pieces of paper!

      Other than the verification aspect, is voting really that much harder a problem? Why can’t we manage to move it out of the crap-designed-by-cranks column?

      * Since he got a faculty job he has had to piss on some of his older writing to make it smell more academic and publish it in “respectable” form.

      • Nornagest says:

        How buggy a piece of software is, is only very loosely correlated with how hard the problem it’s solving is. And ATMs get a lot more hands and eyes on them. If the average person votes every two years (probably an overestimate), and is 100% likely to use voting machine when they vote (definitely an overestimate), and gets cash out of the ATM once a week (no idea, but probably in the right ballpark), then that means their ATM’s getting about a hundred times as much exercise as their voting machine is. The number of voting machine vs. ATM designs probably isn’t anywhere near that divergent.

        It’s also likely that the ATM is getting stricter scrutiny during development, for both legal reasons (more regulation of financial software) and practical ones (the bank cares more, and it’s less vulnerable to the principal-agent issues that often arise with government contractors), but that’s an issue that could theoretically be fixed. This can’t.

        • skef says:

          Stricter scrutiny could also lead to fewer players, which would arguably make the malfeasance problem a bit worse, but could dramatically improve the eyes and hands ratio. This was part of Graeber’s point, that you can slap some half-baked bits on top of Windows and sell that as a voting machine, but you can’t make an ATM that way. Our voting machines say “We don’t take this even remotely seriously.”

      • albatross11 says:

        One problem here is ballot secrecy. An ATM is supposed to log every transaction, so if someone hacks the machine/duplicates my card and PIN and pulls money from my account, that shows up on my bank statement, and I can complain and get my money back. With a voting machine, once I’ve cast my ballot, the machine might do *anything* with that ballot, and there is no way for me or anyone else to know what happened. If tomorrow you go vote on a touchscreen machine with no paper trail, and then it changes all your votes to the opposite party in its electronic records, how would you tell?

        There are clever ways to get around this problem, but they’re not widely used anywhere that I know of[1]. The best widely-used voting technology in the US is precinct-count optical scan–hand-marked paper ballots that get counted electronically at the polling place. To keep that system honest, you need to recount a random sample of the ballot boxes and compare them to the electronic totals from the corresponding scanner.

        [1] The best schemes I know of give the voter a receipt when he casts his vote. The receipt doesn’t let him prove to anyone how he votes, but he can check a public bulletin board for a copy of his receipt, and it proves to *him* that his vote was counted correctly.

        A lot of these schemes get into fairly gnarly crypto pretty quickly. If you want to understand the idea of what people are working on, ThreeBallot is a nice place to start. That scheme has some flaws, but it’s very easy to get your head around, and it is trying to accomplish the same goals as other end-to-end voting systems built on crypto that may be hard to understand without having a bit of background in the area.

        • HeelBearCub says:

          The receipt doesn’t let him prove to anyone how he votes, but he can check a public bulletin board for a copy of his receipt, and it proves to *him* that his vote was counted correctly.

          I don’t think this is true (or even possible.) Unless I am misunderstanding what you mean by this.

          If I can check the receipt for accurately counting my intended votes, I can do it with someone else watching.

          • albatross11 says:

            It’s surprising that this can work, but it really does. A good first cut is ThreeBallot or VAV.

            A sketch of ThreeBallot: (But read the paper I linked–it’s not super complicated!)

            For simplicity, assume we have one two-choice race: you either vote for Alice or Bob. Also, each ballot has a unique serial number.

            Each voter gets three ballots. If she wants to cast a vote for Bob, she marks two ballots for Bob and one for Alice. If she wants to vote for Alice, she marks two for Alice and one for Bob. When she’s ready to cast his ballots, she gets to make a copy of one of them–whichever one she likes. The rest of the system has no idea which one she copied. She casts all three ballots, and they’re only accepted if they follow the rules (two votes for one candidate, one for the other). Once she has cast her votes correctly, she’s given the copy of one of her ballots.

            The voting system posts all the ballots to a public bulletin board. The winner of the election is just done by counting what’s on the bulletin board.

            A voter can check that her ballot was included in the total by checking that her copy of one of her ballots was included on the bulletin board. If the voting system changed her ballot or omitted it, then she will have a receipt that doesn’t agree with the stuff on the bulletin board, and she will know something bad has happened. Since the voting system doesn’t know which ballot she copied, it can alter or replace a ballot with 2/3 probability of success–with probability 1/3, she will catch the tampering. (Ballot-box stuffing is a different attack, and has to be dealt with by checking the number of voters against the number of ballots.)

            Now, imagine I’m trying to pay the voter to vote for Alice. Or maybe I’m telling her she’s fired unless she votes for Alice. She can always vote for Bob, but come back to me with a receipt marked for Alice.

            This scheme isn’t perfect–there are clever and somewhat subtle ways to enforce a vote-buying contract. But it’s pretty easy to understand.

            There are better schemes which are higher security, but they’re usually harder to understand. So understanding ThreeBallot is probably the right way to start.

            All these schemes have this step where the voter interacts with the voting system in such a way that:

            a. She walks away with a receipt.

            b. The receipt is convincing to her, given her knowledge of what happened in the voting booth in what order, that it incorporates her choice.

            c. The receipt can’t be used to prove how she voted to anyone else, at least not without some kind of massive subversion of the whole election system.

            Then, there’s a later stage where those receipts are posted to a public bulletin board and everyone can look at them. (Nowadays, you’d probably incorporate the hash of the bulletin board into a Bitcoin transaction to make it really unchangeable.) And there is a vote counting process that uses what’s published on the bulletin board to determine the election outcome in public. [ETA] The voter can check that her receipt is included on the bulletin board, and if it’s not included or not correct, she has evidence of fraud she can show to the world.

          • acymetric says:

            @albatross11

            I’m not sure the solution to this problem is to make voting wildly overcomplicated, it would certainly introduce new problems and kind of eliminates the benefits of electronic voting to begin with.

          • albatross11 says:

            Well, the existing schemes are actually moderately complicated, too, its just that most people don’t pay a lot of attention to it. (By contrast, some alternative voting methods–ranked choice voting, say–are actually quite complex for the voters.) All-electronic voting machines rely for their security and privacy on the reliability of their software, and it’s not like many voters have any idea how to evaluate that. (And securing software against an adversary who may be one of your programmers is really hard to do. Getting software without critical security bugs is very hard even when nobody but Murphy and human error is against you.)

            In the end-to-end schemes, the voter’s actions are pretty simple–vote, and then check the bulletin board for their receipt. The mechanisms that protect the votes from tampering are more complex, and it’s not at all easy to explain them to nontechnical people.

          • acymetric says:

            @albatross11

            I mean over-complicated for the voter (who doesn’t actually care or need to know about the complicated security involved). I was mostly saying “that seems so highly impractical as to be useless” but I said that before reading the abstract (where they admit this is much more academic than practical) and skimming the paper so my post was maybe unnecessary (for those who followed the link).

            On the other hand, as a developer I am quite familiar with bugs, both security related and otherwise 😉

          • HeelBearCub says:

            She can always vote for Bob, but come back to me with a receipt marked for Alice.

            So, at most they can know that one of their ballots was counted, and in this case the ballot for the person they did not want to cast a vote for, yes? It seems they have literally no idea whether there vote ended up with a net vote increase for the person they wanted to vote for…

            Maybe I should read the paper, but the end results of this seems strictly worse than what we have now (in terms of voter confidence)?

          • John Schilling says:

            It’s surprising that this can work, but it really does. A good first cut is ThreeBallot or VAV.

            This will relatively disenfranchise the sort of people who can’t or don’t read directions and think they are showing maximal support by casting all three ballots for their preferred candidate. And trying to explain that these people have nominated themselves for the short end of a tradeoff, is the kind of thing liable to get the politically clever a bit shortened themselves.

          • There seem to be two different issues here. One is how to prevent vote buying, the other is how to make sure the count of votes is accurate.

            Preventing vote buying at a technical level is a lost cause at this point, because absentee voting is common. If I want to pay you to vote for my candidate, all I have to do is look at your ballot before it is mailed. It’s still possible to try to catch people who are offering to buy votes, but that’s not what fancy voting schemes are doing.

            To prevent miscounting, ideally without letting third parties know how I voted unless I want to let them, all you need do is have a random long number associated with each ballot, give the voter his number, and post all of the ballots by number not by name.

            That still has the problem of fake accusations of fraud. A supporter of candidate X deliberately votes for Y then claims he voted for X. To prevent that you need to provide the voter with a digitally signed copy of his own ballot that he can use to prove how he voted if necessary.

          • Nornagest says:

            I don’t think this is true (or even possible.)

            It’s fairly easy to do it in principle; this sort of thing is well-trodden ground in crypto. A simple approach would be to make the exact encoding of the ballot public, then choose a hash function, make that public too, and publicize a hash of the ballot and a secret credential that’s unique to each voter. Then they can go home and verify that their votes hash to the public value, without those votes being derivable from public information.

            Distributing those credentials is the hard part. If the US had a national ID card, you could use a private key on that, but we don’t. SSNs aren’t secret enough or high-entropy enough to work. Biometrics aren’t consistent enough between scans to use as cryptographic keys, and storing templates centrally would defeat the whole purpose. One solution might be to send a token to each voter with their voter guide for the year.

            Another issue is that the math is pretty hairy. This system is good enough to make the results auditable, which might be all we need, but if you wanted to make votes verifiable to your average voter, then you have a harder problem. It’s more an interface problem than a mathematical one, though.

          • albatross11 says:

            HeelBearCub:

            The voting system, and thus the attacker, don’t know which ballot the voter copied. So if they want to tamper with your vote, they’ve got a 1/3 chance of tampering with a ballot for which you have a receipt, and thus being detected. To change a very close statewide election, they might need to change 100,000 votes. If everyone checks their receipts, that means >30,000 people complaining that their receipts don’t appear on the bulletin board or are wrong somehow. Even if only one person in ten checks, you’ve got 3,000 people complaining, with some physical evidence, that their votes were changed.

            Current voting systems have no way to check statistically whether your vote was included in the final tally.

            A DRE system (computer voting machine with no paper) asks for your vote, says thank-you, and then reports some numbers to the central tally. There’s nothing keeping that system from changing your vote, and no way you could find out if that had happened.

            A precinct-count optical scan system has you hand-mark a ballot and scan it at the polling place, and they keep the ballot. Once you leave the polling place, you also can’t verify that your vote has been included in the final total. An attacker might change the electronic totals and arrange for any auditing of paper ballots not to check that ballot box, or they might tamper with both the electronic totals and the contents of the ballot box. There is no way for you to find out this has happened. (Though in the paper case, the attack requires a lot of people and is hard to get away with if the other party is paying attention.)

            ThreeBallot is more of a thought experiment than a real system, and has some security problems, but it demonstrates that it’s possible to do a lot better than current systems in terms of making it hard for corrupt people running your elections to get away with cooking the results.

            Probably the most well-developed end-to-end voting system I’ve seen described is Scantegrity, which has been used in some local elections. (But I haven’t been keeping up with this area of crypto for the last few years, so maybe there’s something way better out there.)

          • albatross11 says:

            David:

            Yeah, if you don’t care about ballot secrecy, the problem is a lot easier to solve. It’s somewhat similar to the difference between building an electronic payment system with vs without anonymity.

          • HeelBearCub says:

            @albatross11:
            I see what you are saying, it makes it more statistically unlikely that ballot tabulation fraud can swing an election. However, I think that’s a good deal weaker than proving to the individual voter that their vote was tabulated correctly (while keeping it secret from 3rd parties).

            @Nornagest:
            Yes, I agree there are ways to allow a voter to check their tally while choosing to keep it secret. What I was objecting to was the claim that you could do this while also guaranteeing that a 3rd party could not know the information. Indeed, in the scheme mentioned, if you wish to keep the information secret from the the 3rd party, you also lose the ability to check that the vote that you wished to cast was recorded, nor do you ever know that your net vote total was recorded accurately.

          • Yeah, if you don’t care about ballot secrecy

            “Ballot secrecy” is two different issues. One is my ability to conceal how I voted if I want to. The other is my ability to reveal how I voted if I want to.

            Mail in ballots mean that I can reveal how I voted if I want to by showing someone my ballot before I mail it. But it doesn’t let anyone discover how I voted without my assistance.

          • albatross11 says:

            Most of the time, the thing people are worried about wrt ballot secrecy is vote buying or vote coercion. If you publish every ballot with the name of the voter, then both are easy, along with more subtle social pressure to vote the way your friends vote. If you publish every ballot in some scrambled order, but don’t link them to the voters, then it’s still pretty easy to sell/coerce votes when the ballot has a reasonable number of questions on it (I tell you to vote Smith for governor and then give you a precise unlikely list of how to vote on the remaining questions, and then look for that sequence of votes in the published ballots.) But at least the social pressure from boss, coworkers, family, and friends doesn’t work anymore, because nobody can tell who you voted for.

          • albatross11 says:

            Nornagest:

            Just a nitpick, but the scheme you described doesn’t work for resisting vote buying attacks. If I give you my secret credential, I can prove to you I voted the right way. Even if I just verify my own vote in your presence using the secret credential, I can prove it to you.

        • Nancy Lebovitz says:

          Here’s a notion: paper ballots (however generated) are carried away from the voting place while being video monitored. They are posted on walls, still video monitored, and anyone can count the votes on the walls.

          A hard part would be making sure the paper ballots have their order randomized (while being monitored) is some way which makes it impossible to connect a ballot to a particular person.

      • gbdub says:

        How often does an ATM need to reformat its options and menus? Basically never. A voting machine needs to be reconfigured for every election, i.e. multiple times a year, and only gets a few days of live-fire testing for each of those reconfigurations. EDIT: and it’s even worse than that since each municipality has a unique configuration.

        Code designed to allow reconfiguration by end users is going to be a lot more prone to bugs.

        On top of that banks own all the info that gets punched into the ATM. There aren’t partisans at each other’s throats over how and how long the data gets retained, or over its distribution. So I’m guessing banks can more efficiently and effectively audit their transaction history and catch any errors.

      • cassander says:

        The difference is in how much hot water the people involved get in if the ATM puts out the wrong amount of money. If the amount is too low, loud complaints will come immediately. If it’s too high, it will take a while, but those loud complaints will come eventually. Either way, heads will roll at the bank somewhere, and probably at the company that made the ATM. Most votes don’t matter, so counting them wrong (unless the error is on a huge scale) is quite likely to ever be notice, and if it is, the people who bought and installed the machines almost certainly won’t suffer for it.

        In sum, incentives matter. From a purely technical perspective, an electronic voting system every bit as good as the banking system could exist, but none ever will.

        • Matt M says:

          Hmmm.

          It occurs to me that there are private elections of significant consequence that do exist in society. Namely, elections for board of directors members at major corporations.

          Now granted, the overwhelming majority of such elections are uncontested. But some are contested quite fiercely and with a whole lot at stake. How do they count/secure votes? Is it significantly different than the process governments use?

          • Brad says:

            Shareholders send in sworn proxy statements to some lawyers and they are counted up. As a practical matter it’d be really hard to commit fraud because there are typically less than 100 entities whose votes matter and they tend to announce public positions.

      • Lambert says:

        There’s probably a better target than ATMs.
        Easier to hit the backend, or hack/phis (phise?) users. Doesn’t require a physical presence like an atm does.
        But for manipulating elections, there are not as many options. (though there is plenty of attack surface to go round: central vote totalling, dissemination of data to voting machines etc.)

      • Nancy Lebovitz says:

        An ATM shorted me by $200 once. Fortunately, there was evidence in the records, and I got a refund. I think it took a couple of weeks.

    • Thegnskald says:

      Looks like a false flag operation to me, probably by Russian agents in the Ukrainian outsourcing firm the software was written by; apparently Ukrainian software developers have gotten cheaper than Indian or Pakistani developers for outsourcing purposes, among other East European nations.

    • BBA says:

      If this were intentional tampering with the election results, it’d be a lot less obvious.

      Probably a typical bug. There’s a <1% chance that this is intentionally planted and meant to look like a bug, in order to sow maximum discord and distrust in the electoral system. Hail Eris.

    • acymetric says:

      I think XKCD tackled how to handle electronic/software based voting the best (at least for the next decade or so):

      https://xkcd.com/2030/

    • dndnrsn says:

      When it starts changing from “straight democrat” to “bent paperclip” that’s when you know to worry.

    • John Schilling says:

      If it were possible for any random hacker to do this, the odds would be >>1% for both Republican-aligned hackers trying to do it “right” and Democrat-aligned hackers trying to do it just badly enough that it would be caught and blamed on Republicans. But because of the obvious “chance the electronic cash dispenser to dispense free cash whenever I use the backdoor code” threat, people who develop this sort of systems have developed pretty good defenses that limit the real threat to a handful of insiders and a short list of Advanced Persistent Threats. At that point, the odds for the “Republican operative” threat go to 1% or less.

      The most likely explanation is poor UI design with no actual malice.

      And the odds of this being an innocent, independent question on your part, I’m estimating at maybe 10%.

    • Garrett says:

      As someone in the software field, my most likely guesses would be:
      1) Stupid user error. Pressed the wrong “select all” button, didn’t notice until the end.
      2) Odd hardware/software behavior. Eg. pressed the right physical part of the screen but the touch registered elsewhere. This happens on my cellphone a fair bit if my hands are cold/wet/whatever.
      3) Bug, not otherwise specified. Eg. you press one of the buttons 256 times an overflow error causes an unexpected state transition.

      If done intentionally (low probability):
      1) Bug designed to “shake up” specified states in certain areas went from one state to another that happened to be noticeable. If you can selectively cause votes to be randomly-distributed in precincts in which your opponents predominantly live you can change the outcome of the election without having to do something which is obviously supporting your side.
      2) Bug designed to switch from straight (D) to straight (R). If done only rarely, it has the advantage of looking like a user error.

      In either case, why show the outcome to the voter instead of simply showing one result and counting another? If there are paper ballots issued/deposited it allows for the recount to be identical.

    • Gobbobobble says:

      Am I the only one disgusted that a “straight ticket” shortcut exists at all?

      • HeelBearCub says:

        Frankly, I think we would probably be better off if we had a strictly party based system. That is by and large how things actually shake out anyway, and it results in a much greater alignment of voter intent to policy outcome. The choice on the voter end is very simply, cast a single vote for your desired party. The idea of the “independent” political statesman is something of a fiction.

        It’s also how a great deal of the rest of the world does it and I believe ballots frequently can be counted by hand within the precincts on election night with representatives of all parties as observers.

        But, all that said, we aren’t getting that system in the US any time soon.

        • Gobbobobble says:

          Hmm, it might be better than the current state but still far from ideal. The more entrenched parties are the more the platform is a tool to serve the organization* than the reverse.

          The biggest issue, IMHO, is lack of choice. Two-party systems almost inevitably devolve to the proverbial Giant Douche vs Turd Sandwich. It shoves any real influence from the base into the primaries, which, being private organizations, are open to all sorts of undemocratic fuckery (ranging from needing the local party boss’s approval to get on the primary ballot all the way up to superdelegates). Reducing your ballot to straight-tickets-only removes what meager choice is left. [hyperbole]If it were up to me the first step would be to get rid of primaries altogether.[/hyperbole]

          If we can reliably get 6+ relevant parties running then maybe. But I’m still skeptical of measures that transfer power to unelected party bigwigs.

          *Edit: I’m not referring to literally supporting laws (or things like the Presidential Debate Committee that I could call “para-laws” and probably be technically wrong but hope y’all know what I mean) that benefit the party at the expense of the voters, though that does happen. I mean how you get nonsense coalitions like Big Business + Evangelicals because the Party figures out it can carefully mold its platform to appeal to both, or at least piss them off less than the other Party, rather than the platform deriving from any sort of coherent philosophy.

        • albatross11 says:

          How does Bernie Sanders fit into this theory? Or Joe Lieberman?

        • HeelBearCub says:

          @Gobbobobble:

          I think you will find that representative parliamentary systems have their own issues, which is that the overall machinery is particularly beholden to the fringe elements when a major party can’t easily form a majority coalition. Perhaps this is your desire, but it tends not to be the desire of 95% of the population. The general trend is still towards two fairly large parties, with long periods of fractious nothing and frequent called elections when neither is particularly dominant.

          What you don’t get is amorphous ever changing coalitions that let these 60% of people who agree on policy X and those other 60% set of people who want policy Y both getting their preferred policies. At the end of the day, their is only one head to the executive branch, which tends to mean either solidarity in coalition, or breakup.

          And if the G Party has 10% of the vote, it may mean you feel that you are represented, but in the end you tend to still get what you would get out of a coalition that formed before the election, where a party platform and particular policy positions are hammered out as compromises.

          But, YMMV and I don’t think FPTP Presidential elections are in any intrinsically better. I just think politics will always be messy and frustrating.

      • Matt M says:

        Man, check out what’s been going on in New Mexico, where former Republican Governor and former Libertarian Presidential Candidate Gary Johnson is running for Senate as a libertarian, started gaining some pretty good traction in the polls, and then New Mexico immediately re-introduced straight party line voting, which they had eliminated a few years prior.

        But no, it’s “Russian” interference that we need to be worried about…

      • Guy in TN says:

        It probably cuts down on wait times, particularly if the ballot is rather long. There’s a trade-off against the button vs. funding more polling stations.

      • dodrian says:

        I agree with you in principle, but while I took the time to decide who I wanted to vote for as Senator, Representative, and Governor, and at least checked to make sure there weren’t any crazies running for the other big positions (Lt Governor, Railroad Commissioner, Ag Commissioner, etc), by the time I’m voting for the n-teenth judicial appointment I honestly have no idea who these people are and don’t really care – it’s going to the party that I think is more likely to represent my interests.

        At least at the local level this time everyone was running uncontested – they tend to sort things out during the primaries.

        • Gobbobobble says:

          Well I also think voting for judges at all is ridiculous. The various Commissioners too, why the hell is railroads something we have campaigns for? Just appoint people.

          This arguably shakes out to a straight ticket depending on when appointments get updated, but it’s still saner. I suppose the commissions could be used for patronage but it’s not meaningfully different when they’re running under a Party.

    • HeelBearCub says:

      So, why I did I make this post?

      – I happen to think that electronic voting machines are really bad, on many levels, although not because I think they are particularly prone to being hacked or tampered with. If hacking of a machine occurs, it is far more likely to come from “inside” and optical scanners (my preferred method) are just as vulnerable to this, at least initially, as purely electronic machines. Strictly electronic voting is more hackable by outside actors, as voters get to interact with the machines more directly, but it is less vulnerable to this because the tabulation is more diffuse.

      My objections mostly lie along two axes: one being that they increase overall election expense (as buying (and programming) many election machines per precinct is costly) and on net make wait times in highly populated precincts quite long, especially if budget has constrained the number of machines in the precinct.

      The second is perhaps more pernicious, in that they reduce confidence in voting results. Once a vote has been cast, they are essentially unauditable with regards to the voter intent. This is a poor situation and will lead partisans on either side to suspect those on the other if they are in charge of the voting machinery. Given that they also reduce the availability of the franchise via point one, this is untenable. Having some electronic machines available for special needs voters is not objectionable, but this is not the same as having all or the vast majority of machines be electronic only.

      As a point of reference, I worked for a Board of Election creating and maintaining a statewide voter registration system for 5 years. I am not unfamiliar with the territory (albeit in only one state).

      – It also served as a very small scale test of the proposition that asking people to take the contrarian side and argue for a highly unlikely event will naturally lead people to argue for it in ways that the would not otherwise. There were very few attempts to argue for the ways in which someone could do these things, no quoting of sources about how vulnerable the machines are to external manipulation, no real contemplation of the idea that a lone election’s employee might undertake to make it possible or likely to vote incorrectly for a Republican candidate, no searching for conspiracy minded theories on the left, etc.

      What we got instead was a fisking of the idea that this was intentional, divergence into arcane voting constructs and the perennial favorite meta debate about what is fair in the waging of culture war. In other words SSC proceeded to talk about what are among the favorite things SSC talks about, much as it likes to talk about how people on the left are particularly prone to making up fake stories (but aren’t interested at all in the converse).

      Or perhaps it’s because I brought it up, no one wanted to give me satisfaction and forward an argument that would help “my” side. Although, you would think that given that the parallels to the previous were obvious, you could have proven me “wrong” and thus this would have been an incentive to argue for the intentional hacking.

      Regardless, it seems to me that there just isn’t all that much reckoning with the idea of motivated reasoning and biasing of priors, at least not for the in-group.

    • Brad says:

      1% or less is rather large range.

      The excitement over the big lottery has gotten me thinking a fair bit recently about how terrible we humans are at distinguishing low probability events. Introspecting I think I have an intuitive feel for odds down to about in 1 in 10. But beyond that things just get mentally bucketed into a giant “unlikely” category.

  25. Le Maistre Chat says:

    Theism in psychology:

    “That conversation between you was to become the first link in the chain of events that led to the founding of Alcoholics Anonymous”. Wilson specifies further: “My recollection of his account of that conversation is this: First of all, you frankly told him of his hopelessness, so far as any further medical or psychiatric treatment might be concerned. … When he then asked you if there was any other hope, you told him that there might be, provided he could become the subject of a spiritual or religious experience – in short, a genuine conversion. You pointed out how such an experience, if brought about, might remotivate him when nothing else could.” — AA founder Bill Wilson to Carl Jung

    William James, psychologist and founder of Pragmatism when moonlighting as a philosopher, also famously wrote on the utility of religious experience.
    Evidence that genuine religious experience or conversion can solve psychological problems presents a paradox for materialists in the field. In the example of alcoholism, expert consensus is that it’s about 50% genetic. Everything about the mind being reducible to biology is commonly understood as evidence against idealism/God/et al. (I would perhaps be unfairly simplifying Pragmatism to say it attempted to solve such issues by telling us to only care about “practical consequences” of truth claims rather than truth qua truth.)

    Discuss?

    • Protagoras says:

      James himself admits that the behavioral changes supposedly resulting from religious experiences were very frequently not lasting. I’m not hugely familiar with the literature as a whole, but what I’ve seen from the research into alcoholism suggests that whether religion is involved or not does not affect the likelyhood of recovery.

  26. Nancy Lebovitz says:

    SSC won’t let me post a comment, so I’m trying this without the link, which is a Consumer Reports article about 15 supplements they say you shouldn’t take.

    I suppose supplements should be approached with the same caution one gives to prescription drugs rather than completely avoided.

    Pennyroyal can be used to cause abortions. I’m surprised there’s no mention that you shouldn’t take it if you want to continue a pregnancy.

    It’s interesting that red yeast rice, which is taken to lower cholesterol, has some of the same bad side effects of statins where are taken for the same purpose.

    • metacelsus says:

      1) Pennyroyal is seriously toxic (due to containing pulegone) and taking it is a bad idea in general.

      2) Red yeast rice apparently contains some of the fungi from which the first statin (lovastatin) was originally isolated. Therefore it’s not surprising if red yeast rice contains low levels of statins.

  27. Nancy Lebovitz says:

    https://www.bbc.com/news/stories-45971416?ns_source=facebook&ocid=socialflow_facebook&ns_campaign=bbcnews&ns_mchannel=social

    A man’s liver was wrecked by the green tea supplements he was taking. He was very lucky to be able to get a liver transplant in time, but he still has serious health problems.

    Tentative advice: It’s probably safe to take extracts if you limit it to the quantities that would be normally be eaten. On the other hand, the article doesn’t mention how much he was taking.

    I think it’s likely that the amount of liver damage from green tea extract is being underestimated. Would it be identified in people who are heavy drinkers or who are taking prescription drugs which are hard on the liver? Or in people who are suffering moderate rather than severe damage?

    • metacelsus says:

      Link without all the tracking stuff: https://www.bbc.com/news/stories-45971416

      (note that in general you can remove all parts of URLs after ‘?’ and still have them go to the same page)

      Anyway, this is another example of “the dose makes the poison.”

      • Nancy Lebovitz says:

        The article didn’t mention how much green tea extract he was taking, though he doesn’t sound to me like the kind of person who’d be trying megadoses.

        It could be that a normal dose for the vast majority is poison for a few people.

  28. ausmax says:

    Does anyone have a good metric for evaluating judges that are up for election? I feel like I’m generally a fairly informed voter, but I haven’t developed a good system for evaluating judges other than read a bunch of their opinions, which is largely an amount of research that I (and probably the vast majority of voters) don’t have time to do. Anyone have any good strategies? I’m in Los Angeles if you just want to tell me who to vote for and why. 🙂

    • Eric Rall says:

      One factor to consider is that for CA Supreme and Appeals court judges, losing a retention election will create a vacancy that will be filled by an appointment by the incoming Governor. That will almost certainly by Gavin Newsom. So you could use a heuristic of your opinion to Newsom relative to your opinion of whichever Governor initially appointed to judges up for retention. Brown appointed Kruger and Schwarzenegger appointed Corrigan to the state Supreme Court, and I’m not sure who appointed the various appeals court judges who are up for retention in your district.

    • Brad says:

      This probably won’t be applicable to you, because NY state election law is its own special kind of screwed up, but I never vote for any judge that is cross listed on the Republican and Democratic lines. Then I look to see if the remaining ones are affiliated with the local machine. Yep, we still have machine politics in NYC though thankfully it is finally weakening. I may not like Alexandria Ocasio-Cortez’s politics but at least she isn’t part of the machine. If there are any judges left after those steps, I try to find their campaign literature to see if they are kooks.

  29. Nancy Lebovitz says:

    https://slate.com/technology/2018/10/ovulation-research-women-replication-crisis.html

    There were a lot of bad studies about how ovulation affects women.

  30. Vermillion says:

    LET’S GET READY TO RUUUUUUMMMMBBBLLLLEEEE

    Rumble is a game I propose to play starting next comment thread. Or rather, we’ll play by email and I’ll post each round as it occurs. In brief (if you don’t wanna click the thing) Rumble is

    a fast-paced game of superhero combat, with players secretly allocating energy points to attack and defense, attempting to outguess their opponents and knock them out of the game.

    Heroes are imbued with original super-powers invented by the players at the start of each game. Powers can modify any aspect of the game, from minor attack boosts to mind-bending gameplay alterations, with the game’s auction system keeping power levels in check.

    This post is just to gauge
    a) if there’s any interest at all in playing.
    b) if anyone has played before and what they’re experience was like.
    c) if you have any suggestions on how to make it go smoothly.

    For my part a) Yes, obviously.
    b) I played this on another forum oh….9 years ago? And found it pretty dang fun so maybe you all will to. As I recall the winner wound up bidding very low amounts for a wide swath of powers while almost everyone else focused on just a few. Maybe 2/3rds of players were knocked out by round 4. So uh keep that in mind.

    c) My plan is to collect players and powers via a google sheet that I’ll share with everyone who signs up. Every player can select two powers, either from the classic list or of their own creation (subject to my editing and approval) to submit for bids, probably within 1 week of kickoff. After powers are assigned, I’ll post what everyone selected and their current power then it’s off to the races.

    Rounds will take place over a couple days (3?) and anyone who hasn’t emailed me in that time will assign all their energy to defense. Whoever is left standing at the end will be crowned king of the ring, the sultan of smash etc etc.

    YALL READY FOR THIS?!?

    • Randy M says:

      Like I reported last thread, I just read worm, so I don’t think I can turn this down.

    • Thegnskald says:

      Can the non-participating audience suggest powers for the auction?

      • Thegnskald says:

        Sample powers:

        Aura of Vague Familiarity:
        Everybody thinks they have met you, but can’t remember how or why.

        Coveted Covetous Ring of Power:
        Everybody who sees it wants to wear this ring. Whoever is wearing this ring wants to collect more rings.

        Sagan’s Garage Door Opener:
        This device, when pressed, causes the user to become invisible, intangible, unaware, and unable to move under one’s own power for ten minutes at a time.

        • Vermillion says:

          All great pitches and yes, I think opening up powers to anyone who wants to submit one would be fine. You’d still need to be a player to nominate one of them though.

          Also for anyone else with power ideas here’s the basic format they should fall under. First they should be given a name (good job with that) and a type:


          Offense - Powers that mainly deal with adding damage to Attacks or damaging opponents.

          Defense - Powers that mainly deal with modifying your own Defense or mitigating others' Attacks.

          Mixed - Powers that have a combination of Attack and Defense bonuses or effects.

          Support - Powers that restore Energy, target other players with non-Attack effects, or modify the way you Attack or Defend without directly fudging your Attack or Defense numbers.

          Special - Powers with abilities that deal with things like the bidding, allocation, and resolution phases, as well as any other meta-game elements.

          And then for the power description you’d want to think in terms of how they’ll affect the combat, or the bidding process or whatnot. So for example I’d make your powers something like this:

          Name: Aura of Vague Familiarity. Type: Defense

          Everybody thinks they have met you, but can’t remember how or why. Spend 5 Energy to reduce all attacks against you by 25% this round.

          Name: Coveted Covetous Ring of Power. Type: Special

          Everybody who sees it wants to wear this ring. Whoever is wearing this ring wants to collect more rings. During bidding any player may bid any amount of energy on this power and will receive that number of rings. Any attacks against another player with rings will be increased by their ring count while their defense is reduced by your ring count, so long as ring > 0. If that player dies their rings are divided amongst whoever attacked them that round.

          I like this power, I think it could get pretty wacky if a lot of people are vying to be the ringmaster. Edit: On reconsideration why would anyone take the power? Maybe there should be a secondary effect like every ring in your possession gives you 1 energy or something.

          Sagan’s Garage Door Opener is left as an exercise for the reader.

          • Thegnskald says:

            As for why someone would want that power…

            If I start with 30 rings, would -you- want to attack me, knowing you are stuck with them afterwards?

    • Gobbobobble says:

      Sounds fun!

    • Subject4056 says:

      Count me in. How specifically would you like us to register?

    • FLWAB says:

      I’m up for it.

  31. rlms says:

    An interesting analysis of the causes of Trump’s electoral victory. Personally I’m not wholly convinced, but food for thought nevertheless.

    • Randy M says:

      How droll.

    • Deiseach says:

      How very American 🙂

      Though it’s fun to compare this with the Wiccan/pagan spell-casting both immediately post-election and the one during and after the Kavanaugh confirmation hearings/investigation. If we take this seriously oh who am I kidding, if we take it in the sports sense of “who will win the league”, then in Christian prayer versus witchcraft spells, the prayer wins!

  32. johan_larson says:

    Your Halloween costume is “sexy” + the last thing you searched for on Google

    For me, that is “sexy Goldman Sachs interns”.

    I’ll just stay home.

    • bean says:

      Mine is even worse. Sexy G3 battlecruiser.

      • Nornagest says:

        I don’t know if they’ve gotten to the G3 battlecruisers yet, but sexy warship girls are unfortunately very popular in some parts of the Internet right now.

        • dumpstergrad says:

          some of us like big guns

        • Le Maistre Chat says:

          Dear Japan, stop making cartoons. (/Mystery Science Theater 3000)

        • bean says:

          I know about Kantai Collection, but no, they haven’t gotten to the G3 and probably never will. Currently they have one American and two British battleships. (They did display excellent judgement in selecting Iowa for that, which confused me for a while. I would have expected Missouri, until I realized that Might Mo probably had a very different connotation among Japanese who play warship games than it does here.)

          • helloo says:

            Azur lane has some from the Benson class. Not sure if that could be considered G3s. No idea about WSG (the other OTHER ship girl game).

            One thing about those two is that their developers are Chinese/Taiwan based rather than Japan based which could have allowed their more lenient approach to add US ships.

          • Matt M says:

            I’ve been getting FB ads for Azur Lane quite frequently for some time now.

            I sort of got why I might be getting targeted ads for sexy anime girls, but was confused as to why they thought I might be interested in battleships.

            But it all makes sense now. bean is to blame.

          • bean says:

            I did not know of Azur Lane, but the only Benson-class I know of is the US destroyer class, which is entirely unlike the G3 battlecruisers. (Those are British, cancelled in 1921.) I have no idea what WSG is.

          • helloo says:

            Sorry I meant to say Nelson. Not sure why I confused it with Benson considering I just looked at the wiki for G3 to find Nelson.
            And apparently, Kantai also just added Nelson though not Rodney.

            WSG is short for Warship Girls.

    • johan_larson says:

      How the heck does one dress like a Goldman Sachs intern, anyway? I’m picturing a young man in a short haircut, khaki pants, glasses and an understated sport shirt.

      • Chalid says:

        What you described is a bit too casual, at least for the main NYC office. Dress pants and a dress shirt is typical. Suit and/or tie or female equivalent is overkill unless they’re in a client-facing role, in which case it might be mandatory.

    • Nabil ad Dajjal says:

      Sexy twilight imperium. So I guess I’m going to as a shirtless guy with a lion head?

    • Randy M says:

      Sexy defenstration of Prague.
      Going to require some interesting reinterpretation.

      • Lambert says:

        Assuming you mean the second one, isn’t it just ‘sexy clergyman’ but you have to keep jumping out of windows?

    • Statismagician says:

      ‘List of measures of auto-correlation.’

      This is going to be a weird Halloween for me.

    • The Nybbler says:

      But I’m using Bing nowadays!

    • rahien.din says:

      Sexy motivational interviewing

    • dodrian says:

      Sexy Chinese Chicken

      Definitely going to fall, ahem, afowl of the culture police that one…

    • Chalid says:

      Sexy find recent google searches

      Before that, sexy list of political false flag

    • dumpstergrad says:

      Sexy SR-71.

      Endorsed.

    • Le Maistre Chat says:

      “Sexy how to retrieve ring from air duct.”

    • baconbits9 says:

      Sexy Governor of either Florida or Georgia.

    • AG says:

      sexy excel vba csv extra commas

      uh

      • albatross11 says:

        YKINOK

      • Deiseach says:

        I don’t know about the sexy, but work-related I too recently had reason to look up about removing extra commas after converting Excel files to CSV format!

        So you’re not alone, AG 🙂

        • AG says:

          It’s fucking stupid.

          Apparently Excel automatically adds a comma on the end no matter what because something something Oracle needs it for uploads something, and they used to have the ability to toggle that off in pre-2007 versions of Excel, but because Microsoft Office’s creed is now “our users are BABIES and can’t think for themselves” you can’t customize that anymore.
          And then, on top of that, it uses the UsedRange aspect every 16 rows to determine how many columns you have, so you get commas to fill out the max number of columns even on empty rows, even if you do a full clear of the empty cells.

          The only viable solutions I’ve seen are to run a non-Excel VBscript afterwards to read and edit all generated csvs to remove the commas, or to have the macro write the csvs fresh instead of using the export function.

    • CatCube says:

      “Sexy House Document 531”


      That would be both weird and culture-warry in a bunch of different directions.

    • Nornagest says:

      “Sexy Calvin and Hobbes snowmen”.

    • A Definite Beta Guy says:

      “Sexy Bulls Hornets”
      Hmmmm…….

    • arlie says:

      sexy ‘stumbling on happiness’ ?! I guess I’m dressing as a book.

    • metacelsus says:

      Sexy fast multipole method

      Mmm, computational physics algorithms. Great for the many-body problem.

    • fion says:

      Sexy natural gas vs coal

    • Nick says:

      I’ve been reading the thread chronologically, so sexy axis and allies.

      You folks can pick the power, I suppose.

    • andrewflicker says:

      Sexy dual n-back.

      Hmm…

      This is harder for me to come up with a good design than I expected. Something with simultaneous auditory and visual “stimuli” seems like it’s a gimme.

      • helloo says:

        I can sort of imagine some kind of demonic enemy that can only be hurt when it repeats an action and some weird two-faced or multi-armed boss hybrid that attacks separately using its multiple faces/arms or with a ranged/melee mix.

        No idea on how to make it visually representative though.

  33. Nabil ad Dajjal says:

    Board game thread:

    Twilight Imperium is a very odd board game, because it’s the first game I’ve ever played which seems to be actively trying to distract the players from their own ultimate goal.

    The core of the game is actually dirt simple. Fulfill objectives to gain victory points, first one to ten points wins. There are a few extra rules to cover corner cases like the turn order in the status phase to prevent multiple people from getting to ten at the same time. But that isn’t too complex in itself.

    Layered over that, though, are about a half-dozen board games worth of rules for trade, exploring and colonizing planets, warfare, researching technology, and eventually passing laws in the galactic senate. The game even comes with a full booklet of backstory for the fictional universe of the games and each of the eighteen different species.

    Over the course of a 6-12 hour game keeping track of all of that stuff gets really distracting and it’s easy to lose sight of your ultimate goal of collecting ten victory points. You need to maintain a clear head to keep from getting sucked into the minutiae.

    Have other people had the same experience, either with TI or another game?

    • Randy M says:

      The board game Dominion consists of treasure, victory, and action cards. Victory cards are all that matter for winning (usually) but slow you down before the end of the game; treasure cards are used to buy more cards of all types, and action cards are used to accelerate by breaking the rules in various ways (removing cards from your deck, drawing extra cards, attacking opponents, etc.). The action cards used in the game vary according to player choice at set-up, with many being introduced in expansions.
      With some set-ups, a competitive strategy is to simply buy treasure cards, ignoring the more interesting but niche action cards (partially because usually only 1 action card is playable in a turn and your had size is limited). Even apart from that, you are probably better off focusing on a strategy and not being distracted by the variety of action cards.

      It’s a flaw in game design if the fun sub-systems actually detract from winning, but if all of them are useful at certain times but not others, that adds depth. I think this is the case with TI, but it’s also true a lot of it is there for immersion. A lot of the politics cards that I’ve seen, for instance, often don’t do much, but it emphasizes the “deteriorating galactic empire” aspect of the setting.

      • dick says:

        Never played TI, but re: Dominion a lot of the action cards do things like giving you more actions, and I don’t think it would ever be a winning strategy to not get them. But usually of the ten action cards available, there might be 2-3 that you don’t get any of. Good game by the way.

        • Randy M says:

          a lot of the action cards do things like giving you more actions, and I don’t think it would ever be a winning strategy to not get them

          You mean get them along with other (“terminal”) action cards. While you are setting up this engine, someone else just buying silvers, then golds, then provinces may be able to out pace you.
          Usually not, thankfully; it’s almost always better to mix in some action cards for card draw or filtering, which is why I said competitive and not dominant, which is close enough to a pun that you can be sure I would have gone for it if applicable.

      • Nabil ad Dajjal says:

        Dominion sounds interesting, I’ll check it out.

        A lot of the politics cards that I’ve seen, for instance, often don’t do much, but it emphasizes the “deteriorating galactic empire” aspect of the setting.

        Some of the agenda phase cards are boring, but in the games I’ve played there’s always one potential game-changer card per agenda phase. Usually because an obscure card plays into one or more races strategies.

        I don’t think that TI does immersion very well though. It doesn’t feel like you’re barbarians fighting over the corpse of a dying empire so much as explorers populating an empty galaxy. If I wanted to get that feeling I would either add a seventh player to play the Lazax remnants or at least put some kind of tokens on empty planets to make them seem inhabited.

      • Gobbobobble says:

        For a while “Is it better than Always Buying Silver?” was the first hurdle a potential strategy had to clear. Seems like the expansions have since lowered that bar considerably.

        Edit: Ack, ninja’d!

    • Machine Interface says:

      Can’t say I’ve experienced that. The heaviest game I’ve played is Scythe, which is a 4X-ish game (explore, expand, exploit, exterminate — “-ish” because there’s no real extermination, in a battle the defeated units are just sent back to their starting base), and while there are many different things you can do during the course of a session — acquire ressources, explore the map, build buildings, build mechs, fight other players, develop your technology, develop your administration, develop your worker population, gain reputation, etc, pretty much all of them are conductive to gaining victory points.

      • Nabil ad Dajjal says:

        I don’t want it to sound like building fleets or researching technology is completely useless to winning in Twilight Imperium. It’s more that you can easily go overboard.

        If you’re playing Hacan, it’s to your benefit to build up a huge pile of trade goods because you can spend them on achieving objectives. But being the richest player takes work and is rewarding in itself, which means that they can end up focusing more on getting rich than thinking about what to spend their money on.

    • A Definite Beta Guy says:

      I mean, over the 12-18 hour marathons of Axis and Allies, occasionally someone gets obsessed with one particular territory that ends up blinding them to everything else. Not exactly the same though…

      • Le Maistre Chat says:

        12-18 hours?! Is this the 1940 edition, where you buy the European and Pacific boards separately to have twice as many territories and units?

        • A Definite Beta Guy says:

          1984 edition. It was….a slog. Our shortest game was 12 hours because I accidenally under-defended Japan and America got really lucky with an amphibious invasion. We just forfeited after losing Japan.

    • Randy M says:

      Here’s another example–Magic the Gathering. Magic has 5 colors with different strengths, and in each set the relative strengths of the colors fluctuates a bit. Sometimes the colors that look to control and win in the end game excel, but sometimes the aggressive colors are stronger, even to the point that the special mechanics of the set are basically traps.

      I think this was the case with the Amonkhet expansion, at least in limited (draft) format–expecting to play cards with cost more than 4 or so meant death to Boros Exert aggro.

      Players encountering a strong red-white deck will mutter “Oh great, looks like the fun police are out again.”

      • Gazeboist says:

        Part of the problem with Magic is that the designers only ever seem to design mechanics that are friendly to out-of-block cards by accident. The result is that you can’t make a $mechanic deck in an eternal format unless $mechanic has enough support over 1-3 sets to build a deck competitive with 20-30 years of cards, or $mechanic sort of accidentally fits with an older mechanic.

    • helloo says:

      Many games where expansion needed to gain power isn’t the same as winning –
      ie. Dominion or Splendor, where some people concentrate/distracted by getting power rather than points.

      Sometimes people concentrate on one victory condition and completely forget about others.

      Games where you can sacrifice battles/pieces to win wars – Go, Chess, etc.

      I haven’t seen much of games where it’s simply too complicated and the goal is lost – generally just when the end goal was distracted by a more immediate success.

    • Gazeboist says:

      I’ve always told new TI players this:

      “Remember that your goal is to claim the Imperial Throne, not to hold the Imperial Throne for more than twelve minutes.”

      I find that it works well to flavor the game.

  34. Nabil ad Dajjal says:

    In an academic context, the questions are essential. It’s not exaggeration to say that I often learn a lot more from a single good question than from ten minutes of a presentation. If I can’t think of anything worth asking after hearing someone speak for a half an hour it’s clear that I didn’t understand it.

    For a pop science or entertainment talk, skipping the Q&A might help speed things along but to the extent that anyone is there to actually learn it’s totally invaluable.

    • quanta413 says:

      I agree with Nabil about the importance of questions in academic talks.

      I almost never wish the speaker had more time. I not rarely wish that somebody would interrupt the speaker to question them. Too many interruptions can be a problem. Questions at the end is an acceptable compromise with a big audience.

    • The ideal academic version, in my opinion, is the Chicago style workshop, which is almost entirely questions. The base rule is that everyone attending is supposed to have read the paper. The presenter gets to talk about it for fifteen minutes or so, after which it is open season.

      When it works, it is the nearest thing to thinking in multiple brains that I have observed.

      • Nabil ad Dajjal says:

        That just sounds like a normal journal club, and runs into the issue that it’s impossible to present unpublished findings this way without sending out dozens of copies of the manuscript.

  35. johan_larson says:

    There is an interesting new book out about parasites that strongly affect the behavior of their hosts for their own benefit, such as fungi and wasps. The book is “Plight of the Living Dead” by Matt Simon. You can read a review and discussion with the author here:

    https://gizmodo.com/real-life-zombie-animals-walk-the-earth-thanks-to-thous-1829657886

  36. albatross11 says:

    Megan McArdle column on immigration.

    This seemed broadly correct to me. It seems to me that immigration policy is debated as a culture war/symbolism kind of policy. (Build The Wall and No Human is Illegal are both about symbols. Hell, separating kids from parents is symbolic, but with some people liking the symbolism while most (I think) don’t like it.)

    If our immigration policy really mattes to our country’s future, then it should be debated as a kind of technocratic issue–which immigrants will add the most to the country’s future, what are the costs, what criteria should we use to grant visas, etc. But that essentially never happens in any mainstream venue I’ve seen. In many ways, Trump has been successful in raising the immigration issue because he’s very effective at making the symbolic arguments. (Universal e-verify plus fining the hell out of employers of illegal immigrants would probably do more than building a big, beautiful wall, but it’s not much of a symbol.)

    • Matt M says:

      But that essentially never happens in any mainstream venue I’ve seen.

      Eh, I think it happens plenty in right-wing circles. I go pretty far right and I know nobody whose position is “No immigrants, ever, period.” There’s a whole lot of discussion about criteria, costs, benefits, etc.

      “Build the wall” notably applies to illegal immigrants only, and even then, is largely used as a symbol (as you say) meant to drive a hard bargain against the other side.

      • hyperboloid says:

        There are many people who’s position is no non white immigrants, and many more who’s position is no immigration that would have a net effect of reducing the white share of the electorate.

        • engleberg says:

          Heck, there were enough voters who wanted lower immigration and higher wages to elect Trump.

          • Matt M says:

            “Lower immigration” is quite different from “no immigration.”

            There are many people who’s position is no non white immigrants

            I’ve never heard of any. Can you name one. Freaking Chris Cantwell has conceded that minorities would be technically allowed to immigrate into his imagined utopian ethnostate (under certain conditions, of course).

        • Mr. Doolittle says:

          While I’m sure there are some who would make such an argument, can you support “many” and “many more” in some way? Preferably sources not about actual KKK or Stormfront types, which are by no means “many.”

      • BBA says:

        I’ve seen some praise around here for Japan’s immigration policy, which isn’t quite “no immigrants allowed” but close enough. In particular, that they’ve chosen automation over importing cheap foreign workers, and although their economy is declining it’s still doing well on a per capita basis because the population is declining too.

    • Statismagician says:

      Yep. There is no coherent policy proposal, or even [publicly-available] conceptualization of the issue, on either side. There certainly hasn’t been in decades and very probably never was.

      • Matt M says:

        Trump made a coherent policy proposal months ago. It was rejected for being clearly a racist and white supremacist policy because it instituted a sort of points system.

        • Eugene Dawn says:

          Actually, as I pointed out to you not long ago, the fact that it instituted a points system was not objectionable, as evidenced by the fact that Democrats and moderate Republicans tried to pass their own version of a points system proposal five years ago; an attempt that was rejected by conservative House Republicans.

          • Matt M says:

            That does nothing to address the claim at issue – which is that “no coherent policy proposals have been made.”

            You seem to not only confirm my counter of “yes they have” but to further validate it by adding “more than once.”

          • Eugene Dawn says:

            It counters the claim that

            It was rejected for being clearly a racist and white supremacist policy because it instituted a sort of points system

            I agree this doesn’t affect the broader point you’re making; I’m only interjecting because this is the second time in recent days I have seen you making the exact same claim, even after having already given you evidence that your claim is wrong and a mischaracterization.

          • A Definite Beta Guy says:

            This is not even remotely logical. The Senate plan was a result of political compromise. You might as well argue that the GOP doesn’t object to tax increases on rich people because the 2012 tax act let the Bush tax cuts on upper income earners lapse.

          • Eugene Dawn says:

            I can’t tell who you’re responding to or which Senate bill you’re referring to, but if you’re talking to me:

            obviously the 2013 Gang of Eight bill was a result of political compromise–but it was a compromise that was accepted by Democrats; this even though it had a points system. So, clearly, the presence of a points system is not enough for people on the left to dismiss a bill as racist and reject it since we know of at least one bill containing a points system that they did not reject on those grounds.

            The passage of the 2012 Tax Cut extension is at least evidence that, if a bill fails to gain GOP support, it can’t be because they will never accept any compromise with a tax increase.

            Also, the Tax Relief Act failed among House Republicans 85-151; it’s clear that Republicans were not totally happy with its provisions. In the case of the Senate immigration bill, all Democrats voted for it in the Senate, while Republicans opposed it 14-32; it’s clear that disapproval of the provisions was stronger among Republicans than Democrats. There’s no evidence here that a points system was toxic to Democrats even if it wasn’t their preference.

          • A Definite Beta Guy says:

            So you’re adding details now, but you’re selectively adding points. The 2013 bill did not create a points system, it created a points system for one subset of visas that was being offered in addition to the existing immigration system. A replacement of the current immigration system with a points-based system would be entirely opposed, and the preference for left-wing groups is the expansion of stuff like refugee programs and legalization of current illegal immigrants.

            The points based system is not objected to, the points based system is objected to because it was intended as a replacement for the current family system. Democrats make disingenuous arguments that large numbers of current Americans would not be able to enter America under the current standards, but that of course is the POINT: we expect immigrants to be BETTER than us, unless you strongly support increased immigration or open borders entirely.

            Democrats will never accept a points-based immigration system, they will accept some visas that are issued on a points-basis. These are different things. And even then, they will still make aggressive attacks if they are in the minority, because that is how politics works. The more political points they score, the more they can meet their other priorities.

            So, yes, on net, people were objecting to Trump’s immigration policy because it was racist, though this was just one of their objections.

          • Eugene Dawn says:

            The points based system is not objected to, the points based system is objected to because it was intended as a replacement for the current family system.

            Yes, this is what’s at issue: it’s not a points-based system that’s the problem, it’s the removal of the family reunification stuff, the diversity lottery, and the reduction in refugees. Not because it “instituted a sort of points system” which is what the comment I responded to said.

            It’s also worth noting that often Canada is held up as an exemplar of the points system, including by Matt M the last time this came up, and Canada also does not have a points system for all of its visas, has family reunification visas (about a third to a fifth of all immigrants), and lets in 50 000 refugees (the same number as the RAISE act allows, despite Canada being 1/10 as populous).
            So, Matt M’s usage of the term “points system” is consistent with how I have used it, meaning prioritizing a points-based visa but allowing for other visas to coexistence at a meaningful level.

        • dndnrsn says:

          Didn’t it also cut legal immigration significantly?

          Let’s say that the proposal is this: border enforcement that just works because magic, plus Canada’s immigration system, plus the same per-capita rate of legal immigration as Canada has (over double the US’, I think).

          Would that proposal go over well with Republicans?

          • Nabil ad Dajjal says:

            If we also deport the >12 million illegal immigrants currently here, it would almost certainly go over well.

            Assuming that the deportations and border enforcement actually seemed likely to happen that is. Not just an insincere promise.

          • Mr. Doolittle says:

            That depends on the subset of Republicans, but it would be very popular with almost all of the Republicans I know.

            Business Republicans and Libertarians would approve of the increase in legal immigration, and the primary problem for most of the blue collar group would be handled.

            Doubling the US’s legal immigration might be a harder pill to swallow for those who think the total ratio of 1st generation immigrants is too high, and may ask for a lower amount until current levels subside over time. That, or deport as Nabil says, but that sounds politically and logistically implausible.

          • A Definite Beta Guy says:

            It has the effect of cutting immigration because it cuts family immigration and refugee classes without offsetting increases in work-based immigration.
            The GOP as a whole will not pass a bill like this because it would involve higher family based immigration. A quarter of Canadian immigration is still chain migration.
            Plus, 3 million immigrants a year is not something most GOPers will stomach. You might get it if you have Dem Senate and House again, though, because you could convince enough moderate GOPers to go along with it.

          • Brad says:

            Even employment based immigration in the US is largerly family based immigration. Every primary applicant can bring along his immidate family. Less than half the employment visa quotas, which is only about 15% of the total number of greencards issued every year, go to people actually selected on an employment basis.

            How do point system countries deal with families? Adding the the family points together and then dividing by the number of people, and making that average pass the threshold seems to make some sense. Maybe with some exception for children under five.

            Although I’m far from a Trump supporter, I do think there are things to like about his immigration proposal. The DV and F4 *should* go away and parents of USC need a lot more scrutiny to show that they aren’t going to be public burdens. Probably they should be required to buy into medicare (with ten years of premiums due up front.)

            But cutting refugee numbers and moving none of the cut quota allotments to employment based made the “deal” a non-starter as far as I was concerned.

          • Hoopyfreud says:

            @Brad

            Why children under 5? All children seem likely to tally up to whopping zero points.

            Also, insofar as intact families are desirable, this seems likely to create bad incentives.

            Also, demanding a $22,000 per-person fee for each immigrant is not likely to go over well, IMO. Immigration costing a family an additional $100,000, all of which they’ll never see again, shuts the door to a lot of qualified/qualifying people from poor countries.

        • Statismagician says:

          Fair enough, I wasn’t aware of that. I think I still stand by no side having a concrete, internally-consistent position, or at least being unwilling to publicly say what it is.

    • Conrad Honcho says:

      Eh, I do not want a “symbolic wall.” I want a wall that stops/slows illegal border crossings*.

      Also, e-verify** has been part of Trump’s immigration proposal from the beginning, and raids on illegal employers have increased markedly, and the owners are being prosecuted to the extent they can be.

      None of this has anything to do with cultural symbolism. What I want is as close to zero illegal border crossings as possible.

      * I understand many people do not believe a wall will work. I file these arguments in the “arguments from thing I don’t want won’t work anyway” bin. No one is convinced by such arguments.

      ** I understand there are problems with e-verify, which should probably be addressed in a technocratic manner.

      • John Schilling says:

        I file these arguments in the “arguments from thing I don’t want won’t work anyway” bin. No one is convinced by such arguments.

        How do you expect to ever find out that the thing you want, won’t work if you do it the way you are currently trying to do it but might work if you did it a different way?

        Which I am fairly certain, is the case here, but you’ve just pretty much announced that you are going to put your fingers in your ears and hum real loud if anyone tries to tell you that.

        • Conrad Honcho says:

          I would take such arguments from someone who I believed legitimately wanted zero illegal immigration and had a workable solution. Instead I get the same tired responses about ladders and tunnels (both of which can be defeated with sensors) and “walls don’t work” even though the one in Israel sure does. Nobody’s saying you don’t also patrol the wall.

          Arguments against cost seem motivated to me, particularly when coming from people who are okay with multi-trillion dollar proposals like UBI or universal healthcare. $20B over 5-10 years for a wall is peanuts. The fierce opposition to the wall does not make much sense.

          So what I would like to do is build the wall and patrol it, and see what happens to illegal immigration. I predict it would go down considerably.

          • John Schilling says:

            So what I would like to do is build the wall and patrol it,

            Politics being the art of compromise, you almost certainly can’t have both. And your champion is proposing the one that is least likely to function on its own, least likely to endure his inevitable departure from office, and two years in still hasn’t even broken ground.

            and see what happens to illegal immigration. I predict it would go down considerably.

            You’ll never know. Instead, you’ll get to see what de facto open borders looks like. Unless the wall actually is completed, in which case for a few years it will be “open except the immigrants have to bring a ladder”, which I do not see making a big difference.

            I’m not a proponent of open borders, but I can see that’s where we are headed. And I’m also not a proponent of zero tolerance anything, which lack of fanaticism I expect disqualifies me from your list of advisors. Have fun defending the castle.

          • Lillian says:

            Half of illegal immigrants are Visa overstays, so the absolute upper ceiling on a wall’s effect on illegal immigration is cutting it in half. In practical terms, it’s probably going to be much less than that. As such it’s not unlikely that there may be other more cost effective methods like cracking down on employers who hire illegals.

            Personally my position is that we should crack down on illegal immigration, and then massively expand legal immigration. Basically i want people coming over, but i want the US government to be able to control the rate. It would make me much happier if the bitchfights in Congress were about how many immigrants to allow, rather than about what to do about all the illegals. Unfortunately decades of indifference about actual illegal immigration has made this position politically orphaned.

      • albatross11 says:

        Most illegal immigrants come here for economic reasons–they need work, and there’s work here. If we make it uneconomical for American employers to hire a lot of illegal immigrants, then a whole lot fewer will come/stay here. To the extent that your goal is seriously decreasing the number of illegal immigrants, making it uneconomical to hire illegal immigrants probably accomplishes that goal.

        I haven’t tries to work out the economics in detail, but a wall has always seemed to me to be more of a symbol than a real thing that stops illegal immigrants. But mainly, my assumption is that illegal immigration is overwhelmingly about economics. According to the best statistics I’ve seen (the Pew Center), we actually had zero net illegal immigration during the post-crash recession. That’s exactly what you’d expect from economically-motivated migration.

        • Conrad Honcho says:

          I agree, but “let’s crash the economy so no one wants to come here” is not a viable solution to illegal immigration. I want both a functioning economy, and little to no illegal immigration. I believe this is possible and a primary duty of the government and can be accomplished through multiple coordinated efforts including a physical barrier, sensors, patrols, economic incentives, and law enforcement action.

          ETA: Legal immigration is a completely different debate.

          • HeelBearCub says:

            I don’t think legal and illegal immigration can really be separated. One of the primary reason we have so much illegal immigration is that people aren’t willing to come to grips with legal immigration. Much like drug prohibition, this is what you get, but it’s like drug prohibition where you never go after the sellers.

          • Matt M says:

            I agree, but “let’s crash the economy so no one wants to come here” is not a viable solution to illegal immigration.

            I seem to recall some right wing radio chatter during the Obama years that illegal immigration was down because the economy was so shitty, Mexicans no longer saw any value in coming here. Not really sure if the statistics bear that out or not…

          • quanta413 says:

            I don’t think legal and illegal immigration can really be separated. One of the primary reason we have so much illegal immigration is that people aren’t willing to come to grips with legal immigration. Much like drug prohibition, this is what you get, but it’s like drug prohibition where you never go after the sellers.

            I agree this is part of it. Many times illegal immigrant are hired partly because they have a weak bargaining position. Employers can pay them less, threaten to call the authorities on them if they get in a dispute etc.

            If illegal immigrants had to be hired following all relevant law that applied to hiring people here legally, that would discourage illegal immigration.

            But if what I say is true, then legal immigration and illegal immigration aren’t perfectly exchangeable. Even if you raise the number of legal immigrants, there are real benefits to an employer to hiring illegal immigrants instead unless you increase the number of legal immigrants so drastically that you manage to push the cost of employing a legal worker down to that of employing an illegal worker.

          • Conrad Honcho says:

            I would definitely like legal immigration reform so we can streamline the process for both people we may want to move here permanently (doctors, engineers, etc) and people we may want to be able to work here temporarily (migrant farm workers).

            But at that point yes we will still have the problem of how to handle illegals because as long as we’re not doing open borders (which I believe HBC you have said previously you are not in favor of) there are going to exist people to whom we say “no, you do not meet the criteria for legal immigration, you may not come” and they respond “screw you gringo I’m coming anyway” and we have to deal with that. That is the thing I would like handled in a comprehensive way, including physical barriers and law enforcement action.

          • Legal immigration is a completely different debate.

            For once I agree with HBC. Most obviously, if you have open borders you have lots of legal immigration and no illegal immigration–the situation of the U.S. for a large part of its history.

            Short of open borders, anything that makes legal immigration easier for the sorts of people who might otherwise come illegally reduces illegal immigration.

          • Matt M says:

            Short of open borders, anything that makes legal immigration easier for the sorts of people who might otherwise come illegally reduces illegal immigration.

            This doesn’t strike me as obviously true.

            It seems to assume that every illegal immigrant would rather be a legal one – but we can certainly imagine scenarios in which this would not be the case – someone who is fleeing the law, someone who intends to work exclusively off the books and not pay taxes, etc.

      • Brad says:

        Also, e-verify** has been part of Trump’s immigration proposal from the beginning, and raids on illegal employers have increased markedly, and the owners are being prosecuted to the extent they can be.

        My strong hunch is that if there was an immigration bill e-verify would end up with the same fate as eliminating the carried interest loophole.

        As for the second part, I do have the impression that splashy workplace raids are up but not that indictments of owners/executives are. Do you happen to have stats?

    • John Schilling says:

      If our immigration policy really mattes to our country’s future

      Then you have already presumed an answer to the most important question in the debate, and it is an answer at odds to the one held by half the debaters. That question is, “should we make this decision based on what is best for our country’s future, or on what is best for the set of people including our countrymen and all prospective immigrants?”

      Or, more cynically, everybody cares mostly about our country’s future, but the most important question regardng that issue is the one you (and McArdle) are still not asking: “Will these immigrants likely vote for Democrats or for Republicans”?

      • albatross11 says:

        My guess is that most people haven’t thought it through carefully, but among those who have, I would be quite surprised to see a large fraction arguing for the approach of summing up the utilities of the people inside the country plus the immigrants. That’s mostly libertarian economists, I think.

        The second question is more plausible for political actors to be thinking about, but it requires thinking a lot further ahead than most politicians seem to think. A big wave of immigration today will probably start changing electoral outcomes in a decade or two.

        • John Schilling says:

          My guess is that most people haven’t thought it through carefully,

          It’s mostly not a think-it-through-carefully sort of position, but more “look at those poor suffering telegenic children on television; what sort of Scroogian monster would leave them to starve outside a locked gate?” position.

          Not that the alternate position is usually thought through very carefully either; they just have a different set of telegenic victims to look at.

    • A Definite Beta Guy says:

      The argument is somewhat technocratic at the political level, but there are a few sticky issues that you just cannot reason your way out of. Specifically, how to handle already present illegal immigrants is a touchy issue akin to abortion: there’s not a right or wrong, there’s just a values question.

      To what extent you want to accept refugees or chain migration vs. economic migration is also a values question, not just a technocratic question.

    • arlie says:

      Interesting. I’m Canadian, and one of the odd things about living in the US is the combination of hearing lots of routine anti-immigrant rhetoric (via media), while working with lots and lots of my fellow immigrants. Another and sadder aspect is that I don’t personally get any flack, with my white face and obviously Canadian accent, whereas other immigrants I know who lack those advantages clearly do.

      In Canada, one clearly articulated political position is that those immigrants are going to save our country from the demographic time bomb of below replacement level birthrates and baby boom retirement. The immigrants we welcome today (and have been welcoming for decades), and their children, will be paying into [our equivalents to] social security – not to mention some of them working in health and elder care, when we’re no longer able to care for ourselves.

      Not everyone agrees with this, and even if it did that wouldn’t stop us from arguing about how many immigrants are enough, and which ones we prefer. But it’s a normal, respectable position, which I routinely hear on (podcasts of) boring radio talk shows, or from family members still in Canada. Whereas I can’t recall hearing anything of the sort in the US, where on the one hand “social security is going broke; you’ll be left with nothing” and on the other hand “immigrants are taking all our jobs”.

      The other thing I notice is that while some folks in the Canadian debate happily conflate legal and illegal, skills-based and family based, and wind up talking as if all immigrants are illegals with 3rd grade educations – that’s kind of rare, and interviewers mostly challenge them to explain which they really mean – something I would be (pleasantly) surprised to see in US discourse. (And conflating things the other way isn’t especially helpful either – though as a legal, skills based immigrant myself, I tend to resent that less.)

      • Deiseach says:

        I think Canada has a different method of immigration than the USA. On Canada’s southern border, you are not really likely to get streams of USAians trying to cross over into the Great White North (despite all those who swear that they’re going to pack up and leave if the Wrong Side win the next election) and seems to have a reasonably tough requirement for legal immigration from other countries (though recently that does seem to have eased, it looks to be a lot easier to emigrate from Ireland to Canada now than even as recently as 2014).

        According to Wikipedia, there are over 300,000 American-Canadians living in Canada, and up to 2 million American-Americans living/working/studying there as part or full time residents.

        By contrast, there are 11 million Mexicans living in the USA, out of 44 million immigrants. I do have to wonder if the Canadian immigration policy would change in the face of something equivalent to the Honduran caravan?

    • fortaleza84 says:

      One elephant in the room which is not mentioned in the article is the expected voting patterns of prospective immigrants.

      Here is a thought experiment: the compromise is that the Right will go along with a loose borders policy so long as 60% of the newcomers are Christian or Jewish and from either Eastern Europe or Israel. Would the Left go along with that so they can get the warm fuzzies of helping people — white, brown, or black — to live better lives? I highly highly doubt it.

      I think it’s pretty clear the Left wants to open the doors wide to enhance their political power. And why there is very little room for a technocratic compromise.

      • Hoopyfreud says:

        Would the Left go along with that so they can get the warm fuzzies of helping people — white, brown, or black — to live better lives?

        I strongly believe they would in a vacuum, but because it’s impossible to wave a magic wand and arrange people’s preferences in this way, it seems more likely that you’ll be instituting open borders for Europe and quotas for everywhere else, which is the exact opposite of what I think of when I think of “loose border controls.” Also, a demographic breakdown of immigration suggests that the absolute number of non-white-christian-European immigrants is likely to fall, not rise, as a result. So no, the lefties wouldn’t agree because that’s a system that helps fewer people live better lives.

        • fortaleza84 says:

          it seems more likely that you’ll be instituting open borders for Europe and quotas for everywhere else

          You’re fighting the hypothetical a bit here I think.

          Also, a demographic breakdown of immigration suggests that the absolute number of non-white-christian-European immigrants is likely to fall , not rise, as a result.

          I don’t see why, as there are plenty of people all over the world who would like to move to the United States. So the total number from Latin America; Asia; and Africa could be increased a bit while the total number from Eastern Europe and Israel is increased a lot. The net result is more warm fuzzies for the Left, especially if they like the idea of improving the lives of people from Eastern Europe and Israel.

  37. zinjanthropus says:

    This post is quite self-serving (or hopefully it will serve a close relative). There is a theory out there that treatment of young patients with antidepressants can bring on bipolar disorder. Can anyone point me to any guidance on this?

    I’ve read the following abstract:

    In 51 reports of patients diagnosed with MDD and treated with an AD, the overall risk of mood-switching was 8.18% (7837/95,786) within 2.39±2.99 years of treatment, or 3.42 (95% CI: 3.34–3.50) %/year. Risk was 2.6 (CI: 2.5–2.8) times greater with/without AD-treatment by meta-analysis of 10 controlled trials. Risk increased with time up to 24 months of treatment, with no secular change (1968–2012). Incidence rates were 4.5 (CI: 4.1–4.8)-times greater among juveniles than adults (5.62/1.26 %/year; p<0.0001). In 12 studies the overall rate of new BPD-diagnoses was 3.29% (1928/56,754) within 5.38 years (0.61 [0.58–0.64] %/year), or 5.6-times lower (3.42/0.61) than annualized rates of mood-switching.

    https://www.sciencedirect.com/science/article/pii/S0165032712007306

    There is an obvious confounding factor — I assume that juveniles diagnosed with depression and being treated with antidepressants have more severe symptoms than juveniles diagnosed with depression but not being treated with antidepressants. But, my brother was bipolar, and eventually committed suicide, while my relative has been diagnosed with depression and is being treated with antidepressants (and lithium) following a suicide attempt. So this theory is a concern.

    Relatedly, schizophrenia and bipolar disorder seem quite similar at the severe end, where my brother was.

  38. A Definite Beta Guy says:

    Your strategy for enduring mind-numbing work?
    Right now I am reconciling an entire week’s worth of punch cards and scheduling for one of our lines (54 people per day for 6 days with substantial overlaps between days). It’s something we supposedly audit every single day, but our sampling indicates a lot of errors in the past. So I’m stuck doing a full recon to find all the errors and the dollars lost.

    Unfortunately this bores the hell out of me.
    Right now I am posting this comment to SSC, and listening to random history videos on Youtube.
    How do you deal with this?

    • dodrian says:

      It’s often helpful to me to break it up into smaller chunks that I can count off as I work. If the work is quick enough that you can count it off in fives or tens as you work, that’s great. If it’s not so quick but very methodical you can create a little song or mental checklist to work through as you do each one. Setting goals for yourself (most I can do in the next half hour), or racing against other people helps if you can trust yourself to still do the job properly when you are trying to work fast.

      Setting yourself little rewards for making it through chunks also helps (five more and I can take a break to argue with someone on SSC!)

    • DragonMilk says:

      Put on music or podcast – if something requires more attention, use music, if less, go podcast

    • Well... says:

      1. Break the work down into the smallest possible steps.

      2. Complete the first step, even the first cycle of the first step.

      3. Cue inner drill sergeant: “See? Not so bad. Now complete the next step/cycle. You’ll get a break when I feel like you’ve really accomplished something!”

    • Incurian says:

      Music and sunflower seeds.

    • lazydragonboy says:

      Audiobooks. I once had a job buffing CDs where when I got the job my boss said I would need headphones to either listen to audiobooks or music while I worked. I finished all of Robert Caro’s The Power Broker and learned a good deal about the history of New York.

    • Nick says:

      Podcasts. You’ll never ask for a transcript again!

    • dumpstergrad says:

      Chunk up my work into fifteen minute intervals. Take a break, remove my laptop from wherever it is moored, take my power cable, and relocate every hour. If that fails, drink coffee until every single thing is interesting.

      Take a vacation when you get above two french presses worth of productivity.

    • A Definite Beta Guy says:

      After reconciling the time cards, I discovered:
      25 hours for people who actually worked at different lines
      30 hours for people training
      120 hours for people on “light work”
      25 hours of overclocking (your shift lasts 8 hours, and you clock out at 8 hours and 5 minutes)

      We budget at $60/hour, so this is over $10,000 of problems. All issues I would expect to be found and tracked and if 4 people are supposedly auditing this every single day.

      Unbelievable. I particularly do not understand the people on light work. The line is fully staffed without the additional workers. If you’re going to inflate your labor by 6%, you better have a damn good reason, and actually use that labor for something useful. Something tells me they aren’t the ones cleaning the lines, since they are not trained to do the deep cleaning, and…you can’t actually do deep cleaning while the machines are running.

      • Mr. Doolittle says:

        Are they on light work because of Worker’s Comp claims? It’s quite common to have people who would otherwise be collecting at home to come to work and do some kind of work, even if it would be considered wasteful otherwise. 6% of the workforce does sound pretty high for that, though.

      • acymetric says:

        Overclocking could also be related to line management policies regarding what you should be doing at your “start time”.

        When I was working on a shop floor, there were some managers that wanted you to punch in/out at your exact scheduled 8-hour start/stop times, but others that said “start work at 6 a.m. means you are at your station starting to work at 6 a.m.” which of course means you will be clocking in a few minutes early and punching out a few late each day. Granted, overtime was more or less assumed here so nobody was really concerned about this either way, but something to think about when assessing the cause for the overclocking.

        Training could possibly be a time-dumping ground for “overhead” or non-productive time that is necessary but not easily classified or charged to a production task. In my place we had “organizational time” (although we were constantly told to reduce the amount of it).

        For people that worked different lines, is there any interaction between lines? If someone takes 15 minutes to go to the other line and say “hey guys, we need this” and have a discussion about it, I could see it being charged that way and possibly the time being rounded up which would inflate it a bit. This could either be that they think this is the correct way to charge the time, or as a “dammit if I have to go tell them to do what they’re supposed to do I’m charging the labor to their department” screw them type of move.

        • A Definite Beta Guy says:

          Overclocking: not really our concern. It averages out to something like 2-3 minutes per day per employee. Our policy says we do not investigate anything under 10 minutes in a shift. However, we do have one employee who seems to be hitting 10 minutes, every shift. That will be reported and correct.
          The bigger problem is that we do not budget for overclocking, so overclocking is a loss.

          Training actually is training. Our policy says a line can assign someone to a training cost center for 2 weeks, but after that, it falls on the line’s budget. This line trains people between 6 and 8 weeks. Honestly, the whole training budget here is totally messed up from start to end, but the bottom line is that “training” employees are entered as actual labor, despite not actually doing anything. So it will be overuse of labor at the end of the week.

          For different lines: this movement between lines can happen, but my understanding is that it is usually rare. Either way, the actual employees are not responsible for correctly assinging their costs. The supervisors are responsible for assiging labor to the correct line, and are supposed to review the costs every single day. There’s also not much “turf war” between the lines, because the same supervisors oversee both lines in this particular case.

    • Anonymous says:

      Your strategy for enduring mind-numbing work?

      Subcontract someone on the spectrum.

  39. Nancy Lebovitz says:

    https://www.eater.com/2018/10/19/17991578/zingermans-ari-weinzweig-interview

    (podcast and transcript)

    It’s a complex of businesses– mostly food-related– based on the idea that everyone knows things worth sharing, and that hierarchy doesn’t work.

    It works really well, and I’m surprised there isn’t more of it. It’s possible that this kind of thing only happens if it’s led initially by very rare, inspired people.

    One bit I liked was that trying to focus on only the five most important whatever is destructive because the system is an ecology, and neglecting parts of it doesn’t work.

    • gbdub says:

      Zingerman’s is awesome. Everyone wants to talk about the anarchy when the real key to success is the pastrami. Seriously, no one would give a damn if the food sucked (and no one would pay their prices if it was mediocre).

      It probably doesn’t scale all that well, but that’s okay because uniqueness and scarcity is part of the selling point.

      They also found a niche where expertise is pretty easy for employees to obtain, and the whole thing is very customer facing, so it makes a lot of sense to give your low level employees autonomy – I suspect that’s actually pretty common with other luxury goods (which is what a Zingerman’s sandwich is ultimately), even at shops run more traditionally.

  40. DragonMilk says:

    I think there is general agreement that -isms and -ists are being used too hyperbolically and frequently to silence dissent and leading to loss of meaning.

    Let’s take racism – under politically correct constructions, it’s quite obvious that *everyone* is a racist, but I’ll speak for myself.

    I am east-Asian, grew up in very white neighborhoods, occasionally did charity work in poorer areas, and also have friends who hail from Africa (parents grew up in Africa or are still there, came here for education). So I’ll addrss three “races” if you will.

    All else equal, I think East Asian households place a much larger emphasis on education than the other two groups, though this is rivaled by Africans. East Asians, with that emphasis, tend not to have participated in athletics as much as whites (mainly to be used as a “diversity” and “well-rounded” component for college applications). They still largely be first generation or second generation, where parents work hard in jobs more menial than they would have had back in their home countries and stress education as the way children get to have a better life.
    Whites are less homogeneous and that’s where the family background step is introduced. Is this blue collar? How well are the parents educated? What is family size and profession? Still, The New York Times says median white wealth is about 20 times that of black wealth.
    As for Black Americans, sure they are less homogenized now, but there are still broad swathes that are not – google says 77% of black births are to single moms. And yet look at any sport and you see that despite the historical barriers, 65% of NFL and 75% of NBA players are black vs 12% of overall population.

    So were I to have kids, I would not hold him back from picking a black kid over a white kid, and much less an asian for pick-up basketball if everyone is a stranger. That’s just common sense. I’d just tell him that you have to be very ready to change initial opinions based on observation. If the asian turns out to be Jeremy Lin-like, then obviously you update your assumptions. On the other hand, I don’t think it would be controversial for people to think picking the asian kids for a school project over the other two races is any different. And I’d always emphasize to know the background of the kid, who the parents are, what the values are, and make friends with honest and diligent kids who share the same interests as you. As a side note, if that black school kid were from Africa, my personal bias would be to potentially choose the African over the Asian because Asians may be less ethical.

    To me, that’s not being racist, I’m just being Bayesian. It’s logical to hold generalizations for various ethnicities. If I were not to update my assumptions in light of new information, then it’s arguable I’m racist.

    Yet in NYC, I of course can’t express any of this. I tried playing Puerto Rico for the first time, and someone refused to play, saying it was racist.

    Why? I asked.

    The colonists are clearly slaves!

    Why would you think that?

    They are brown and coming in on ships!

    Oh come on, mechanically they don’t function like slaves at all, you can’t buy them, sell them, trade them, and if they were white, you wouldn’t think this at all. Maybe brown was a cheaper paint color. Just think of them as roombas or something.

    You calling my little sister a roomba??! (Apparently her family adopted a girl from Africa).

    Needless to say I was not able to play Puerto Rico that day and have gone out of my way to avoid this person (childhood friend of a good friend of mine who moved to NYC without a job). But she’s not the only one. How can anyone work to tackle serious issues of single motherhood, terrible public school systems, gang dynamics, and black-on-black gun violence if games of Puerto Rico can’t even be played without insisting brown circles must be slaves?

    • arlie says:

      Damned if I know. My instincts are to avoid making any comments or conclusions about race-based groupings. Don’t talk about single motherhood as a “black” problem – if it’s a problem, it’s a problem for everyone. Don’t talk about black-on-black gun violence – figure out what else it is (neighbours? friends? relatives? poor people victimizing other poor people?) and talk about that.

      It’s a frustrating but easily understandable taboo. Far too many people move from “statistically higher likelihood of x being y”, to “all x are y” and/or “this not-x cannot possibly be y”. If I’m looking for a potential NBA player, I’m going to want someone tall who already plays the game fairly well – not some random person of the statistically more common race. If I’m looking for a software engineer, I’m going to want someone with the appropriate skills, and preferably a track record – not any random member of the group that provides a statistically larger proportion of software engineers (probably Indian males, but I’m not really sure).

      Unless the correlation is incredibly strong, favoring people in the statistically right group is likely to lead to bad choices. You’ll miss some good people, and select others who merely “look right”, but are in fact not so good. (As an example of incredibly strong correlations, if I want someone to act as a gestational surrogate [aka ‘surrogate mother’], they need to have a uterus, and be willing to use it – something not likely to include males, however defined. But that’s a very unusual case.)

    • Machine Interface says:

      On the precise problem of Puerto Rico, the controversy around that game (and other similar games which are themed about colonialism) is not new among board-gamers. While it was probably not the designer’s intent, and while mechanically there is indeed nothing that really explicitely references slavery, the history of the actual settlement of Puerto Rico makes it hard to pretend that the board game can’t possibly have any unfortunate connotation, and so quite a number of people are not confortable with playing that game — it’s fine, there are thousands of board games outhere, including many with similar mechanics as but different themes from Puerto Rico.

      There’s definitely a difference of culture here: Puerto Rico and many other games about colonialism are originally games designed in Germany, where these topics do not carry the degree of controversy that they do in the US (at least among some people — Puerto Rico is still a best-seller in the US and one of the highest ranked games on Board Game Geek). I don’t think it’s a big deal if some people in the US or elsewhere are unconfortable with playing games with even thin themes of colonialism like Puerto Rico, Mombasa or Santa Maria — again, there are thousands of other games to chose from.

      • Gobbobobble says:

        Sure, don’t buy it, don’t suggest it, quietly sit out if it bothers you so much. But it’s awfully thin grounds to veto the game for the whole group. There are thousands of games to choose from: don’t be the asshole who makes us spend another 15 minutes choosing a different game because you can’t resist injecting politics into game night.

      • DragonMilk says:

        So i actually *finally* got to complete a game of Puerto Rico with a subset of that game night two weekends ago, which was also about two years after the initial incident.

        My friend (who brought the outraged person) actually recommended it to my now-fiancee as a board game I’d like, and I got that for my birthday over two years ago.

        I would agree to disagree with anyone who needs to read into implications of the color of circles, and would not mind that person sitting out. My issue is with derailing normal social interactions to make a political point when you are tangentially invited over for board games. No one even thought of it as we tried learning the game until she decided to take a stand of, “those have to be slaves!” I would have thought she’d be ok with sitting out or playing another game as many choose to do, but she decided to make it a good vs. evil drama that resulted in ruining everyone’s night.

        • Machine Interface says:

          Well that seems more of a problem with this person in particular and their understanding of social etiquette. I’ve heard many anecdotes of players who sit out of one particular game because they’re unconfortable with the theme, but it’s the first time I’ve heard of a player making a scene in such circumstances — even if negative reaction to the theme of Puerto Rico is frequent enough that it’s hard to deny that it rubs a large segment of American/Western culture the wrong way — but usually people are courteous about their distate, in my experience.

          • DragonMilk says:

            Alas, things like this make me think, “yeah, NYC is no place to express real opinions that may be remotely controversial.”

            By the way, if she were black, I would have been far more sympathetic/open to her concerns. Since she was white, I was entirely irritated and dismissive.

            Perhaps that proves that I’m a racist!

          • quanta413 says:

            @DragonMilk

            As a blue tribe white* person, I endorse your evaluation of this blue tribe white person. Although it may be unfair of me.

            After many years, I eventually realized that on average** I felt safer discussing these sorts of things or even talking about topics that might skirt these sorts of issues with non-white people in my social circles than with white people.

            *mostly. mixed but look white which pretty much is white outside of a Klan meeting

            **obviously it varies from person to person

    • Hoopyfreud says:

      There’s probably room for some game theory to be applied here, but it seems likely that it’s better to make assumptions that violate weak predictions at least some of the time. The problem with these weak predictions is that it’s very easy to ride them indefinitely, without ever having to resort to giving a marginally-lower-expected-value minority a chance; this seems like a bad end-effect, especially where an individual’s violation of those predictions is invisible. If a black student is both academically gifted and quiet about it, why would anyone pick them for a group project anything other than last?

      • albatross11 says:

        After you’ve interacted with someone for awhile, you’ve got a much better idea of how smart they are than you could get from the average IQ of their racial group. After you’ve worked on a couple projects together, you’ll know a lot about their intelligence, work ethic, ability to communicate, organizational skills, etc.

        • Hoopyfreud says:

          Right, but why would you ever work with them in the first place in order to find that out if you acted based on your weak predictions 100% of the time?

          • DragonMilk says:

            @Hoopyfreud

            Because kids talk and know who the smart kids are – who’s getting better grades, who knows all the answers in class, etc. If that were to actually happen, that would work in great favor of an instance of *breaking* stereotypes and why it’s crucial to update assumptions.

            In your hypothetical, I’d ask, “how are his grades? What does he like to do?” A lot of my college friends whose families recently (generation-wise) immigrated from places like Ghana had no problem finding groups to work with.

          • Hoopyfreud says:

            @DragonMilk

            I spent two years at the top of my classes before I noticed people wanting to work with me; it got a lot better after that, but only as long as I kept going to classes with the same set of people (luckily, this was most of the next two years). It’s very possible that this was down to me seeming unfriendly and not some sort of racial animus, but nobody who I ever actually worked with seemed to mind.

            Whatever the actual reason for my experience, the diffusion of knowledge about competence doesn’t work when you interact with strangers, so being overly discriminating based on demographics can screw people over pretty badly. Again, I’m not suggesting that affirmative-actioning your way to proportional preferences is correct, just that the strategy you’ve described, followed strictly, probably isn’t optimal.

      • DragonMilk says:

        My contention is that people implicitly are doing this anyway, and disguise it as something else to avoid being called racist.

        Willful blindness regarding biases and self-proclaimed unbiased people policing others is a worse outcome for everyone.

        Race/ethnicity is an attribute for every individual, and the culture and upbringing leads to assumption forming. Being unable to address those assumptions and speak to when and when it’s not appropriate to apply them and instead say, “NEVER! DON’T BE A RACIST BIGOT!” just preserves the status quo since some things seem too obvious.

        I think it’s more helpful to say, “well, if grouped as a subset, their ancestors were brought here involuntarily, oppressed brutally, could not form stable families, and to this day has a 77% single mother birth rate, so a person drawn from that subset is drawn from a severely disadvantaged pool. In contrast, look at recent African immigrants; they’re more like Asians when it comes to education. Now given all that, let’s talk about the individual in question…”

        • Matt M says:

          and to this day has a 77% single mother birth rate

          IIRC it was much less, and roughly on par with whites, prior to the creation of the welfare state.

          • SamChevre says:

            IIRC….

            Not quite: the black illegitimacy rate was far higher than that of whites prior to the creation of the welfare state. HOWEVER, the black illegitimacy rate pre-Great Society was lower than that of whites today.

          • Lillian says:

            Found a brief overview of the subject. The really short version is this chart. Interestingly, it looks like black illegitimacy rates were already rising in the late 40s. What the Great Society seems to have done is make the slope much steeper. The white pattern is different, their illegitimacy rates start going up in the late 50s but less sharply, and don’t seem affected by welfare programs at all. The end result is that while blacks start and end having much higher illegitimacy rates than whites, the gap gets considerably wider.

          • quanta413 says:

            @Lillian

            Christ, that chart is depressing.

          • Lillian says:

            It’s not as bad as the chart makes it look. There are plenty of couples who are, for all intents and purposes, married and raising children together, but for whatever reason haven’t bothered to make it official. This is actually an old time complaint from government and church officials. Absent a tremendous amount of outside social pressure, commoners just don’t seem all that interested in getting officially married, even if they otherwise live and act exactly like a married couple.

            From the Telegraph:

            “When the state started taxing marriage in the 1690s, the vicar of Tetbury in Gloucestershire carried out a survey of his parishioners to find out how many had been married in church. He was covering his back – clergymen who failed to ensure that their parishioners were officially married were penalised. He discovered that half of them had not been married in church, but clandestinely, making private vows to each other, or married in a private dwelling by some roving clergyman. They were living in stable, but irregular unions.

            Given the choice, those unencumbered by property preferred to avoid the expense and rigmarole of an official church wedding and spend their money on drinking to celebrate the new partnership. Dodging the newly imposed tax and resentment at the state’s interference in their private business provided further incentives to live in “common law unions” that had no basis in law and did not carry property rights. As long as a couple considered themselves “married in the sight of God” and was “reputed lawful man and wife amongst their neighbours” the forms of ceremony mattered little to them.”

            So over 300 years ago, Gloucestershire in England probably had an official illegitimacy rate somewhere around 50%. Actually higher than that of modern white Americans. Granted their effective illegitimacy rate was very likely lower than ours, but nonetheless, the rate of people who are officially married is not the same as the rate of people who are raising children together as couples.

            Do read the whole article, it’s very interesting. Apparently the rates of informal marriage, bigamy, adultery, and premarital sex used to be considerably higher than we commonly imagine.

          • quanta413 says:

            The article is interesting, and I’m glad you recommended it. I’m less hopeful than you. I think the difference you point out in the effective vs actual illegitimacy rates is important. And the Telegraph article seems to indicate that the Church accepted private vows as binding for quite a while.

            The single parenthood rates from the first post you linked are still about two-thirds the out-of-wedlock birth rates, so I don’t think that people living together unofficially is the primary shift here.

        • Hoopyfreud says:

          I’m not promoting willful blindness. I’m suggesting that it’s probably a more optimal strategy to introduce an element of randomness inversely proportional to the strength of the correlation.

          Consider: there are two groups of candidates for a job – 50 from group A and 10 from group B. The average A is about 10% better than the average B, and the in-population variances are similar. You will be hiring 5 people, and can do a short evaluation with 10 candidates. The other 50 will be screened out.

          How many people from group B do you screen out?

          If the answer is “all of them,” this introduces structural problems, and my intuition suggests that you’ll be shooting yourself in the foot some fraction of the time.

          • Nabil ad Dajjal says:

            Whether or not you’re shooting yourself in the foot depends a lot on the magnitude of both population means and the variance.

            For an extreme but normal example, think about recruiting for an American high school football team. You get one hundred applicants: fifty men and fifty women. Spending any time at all looking at the women’s applications is a waste of resources; the differences in body composition and strength are so huge that there is likely not a single woman alive capable of competing with men at a varsity level.

            Heuristics, including stereotypes, are useful because they allow us to quickly and accurately make decisions like this in conditions of uncertainty. Your proposal would replace those heuristics with expensive statistics for small or no gains.

          • Hoopyfreud says:

            @Nabil

            As you said, that’s an extreme example. There are a lot of cases where the distributions overlap substantially enough (especially among self-selected sub-populations) that you’re probably not actually better off. For example, if someone who knows little about basketball is drafting a fantasy team based only on headshots, they probably don’t have a good reason to pick Jeremy Lin last.

            E: and, in the more likely case that you have one or two women trying out for a men’s sport, I’d argue that they’ve clearly self-selected enough that there’s a decent chance they’re at the extreme tail; unless they’re REALLY freakily built they’ll never be linebackers, but you might have found a Becca Longo.

          • Matt M says:

            Yeah, the way to handle applications to a team is… to have them try-out. Put them on a field and watch them run around for 30 minutes and pick the most athletic seeming ones.

    • John Schilling says:

      Oh come on, mechanically they don’t function like slaves at all, you can’t buy them, sell them, trade them, and if they were white, you wouldn’t think this at all.

      You can procure them by a mechanistic process, and you can assign them to labor in the fields for your benefit with no possibility of their choosing otherwise. And the name of the game explicitly calls out that they are fictional representations for people who historically were often slaves. And they aren’t white.

      It’s not an unreasonable interpretation, and it was a bit tone-deaf of the game’s creator to do this. Of course, he’s German, and they have a completely different set of tonal questions re historic mistreatment of different ethnic groups, so I’m chalking this one up to bad luck and putting Puerto Rico on the list of games to play only with people I know well enough to trust that this will not be an issue.

      OTOH, Sid Meier is American or Can-Am, and Colonization explicitly had the players trading in slaves or indentured servants to their perceived advantage. That one was interesting enough at the time, but seems to have been memory-holed.

      • Gobbobobble says:

        You can procure them by a mechanistic process, and you can assign them to labor in the fields for your benefit with no possibility of their choosing otherwise.

        Also every Worker Placement game ever.

        And the name of the game explicitly calls out that they are fictional representations for people who historically were often slaves.

        It also explicitly calls them “colonists”.

        And they aren’t white.

        They’re tiny wooden discs. I’ll have to doublecheck my set when I get home but IIRC it’s not even brown paint, it’s just a dark wood or woodstain. Hanlon’s Razor implies “cheap components” not “let’s make them brown people”.

        OTOH, Sid Meier is American or Can-Am, and Colonization explicitly had the players trading in slaves or indentured servants to their perceived advantage. That one was interesting enough at the time, but seems to have been memory-holed.

        Nitpick, but there was no explicit slavery in Colonization. This is actually a pretty frequent criticism of the game. In-game you have Indentured Servants (white), Petty Criminals (also white), and Indian [Religious] Converts [of questionable consent].

      • achenx says:

        Despite Meier’s name on the box, Colonization was designed primarily by Brian Reynolds. I heard an interview with Reynolds from earlier this year, where he joked about Colonization and that the theme would be harder to pull off now.. “back then we were just selling to 30-year-old white guys”.

  41. helloo says:

    In US politics, there’s a common callout towards the lack of fervor for nuclear energy from the left shows how they are not really caring about reducing carbon-emissions.

    However, why is there so little attention towards the inconsistency for another source of energy that probably has been responsible for much of the reduction in carbon-emissions so far – natural gas. Or more specifically, one of the biggest sources of the increasing natural gas production – fracking.

    To be “fair”, I feel that the pro-nuclear side is not all that honest either due to things like thorium reactors being only a side front of being a main one indicating that it is still being propped up by the military.

    • arlie says:

      I don’t “get” politics, especially US politics. So damned if I know why someone isn’t pushing a given thing regardless of whether or not it’s actually a good idea.

      But natural gas is a self-limiting solution. It’s less bad than coal. But even if we assume unlimited supplies – replacing all coal with natural gas seems like it would still produce ‘too much’ carbon dioxide, if energy consumption rates stayed the same or kept on growing. It’s kind of like switching from a 20 year old gas guzzler car, to a modern, more efficient but still internal consumption car – better than nothing, but still polluting.

      If it’s expensive to switch to gas – and gas will be better, but not good enough, so will just have to switch again, why bother.

      And that’s before we start talking about externalities caused by various means of natural gas producion/collection. (I know nothing about this, but that seems to be the commonly cited problem with fracking in particular.)

      • helloo says:

        And how is this different from nuclear?

        • The Nybbler says:

          Nuclear IS “good enough” from a carbon perspective.

          • helloo says:

            Natural gas is “good enough” as it’s a crucial part of how many places reduced their carbon emissions already.

            In fact, for most people natural gas is probably “better” than nuclear even if only as an intermediary.

          • arlie says:

            @helloo

            Numbers? What is the current amount of energy used in the whole world (BTUs or equivalent units). How much is that projected to increase with population and improvements in 3rd world living standards?

            How much carbon dioxide is released per BTU from natural gas?

            When you multiply these numbers, how does this compare with the level of cabon dioxide production from before the industrial revolution? From 1950? With the level your favourite scientific commitee suggests would be needed to avoid dangerous levels of climate change? Tthere’s quite a range of those estimates, of course.)

            I haven’t done the research (which is somewhat difficult) or the arithmetic (which is easy), but my intuition says that the amount of carbon dioxide produced would still be too much. Natural gas might be *part* of a solution, but only with a good effective way of re-sequestering (sp?) the excess carbon dioxide – and if that method itself requires energy (as it probably would) you can’t be getting *that* energy from natural gas, because that will just raise the amount you need to re-sequester.

            Put another way – slowing this thing down is unlikely to be good enough, and certainly won’t be good enough if the problem is then seen as being solved – and all switching to natural gas can do, if my intuition is correct, is slow it down a bit.

          • albatross11 says:

            My (limited) understanding is that natural gas plays well with solar and wind power, because it’s pretty easy to start when you need it. Nuclear and coal plants need a lot more time to come online from a cold start. So if we move heavily to renewables, I think we end up with a lot of gas turbines used to keep the lights on when the wind isn’t blowing and the sun isn’t shining.

          • Aapje says:

            Either that or we need a way to store electric energy that is cost effective. That can mean improved storage techniques or improved solar that is so good that fairly high losses aren’t very costly.

        • Nuclear releases zero CO2.

    • Baeraad says:

      Okay, since I previously spoke up against a similar liberal “gotcha” argument (the one about how conservatives supposedly don’t really care about reducing abortions since they’re against sex education), I feel I’ve earned the right to point out that this is just more of the same.

      Listen, there is no contradiction here. Us liberals, we are against polluting the air and we are against irradiating ourselves and we are against blowing up the countryside. We’re against those things due to some pretty closely related principles, actually. We don’t want to avoid doing one of them by doing another. We want to avoid doing all of them. If we end up having to do any of them, that’s a fail state for us.

      Now, you can say with some justification, “but those are the only reasonable options!” But that’s exactly what liberals say about the whole abortion vs sex education thing. And in much the same way as the religious right, we are reluctant to give up hope of finding some additional option that we don’t actually hate.

      • The Nybbler says:

        Listen, there is no contradiction here. Us liberals, we are against polluting the air and we are against irradiating ourselves and we are against blowing up the countryside. We’re against those things due to some pretty closely related principles, actually. We don’t want to avoid doing one of them by doing another. We want to avoid doing all of them. If we end up having to do any of them, that’s a fail state for us.

        For me (and I imagine for many others), a fail state is shivering in the dark without transportation. Which, if you veto all large-scale production of energy because of the side effects, is the only option.

      • helloo says:

        This actually is meant more for the political opponents of liberals.
        As in, why haven’t they tried to point towards the environmental benefits of fracking.

        I feel that issues people have with nuclear are similar to that of fracking, but only one seems to get singled out despite both having good reasons to be beneficial for their end goals.

        And no, I do not see many complaints about switching to natural gas to reduce it’s carbon emissions. Even if they probably should.

      • cassander says:

        Listen, there is no contradiction here. Us liberals, we are against polluting the air and we are against irradiating ourselves and we are against blowing up the countryside. We’re against those things due to some pretty closely related principles, actually. We don’t want to avoid doing one of them by doing another. We want to avoid doing all of them. If we end up having to do any of them, that’s a fail state for us.

        This is perfectly fine, except for the part where irradiating people is a vanishingly unlikely consequence of more nuclear power and a very likely consequence of burning coal for 20 years while we wait for solar power to catch up, which is the only real alternative.

        • albatross11 says:

          As best I can tell, there is no way to get enough energy to run our civilization that won’t have significant downsides, given our current technology. So when we talk about alternative energy sources, I imagine we’re talking in terms of accepting some set of downsides, and trying to decide what the best available set of tradeoffs are.

          • fortaleza84 says:

            I imagine we’re talking in terms of accepting some set of downsides, and trying to decide what the best available set of tradeoffs are.

            That’s if we are actually trying to solve problems. There seem to be a lot of political disputes where one or both sides are simply pretending to be interested in solving some problem when in reality they are virtue-signalling, empire-building, or otherwise jockeying for social status, wealth, power, etc.

          • Aapje says:

            @fortaleza84

            I think that in most cases like that, people do convince themselves that they are fighting to solve the problem and that their selfish biases makes them believe that the thing that benefits them, also solves the problem.

          • fortaleza84 says:

            I agree. Although there may be a few people with enough insight to realize that they go to protest rallies to meet girls. Or that supporting the local historical preservation society is a good way to make connections.

      • WarOnReasons says:

        I previously spoke up against a similar liberal “gotcha” argument (the one about how conservatives supposedly don’t really care about reducing abortions since they’re against sex education)

        Could it be that both “gotcha” arguments are in fact correct? What if deep down neither conservatives nor liberals really care about reducing the number of abortions/solving global warming? Could both sides be simply using these topics to prove their own moral superiority?

        If liberals truly believe that without an immediate massive reduction of greenhouse emissions an environmental catastrophe is inevitable then should not they be ready to get at least a little out of their ideological comfort zone?

        • fortaleza84 says:

          Could it be that both “gotcha” arguments are in fact correct?

          I would guess yes, assuming that sex education and birth control really are effective at reducing the demand for abortions. When it comes down to it, most people are hypocrites and most people don’t seriously believe in the principles they pretend to espouse.

          • Thomas Jørgensen says:

            They are – Going of the data, not so much any form of sex education widely deployed in the us (They are all pretty ineffective, because nobody puts enough effort/hours in) but adopting the Dutch sex-ed curricula would more or less eliminate the demand for non-medically required abortions.

          • johan_larson says:

            adopting the Dutch sex-ed curricula would more or less eliminate the demand for non-medically required abortions

            What are the Dutch doing about sex ed that the rest of us aren’t?

          • Thomas Jørgensen says:

            1: Starting very, very early, – because you are not going to be very effective at reducing teen pregnancy if you only start teaching in the mid teens for obvious reasons.

            2: keeping it up consistently every year, strong focus on consent and the emotional side of things.

            3: A very practical doctrine that the way to avoid getting knocked up is to use two compatible forms of birth control – such as the pill and condoms, because that requires a lot more things to go wrong before you get pregnant.

            Mostly, the biggest thing is that they just spend a whole lot more classroom hours on it than more or less anyone in the US does, which means it sticks. Teen pregnancy rate one eight of the US one, pretty much no adult unwanted pregnancies.

          • HeelBearCub says:

            I imagine it also indicates that knowledge of sex non-taboo, which is not the case in the US. The cultural norm on sex in the US has been, roughly, “There is a very dangerous thing that you will very much want to do and you must never speak of it. It’s shameful to talk about it. Also, everyone does it, but only when they are older and married. Don’t ask adults questions about that.”

            Predictable failure is then predictable.

          • fortaleza84 says:

            but adopting the Dutch sex-ed curricula would more or less eliminate the demand for non-medically required abortions.

            What’s your evidence for this? Did the Netherlands randomly assign some regions to have sex-ed and other regions not to? Did the Netherlands observe a massive drop in abortion demand starting a few years after sex ed was implemented? Or is it simply that the Netherlands has low demand for abortions and you assume this is due to sex education?

            I ask because sex education seems to be a matter of religious fervor for a lot of people so I am quite skeptical of any claims regarding sex education.

          • Thomas Jørgensen says:

            The general European pattern is that sex ed and abortion were part of the same political settlement – That is, abortion was legalized, and schools started being fairly aggressive about teaching kids how to not get knocked up at the same time, explicitly in order to keep the number of actual abortions as low as practical.

            I cannot find any english-language histories of dutch sex ed, but the abortion stats exhibit the usual pattern of “high abortion for three-four years, then drop”, except the drop is far bigger than usual, which fits with dutch sexual education being best known practice.

          • Aapje says:

            @Thomas Jørgensen

            adopting the Dutch sex-ed curricula would more or less eliminate the demand for non-medically required abortions.

            This is false. Health problems are part of the decision in 16-25% of the cases. The most common reason for Dutch women to get an abortion is financial problems.

      • dodrian says:

        I appreciate your response for helping me think about the issue in a different way. I think the issue in making progress comes as much from the nature of politics as it does from people’s views.

        I’ve known environmentalists who have been pro-nuclear, and environmentalists who have been begrudgingly pro-nuclear (“we don’t like it, but it’s better than the alternative”), but they have to form a environmentalist coalition with the no-mining environmentalists, the no-irradiating-the-countryside environmentalists, and the no-nuclear-weapons environmentalists. It’s easier for the pro or pro-ish nuclear environmentalists to align with other groups with similar environmentalist goals than it is to appear to ‘defect’ to the environmentalist cause and join Big Energy, even if it might help their goals more.

        Similarly I’ve known pro-lifers who are practically begging for better sex education in schools, and others who begrudgingly admit that moving away from abstinence-only might help reduce abortions. But to be pro-life they have to join under a banner with the no-contraceptives Catholics, and the sex-is-a-private-matter evangelicals, and the don’t-trust-the-government-to-set-a-sensible-curriculum conservatives.

        This Politico article makes the case that Obama set a goal to “reduce the need for abortion”, but the difficulty associating with Obama appearing to defect to pro-lifers, or pro-lifers appearing to defect to the Democratic President stalled efforts to make progress on policy.

        • Edward Scizorhands says:

          How much do pro-lifers need to coalition with each other, as opposed to coalition under the GOP?

          I’ve argued before that Obama could have told the anti-nuke environmentalists to GTFO, and he had the smarts and charisma to do it. Where else would they go, the GOP?

          I’m less confident about my prior assertion now, reading what is probably the closest parallel I would get about that from the reverse angle. Thank you for the link.

          • SamChevre says:

            Pro-life groups have a common goal, but a fair number of them won’t even attend the same demonstrations as “the others”. The important split is between conservative Protestants (who are often quite committed to the view that the Catholic church isn’t Christian) and conservative Catholics (who think the Protestants are heretics.)

            (I’ve known active pro-lifers on both sides, and they wouldn’t protest at the same time and place as the protesters on the other side.)

          • acymetric says:

            I’m not sure that is an accurate description of the stance of most Protestants, at least modern day. Heck, at least some (maybe a lot of) Protestants place some value on things said by the Pope. Maybe certain denominations that I am not as exposed to are more like what you describe (I was mostly around Presbyterians and Lutherans I think, although with a lot of Catholic extended family).

            I can’t speak as well to the mindset of Catholics generally, but do feel like I know enough to be confident that the general Catholic population doesn’t care in the slightest if someone is a Protestant.

            I think you are describing perhaps the most devout/hard-line parts of the Protestant and Catholic groups, but I’m not sure that approaches anything close to a majority or even significant minority.

          • SamChevre says:

            @acymetric

            Oh–definitely agreed; I should have said that differently on second reading. The hardliners on both sides aren’t the mainstream at all–but the hardliners are over-represented among the people who routinely participate in protests.

          • Jaskologist says:

            Are the pro-lifers you know Irish?

          • acymetric says:

            Are the pro-lifers you know Irish?

            SamChevre already clarified (and I agree with the point being made), but this made me chuckle.

          • SamChevre says:

            Not Irish, Scots: KJV-only Bible-churchers, Covenanters* and the “Vatican II was a bad idea” sort of Catholics are all somewhat over-represented among my acquaintances.

            (Covenanters: “who read the Scriptures carefully, determined that God was crazy, and attempted to imitate Him.”)

        • Your point reminds me of the old libertarian/conservative alliance. Conservatives were aggressively anti-communist and willing to be pro-free market to please the libertarians. Libertarians were aggressively pro-free market and willing to be aggressively anti-communist to please the conservatives.

          In both cases, I think, with some hardline members less willing to go along.

          And going back to nuclear/AGW, I note that James Hansen is pro-nuclear.

        • albatross11 says:

          +1

          I think there are tons of issues where:

          a. The extremists become the main face of the movement because they care the most, show up at the protests, join the anti-X organizations, etc.

          b. There are moderate positions that most of the people in the movement would favor, but they can’t be reached because they’d alienate the extremists.

          c. There are also moderate positions that most of the people in the movement would favor, but they don’t dare show that willingness lest they be salami-sliced into giving up their whole issue.

          • dodrian says:

            That’s part of what I was getting at, though I think I unintentionally undermined Baeraad’s point a bit.

            For any issue there are a wide variety of opinions. Baeraad expressed that they legitimately disagree with nuclear power as an option to reduce carbon emissions – I wouldn’t label that extremist at all, there are definitely concerns to be had about widespread nuclear power! I was trying to make the point that attempting to seek a compromise is going to be a politically difficult task.

            I think the Obama article showed that – even with a sympathetic executive and so much potential for good compromise, efforts to put an abortion reduction bill into motion were floundered by political reasons because there were so many different factions all representing different things.

      • John Schilling says:

        Listen, there is no contradiction here. Us liberals, we are against polluting the air and we are against irradiating ourselves and we are against blowing up the countryside.

        What’s your opinion on causing massive disruptions to industrial civilization by forcing a rapid shift from concentrated power on demand to diffuse and erratic energy sources?

        Because the bit with “irradiating ourselves and we are against blowing up the countryside”, as a result of nuclear power generation disconnected from nuclear warfare(*), is a paranoid fantasy. The risk of radiation release is very small and localized, and “blowing up the countryside” is right out. And this is a far more certain consensus than anything involving global warming.

        So one might uncharitably wonder, given the choices actually on the table, whether some liberals are secretly harboring the agenda of disrupting industrial civilization for their own reasons. More charitably, of course, you sincerely believe that stuff about irradiating the countryside and don’t know that paranoid fantasy should have been put to rest about the same time as the one where burning coal was going to bring about the next ice age.

        Why don’t you know that? Why does the partisan ideology that paints itself as the champions of science and reason and “fact-based politics”, remain so stubbornly ignorant when the science in question is nuclear engineering?

        * The Chernobyl plant was designed specifically to produce nuclear weapons as well as nuclear power, and it was the plutonium-breeding mission that drove the design decisions that made it flammable. Seriously, don’t build your nuclear reactors out of glorified coal.

        • Thomas Jørgensen says:

          The anti-nuclear movement is the unholy spawn of the anti-nuclear weapons movement, fossil fuel cash, and the early environmental movement.
          During the sixties, early 70s, there was a big movement opposing nuclear weapons.
          They got their teeth kicked in – Near and total utter defeat in the arena of politics.

          A whole lot those activists drifted over to the environmental movement which was scoring a lot of wins because, well, people do not much like it when rivers catch fire.

          These people were firebrands, and they hated all things nuclear, because of the bomb. So they started campaigning against nuclear power plants.

          This mysteriously helped their funding a whole bunch – No, really, you can find material from back then where coal mining orgs proudly trumpet that they are supporting the anti-nuclear movement, and they still get money from that corner, though the current sources are a lot more quiet about it, and most of their current funding is from people who have been persuaded that this is a vital issue by decades of propaganda.

          They perfected a set of techniques for making building nuclear reactors very difficult and expensive through legal and pr harassment, and also a whole lot of out-right lying about the dangers of nuclear power, which successfully stopped the logical transition to nuclear in most countries.

          This was all an unmitigated disaster for the environment. But most of those people and their disciples are still active, and they cannot change course and maintain their self-respect – if they admit error, they admit personal responsibility for most deaths from air-pollution between 1980 and 2018. That is a fairly sizable pyramid of skulls. People do not stand up and go “I am a villain out of a horror story” very often, even if they are.

        • brmic says:

          While we’re at it, what’s the pro-nuclear side’s take on safely storing nuclear waste?
          I’m asking because from my POV this is an unholy shitstorm of NIMBYism, politicans kicking the can down the road and wrong financial incentives. I believe -possibly wrongly – that were waste producers actually required to build reserves for the secure storage of their waste, nuclear would no longer be cost competitive.

          • Nornagest says:

            Safely storing nuclear waste is a difficult political problem, but it’s not as much of a technical issue as it’s been billed. It’s nasty stuff, but the volumes involved are comparatively low (per Wikipedia, about 27 tons of high-level waste per gigawatt per year) and could be made a lot lower with breeder reactors and reprocessing. By comparison, the WIPP — which was an expensive pilot project — cost about 19 billion dollars and is good for about 72,500 cubic meters of waste. Even at current levels of waste generation, if a cubic meter’s about a ton, then those 19 billion dollars would serve the needs of 134 gigawatt reactors for twenty years. I don’t think that works out to a substantial fraction of their lifetime cost.

            And once again, the alternative here is the waste and environmental degradation coming from fossil fuel power generation. Which is substantial. And which has a pretty strong NIMBY argument against it as well: not many people want the view from their retirement home to be ruined because the next mountain over happens to be full of coal.

          • Lillian says:

            The worry about nuclear waste always struck me as a little weird. We got the radioactives out of the ground, what’s wrong with putting them back in the ground?

          • Nornagest says:

            To be fair, some of the stuff that conventional nuclear fuel cycles spit out is very much more radioactive than U-235.

          • brmic says:

            @Lilian
            Does that apply to petroleum and it’s derivatives as well? Lead?

          • Lillian says:

            @Nornagest: The stuff that is more radioactive stops being radioactive fairly quickly. The stuff that’s radioactive for a long time, which is what everyone panics about, is usually about as radioactive as naturally occurring substances.

            @brmic: Putting petroleum byproducs in the ground seems much better than dumping them into the air we breathe, yes. It is my understanding that we already dispose of lead by either recycling it or burying it.

          • John Schilling says:

            To be fair, some of the stuff that conventional nuclear fuel cycles spit out is very much more radioactive than U-235

            Much more radioactive, but much shorter lived and much better contained – and since much of the expressed concern is over the effect of buried radioactives on our umpty-hundred-generations-removed descendents, I'm pretty sure a rational assessment would favor running the stuff through a reactor and then welding it into a steel container before burying it back where you found it (or someplace even deeper).

            But you'd have to define an explicit utility function for future vs. present lives to make that calculation, and you'd need a range of estimates for future human population and industrialization over coming millenia.

          • Nornagest says:

            The stuff that is more radioactive stops being radioactive fairly quickly. The stuff that’s radioactive for a long time, which is what everyone panics about, is usually about as radioactive as naturally occurring substances.

            The hottest stuff we deal with from the nuclear fuel cycle has a half-life of years to decades, true, and it’ll be more or less inert by the time our weird Morlock descendants crack open the WIPP, ignore all the “this is not a place of honor” inscriptions, and start worshipping the fuel casks. But on the other side of the spectrum, U-235 fission also puts out a bunch of fission products with half-lives in the range of 150K to 15M years (for comparison, U-235 itself has a half-life of 703M years). Those are hot enough to cause problems that uranium doesn’t, but long-lived enough to still stay more or less intact over the 10,000 year design lifetime of deep geological containment. They’re not crazy dangerous like, say, cesium-137, but you probably wouldn’t want to have a cask of them in your living room.

            I’m still in favor of nuclear power in general and deep geological containment in particular, for reasons given above, but it doesn’t make sense to say that these are just like the original fuel.

          • brmic says:

            @Lilian
            Yes, and if you go one step further, you’ll find that among other things, the concentration of the stuff is a crucial factor, as is the safety and security of the storage. Whether it’s ‘from the ground’ and ‘returns to the ground’ doesn’t matter.
            Apart from that, who is the ‘everybody’ that panics?

            @Nornagest
            The WIPP page leads me to the pages for Morsleben and Asse. Asse suffers from water influx, so presumably (a) the pumps needs to be maintained and (b) there’s a bit of finger crossing involved. Morsleben has to be pumped full of concrete to stave off collapse. In both cases, the ‘unexpected’ happened, and WIPP had problems in 2014. Which is all well and good, these things happen. But at the same time, if I’m expected to trust these solutions will hold for 10000 years, my confidence is low, given the track record and the simple fact that no one who argues for that solution will be around to face consequences for errors.
            At the very least, it would seem that permanent storage is a feasible solution for countries like the US, with large land mass and low population density. However, for Europe and Japan this is more difficult.

            On the cost side https://en.wikipedia.org/wiki/Nuclear_decommissioning

            In 2016 the European Commission assessed that European Union’s nuclear decommissioning liabilities were seriously underfunded by about 118 billion euros, with only 150 billion euros of earmarked assets to cover 268 billion euros of expected decommissioning costs covering both dismantling of nuclear plants and storage of radioactive parts and waste.

            I expect, (though haven’t bothered to search) that even now proponents declare the funds sufficient and that 10 years ago concern of insufficient funds was branded as paranoid conspiracy theory by advocates of nuclear power.

          • Eternaltraveler says:

            Most high level nuclear waste is otherwise known as nuclear fuel. What’s left after fully burning up all that can be reasonably burned up needs to be stored for a few centuries before it’s about as radioactive as high grade uranium ore.

            Solar/wind energy requies us to set aside vast tracks of land.
            Meanwhile all high level nuclear waste in existence would fit in a large building (and and all the transuranics in there are valuable fuel). Nuclear waste is a political problem, not any other kind.

          • But at the same time, if I’m expected to trust these solutions will hold for 10000 years, my confidence is low

            Do you think it makes any sense at all to base present decisions on our guess about the situation in 10,000 years? Try to imagine people 10,000 years ago doing it–or even 500 years ago.

            This whole line of argument has long struck me as crazy. The world is changing much too rapidly for us to have any reasonable guess about the circumstances of our descendants that far in the future. The human race may have wiped itself out by then. It may have gone to the stars. We may have all been uploaded to hardware.

            About the least likely projection is that things will be about like they are now, plus a few details such as flying cars.

          • Thomas Jørgensen says:

            General solve: If you have a public policy problem, look up what the Swedes are doing. And, gosh, yes, they do have a good solve for this one, too.

            http://www.skb.com/future-projects/the-spent-fuel-repository/our-methodology/

            This seems extremely reliable and easy to copy, too.

      • fortaleza84 says:

        I think the contradiction comes from the argument that mankind’s CO2 emissions, if unchecked, will lead to total disaster — hundreds of millions of deaths due to starvation and flooding; entire countries submerged; etc.

        If such a catastrophe were truly a significant possibility, a Chernobyl or two every few years would be a price well worth paying to avoid it. So I do think there’s a contradiction.

        One possible way out of it is to admit that these global warming doomsday scenarios are wildly exaggerated for the sake of effect, but even that is something I’m skeptical about. I myself am a conservative and I think CAGW is bunk but I would get on board with strict CO2 emissions caps if they were combined with policies that I want, such as a drastic reduction of third-world immigration to the United States; national concealed carry; total bans on affirmative action; etc. But you almost never see environmentalists proposing these sorts of compromises in order to attract conservative support.

        I guess the bottom line question is what sacrifices the Left is willing to make in order to achieve reductions in CO2 emissions. If they aren’t willing to make painful sacrifices, it’s reasonable to conclude that they are hypocrites who don’t actually care about CO2.

        • brmic says:

          I guess the bottom line question is what sacrifices the Left is willing to make in order to achieve reductions in CO2 emissions. If they aren’t willing to make painful sacrifices, it’s reasonable to conclude that they are hypocrites who don’t actually care about CO2.

          By that logic, anyone going la-la-can’t-hear-you when you tell them of a problem can then extract arbitrary sacrifices from you for their vote for your solution to the problem. In strictly tribal terms, sure, that’s true, but in a democracy it’s hopefully eventually limited by non-mindkilled voters.
          You appear to be modelling CO2 emissions caps as a slightly eccentric preference (of pretty much every major political party in the western world except for US Republicans) whereas from the other side it’s more like the opiod epidemic: A new-ish problem which both sides try to solve albeit with partisan preferences for particular solutions.
          Imagine, if you can, a political party approach the evidence for the opiod epidemic in the same way the cigarette industry approached science linking smoking to cancer. After having stonewalled you for a decade, they then say ok, we agree to do something about it, but only for abortion-on-demand up to birth. If you’re not willing to give that up, you’re abviously a hypocrite not really concerned about opioid overdoses.

          • cassander says:

            (of pretty much every major political party in the western world except for US Republicans)

            The republicans might talk about global warming differently from other major western parties, but they don’t act any differently. No where is any political party rushing to embrace serious reductions in carbon emissions. the closest you get is somewhere like Germany, where they could shut down their own plants confident that they could import power from abroad

    • Edward Scizorhands says:

      Natural gas via fracking had a fair effort to go at it with lobbying, but it was an existing industry that could fight back and employ people and provide energy right now. Not building nuclear plants maintains the status quo. Banning natural gas wrecks the status quo.

      • Eternaltraveler says:

        Not building nuclear plants maintains the status quo

        After Fukushima environmentalists around the world are successfully shutting down existing nuclear plants, rapidly shifting the status quo away from nuclear energy.

        This is despite the fact that zero people died from radiation exposure (~20,000 died due to the earthquake/tsunami), and modern plants are pretty meltdown proof (Fukushima was not one, it was built in 1971).

        People who care about climate change are almost entirely part of the anti-nuclear coalition because it’s associated with the wrong tribal affiliation. While the other tribe is largely enough pro nuclear. You can therefore blame people who care about climate change almost entirely for whatever doom ensues.

        On the other hand if we can ever manage to successfully switch to a nuclear energy based economy we will be good on energy until the heat death of the universe.

        Luckily China doesn’t seem to care about the anti nuclear frenzy that gripped most of the world and is building nuclear plants at a cost of about 1/5th of an identical plant in the west.

        • Nornagest says:

          On the other hand if we can ever manage to successfully switch to a nuclear energy based economy we will be good on energy until the heat death of the universe.

          There isn’t that much uranium out there. It’d last longer than fossil fuels would, but only by a couple orders of magnitude, and that’s with widespread use of breeder reactors — with the once-through cycles that most current reactors use, known reserves don’t even have the total energy that fossil fuel does.

          Thorium’s much more abundant, but we need to figure out how to make thorium reactors before we can use it. Same goes for fusion.

          • Eternaltraveler says:

            There isn’t that much uranium out there

            Well technically it may be true that uranium itself may not be as viable of a energy source deep into the heat death as fusion or the penrose process (U 238 has a half life of only 4.5 billion years after all, and there are stars that will last for 10 trillion years).

            However, uranium is a more than viable energy source for 100% of the world’s energy supply at present consumption rates until the sun expands to it’s red giant stage (when it may or may not swallow the earth). This is because, especially when taking breeder reactors into account, the contribution of the cost of uranium to the cost of nuclear energy is almost nothing. That makes extremely expensive uranium viable.

            I do agree with you about thorium; we can use that a lot longer than uranium, mostly because it’s half life is 14 billion years. So we can get a few more earth lifetimes of energy usage if we also use thorium. Sometime in this 25 billion year time-frame it would be helpful to work out fusion which gets us well into the heat death. Considering that net energy was generated from fusion in 2013 I’d say we are on track to achieve economic viability in this time frame.

          • Nornagest says:

            I’m going by Wikipedia’s page on world energy reserves. All bets are off if we get into asteroid mining, of course.

            Speculation about the deep future is speculative, but we have to get there first.

          • Eternaltraveler says:

            I’m going by Wikipedia’s page on world energy reserves

            The primary source there is the IAEA which seems to say that easy uranium is available to meet demand for at least the next 2500 to 20,000 years (depending which grade ore you count). After that it becomes harder to get, but not that much harder. At a cost of only $660 per kilogram uranium can be directly extracted from seawater right now (which is about 10 times the present spot price). Ultimately the earth’s continental crust is about ~1.5 parts per million uranium, and most granite’s are around 5 ppm. To put things in perspective gold is presently economic to extract from as little as 0.25 grams per ton (250 ppb) of ore if certain conditions are met (1 gram per ton is almost always economic, which is 1 ppm).

            There is a lot of slight of hand by anti nuclear groups because they don’t want to have nuclear classed with renewable energy and they make assumptions like we will throw away 99.5% of the fuel (by not using breeder reactors), and they ignore 99.9% of the uranium that we can extract even today (at higher costs) etc. In any case I’d say we have more than enough time to develop fusion or build a Dyson sphere. Uranium or thorium are not going to run out in a time scale that is relevant.

            In the long term energy is practically free and limitless.

        • The Nybbler says:

          Unfortunately, China probably doesn’t care about nuclear safety either, so we’re likely to see a meltdown which will make continuing use of nuclear power even less likely in the West.

          • Eternaltraveler says:

            Unfortunately, China probably doesn’t care about nuclear safety either, so we’re likely to see a meltdown which will make continuing use of nuclear power even less likely in the West.

            Luckily most of China’s reactors under construction are pretty modern. China didn’t really get into the nuclear energy game until the 90s. So whether or not they put as large an emphasis on safety, they benefit from more modern safe designs (in contrast most US nuclear reactors in operation today were built in the 70s and early 80s, and those are still amazingly safe in comparison to fossil fuels)

    • Well... says:

      I won’t speak to the political angle (because right now I just don’t really care) but you should know that different energy sources are not identical/interchangeable: nuclear energy doesn’t produce a lot of carbon emissions, but it produces nuclear waste which is its own can of worms. Natural gas is very inefficient compared to coal or nuclear. Fracking is a way to get at energy-dense fossil fuels that conventional mining can’t, but it’s environmentally hazardous and causes earthquakes. Etc.

      (BTW, there was a lot of furor about fracking. IIRC there was even a pretty high-profile documentary made about it about around 2009 or so.)

      Everyone knowledgeable on energy seems to agree it makes the most sense to keep a portfolio of different energy resources (including coal and nuclear) and gradually — as the technology and market permits — shift towards renewable/greener sources.

      • fion says:

        Nitpick: “produces nuclear waste which is its own can of worms”

        It’s a very small can of very small worms compared to most of the others on offer.

    • helloo says:

      People seem to be misreading what I am trying to describe.

      Some things I am inferring
      * Just as there is rhetoric against the use of nuclear, why isn’t there any against the left for trying to stop/shut down fracking despite the fact it probably is a significant factor in the reduction of carbon-emissions in the US.
      * Shouldn’t there be more pressure from the left against the trend of moving to natural gas to reduce emissions as this is counter productive to their efforts to reduce/stop fracking.
      * I AM NOT DIRECTLY CALLING OUT AGAINST THE LEFT FOR NOT SUPPORTING FRACKING. I can understand why they would be against both nuclear and fracking- just that this inconsistency doesn’t seem to be really noted on.

      Some things that could be argued is that the proponents of fracking REALLY do not want to feel they are helping with the efforts of carbon-emissions as this belies the fact its a real problem or something.
      Or that nuclear proponents are not opponents of the left and rather a minority ingroup and these arguments are from them and noone really cares enough about natural gas as it’s not being hindered much.

      • Mr. Doolittle says:

        I think the main reason conservatives don’t make the same argument about natural gas as nuclear is that the argument for natural gas has a really easy answer for a liberal:

        “That still produces too much carbon!”

        That may or may not be factually true, but it pulls the discussion into an academic debate about the relative levels of carbon production that the world can sustain and how well natural gas fits into a long term plan.

        Since nuclear doesn’t produce CO2, then it’s a much easier fight for conservatives on a rhetorical level. Imagine the following argument:

        1: We need to eliminate CO2 production!
        2: How about natural gas?
        1: That’s still CO2!
        2: Yes, but not as much.
        1: Any is too much! and the world is going to die if we don’t get rid of CO2!*
        2: (On the defensive in an academic debate about CO2).

        verses

        1: We need to eliminate CO2 production!
        2: How about nuclear?
        1: Well, that’s not a particularly good option [lists downsides].
        2: But you said any CO2 is too much, and this option avoids world-death. If the world is on the line, then clearly we must select this option if it will save us.
        1: (On the defensive in an academic debate about nuclear).

        *-This is a weakman argument, but it’s generally the level of argument you are talking about in the OP.

      • Edward Scizorhands says:

        why isn’t there any against the left for trying to stop/shut down fracking

        Because they aren’t successful at it. Hollywood can make all the anti-fracking movies it wants, but fracking happens and will continue to happen. I am sometimes amazed at how brazen the frackers are, but that’s because they are incredibly secure.

        It is hard to build up a head of steam about dem liberalz wanting to stop the natural gas when they are ineffective at it.

        • Matt M says:

          I am sometimes amazed at how brazen the frackers are, but that’s because they are incredibly secure.

          I mean, it helps that most of the fields worth fracking are located in places like West Texas, northern Alberta, Wyoming, rural Pennsylvania, etc.

          If all the prime deposits were located in Northern California, this would probably play out much differently…

    • Nornagest says:

      California’s switched almost entirely from coal to nat gas over the last fifteen years or so, and while I don’t know for sure if there’s environmental reasons behind that it seems like a safe bet. So at least one place seems to be acting consistently (somewhat to my surprise, given what I usually think of California politics).

      • helloo says:

        Not really – how does it treat fracking which probably produces much of the natural gas it consumes?

  42. INH5 says:

    This study which claims to show that personality differences are larger in more developed countries has been passed around for a while now. Last year, I made a post about some issues that I had with it, including possible signs of p-hacking and potential regional confounds.

    I took another look at the paper again a few days ago, and I think I may have found another serious problem.

    The introduction discusses three hypothesis, the third of which is that the cross-cultural patterns may be driven by methodological artifacts. The last paragraph is particularly interesting (page 170):

    Finally, it is plausible that differences in personality traits are masked by measurement error. One might expect, for example, that in countries where people are better educated and more literate, overall internal consistency of personality scales is higher. In countries where access to education is more restricted, differences between men and women in personality traits may still exist, but these differences are attenuated due to a larger response inconsistency. Indeed, cross-cultural studies have observed that average Cronbach’s alpha across all personality traits tends to be higher in prosperous and well-educated countries than in countries where access to knowledge and education is more constrained (McCrae et al., 2005).

    This is something that I’ve been wondering for a while: whether results like this could simply be because measurement tools like the Big Five personality tests were “calibrated” for WEIRD subjects, and they don’t capture as many relevant variables as well in different cultures. If I understand it right, Cronbach’s alpha is supposed to measure the reliability of a scale, so it seems like a decent starting point for testing this.

    A side note: I did some digging a while back, and it turns out that almost all of the “cross cultural” data used in this study came from college students (pages 180-181). That means that access to education, specifically, probably won’t be much of an issue but it raises other questions about whether these samples are representative and whether a student sample in one country might be unrepresentative in different ways than a student sample in another country. But let’s leave that aside.

    The results section (page 177) discusses an attempt to test the measurement artifacts hypothesis:

    Finally, we found no support for the notion that measurement artifacts influence the degree of sexual differentiation across cultures. Although Cronbach’s alpha, r (54) = .41, p < .001, and negative item bias, r (53) = .32, p < .016, significantly correlated with size of sex differences in personality traits, their contribution vanished when level of human development was controlled.

    Forgive me if I’m missing something, but isn’t this the exact opposite of what you would want to do? The hypothesis would be: country-level variables that happen to correlate with HDI -> the personality tests are less reliable in low HDI countries -> sex differences in the test results are smaller in those countries. Presumably, then, you would control for Cronbach’s alpha and see if the HDI correlation was still significant, right?

    I might actually be able to do that myself in Excel (a Google search tells me that Excel can do multiple regressions), since the paper provides the Cronbach’s alpha data for each country and I’ve already copied the personality data. But I wanted to run it by people here first, to make sure that I wasn’t missing anything.

    • Edward Scizorhands says:

      My default assumption is that for a given thing there will be more variations in bigger and richer countries where people can spend more time worrying about those things.

    • Statismagician says:

      I swear, when I am Dictator of the World I will force all new social science PhDs to write ‘Data derived from college student are not generalizeable’ ten thousand times on the blackboard before giving them their diploma.

      The statistical issue is actually much worse than you mention, because HDI and sexual disparities are negatively autocorrelated, mediated through life expectancy and education off the top of my head. They should have used a more discrete level of aggregation and non-composite variables, at the very, very least.

  43. ADifferentAnonymous says:

    Your mission, should you choose to accept it, is to be an Effective Action Hero.

    The action hero part is given to you: no weapons can harm you, no trap can hold
    you, and you can best any foe you face in direct combat. However, you
    cannot pump any physics violations involved in this for economic purposes–
    your special abilities are pretty much only good for beating people up.

    How, then, do you make the world a better place?

    • alexkidd says:

      trick question! anything i did would have economic effects

    • dndnrsn says:

      Become a simultaneous superstar in every combat sport and use the resulting soapbox to promote good causes, then parlay that into a political career, and do same.

      • albatross11 says:

        Yeah, I’m kinda wondering if the answer isn’t something like “make a gazillion dollars as the world boxing/MMA champion, and use it to buy DDT-soaked bednets and ship them to regions with a lot of malaria.”

    • Edward Scizorhands says:

      I could do regime change anywhere with a normal functioning government. Even the USG would have a tough time operating if I could walk anywhere and destroy what I find.

      North Korea is seems to be the place that I would most likely to help and least likely to harm. There are all sorts of civil wars and proxy wars being fought in Africa and the Middle East in which I honestly do not know which side is better, but I would start researching that.

      In theory I could force first-world governments to modify their policies, too. In terms of maximizing my impact this is how I operate, but there is too much chances of seriously wrecking things I do not understand.

      (I assume that I cannot be economically bought off, given the constraints of the problem.)

      • Matt M says:

        As soon as North Korea becomes aware of your existence, can’t they say “If you lay one hand against us, we immediately launch our entire arsenal at Seoul/Tokyo”? Would that not deter you?

      • ADifferentAnonymous says: