Open Thread 73.5

This is the twice-weekly hidden open thread. As the off-weekend thread, this is culture-war-free, so please try try to avoid overly controversial topics. You can also talk at the SSC subreddit or the SSC Discord server.

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

331 Responses to Open Thread 73.5

  1. Admin says:

    Pretty sure the Discord server fried my laptop so I’m posting my Tom Swiftys here from now on:

    “This disease is causing me to lisp,” Tom said thickly.
    “This isn’t enough medicine,” Tom fumed.

  2. Emma the trust fund baby says:

    Hey everyone, I have a sort of request for educational outcome advice. I read Harris’s “Nurture Assumption” and Caplan’s “Selfish reasons to have more kids” books. Caplan’s book basically says that child outcomes are 50% genetics, 50% non-shared environment;
    Harris’s book says 50% genetics, X% peer group, Y% non-shared environment. I forgot the exact figures, but peer group was a non-negligible amount.

    The relevant SSC article is here.

    Anyway! So I have a baby and I’m wondering whether to let him go to whatever random school, or spend some effort to get into a private school. If I follow Caplan, I shouldn’t bother, and spend the effort on improving my own outcomes so the baby has a happier home environment. If I follow Harris, then I should spend effort to go into the private school, so that the baby has a good peer group influence.

    What is the right way to understand this research? Please help me out if you can.

    • rlms says:

      Not specifically about the research, but quality of non-private schools (and that of private schools) varies significantly.

      • Chalid says:

        Yeah, you can ask a similar question about whether you should buy a house in an expensive neighborhood to get access to better public schools.

    • IrishDude says:

      Caplan’s book basically says that child outcomes are 50% genetics, 50% non-shared environment

      I haven’t read the book, but I’d think that some outcomes would be more or less related to genetics. Does the book discuss this?

      Caplan also home-schools his kids, so it seems he thinks where kids learn matters.

      • onyomi says:

        Caplan also home-schools his kids, so it seems he thinks where kids learn matters.

        Or that his kids prefer to be homeschooled and/or he prefers to spend more time with them, and, since he doesn’t think it matters what type of schooling they receive, it might as well be of the sort they enjoy.

        • IrishDude says:

          That could be true. It would be surprising to me though if kids learning more about the things they enjoy, in a manner they enjoy, didn’t have an impact on their outcomes later in life. Perhaps this is discussed in Caplan’s book and there’s convincing evidence against this.

      • Chalid says:

        Caplan says:

        While the power of nurture to change kids’ adult outcomes is indeed vastly overrated, it is well within my power to give my sons a better childhood.  My kids prefer a challenging academic curriculum.  I can give them that.  My kids hate music, dance, art, and group projects.  I can spare them these indignities…

        More speculative: I suspect – though I’m far from sure – that the Caplan Family School is such an exceptional experience that ordinary twin and adoption evidence isn’t relevant.  For example, my sons are plausibly the only 12-year-olds in the nation taking a college class in labor economics.  Perhaps it really will forever rock their worlds.  More obviously, their peer group now includes Robin Hanson, Alex Tabarrok, Tyler Cowen, Garett Jones, and Nathaniel Bechhofer.

        Caplan’s belief that his homeschooling might be a truly exceptional experience might also be true of some extremely high (and low) end schools as well, of course.

        • Deiseach says:

          I’d be more convinced by that if it didn’t sound like “By sheer coincidence, my kids love the subjects I am uniquely qualified to teach and hate the things I am no good at – who would have thought that?” 🙂

          If he was lamenting how he’d love if they shared his interest in Northern European late Mediaeval art so they could visit galleries together but all they wanted to do was learn economic theory so he’s chained to the kitchen table teaching them, I’d be a lot more “Ah yes, that sounds like the kids really do want to learn this and not like Dad is subtly guiding them in ‘this’ll be such fun, Junior!'”

          I’m not trying to say his children are not intelligently interested in the subject and aren’t clamouring for “a challenging academic curriculum” but that 12 year olds would be natively and with no hint of urging from a parent passionate about labour economics? Well, maybe! (I also wonder when they are 16 or 17 if perhaps they may not go “I kinda wish now Dad had made us do the things we didn’t like, like learning to dance – apparently girls go for that kind of thing, who knew?”)

    • Urstoff says:

      Seems like you’re making a huge assumption and how particular kinds of peer groups affect personality. We may know that peer groups affect personality development, but that doesn’t give us any insight into what elements of particular peer groups affect what specific personality traits.

    • Wrong Species says:

      I’m pretty sure that Harris didn’t separate out peer group vs non-shared environment in percentage terms.

    • Loquat says:

      As a fellow new parent, I’m currently going with the theory that peer group should be accounted for, but mainly in terms of avoiding bad outcomes rather than ensuring good ones. If Junior falls in with a crowd where violence, substance abuse, or teen pregnancy are normal or even cool, that’s probably not good for future prospects, and from personal experience I wouldn’t advise moving a half-grown child into a school where there’s no turnover and everyone else has been set in their cliques for ages, but if your public school doesn’t have particular problems in any of those areas then public versus private peer group probably won’t make a big difference.

  3. onyomi says:

    Putting this in “culture war-free” zone since I’m interested in discussing it in as value-free a manner as possible, but I won’t be offended if it doesn’t work out and Scott wants to delete it or put it elsewhere:

    What is the opinion on Austrian Business Cycle Theory? Links to any good rebuttals of it? Most of the critiques I’ve read feel pretty weak to me. I am, unsurprisingly, positively inclined toward it and it makes a lot of sense to me, though not all libertarians subscribe to this view (I’m pretty sure David Friedman and Bryan Caplan do not, for example; I do not know, however, of anyone who likes ABCT but isn’t a libertarian, though they could exist).

    In a nutshell, for those who don’t know, it’s the theory that the business cycle (boom and bust) is caused primarily by money supply inflation, which distorts interests rates (generally by lowering them below their “natural” levels), misleading investors, and entrepreneurs about supply and demand, causing them to systematically err (usually by piling into some particular “hot” sector, like tech startups or housing, or in the 20s, stocks in general), leading to misallocation of factors of production (the boom=more investment in the hot sector than the fundamentals justify) which must eventually be liquidated when the underlying problems are revealed and the current lines of production are shown to be unsustainable (the bust).

    This seems a very elegant and plausible explanation to me (a non-economist); what’s wrong with it? I’m guessing part of the answer from non-Austrians will be that “the data better supports other interpretations” but how did we pick which data was relevant for deciding whether or not this interpretation was right? (I’m aware the general criticism of Austrians is that they are too a priori, too “armchair,” but it seems one must always have some theorizing in order to determine which data, of an infinitely large possible set, to examine).

    • What’s wrong with it as theory is that it assumes that investors are too stupid to realize that a temporary cause will have only a temporary effect on interest rates, and so assume that the low rates are permanent–and not only do it once but make the same mistake over and over again.

      I don’t do macro, so am not competent to judge how well the theory predicts. My father’s view, as best I recall, was that the theory was not logically impossible but was not consistent with past data.

      • onyomi says:

        Regarding investor stupidity, two questions:

        Couldn’t it be the case that, though investors can see interest rates are lower than they would be if the government weren’t e.g. engaged in quantitative easing, nonetheless, they can neither predict when the easing will end, nor which sectors have been artificially “pumped up” (that is, isn’t it probable that some sectors will diverge more from the fundamentals in response to e.g. monetary stimulus than others?) with the result that this knowledge does them little practical good?

        That is, since bubbles are generally (inherently?) not widely recognized until after the fact, knowledge that monetary stimulus might be causing bubbles doesn’t help one in recognizing where, if anywhere, they are?

        Second, aren’t there probably a lot of cases where, even when investors might suspect x is in a bubble, they nevertheless hope they can “ride the wave” and get off before it bursts/feel a lot of pressure not to avoid the “hot” sector since they have to compete with e.g. other fund managers who are in the hot sector? That is, if you are a fund manager, it may be more important for you to keep your job now than to eventually be proven right in a few years when the bubble bursts; i.e. some variant of “the market can stay irrational longer than you can remain solvent”?

        • Matt M says:

          Second, aren’t there probably a lot of cases where, even when investors might suspect x is in a bubble, they nevertheless hope they can “ride the wave” and get off before it bursts/feel a lot of pressure not to avoid the “hot” sector since they have to compete with e.g. other fund managers who are in the hot sector? That is, if you are a fund manager, it may be more important for you to keep your job now than to eventually be proven right in a few years when the bubble bursts; i.e. some variant of “the market can stay irrational longer than you can remain solvent”?

          This has always been my theory. I knew plenty of house flippers back in 2006 who would readily admit that housing was in a bubble and that it couldn’t last forever – but each individual one of them was convinced that they would be able to time it right and get out ahead of any major collapse. (None of them succeeded in this to any significant degree).

          And I think this is compatible with observed investor behavior generally. Studies have virtually proven that it’s nearly impossible to “beat the market” through active trading on a consistent basis. Most professional fund managers don’t beat the market, yet their job is fundamentally premised on the notion that they can.

          If investors aren’t “smart enough” to realize that active trading is a net loser and settle for S&P Index Fund shares, why should we assume they’re smart enough to spot and avoid bubbles in specific assets (many of which have far less “efficient” markets than the stock market)

      • Urstoff says:

        Is there a book or source that explains the various current macro schools and where/why they disagree? Also, is Romer’s textbook still the standard advanced macro textbook?

    • Wrong Species says:

      I know that Austrians are philosophically opposed to empiricism in economics but their predictions have simply not panned out. Where is the inflation we’ve been promised? Why hasn’t another recession, worse than the last one, happened after the Fed has begun the process of increasing interest rates when QE and low interest rates were supposedly a forerunner to a worse calamity?

      Lets look at the last 5 years. What events have constituted evidence for Austrian economics?

      • onyomi says:

        I don’t think it’s quite fair to say that Austrians are opposed to empiricism, though opinions on the proper relationship of data and theory vary within the “school.”

        My personal understanding of the “orthodox” Austrian take on theory v. data is that theory must guide the selection and interpretation of data because in economics, there can be no controlled experiments.

        As we’ve seen, for example, an economist can argue that the best thing to do in response to a crash is monetary or fiscal stimulus and another might argue that the best thing is for the government to cut spending and let the money supply tighten. Neither of them can ever really claim victory and both can claim to have been vindicated because they can’t ever re-do the same recession and try a different policy.

        If, as seems to be the case since the most recent crash, monetary and fiscal stimulus are tried but produce lackluster results, those who were in favor of monetary and fiscal stimulus can always say “well, thank god it wasn’t even worse, but we could have recovered faster if only people had listened to me and done more monetary and fiscal stimulus.” And the opponent of monetary and fiscal stimulus can always say “see, the recovery was really anemic even though we did monetary and fiscal stimulus, which means monetary and fiscal stimulus don’t work” (I think this is one prediction many Austrians have been pretty good on).

        Also, though we can’t run a controlled experiment on any given recession, if we assume that most recessions have some things in common (may not be true, but seems likely), we can compare how the government reacted to various historical recessions and how things turned out.

        Austrians like to point to a crash in 1919, for example, in reaction to which spending was slashed, and from which we recovered much faster than the 1929 crash or, indeed, the most recent crash, both of which were reacted to with fiscal stimulus (Austrians like to argue that the conventional wisdom that Hoover was a “do-nothing” in response to the crash is ahistorical, since he did more than any previous president in response a crash; difference being he was hugely eclipsed by Roosevelt, who took even more unprecedented measures).

        Regarding more recent predictions: it probably is fair to say that Austrians, in general, being generally anti-state in their thinking, and being somewhat like the person who sees nails everywhere when armed with a nice hammer, tend to see upcoming inflationary bubbles and crashes everywhere, perhaps eager to be the next person who called the next bubble, which are, in reality, still hard to spot without industry-specific knowledge, and even harder to predict in terms of when they might pop.

        To cite a prominent example, Peter Schiff, an investor with an Austrian perspective, famously predicted the housing crisis, not just in a vague way, but including a lot of details. However, he was also called a crank for a few years before it finally happened. He rightly gained a lot of credibility after things played out as expected, but has since lost a fair amount since he has, since then, continued to predict a crash in the value of the US dollar which has, thus far, not materialized.

        A few points on that: it could be that again, he’s right, but early. Second, I do think inflation is a lot higher than most realize because of what I was talking about in a recent thread about airlines, e.g. price stickiness: many companies will sooner make the product crummier or simply not lower prices as they might have otherwise as a result of a monetary inflation with the result that these things don’t show up in e.g. the CPI, even with hedonic adjustments, etc. Certainly many Americans seem to have a subjective impression that it’s still much harder to make ends meet since 2008, and this is largely because the prices of things, while not skyrocketing, still seem to go up, while wages remain stagnant. This would be largely consonant with the Austrian prediction of failure of monetary and fiscal stimulus, though obviously those who predicted runaway inflation were wrong. But I think 5 years is really a very short time horizon on which to evaluate macro predictions. Getting the timing right is a lot harder than getting a general theory right.

        As for why, at least hypothetically, Austrians could be right about the big picture but wrong about the extent of price inflation, there are many possibilities: the most obvious may be that, again, being eager to use that hammer, Austrians might tend to ignore many countervailing trends: technological and efficiency advances (ironically, the Austrians, who praise the ingenuity of the free market the most, tend to underestimate how nimble the free market can be in dealing with government-imposed distortion), world events (China continuing to buy our currency when we thought maybe they wouldn’t keep buying so much), typical mind fallacy: Austrians have little confidence in the government and so underestimate just how much confidence is still out there in e.g. treasury bonds, and so on.

        • Wrong Species says:

          Peter Schiff made some very specific predictions

          Schiff has long maintained that the Fed cannot continue to raise interest rates without sparking a major market crash, and urged investors to instead buy gold. That made Nations accuse him of trying to “scare” investors into buying the commodity for his own benefits.

          “I recommend what I think is going to make investors money. Gold is outperforming the U.S. stock market this year by triple,” the Euro Pacific Capital CEO responded. “Buying gold and having gold in your portfolio has been a wise choice for investors.”

          In 2013 Schiff notoriously called for gold to hit $5,000 per ounce, a prediction that hasn’t quite panned out with gold only trading near $1,300

          It certainly looks like he got lucky the first time around more than anything else.

          Certainly many Americans seem to have a subjective impression that it’s still much harder to make ends meet since 2008, and this is largely because the prices of things, while not skyrocketing, still seem to go up, while wages remain stagnant. This would be largely consonant with the Austrian prediction of failure of monetary and fiscal stimulus, though obviously those who predicted runaway inflation were wrong. But I think 5 years is really a very short time horizon on which to evaluate macro predictions. Getting the timing right is a lot harder than getting a general theory right.

          That subjective impression is consistent with a lot of explanations, including ones that don’t have a lot of wrong predictions as well(Not to mention the fact that people’s subjective impression can be out of touch with reality). Getting the timing right is not an easy task but at some point we have to hold them accountable. They’ve predicted two things which haven’t come to pass, high inflation and a catastrophic recession. How many years should pass before we can discredit them? And how catastrophic does the next recession need to be before we can say the theory is proven right? What can we do to distinguish the predictions of Austrians from others in a way we can clearly see who is right and who is wrong? The only time I’ve seen them put their money where their mouth is was when Bob Murphy bet Bryan Caplan that we would have double digit inflation by 2016 and of course he was wrong.

        • HeelBearCub says:

          If, as seems to be the case since the most recent crash, monetary and fiscal stimulus are tried but produce lackluster results

          We can’t re-run the recession, but we can look at what happened to various countries/regions who tried different approaches in the same recession.

          Europe went the austerity route, the US went the stimulus route, the US came out better. It’s not dispositive, merely one data point, and of course I am making a very simple statement about a complex subject.

          Still, I believe statement, general as it is, is broadly true.

          • onyomi says:

            From what I understand, though, most of the “austerity” Europe has tried isn’t the same sort of austerity most Austrians would recommend. Austrians would, first and foremost recommend cutting government spending, and, ideally, also taxes (they would probably recommend that at any time, but especially during a recession, when the private sector is struggling).

            Europe’s austerity seems primarily aimed at restoring governmental fiscal solvency, often by raising taxes to pay back creditors and keep the government pension plans afloat. Raising taxes so the government doesn’t have to cut spending is pretty much the opposite of what Austrians would recommend.

            So Europe v. USA in the wake of 2008 may be a good test case of something, but it isn’t a good test case of Keynesian vs. Austrian approaches to recessions. Seems like USA in 1919 and USA in 1929, or USA in 1919 and USA in 2008 are closer.

          • HeelBearCub says:

            @onyomi:
            I think Greece and Ireland have both cut spending (click on the 10 year button).

          • Matt M says:

            Europe went the austerity route, the US went the stimulus route, the US came out better.

            West Germany went the capitalist route, East Germany went the socialist route.

            In any case, IIRC Mises rejected even these types of comparisons. I think his opinion was something like “You can study what happened to the price of wheat in year X in country Y following the implementation of policy Z, but this teaches you nothing about economic principles, it only teaches you about the price of wheat in year X in country Y – this is economic history, not economics proper”

          • Brad says:

            In any case, IIRC Mises rejected even these types of comparisons. I think his opinion was something like “You can study what happened to the price of wheat in year X in country Y following the implementation of policy Z, but this teaches you nothing about economic principles, it only teaches you about the price of wheat in year X in country Y – this is economic history, not economics proper”

            To me this is just nuts. What would we think of a climatologist that refused to look at data but instead just sat in a room with a blackboard and looked inside himself to discover the a priori principles of climate?

          • onyomi says:

            To me this is just nuts. What would we think of a climatologist that refused to look at data but instead just sat in a room with a blackboard and looked inside himself to discover the a priori principles of climate?

            An important difference is that the climate has nothing to do with human psychology, so introspection would be of no use. Economics has a lot to do with human psychology, so introspection is likely of at least some use.

            The degree to which positivism is useful for study of human behavior, especially complex group behaviors, however, is debatable.

          • Matt M says:

            Right, that’s the whole point of the action axiom, which is the basic and most fundamental axiom underlying all Austrian assumptions. Humans are independent actors. They choose to do things. They are not inert objects which are acted upon. Thus, you cannot set up a sufficiently controlled experiment to use the same type of deductive experimentation you can use in physics or chemistry, or even climatology. (Although given the failure of recent climate models, there probably is an argument to be made that climatology is closer to economics than it is to physics in terms of complications and multi-variable factors).

          • Brad says:

            If a strong version of that is really the case, then the logical consequence is that economic theorizing is impossible. It makes no sense to go from that axiom to claiming a simple model based on spherical cows is more accurate than all the efforts of people that at least *try* to engage with messy reality.

            By the same token that one could in theory introspect on his own psychology and derive economics, the room with the blackboard also has objects in them which obey the laws of physics. Maybe our would be climatologist can discover the laws of gravity with a piece of chalk, electricity by rubbing his hair, and thermodynamics with his coffee. And climatology is just applied physics, right?

          • rlms says:

            I never knew the Austrians were so sympathetic to the “well, the USSR seemed pretty bad but who knows how it will turn out next time?” school of thinking.

          • John Schilling says:

            Europe went the austerity route, the US went the stimulus route, the US came out better.

            Starting from the same 65% baseline, Europe paid for its “austerity” by kicking its collective debt-to-GDP ratio up to 91%, whereas the US went up past 104%. Unless you believe that governments can borrow literally infinity money without consequence, the United States paid rather more out of its ill-defined but ultimately finite reserves for that better outcome. I think pretty much all economic theories would agree that you can get more if you are willing to pay more.

          • Whatever Happened To Anonymous says:

            I never knew the Austrians were so sympathetic to the “well, the USSR seemed pretty bad but who knows how it will turn out next time?” school of thinking.

            Indeed, the Austrians’ criticism of communism is much more theoretical.

          • Matt M says:

            Right, the Austrian view of the USSR isn’t “it might work next time” but rather “Of course it didn’t work, we told you ahead of time it wouldn’t work because of the difficulty of economic calculation under socialism”

          • HeelBearCub says:

            @John Schilling:

            I think pretty much all economic theories would agree that you can get more if you are willing to pay more.

            #1 – Japan would like to have a word with you.

            ii. We are talking about the effects of counter cyclical spending and increase in the monetary supply on a recession. Stating a bare fact about debt-to-GDP ratio doesn’t really say much unless we start positing that debt to GDP won’t shrink again as the economy grows.

            C) It is my understanding that the Austrians predict that increasing spending and monetary supply will prolong a recession and lead to stag-flation, so I believe that they do disagree.

          • Iain says:

            @John Schilling: The Euro area includes a lot of countries who took a variety of different approaches. If you drill down into the PIGS countries that really went hard on austerity (Portugal, Italy, Greece, Spain), they don’t look so hot. Ireland looks a little better, but that’s not really helpful to your case, because if you compare it to their government spending, their debt started getting going down right around the time that spending started going back up. Portugal has a similar inverse correlation, minus the improvement at the end.

            Krugman has a good chart showing GDP growth vs the change in cyclically adjusted primary balance. Search for “chart 2” here.

          • John Schilling says:

            Stating a bare fact about debt-to-GDP ratio doesn’t really say much unless we start positing that debt to GDP won’t shrink again as the economy grows.

            And, indeed, debt-to-GDP has begun to shrink in the Eurozone. But not in the United States.

          • HeelBearCub says:

            Ummm, that doesn’t prove anything about whether debt to GDP can shrink.

          • John Schilling says:

            Nobody has ever denied that debt-to-GDP ratio can shrink. The important question is whether it will. This depends in large part on A: how large you let it grow before trying, and B: how eagerly you embrace debt as the solution to transient unpleasantness. If you wish to argue that you have achieved a positive outcome before you have actually shrunk your debt back to its original size, I’m goiing to want more than “it is strictly possible for debt-to-GDP ratio to decrease”.

            Europe as a whole, at least appears to have established an appropriately negative trend. Of course, it also appears to have established a trend towards European non-wholeness, so we’ll have to see how that turns out.

          • IrishDude says:

            @Iain

            The Euro area includes a lot of countries who took a variety of different approaches. If you drill down into the PIGS countries that really went hard on austerity (Portugal, Italy, Greece, Spain), they don’t look so hot.

            My understanding is those countries you list had the weakest economies, so seeing them not do so hot after going hard on austerity, if that’s what they did, still doesn’t give convincing evidence one way or the other about austerity. They’d lived above their means and had to pay the piper for this, with no macro economic intervention allowing them to avoid the pain due to their decisions of the last several decades.

            This seems to come back to the Austrian point that there’s no controlled macro experiments, as the countries you list above weren’t a random sample of Euro countries but were systematically different.

          • HeelBearCub says:

            @IrishDude:
            The Austrians made predictions, and produced highly flawed analyses that turns out not to have actually been true, about what they believed would/did happen post-recession.

            That should be considered evidence against their theory.

          • onyomi says:

            It strikes me that evaluation of economic theories in terms of history is somewhat like the issue of “scientific consensus” discussed in a recent thread, but progress in economics is still a lot slower than progress in science, perhaps because the only experiment that matters is always unfolding at a human lifetime-scale pace.

            For example, Mises predicted in 1920 that the USSR must eventually collapse due to the economic calculation problem. And the USSR eventually did collapse, arguably because of the economic calculation problem.

            But they lasted much, much longer than Mises thought they would–like fifty years longer, probably because he didn’t adequately take into account the power of black and grey markets (underestimating the robustness of the free market they champion is a common Austrian mistake).

            So if Mises, in 1920, thought the USSR could last say, 10 years, tops, because of the economic calculation problem, then we might continually say he was “wrong” every year between 1930 and 1991. But he wasn’t wrong about the theory. His predictions based on the theory just didn’t take into account other factors and so were off in terms of their timing, if not their fundamental evaluation of the economic forces at play.

            If we accept, for the sake of argument, that Mises was right about why the USSR would fail, but merely got the timing way off, then that means his theory was right all that time. But if we had based our evaluation of his theory on the predictions he based on that theory (the problem being that one could have shared Mises’s theory about the economic calculation problem exactly in 1920 and yet offered widely divergent predictions about how long the USSR would last on the basis of differing evaluation of other factors), we would have said he was wrong for sixty years. But that would have been (in this hypothetical, at least), wrong. The proper way to evaluate his theory then, as now, was on the logic of it.

            Therefore, to me it seems the best way to approach evaluation of ABCT is a priori logic, since that is what it is mostly based on. However, I’m not doctrinaire; it’s conceivable to me, if maybe not some Austrians, that facts could tend to discredit it as well. However, if we want to use data to discredit something like ABCT, it seems we have to do more than say “some guys who believe in ABCT predicted x, but x hasn’t happened yet.” We have to point to facts which are inconsistent with the theory of ABCT, which might not exist, because it might be right–as no facts to prove Mises’s calculation argument wrong existed between 1930 and 1990, because he was right all along about the principle, if not how the principle would manifest in real life.

          • Brad says:

            The proper way to evaluate his theory then, as now, was on the logic of it.

            If a theory has no predictive power then it is useless for decision making. Economic theories should be judged on their predictive power rather than their internal logic because that’s how they pay the rent.

            We have a field that looks for beautiful theories that are internally consistent and doesn’t insist that they correspond to anything out in the world — mathematics.

          • onyomi says:

            @Brad

            I may have ninja edited my last paragraph in while you were writing your response, so that is part of my answer to it.

            I’ll add, however, that I am not saying that Austrian theories are useless for making predictions; I am saying that it is very difficult to evaluate theory in terms of historical facts, because there are so many confounding factors and you may be wrong for years before you are right for reasons unrelated to the validity of the theory.

          • Matt M says:

            If a theory has no predictive power then it is useless for decision making.

            But it does have predictive power, just only in ceteris paribus situations which do not actually exist in real life.

            To the extent that other schools of thought have better predictability but are not ideologically consistent or rationally deduced, they’re either getting lucky, or (much more likely), the school of thought is designed to fit the data – which might seem like a good idea, except the data are incredibly limted in scope and subject to change.

            Matching your thought pattern to fit the data is what caused highly sophisticated banks to rely on predictions like “Housing markets are local, they won’t all possibly collapse at the same time.” Prior to the crash, that theory “fit the data” significantly better than the Austrian theory that said “Housing is highly correlated with the money supply due to how it manipulates interest rates.” So yeah, alternative theories fit the data better, right up until they don’t….

          • HeelBearCub says:

            @onyomi:
            That’s all well and good.

            Aside: Are you comparing Mises to all other economic schools of thought as of 1920? Because that is what is really necessary.

            But the point here is that the Austrians are making short-term predictions about what we’ll happen differentially between two economies that follow austerity and stimulus once they have entered into recession. Your logic doesn’t apply in that case. And, to my aside, we can compare them to the predictions of the economic schools of thought on the time scale of their prediction.

          • Brad says:

            There seems to be two different contradictory points being woven together. I don’t know if the arguing in the alternative is on purpose or not.

            One is that it is simply impossible to measure the predictive power of a theory in economics. The logical implication of that would be that economics is not and can never be a science, social or otherwise. We should just throw up our hands. I don’t see how it can instead imply that we should accept Austrian economics. If it is impossible then it is impossible.

            Another is that it is difficult to measure the predictive power of a theory in economics and we should be leery of simplistic appeals to “this Austrain economist got these specific predictions incorrect and therefore the whole thing is bunk”. On the flip side of the coin we should also be leery of theories that backtest well because of the danger of curve fitting. I can well accept these points, but I don’t see how they offer positive support for Austrain economics in general and I certainly don’t see how they can be used to justify rejecting empiricism.

            I guess what I’m saying is the critiques may well be valid and undermine non-Austrian economics, but the principle arguments *for* Austrian economics aren’t the types of argument that could conceivably justify it. It is a category error.

            I want to know if it is predictive, you say it is, but instead of showing me that it is predictive you want me to judge it on the consistency of its internal logic. Why would I agree to that? It doesn’t get me an answer to my question.

          • Matt M says:

            The logical implication of that would be that economics is not and can never be a science, social or otherwise. We should just throw up our hands. I don’t see how it can instead imply that we should accept Austrian economics. If it is impossible then it is impossible.

            What I would say is that Austrian economics offers insights, it does not offer predictions.

            A lot of people try, for various reasons, to translate insights into predictions, with varying levels of success. But they are, in fact, different. If what you specifically want is predictions and you do not see any particular value in insights, then fine, Austrian economics is certainly not for you and you probably shouldn’t bother with it.

          • Paul Brinkley says:

            One of my main impressions of an Austrian insight, however, was that no economic system would give you practical prediction, as illustrated by The Fatal Conceit. There was too sensitive a dependence upon initial conditions – some specific preference by some individual would overwhelm whatever end condition one desires, if one sought to exert any significant control over that system. One could never know what that specific preference would be, because there are too many of them to search in the time allotted; it would only ever be evident in hindsight, at best. So it was almost always easier to let that preference be constrained naturally by other specific preferences, according to Austrian thinking.

    • Mark V Anderson says:

      I am kind of philosophically opposed to elegant theories of economics. Things aren’t that simple. It has always seemed to me that recessions happen when there are bottlenecks. The economy runs out of various raw materials it needs to continue growing, or low on some kinds of skilled labor, or too much capital is allocated to the wrong place. Certainly when the money supply is changing in complicated ways that aren’t well understood by the players in the economy, that will cause many poor allocations, so the Australian theory likely explains some recessions. But I can’t believe it is the case for all.

      Or probably a better explanation for recessions is a combinations of various bottlenecks bad enough to cause an economy to go into reverse, so it is possible that money supply could be a contributing factor to all.

      • onyomi says:

        I am kind of philosophically opposed to elegant theories of economics. Things aren’t that simple.

        What about, say, the law of supply and demand? That’s pretty elegant, right? And one can find lots of weird cases where it seemingly didn’t apply in real life as we’d expect; yet (almost?) everyone still accepts that it’s a broadly valid principle.

        • Mark V Anderson says:

          No I don’t think supply and demand is so elegant. Simple perhaps — more demand drives up prices, more supply drives it down. Deceptively simple perhaps. But maybe you could say the same for the Austrian business cycle.

          One should always look to simplify economics, just so it can be understood. But I say be very wary of those theories that imply that real life is really that simple. I guess that’s what I think of as an elegant theory, one that is both simple and comprehensive (which I think won’t match real life). Maybe that’s not what you meant by elegant.

          • onyomi says:

            I guess what I like about the Austrian effort to distill axioms based on e.g. contemplation of human nature (“if someone paid x dollars to buy y, it means that, at least at that moment, he valued, or prospectively valued y more highly than x dollars,” etc.) is that I think that, of the things economics can tell us, they tend to be more useful, especially for average joe voter and average joe politician, who are rightly more interested in general guidelines than detailed autopsies of cases which will not, in any case, ever exactly repeat themselves.

            Which is not to say that theories should never be made to “pay rent” in expected experiences; if a theory seems constantly to fail, one needs either to figure out the confounding factors and/or reexamine the logic of the theory.

            But predicting the future is really hard. To take the fairly well accepted supply and demand, one could have predicted that “Cash for Clunkers,” which intentionally reduced the supply of older cars, would raise the price of used cars. And it did. That prediction wouldn’t be because we closely examined the data relating to all the times in history the government bought up and destroyed old stuff (though I guess we might have been able to find some relevant cases); rather, it’s because we know it’s only logical that when there’s less of a thing people want, they’ll pay more to get it.

            Divining accurate guidelines for more complex cases may be a lot more difficult, and maybe the Austrians still have work to do on that score, but even if they do that work and come up with more, valid principles of the nature of the law of supply and demand, evaluating the validity of the individual predictions of individual Austrian economists will not be a good way to test the validity of the principles, especially on a short time horizon. Rather I’d say we should first survey history to see if it’s supportable, and, if so, tentatively accept the idea if it makes good logical sense.

            Where I think I tend to agree with the Austrians is that they aren’t going to assign equal probability to an interpretation of the data which doesn’t make sense in terms of human nature as to one that does, even if the raw data supporting both conceivable interpretations is equally strong.

            Somewhat related: is there any data on how well economists fare as investors? I have the impression that economists are not more successful investors than anybody else, which would tend to indicate that prediction, still, is not their forte. (And on the flipside, people with very different views of politics and/or macroeconomics can both be very successful investors, probably due to other personal characteristics, knowledge, talents, etc. Take Jim Rogers and his old partner, George Soros).

          • Matt M says:

            I think that the real problem here is the ceteris paribus assumption. Austrian theory does not really predict “There will be 20% inflation in 2014.” What it actually predicts is, “All else being equal, a country that engages in quantitative easing will experience more inflation than a country which does not”

            The problem is that economies are SO complex with SO many variables that this assumption never actually holds in real life. North Korea/South Korea is close, but not sufficient. Even minimum wage studies that examine counties across a river in neighboring states still misses some important things.

            The entire exercise of Austrian economics is premised around the idea that a “controlled experiment” is essentially impossible, and that most attempts to try and learn from real life comparisons/data are just as likely to mislead as they are to inform.

  4. jasongreenlowe says:

    Linking to a HuffPost article on drug addiction that I liked because of the way it juxtaposed two ideas:
    (1) rats who are isolated in boring cages always choose to drink cocaine-laced water, but rats in a utopian Rat Park with plenty of tunnels, friends, good food, etc. voluntarily avoid cocaine-laced water, and
    (2) veterans and hospital patients who return from war or recover from their injuries voluntarily stop using painkillers that are extremely similar to heroin.

    These stories are inconsistent with the “classical” understanding that people use street drugs to excess because the drugs are chemically and physically addictive. Instead, they suggest that people use narcotics as a way of coping with painful or stressful circumstances like war, hospitalization, and isolation. It’s not that people don’t have cravings or experience withdrawal symptoms — of course they do — but the cravings, etc. may not be the most important “explanatory factor” that predicts excessive drug use. Perhaps the most efficient way to reduce opioid use is not to search for less addictive painkillers, but rather to provide therapy and life coaching so that people will feel less stress and isolation.

    • The Nybbler says:

      rats who are isolated in boring cages always choose to drink cocaine-laced water, but rats in a utopian Rat Park with plenty of tunnels, friends, good food, etc. voluntarily avoid cocaine-laced water

      Oddly, this is similar to the way Sherlock Holmes used his seven percent solution of cocaine — only when he was bored.

    • Well... says:

      FYI Rat Park rather famously failed to replicate; turned out some of the rats were more or less genetically predisposed to addiction.

    • Evan Þ says:

      If that’s the case (despite what Well… says), that’d also line up with some reports from heroin addicts in small-town America who say they use because they don’t have any purpose to life otherwise.

  5. ZeitPolizei says:

    Can we get/do we already have more data on ssc reader locations, i.e. the # of people in cities with less than 10 people interested in meetups?

  6. CatCube says:

    Is anybody else using Bakkot’s new thread-collapsing app surprised by just how few top-level comments there seem to be in the OTs? There’s only about 20 for 200 total comments. It always felt like more.

  7. Nancy Lebovitz says:

    I haven’t been able to follow these clearly enough to have an opinion about them, but if the white working class in the US isn’t as badly off as we’ve been told, it matters, so I’m posting the links here.

    https://psmag.com/the-death-of-the-white-working-class-has-been-greatly-exaggerated-1c568d3e6b8c

    http://www.slate.com/articles/health_and_science/science/2017/03/is_white_mortality_rising_not_really.html

    Links found here.

  8. Anon. says:

    The French polls are getting really interesting, it’s actually a 4-way race. Even a Melenchon-Macron run-off seems possible at this point.

    Latest numbers:

    Macron (Center) 23%
    Le Pen (Right) 23%
    Fillon (Center-Right) 20%
    Melenchon (Left) 18%

  9. Aapje says:

    Meteorites are all around us (click the link, pretty pictures)!

    A researcher found micrometeorites in the dust collected off rooftops. Elsewhere he said that it takes him 1 hour to work through a gram of dust and that 1 teaspoon of dust has 1 micrometeorite on average. Assuming 6 grams of dust in a teaspoon, that means 6 hours to find one.

  10. AnonYEmous says:

    so culture war free but

    https://www.cato.org/blog/response-scott-alexander

    not a fan of the criticism personally, they seemed to miss that this tactic aims at the middle rather than the offended political wing

  11. Jaskologist says:

    “The Reformation, inwardly considered, was just the ultimate triumph of Augustine’s doctrine of grace over Augustine’s doctrine of the Church.” -Benjamin Warfield, Calvin and Augustine

    Welcome, brave readers! For the next few months, we’re going to plunge into the autobiography one of the most significant thinkers in the West: The Confessions of St. Augustine. You’ll get a healthy dose of life experience, philosophy, and theology.

    What to expect

    We’ll be reading through about 2 chapters (usually called “books”) before every 0.5 Open Thread. If you miss a week, just skip ahead.

    The first nine chapters are Augustine’s autobiography, and one of the very first examples of the genre. The last 3 delve into philosophical issues, such as the nature of time, memory, and the creation of the world. Each chapter begins with a prayer to God, which may seem a little unusual to our modern sensibilities. In fact, the whole book is really a prayer to God (hence the title; Augustine is confessing to God and we’re just listening in).

    Choose your text

    If you want a hard copy, this abridged, modern English version is a good choice.
    Or, read along online.
    Maybe you prefer the original Latin? If so, you get extra credit.
    Or maybe you just don’t like reading, in which case here’s a free audio version.

    Homework

    Read the first 2 chapters. In the abridged hardcopy, that means read up to page 30.

    • Jaskologist says:

      Why should I read Augustine?

      If you are a Christian, the answer is easy. Augustine is the most influential Christian thinker after Paul. All of Western theology is really just a series of footnotes to Augustine. He predates the major controversies which still split the church today, so nearly every denomination from Protestants to Catholics are happy to claim him. If you find an idea in Augustine, it may well be wrong, but it’s probably not heretical.

      Maybe you’re an atheist. You’re heavily influenced by Augustine. So many Western ideas flow from his thought. Seperation of church and state, free will, determinism, even existentialism are built on top of concepts he planted in our minds. He’s the original philosopher and investigator of the innner self, and even invented the word “soliloquy.”

      Maybe you’re an SJW! Augustine was an African. Take a break from reading this white male codex and add some diversity to your reading list.

      Augustine’s influence stretches throughout history; some even credit him with laying the groundwork for the Reformation. Martin Luther was himself an Augustinian monk, and both he and Calvin quoted Augustine more than any other theologian. In turn, during the Counter-Reformation, St Teresa of Avila was inspired by her reading of Augustine’s Confessions to become a Carmelite nun, an order which she then undertook to reform. The philosophy he laid out still shapes our thinking today, and we can all stand to gain from reading his insights first-hand.

    • Jaskologist says:

      Background

      Augustine lived from 354-430AD. He was born less than 50 years after Emperor Constantine proclaimed religious tolerance, ending the extremely vicious persecution that had been prosecuted against Christians before then. (For comparison, if you were born in the 1980s, about the same length of time separates you and the Holocaust). He was born in Africa, in what is now Algeria, back when it was still part of the Roman empire, and was of Berber descent. His mother was a Christian; his father was a non-observant pagan (although he converted shortly before his death). The family was middle class, with enough money to give Augustine a good education (he later went on to become a teacher).

      At this time, northern Africa was basically Bible belt of the Roman empire, full of the ancient equivalent of rock-ribbed fundamentalist Christians. It was also the bread basket of the empire. Picture Arkansas and you won’t be far off.

      In the middle of that is Augstine, a snot-nosed brat who is smarter than everyone else (and knows it). He longs to get out of this backwater and advance to the big city (Rome) where the sophisticated people are.

    • keranih says:

      Sideways. but perhaps context:

      I was taught that there were four main patterns to (Catholic?) Christian observance – not the steps, so to speak, but the rythum.

      Augustine – reflective, authoritarian, contemplative
      Thomasian (Thomas of Aquanias) – inquisitive, knowledge-based
      Ignatian (the Jesuits) – rule bound, obedient, unyielding
      Franciscan – glorifying, exhalting, unfettered

      Each method is focused on God, and strives to serve God, but finds a different niche in the human soul, so to speak. It made me much happier in the Church, once I realized that my path to deeper faith didn’t have to match everyone elses.

    • Le Maistre Chat says:

      It’s been awhile since I read this. I remember the outline (childhood, philosophy education, Ambrose of Milan, Mom dies happy, memory and other philosophical topics) but I’m sure I forgot many details.
      Does he pull the same thing he does in The City of God where he says “I’m addressing the Platonists here, because they’re closest to the Truth”? =)

      • Jaskologist says:

        I don’t recall him addressing them directly, but Plotinus was certainly crucial to his intellectual development, and even provided him with the answer to the Problem of Evil that he ultimately accepted. I’d consider Augustine a Platonist even post-conversion.

    • Winter Shaker says:

      Not sufficiently interested in religion to follow along, I’m afraid, but I did out of idle curiosity click through to the Latin version, having studied the language at school years ago, and knowing a fair bit of two of its modern descendants, French and Portuguese.

      What a weird experience – the gulf between how much I can make sense of so many of the individual words yet how little sense I can make of the overall meaning of the sentences.

    • Jaskologist says:

      Things to expect in this reading:

      Augustine starts his biography as early as possible. Though he doesn’t remember his own time as a baby, Augustine uses his observations of other children to determine what he must have been like. Note Augustine’s view of a baby’s “innocence.” He takes a dim view of human nature even from birth, which is a significant part of Christian theology.

      Augustine recounts a very famous incident where he stole pears from a neighbor’s tree, which he treats as a prototypical sin. This was significant to him because he stole them not because he was hungry or they were especially good pears; he stole them solely because it was wrong to steal them. This also informs his views on the fallenness of human nature.

  12. Nancy Lebovitz says:

    Man convicted for a murder he didn’t commit on the basis of coerced evidence from an “eyewitness” who could barely see. Other relevent evidence was also withheld.

    What incentives make this sort of thing happen? How could things be structured to make it unlikely? I’m very dubious that punishment works especially well, especially for bad behavior which is easy to conceal, so if you want to vent about the culprits deserving to punished go ahead, but I’m more interested in other strategies.

    • Matt M says:

      Seems like a tough problem. Almost everyone involved (the police, the victim/family, society in general) are either highly incentive or at the very least strongly desire to have a suspect named and punished. It gives everyone a sense of closure and faith that crime is adequately punished.

      • hoghoghoghoghog says:

        Aside from defense attorney. Which makes me wonder: what are the incentives in inquisitorial justice systems?

        • Matt M says:

          A defense attorney can’t make any money at all if a suspect isn’t named and arrested. A world in which cops are more willing to shrug their shoulders and say “Well, this one will never be solved” doesn’t help defense attorneys that much!

    • Fossegrimen says:

      I’m in the camp that says all eyewitness testimony is so unreliable that it shouldn’t be allowed.

      • Nancy Lebovitz says:

        I feel the same way about confessions.

        • Brad says:

          Life isn’t a police procedural, there’s rarely slam dunk physical evidence. Are you okay with far fewer convictions at far higher cost?

        • Matt M says:

          I wouldn’t go this far, but I’d probably be fine with something like “confessions to a crime are immune from perjury charges later.”

          It seems to me the general pattern of “false confessions” is some form of coercion, followed by constant threats of “well you told us X before and if you say anything different later we can throw you in jail for lying to us”

          • Matt M says:

            More generally, I find it absolutely disgusting that police are allowed to lie to suspects, but that you are not allowed to lie to them.

            Given that we probably can’t reasonably prevent the latter, I’d be fine with “lying to the police is not a crime.”

          • The Nybbler says:

            There’s all sorts of forms of false confession, ranging from “Your buddy implicated you; if you confess we’ll give you a much lighter sentence in exchange for not going through the trial” to “If you don’t confess, we’ll assume your wife is involved too, arrest her, and put your kids in foster care”.

          • Matt M says:

            Nybbler,

            Right. My point is, AFTER someone confesses in that sort of manner, the main thing preventing them from coming back and saying “Wait, that’s wrong, I didn’t actually do it, I just wanted to keep my wife out of jail” is the knowledge that they confessed under oath and if they change their mind they’ll be in even MORE trouble for lying to the authorities.

            People should have the right to take back a confession at any time. At least until we’re confident that the cops are not lying to, intimidating, or coercing people.

          • Eltargrim says:

            @Matt M: That’s not my understanding at all. I’m unfamiliar with any retracted confession that has lead to perjury charges, and in general perjury charges are quite rare.

            My understanding is that a successful appeal requires something more than just a retracted confession, ie evidence of an improperly obtained confession and that the confession was crucial for deciding the case at hand. The barriers to the appeals process are substantial, and far more likely to influence pursuing an appeal than a hypothetical risk of a perjury charge.

            Basically the only place where you’ll really get nailed on a false confession would be lying to federal authorities, and that’s not the bulk of the criminal system. Popehat has a number of good examples of how lying to the feds can end extremely poorly, but AFAIK this isn’t representative of state governments.

          • Brad says:

            The issue isn’t the threat of perjury, the issue is that the confession can be presented to the trier of fact even if it is latter withdrawn. See e.g. the recent conviction in the Etan Patz case.

          • Eltargrim says:

            @Brad: while absolutely a huge issue with confessions in general, the fact that a confession can be used after being retracted provides no disincentive to retract the confession. It’s a very strong incentive to not confess, but if you’ve already confessed there’s nowhere to go but up. Either your confession will be suppressed, or it won’t, but you’re already in the latter state before retracting it.

            If there was a pattern of charges being enhanced or new charges being laid after retracting a confession, that would be a separate, yet also important issue. But to my understanding this basically doesn’t happen.

            There are many, many issues with confessions, how we obtain them, and how we use them, but again I’m unfamiliar with cases where a retracted confession has lead to an enhancement of charges specifically related to “lying under oath”, as Matt M put it.

          • Brad says:

            Eltargrim:
            I think we are in agreement.

          • Eltargrim says:

            @Brad: Probably, just wasn’t sure who you were responding to. Cheers.

  13. Nancy Lebovitz says:

    Who’s doing good hard science fiction these days?

    My definition of hard science fiction is fairly broad– stories which are strongly dependent on current science. In other words, there can also be speculative or even false (like ftl, most likely) elements as long as there’s a good bit of real science.

    Hard– or hardish– science fiction I’ve liked: Hal Clement, Poul Anderson, Greg Egan (though some of his more recent work required more science than I’ve got), Kim Stanley Robinson (yes, I know about the windmills, but most of his sf isn’t like that– I was especially fond of the variety of habitats in 2012.

    • Iain says:

      Have you read Blindsight by Peter Watts? It’s reasonably hard; he’s a former marine biologist, and he spends quite a bit of time reading journal papers for inspiration.

      • Nancy Lebovitz says:

        Yes, but too bleak for my taste.

        When I read it, I thought it was just nasty to have humans torturing an alien for no good reason, but I’m more cynical about people now.

      • Chalid says:

        It looks like Blindsight has been recommended (or at least discussed) in in 23 threads on SSC. Does any other book come close?

        • Iain says:

          Thinking Fast and Slow?

          • HeelBearCub says:

            Thinking Fast and Slow is 29 mentions.

            The Machinery of Freedom is 30.

          • Chalid says:

            Albion’s Seed gets 54!

            And of course HPMOR has 99 threads (now 100 I suppose).

            The Blindsight mentions are almost all recommendations though. The mentions of the other books aren’t.

        • rlms says:

          Better Angels Of Our Nature and The Righteous Mind are both 40 mentions. The Bible is 303. The Koran (Koran or Quran) is 79, Torah is 63 and Talmud 43.

        • DrBeat says:

          I hate that goddamn book so much, it’s not hard SF, it’s just relentless cheating

    • Zodiac says:

      I’m uncertain if it’s sufficently hard for you but I very much enjoyed the Expanse series by James S. A. Corey (alias Daniel Abraham, Ty Franck).

      • Nancy Lebovitz says:

        I like science fiction of various degrees of squishniness, too. However, hard sf isn’t all that common and I haven’t been seeing much of it lately.

        I read the first Expanse book, and got bored. I’m not sure why, and I get the impression I’m the only person who didn’t like it.

        • hlynkacg says:

          Leviathan Wakes, the First Expanse book, definitely suffers from pacing issues and “first novel syndrome” that said, the follow-up was much better and I’ve been really enjoying the TV adaptation.

          • LHN says:

            I at least temporarily bugged out of the series as of Nemesis Games, because I’ve about had my fill of books which [spoiler rot-13ed] qrfgebl gur rnegu.

            (The story seems as if it’s shaped somewhat differently in the TV series, so I’m interested to see how things play out there.)

            I’d actually sort of like to see a series that stays on the scale that Leviathan Wakes starts at. SF has a tendency to go straight to interstellar, where a solar system developed enough to have settlements and rival powers across the system, with the stars still out of reach for the foreseeable future, seems comparatively underexplored.

          • sandoratthezoo says:

            It’s not a first novel, though. Not even close.

            I mean, I liked Leviathan Wakes well enough, but I don’t think that anyone should expect its problems to go away later.

          • Fossegrimen says:

            Isn’t that the one where the whole plot revolves around water shortage while sitting on 200 million cubic kilometres of water ice, i.e Ceres?

          • Zodiac says:

            @Fossegrimen
            No, not really. I think you’re confusing it with some other book.

          • sandoratthezoo says:

            @Fossegrimen

            They bang the “water shortage” drum a lot more loudly in the TV show than in the novel, as I recall it. Also, in all fairness, when they wrote Leviathan Wakes, we knew considerably less about the structure of Ceres than we do post the 2015 Dawn mission.

        • Brad says:

          I liked it well enough, but I don’t think it is particularly hard, especially later on in the series.

          The Nexus trilogy, which I *think* I may have discovered here, was enjoyable and near-term fairly hard sci fi.

  14. FacelessCraven says:

    CW: Kaiju, artillery, anti-missile/rocket bias

    @HlynkaCG – I’ve gotta say, your writeup in the last thread was spot on. Thanks much!

    “Otherwise with the kaiju down, the batteries begin a systematic bombardment of it and it’s immediate surroundings with thermite or white phosphorous to finish it off and “sanitize” the environment. (Kill it with fire)”

    Wouldn’t this be easier/cheaper to do with mortars/low-grade rockets/conventional light artillery, while saving the heavy cannon’s barrel life for AP shells? I have considered using multiple AP shots on corpses to break them up/confirm the kill/make decontamination easier, though; the idea being that dropping them is one thing, while damaging them enough that they go fully inert and start decaying is another.

    Decontamination is one of the areas I haven’t thought much about yet; I’m assuming WP/Thermite/Thermobarics/conventional incendiaries are a big part of it, and likewise some sort of chemical approach. I’m assuming the Kaiju’s internal conditions and immediate area of effect are too extreme for complex organic molecules like nerve agents to survive intact, but the ecology is probably a lot more vulnerable. Mustard agents or similar that take effect by direct chemical action/corrosion might be a better bet as well.
    Fluorine seems like a really good option for burning out/breaking down corpses and ecology, but it’d likely be tough to produce enough, and pretty clearly a cast-iron bitch for transportation/handling. I’m also considering radiation, either in the form of irradiating the landscape with some sort of emitter or scattering isotope dust/pellets. Your mention of “salting the earth” is pretty much spot-on.

    “Let’s be honest, there aren’t many places a kaiju-sized target can hide on the great plains 😉”

    Dust storms and weather might affect this some, but yeah. really flat ground seems like it’d be easier to scout and easier to transport, so fewer assets can cover a lot more ground much more reliably.

    “Guns wear out, if you can’t replace them don’t bother deploying them. Rather than being irreplaceable commissioning a new gun battery should be an investment on par with commissioning a new warship. This allows you to have each gun or battery be a “named individual” with a storied history while still being able to realistically risk them in battle.”

    This is *exactly* the feel I’m going for, and the biggest reason I want to go superheavy guns rather than rockets.

    “The kaiju are biological, is there anything in particular that they are allergic or attracted to? if so this could be used to bait or herd them into a kill zone. I always felt like “pepper spraying” monsters was a missed opportunity.”

    Biomass, energy concentrations, possibly really, really strong electromagnetic fields, very possibly human brainwaves/computational complexity/information-density. I think John Schilling asked about patterns to kaiju movement, places that seem to get hit more often than others, and the idea intrigues me. As for pepper-spraying, I’d originally been against this, but Civilis’ comments changed my mind; I definately want harassment/weakening methods, whether chemical, explosive or incendiary, to be a thing.

    “On further thought I like the idea of the alien ecology having an aversion to salt. It explains why they don’t swim around the defense line and there’s a certain pathos to the KFF “salting the earth”, destroying their own land to ward off the monsters.”

    If it were that easy, they wouldn’t be nearly so much of a disaster! I’m assuming that for an ecology orders of magnitude tougher, more adaptive and more aggressively expansionist than earth’s native species, the oceans were pretty much a write-off from the start; too much biomass available, too hard for humanity to spot threats or hit back. The oceans are leviathan territory now. I think there’s some disconnect or divergence between the sea-based and land-based ecology though; the coasts are definately hosed, but the ocean ecology doesn’t push across the tideline as hard as land-based ecology does. Still, places like Florida, Louisiana, and much of central america are just not defensible.

    Out of curiosity, have you read any of the War Against the Cthorr series?

    • Eltargrim says:

      Regarding decontamination, a big question is “how are the kaiju contaminating the landscape?” The appropriate approach depends significantly on the nature of the contamination. For example, if the contamination is prion-like, fire or a strong base would do the work. If the contamination is radioactive, there’s basically nothing to do but wait. I’d be happy to try and provide suggestions if you can narrow down what you’re thinking of.

      Also,

      Fluorine seems like a really good option for burning out/breaking down corpses and ecology, but it’d likely be tough to produce enough, and pretty clearly a cast-iron bitch for transportation/handling.

      Fluorine gas is not ridiculously hard to produce, but you might have better luck with aqueous hydrofluoric acid. It’s produced through a simple reaction of hot sulfuric acid and fluorite (an abundant mineral), is less of a cast-iron bitch for transportation, and is also a precursor to the industrial production of the gas.

      HF has the reputation for being ridiculously toxic (thanks Breaking Bad!), is genuinely quite reactive, and produces a large amount of the fluoride ion. It’ll kill things dead pretty damn good.

      • FacelessCraven says:

        Eltargrim – “Regarding decontamination, a big question is “how are the kaiju contaminating the landscape?” ”

        The Kaiju ecology is a very wide range of organisms, from the micro-scale right up to the kaiju themselves. The main thing they have in common is that they heavily outcompete native species, and tend to have highly complex and in many cases interlocking or symbiotic life cycles. Mold and fungi analogs grow from spores, achieve a certain level of complexity, and might then convert themselves into eggs for a fly or wasp analog for example. Critters range from vermin to elephant-scale megafauna; usually with the larger forms emerging as the ecology matures through a number of distinct stages. All life forms tend to spread the base ecology seeds/spores wherever they go. Both the flora and fauna are considerably less complex and exotic than the Kaiju, so defoliants and chemical agents are a lot more effective at wiping them out.

        tl;dr – mold spores that incubate into kudzu that hatches wasps out of its flowers, that inject cat eggs into their victims that form into trees when they get old.

        Since I lack much of a biology background and am basically making this stuff up HP Lovecraft style, it’s hard to nail down specifics on how all of this should work; The laws of physics being what they are, I’m pretty sure there’s only so many ways to set up the sort of complex mechanisms that living creatures need to operate, but I have no idea what, say, a xeno-biological equivalent of nerve gas would be. I *am* sure that I don’t want decontamination to be quick, easy, or cheap; salting the earth is the guiding metaphor.

        “HF has the reputation for being ridiculously toxic (thanks Breaking Bad!), is genuinely quite reactive, and produces a large amount of the fluoride ion. It’ll kill things dead pretty damn good.”

        This is good data! I’m picturing big tanker-trucks arriving at a downed Kaiju, drilling into the corpse at multiple locations and then pumping in massive quantities of hydrofluoric acid to break down and neutralize the internal organs, fluids and structures.

        • Eltargrim says:

          Easy one first:

          I’m picturing big tanker-trucks arriving at a downed Kaiju, drilling into the corpse at multiple locations and then pumping in massive quantities of hydrofluoric acid to break down and neutralize the internal organs, fluids and structures.

          HF is basically the perfect chemical for this purpose in an environment with truncated industrial capacity. There are a couple that would be better in terms of reactivity (chlorine trifluoride and fluorine peroxide), but would be more difficult to produce, handle, store, and deliver.

          For narrative purposes, note that the real mechanism of HF will look nothing like Breaking Bad. Realistically, even with high-pressure delivery and large volumes, expect the process to take hours to days. Possible plot hook: defend the corpse of the megafauna against the subfauna while the decon is happening?

          Now, the harder one: how to guard against a complex biological cycle that is explicitly non-terrestrial in such a way that decontamination is effective, but can’t be used to kill the critters outright as a weapon of war.

          I caution against explicitly saying that the kaiju aren’t carbon-based. Carbon really is uniquely good at bonding in ways that you can’t really replicate in an earth-based environment. The statement about different physiology is enough of a hand-wave.

          I propose a three-step process for decontamination: bullets, fire, and radiation. Bullets kill mobile fauna (which will evade the fire and radiation), fire burns the corpses and the flora (which can survive radiation for too long [not enough cell division]), and the right kind of radioactive particle kills the monocellular life (which could be heat-resistant, but too much radiation damage inhibits cell replication).

          The trick is finding the right kind of radiation source. Priority one is that it is intense enough to actually kill things. A close second is that it lasts long enough to be sure you got everything. Unfortunately, these two are at odds for a single isotope, so you’ll probably need a blend of two or more. You also probably don’t have time to figure out the best option, what with the kaiju knocking down your door.

          I’d go with nuclear waste. “Fresh” nuclear waste has a ton of highly intense radioactive isotopes that give off a good mix of alpha, beta, and gamma radiations, one of which may well be effective. You don’t necessarily even have to know which isotopes are really doing the work. It will also have long-lived fission products that will render the landscape uninhabitable for decades, if not longer. Treatment is a potential option, if you don’t have kaiju breathing down your neck. Short-term, though, it won’t be so toxic as to prevent mobile alien fauna (or humans, for that matter) from getting through the zone.

          Nuclear waste is also good because it’s straightforward in it’s production, but competes against power generation. Maybe fresh waste that has been shown to be effective, so you want to pull fuel rods out of reactors after a short burn cycle to get at certain products; but that shuts down a major electrical source for months, which impacts both industrial production in other sectors and civilians.

          I realize that none of this would be particularly effective against bugs, so there will probably have to be another step in there to account for them. Or not; maybe the bugs are an unsolved problem, leading to an extremely high priority on maintaining a dead zone before the bugs can be spawned.

          This is a fun problem. I think the trick will be finding the right balance of hard science and hand-waving. I’d tackle that by avoiding explaining why things don’t work, only that they don’t, and you can’t exactly get a kaiju sample to a lab to run tests without risking the entire safe zone.

    • cassander says:

      Is there any reason not to use a highly lethal, not particularly long lived, chemical poison? 10 milligrams of VX is enough to kill a 70kg human, the internet tells me that kaiju weigh 7000 tons, so assuming everything scales, that means you need to hit him with 1kg of nerve agent. That seems WAY easier than any other solution that’s been offered, even if I’m off on the lethal dose by a couple orders of magnitude.

      • Protagoras says:

        The kaiju have already been described as having weird alien physiology. There’s no reason they’d use the same neurotransmitters we do, and so no reason VX would have the same effect, or any effect at all, on them.

        • cassander says:

          Right, but presumably there’s some other substance that’s as poisonous to them. And you don’t even need something as lethal as VX, it could be 4 orders of magnitude less lethal and still pack a lethal dose into a 2000lb bomb.

      • FacelessCraven says:

        VX and similar nerve agents operate by inhibiting the enzymes that enable motor neuron function. This works great for terrestrial, carbon based life forms that use those particular enzymes. The Kaiju aren’t terrestrial in origin, probably aren’t carbon-based, and so don’t use those enzymes; their physiology is sufficiently exotic that no chemical weapon with a similar effect has been found, and their internal conditions can probably break down or denature complex biological molecules in any case. Check the art, note, how the Kaiju’s feet are igniting foliage in its footsteps, tossing arcs of lightning, and generating some sort of EM-refracting field at a significant radius; these aren’t just Godzillas, they’re more like Cthulhuoids.

        Stuff like mustard gas, which operates much more like a chemical burn seems like it would be a much better bet against truly alien biology, and also fits the harassment role a lot better as well. On the other hand, VX might be effective against some of the wildly varied ecology organisms, so it probably has uses as a decon agent.

        [Edit] – ninja’d. 😛

    • hlynkacg says:

      Wouldn’t this be easier/cheaper to do with mortars/low-grade rockets/conventional light artillery, while saving the heavy cannon’s barrel life for AP shells?

      Remember, the more rounds your guns spend lob down-range the easier it is to justify their existence. Using your guns for the bombardment helps justify their existence. Besides, the primary driver of barrel wear isn’t the shell so much as the charge behind it. Using a reduced charge to lob a relatively lightweight incendiary or chemical shell in a high arc is going to be a lot easier on both breach and barrel than using a double bag of fast burning powder to throw a heavy penetrator in a low flat trajectory.

      This is *exactly* the feel I’m going for, and the biggest reason I want to go superheavy guns rather than rockets.

      I get that, but this cuts both ways. As I said in the previous thread, unless you give your humans access to the same square-cube law defying magic the Kaiju have, a HET or similar vehicle is pretty much the biggest thing you can drive down a winding mountain highway at anything approaching Kaiju pursuit speed. Your guns can’t acquire that “storied history” if they never reach the firing line.

      I definately want harassment/weakening methods, whether chemical, explosive or incendiary, to be a thing.

      This is part of what I was thinking when I laid out my idea of a “typical intercept”, smack them in the face to blind them, AP to cripple them, and then sustained bombardment to finish them off and sanitize the area.

      Out of curiosity, have you read any of the War Against the Cthorr series?

      No, but reading the synopsis on tvtropes I’m getting an image.

      • FacelessCraven says:

        “Remember, the more rounds your guns spend lob down-range the easier it is to justify their existence. Using your guns for the bombardment helps justify their existence. ”

        If you’ve got it, use it, I suppose? I guess that works.

        ” get that, but this cuts both ways. As I said in the previous thread, unless you give your humans access to the same square-cube law defying magic the Kaiju have, a HET or similar vehicle is pretty much the biggest thing you can drive down a winding mountain highway at anything approaching Kaiju pursuit speed.”

        On the one hand, this feels like an engineering challenge. HETs have a capacity of 70 tons, so if you used two in a fore-and-aft configuration the way the m65 set up its prime movers, shouldn’t you get a vehicle with a 100+ ton capacity, possibly more since they effectively only have one trailer between them? Ultra-class haul trucks get well over this point, with the Caterpillar 797F having a 360-ton capacity, and something like an 80psi ground pressure. With that capacity, you could spare 150-200 tons toward minimizing the ground pressure lower via additional axles, doubling up wheels, etc, and still have enough for a composite 16-inch gun, mount, and gear.

        On the other hand, the 16-inch guns were a greedy idea to start with, and HET/TEL and the m65 was about the scale I was envisioning anyway. According to the wiki page, the army lists the HETs as having a 70-ton payload, which honestly should be more than enough. The m65 11-inch gun comes in at 50 tons. The 12″/50 caliber mark 8 comes in at 61 tons with a half-ton shell; assuming a 33% reduction via advanced composite construction, that’s around 40 tons for the base gun, which would leave plenty of mass for powered gun-handling and some form of powered-loading/autoloader. This would be a heavy anti-Kaiju weapon, with 8-inch guns, possibly in a dual mount (34 tons for a pair, 22.5 tons with 33% reduction ) being a “light” weapon and probably be pretty effective for ecology bombardment.

        Sound about right?

        …Grah, can’t resist. HETs have a 70-ton capacity, right? If you set up two HETs the way the m65 used dual tractors, could you get capacities north of 100 tons suspended between the two?

        “No, but reading the synopsis on tvtropes I’m getting an image.”

        The ecology part of the story is incredibly fascinating. The psychological portion of the story would be a lot better if the main character didn’t find himself in situations where it was very important that he screw children. : /

        • CatCube says:

          I’m running to the airport, so I might be missing some details of your question, but for the HETs bridge capacity will be an issue if you try to use two in tandem. The STRAHNET is guaranteed to have one road that will take one HET, but it’s virtually impossible to find a bridge that will be designed for twice that load. (Plus vertical clearance will probably become an issue)

  15. rlms says:

    People/person (if there are more than one of you, you all have the same first name) who filled out the Cambridge UK meetup form without giving an email address: the meetup is definitely going ahead at the proposed date and time (Monday 24th April, 6pm). The venue (CB2 bistro) may change if someone knows of a bigger one, as so many people are interested (17 yeses, 10 maybes). If it does, I’ll post the new details in the most recent open thread and the list of meetups here. Otherwise, I’ll hopefully see you in the upstairs of CB2 (I’ll have an SSC sign).

  16. Deiseach says:

    Okay, I am in dire need of advice.

    Is this serious? Is it some clever Russian company pulling a hoax? Is this a leg-pull? Or does Russia really have FEDOR robots for space exploration?

    I mean, the first thing that came to my mind was it looks more like a Russian games developer putting out some kind of publicity material about the content in their upcoming game than real Russian army/space programme robots, but in at least some media it’s being treated as “Russian robots taught to shoot”. Though I have to admit, the claimed excuse is hilarious: “teaching our robots to shoot helps develop their motor functions and decision-making abilities!” Gun nuts can now use this one for teaching their three year olds to shoot an M16 🙂

    Can somebody clear this up for me? The real world is currently in such a state I no longer have any metric for “this is too outrageous and obviously fake to be true”.

    • 1soru1 says:

      Institutions and companies involved are legit. No reason why Russian research institutes would be behind everyone else in this area. Just their source of funding is different, so they hype differently.

      An AI that can drive 99.9% safely can shoot with 99.9% accuracy. Which seems like it would put a soldier out of work well before it did an Uber driver.

      Legal and technical challenges apply in both cases, and maybe the whole area will flop for a decade or two. But you are not likely to get one without the other.

      • Iain says:

        An AI that can drive 99.9% safely can shoot with 99.9% accuracy.

        What is your basis for this claim? This seems akin to claiming that an AI that can beat a human at chess should be able to beat a human at go. There may or may not be a general factor of intelligence for humans, but no such thing exists for our current purpose-built AIs.

        • 1soru1 says:

          To be clear, I didn’t mean literally the same AI, I meant a level of AI technology; a given measured computational power and (more subjectively) ability to harness it.

          There may exist a military that is more worried about unintended civilian casualties than a taxi firm; the Russian armed forces don’t give the impression they consider that a goal.

          • Iain says:

            Okay, but the same question stands: what is your basis for claiming an equivalence between driving and shooting?

            There is no necessary connection between our “ability to harness” AI for two different tasks. To the best of my knowledge, the work that IBM did for Deep Blue was mostly irrelevant to the work Google did for self-driving cars. It’s possible that the vision components of Google’s cars might be useful for a military robot; on the other hand, this description claims that they are only accurate out to about 30 meters.

            The concept of a single level of AI technology is (at best) misleading. We have a grab bag of techniques, each with its own strengths and weaknesses, which have had various amounts of money poured into them to be tuned for specific tasks. Big projects like self-driving cars require a tremendous amount of engineering effort, and an absolutely massive pile of training data, neither of which is directly transferable to other problems.

            Modern AIs are domain-specific. There are a few areas — increased computational power, better algorithms for neural networks — where progress is likely to apply across the board, but you still need a lot of work to make an AI for any particular task, and the difficulty of any given task is not always intuitive to humans. Self-driving cars in no way imply robot soldiers.

          • Protagoras says:

            @Iain, Getting a computer to hit a correctly identified target is largely a solved problem; they’ve been doing that with ever increasing success since the fire control systems for battleships Bean talks about. The problem is identifying a target. If a car can tell the difference between what it is absolutely not under any circumstances to collide with and what it might not be as bad to collide with (as required for any adequately safe driving), that is actually somewhat similar to getting the gun to at least figure out what it can’t shoot at. So there seems to be some reason to think we are at least talking about the same ballpark of difficulty. I don’t think 1soru1 was necessarily claiming that the same computer can do both jobs, just that if somebody has built an AI that can do one of those things, somebody else building an AI that can do the other likely either has already happened or is not far off.

          • John Schilling says:

            If a car can tell the difference between what it is absolutely not under any circumstances to collide with and what it might not be as bad to collide with (as required for any adequately safe driving),

            Is it really?

            Adequately safe driving is, in my experience at least, best achieved by not ever colliding with any solid object of non-trivial size. And this is not an unachievable goal. I don’t know if you drive, but if you do and your recipe for “adequate safety” involves actually colliding with less-bad objects based on your skillfully-created value function, I’m pretty sure you’re doing it wrong.

            If every otherwise-adequately-safe driver who nonetheless found themselves facing an inevitable collision, decided at random which object to collide with (or more plausibly, always chose to collide with the smallest candidate object), they would still be adequately safe. The circumstances where this distinction might hypothetically matter, are really quite rare for safe drivers.

          • Matt M says:

            When it comes to stuff like self driving cars, people ARE going to freak out about the worse-case scenarios. Like yeah, 99.9% of drivers are never going to be faced with the “do I run over this baby or swerve away into this heavy metal pole” scenario.

            But it’ll only take ONE driverless car plowing over a baby for huge controversy to result in all self-driving cars being banned by Congress, and Google being sued for $500M for killing a baby…

          • Corey says:

            @Matt M: Maybe. Currently, even though on most things we as a society tend to be overcautious relative to a rational assessment of the risks, when it comes to cars/driving we tend to be UNDERcautious. So I’m optimistic that even in these emotionally charged scenarios, we’ll land at a roughly reasonable risk-avoidance tradeoff.

            In other words, babies get ran over occasionally already, and we haven’t mandated cow-catchers, fences between schools and parking lots, etc. (TBF we have mandated backup cameras (or proximity alerts) starting in a year or so, partially for this reason).

          • Matt M says:

            So I’m optimistic that even in these emotionally charged scenarios, we’ll land at a roughly reasonable risk-avoidance tradeoff.

            In other words, babies get ran over occasionally already, and we haven’t mandated cow-catchers, fences between schools and parking lots, etc.

            We generally have more trust in, and are much more sympathetic towards errors in individual humans than we are in machines (or the corporations who will construct/program them).

            If a driver swerves out of the way of an oncoming truck and runs over a pedestrian, our initial thought is probably empathy-driven. “How horrible. I’m glad that didn’t happen to me. I likely would have made the same decision. No need to punish him excessively.”

            If Google programs its car to do that, there is no empathy with the car. The response would be “How horrible. Why didn’t Google program it’s car better? Google must pay for this!”

          • Iain says:

            Getting a computer to hit a correctly identified target is largely a solved problem; they’ve been doing that with ever increasing success since the fire control systems for battleships Bean talks about. The problem is identifying a target.

            This is akin to saying that we’ve perfected a mechanism for robots to move physical chess pieces, so all that remains to be done is teach the robot to play chess. It may be true, but it is conveniently omitting the hard part of the problem.

            Again: Google’s self-driving cars are a tremendous engineering feat. But driving is a domain in which everything you care about is relatively close and non-adversarial. Nobody is deliberately trying to confuse Google’s algorithms with weird paint jobs. (Well, almost nobody.) The simplified universe that self-driving cars have to model is very different from the simplifications necessary for self-aiming weapons, and it is not safe to just assume that techniques that are effective in one case will work equally well in the other.

            (If you just want a system that will shoot anything that moves, you can probably do that today — but that’s not even in the vicinity of putting soldiers out of work, as 1soru1 claimed.)

      • AnonYEmous says:

        agreeing with Iain

        even if a driving Ai can perfectly calculate trajectories and adjust this ability to bullet shooting, how will they know what to shoot? And what happens when people start messing around with the recognition system with different types of camouflage?

      • Nornagest says:

        Yeah, I agree with the others here. The actual shooting is an easier problem than driving (a humanoid platform makes it harder, but there’s no reason a production Terminator has to be humanoid), but target recognition is much, much harder. Negotiating roadways and recognizing obstacles is already a tricky problem for AI; differentiating enemy from friendly soldiers or from civilians, in a wide variety of chaotic environments, while the enemy’s trying not to be seen, is a whole ‘nother issue.

        You probably could put semi-autonomous tank-like weapons on the ground in the next ten years or so, though, capable of autonomous maneuvers but with humans in the loop to fire weapons. But that doesn’t offer you much advantage over a regular tank crew.

        • John Schilling says:

          differentiating enemy from friendly soldiers or from civilians,

          Or any of the above from, e.g., vaguely human-shaped bushes. Never mind murals in urban environments, and let’s not even talk about the decoys people will start making if autonomous combat robots enter service.

          ED-209: “Those humans have freakin’ Terminators, man! I emptied two hundred rounds of 7.62mm into one of them, dead center, and it didn’t go down! It just stood there, laughing at me! Game over, man! Game Over!”

          Murphy: “It was a scarecrow, Ed. A literal scarecrow.”

    • Matt M says:

      Even if true, how exactly is this different than a aerial drone? (Other than being sufficiently anthropomorphized that it looks more scary?)

      • cassander says:

        Technically, it’s a lot harder to pull off than a drone, and drones, at least in the US, don’t have autonomous weapons release authority, that’s always done by a person. the only place the US lets robots shoot things is at sea and even then I bet it’s turned off most of the time.

        • John Schilling says:

          It’s not just weapons release that requires human intervention for US drones; the actual targeting is done by a human hand on a joystick putting the crosshairs on the thing to be destroyed. As you note, air defense is about the only place where robots can generally handle that, on account of even robots can recognize the difference between “air” and “solid object” and we can define volumes of airspace in which we’d prefer there be not solid objects bigger or faster than birds.

          Similarly, the self-driving car analogy doesn’t really hold up because a self-driving car does its job if it refrains from crashing into all solid objects categorically; it doesn’t have to distinguish between “avoided crashing into mailbox”, “avoided crashing into small child”, and “avoided crashing into fleeing criminal”. FEDOR, or ED-209, is going to be counted a failure even if it “only” wastes all its ammunition shooting at mailboxes.

          • ThaadCastle says:

            I don’t think that is 100% fair to the self-driving car designers.

            I have a friend who works at Google Car (AKA Waymo) and he says a big issue was teaching the cars the difference between a possum/raccoon (try to avoid but don’t crash into other cars or something to avoid) and a crawling child (avoid at all costs, including crashing into mailboxes/other cars)…Although I guess the best of all possible worlds is not to crash into anything.

          • John Schilling says:

            Certainly Google is going to try, but I’m pretty certain any successful self-driving car is going to be better than 99% “always drive slowly and cautiously enough that you will never need to decide which solid object you crash into”. Disappointing though that will be to all the philosophers who thought their Trolley Problems might suddenly find real-world application.

            The Terminator, can’t help dealing with the fact that it’s got to shoot Sarah Connor and it doesn’t have enough ammunition to shoot every vaguely bipedal solid object in Los Angeles. That’s maybe like designing a self-driving car for an owner who wants to maximize roadkilled raccoons without ever hitting a crawling child or a fender-damaging mailbox.

      • Deiseach says:

        Well, this is where I think the real risk of AI threat comes in; not that one day the sufficiently advanced AI will wake up and decide to turn us all into paperclips or computronium, but that we’ve been training its predecessors all along so that its base programming is “What is the best way to solve this problem? WITH HOT LEAD!!!!!!” 🙂

        • Matt M says:

          “What is the best way to solve this problem? WITH HOT LEAD!!!!!!”

          Well do you want an AI that “thinks like a human” or don’t you?

    • FacelessCraven says:

      I’m something of a weapons enthusiast, and I’ve noticed a really weird trend online of what I guess I’d describe as Russian military hype videos. These seem really strange to me; the production values are high enough that it doesn’t look like amateur work, but the sims seem so obviously fanciful and jingoistic that it doesn’t look much like real defense contracter material either. My assumption is that it’s propaganda aimed at a domestic audience, much like the vid you linked is. It’s probably there to make joe blow on the street think “wow, my government has such amazing weapons!”

    • Nornagest says:

      The drone’s probably legit, but I don’t think it was carrying those charges. But I bet you anything the Terminator is an early technology demo, not a combat-ready robot. It’s moving really stiffly, which tells me it hasn’t solved any of the problems with e.g. balance that actually make this sort of thing hard. Contrast Boston Dynamics’ robots, which move in a very fluid, almost creepily organic-looking way. And those are quadrupeds; a bipedal platform would be harder.

  17. onyomi says:

    I didn’t have the patience or fortitude to watch the whole film, but from what I gather, the story goes: “guy has a bunch of digestive troubles; guy attempts to dramatically change his microbiome by taking a bunch of antibiotics and then doing a poop transplant; guy claims big improvements in digestion, as well as a change in his food preferences (developing a sweet tooth)”

    Is using too many antibiotics (or other things which cause our gut flora to be unbalanced somehow, such as low vitamin D or changes in diet) what is making us all fat and/or causing the rise in autoimmune problems, especially of an “inflammatory bowel” sort?

    A kind of “hygiene hypothesis,” but more focused on the end of sterilizing our guts through drugs, as opposed to, you know, not letting kids roll in the mud enough?

    • Corey says:

      Poop transplants from thin to fat mice make them thin, and vice versa, with the same food availability, so there probably is something there.

      Cracked (who else?) has an article with the details of getting a transplant, from a woman who needed one to get rid of a persistent and life-threatening C.Diff infection.

      • onyomi says:

        Why did everyone’s poop bacteria turn into fat person poop bacteria recently?

        • The Nybbler says:

          Chemtrails? Fluoride in the water? Micronutrient fortification of foods? A strain of cold virus which has become endemic and tips the balance? HFCS?

          • onyomi says:

            What about antibiotics? Have there been any studies on frequency/kind of antibiotic use and trends in obesity?

        • Nornagest says:

          Some wild speculation: fat-people bacteria are more resistant to antibiotics? There’s no specific strain of fat-people bacteria and what actually matters is having a robust microecology, which is harder to get now that we’re cleaner and move around less? There’s no real difference; fat-people bacteria are just what happened to colonize (har) the food supply or some other common vector?

          • onyomi says:

            If fat people bacteria do tend to be more resistant to the types of antibiotics we use then that alone would do it: every time you use antibiotics you kill of some of the thin-person bacteria, giving the fat-person bacteria room to grow (by killing off its competitors).

        • Corey says:

          There’s feedback with the foods you eat also, so my guess: other factors caused people to eat small amounts of fat-bacteria-producing food, increasing the fat bacteria, making them crave more such food, and so on.

          • onyomi says:

            This seems likely, especially if it’s true that a poop transplant can change a person’s taste. That is, in a world where parasites which need cat intestines to reproduce can make mice less afraid of cats, it seems likely that the bacteria which live off certain foods might be able to release chemicals which cause us to crave those foods, and would probably do so in proportion to their numbers, which would be, at least to some extent, a result of our past history of eating those same foods.

            It might, however, contradict a bit of diet wisdom which has always seemed “off” to me, namely that losing weight gradually is more likely to result in keeping it off than a “crash diet.”

            Example, Penn Jillette lost a good deal of weight by eating nothing but potatoes for two weeks, and then kept the weight off on a low-fat vegan diet. He reports his tastes had changed–namely that things like corn without butter, which had previously tasted bland, now taste good.

            This was probably, to some extent, a result of eating super bland food for two weeks with a possible resetting of one’s baseline “flavor expectation” (if you’re used to eating butter and salt on everything, you can become “addicted” in the sense that things will taste bland without butter and salt which might not taste so to one without such a habit).

            But conceivably, eating nothing but potatoes for two weeks–a kind of “crash diet” resulted in the bacteria which live off potatoes, say, making inroads while the bacteria which live off, say, fat, lost ground (probably vastly oversimplifying how bacteria really work), with the result of breaking the “crave fattening foods” cycle.

            Of course, if this is true, one might also possibly be able to change the diet and microbiome more gradually, but that might be more psychologically difficult, at least for some. And I’ve seen at least one study to back this up (at least the idea that rapid weight loss is not associated with greater difficulty “keeping it off,” and that it may even be the reverse).

          • MNH says:

            @onyomi, it’s my impression that people fail to keep weight off after crash dieting because it involves two difficult changes in eating habits when they only expect one. Namely, people are ready for the shift to the crash diet, and it’s easy to expect that that’ll make you hungry so everyone knows that they’ll have to fight against their hunger and prepares. But then once crash dieting is over, there’s another hard transition, where you have to switch from it back to a normal, sustainable diet, but with a much lower calorie budget than was available pre-crash. And people don’t think ahead about this part (expecting to declare victory when they hit their target weight), they might have no experience eating sustainably at this calorie level (certainly no recent experience), and they just spent a long time resisting hunger and are now allowed to succumb a little, which it’s easy to run off the rails doing.

            Contrast all this with slow dieting, which is slower, but dodges the second transition and lets people build their post-diet eating habits gradually throughout the diet (naturally tightening margins are a very convenient structure for this).

            Source: I have no hard evidence for this, but have made intentional changes to my weight very, very often as a weightlifter with a mild eating disorder.

  18. Zodiac says:

    Question of fairly new commenter: What topics are considered too controversial for here?
    Also is the “culture war” we are to avoid the same as this one (the US version)?

    • 1soru1 says:

      I’ve not seen any discussion of this culture war, so I assume that’s the one everyone avoids.

      • Zodiac says:

        So we may not discuss the most cost-efficient method to create supernovae to eradicate the heathens? Never thought Scott was such party pooper.

      • Anatoly says:

        You can discuss it in broad terms, but please take care not to spoil the outcome for the locals – remember they only get their news by the speed of light. Heed this warning for the next 3228 years, and you’re fine.

      • Nornagest says:

        Consider Phlebas was depressing as hell and frequently about as disgusting, but I would have liked to see more of the Culture from that perspective.

    • Aapje says:

      @Zodiac

      Yes.

      You are also not allowed to discuss whether opera is better than cinema, whether hip hop is better than classical music, etc. 😛

      • Deiseach says:

        And no koumiss versus yoghurt threads, either 🙂

        • Zodiac says:

          Next you’re telling me I’m not allowed to declare basmatis supremacy over jasmin rice.

          • Eltargrim says:

            Heathen!

          • dndnrsn says:

            No rice or ginger in the open threads.

          • Protagoras says:

            What, I can’t talk about how attractive I find exotic mixed rices? Censorship!

          • quanta413 says:

            Modern rice-progressive nonsense. One must choose the proper rice to match the meal according to centuries of tradition that embody the wisdom of generations of humans embedded in their local cultures. Ideally, of course one is far more specific than just basmati. Of course, some leeway is in order when obtaining the proper strain of rice is very difficult.

            You can’t just choose your rice in a vacuum. You don’t make sushi with Basmati and you shouldn’t put Chana Masala on top of Arborio.

          • dndnrsn says:

            The division of rice into different kinds of rice is completely ascientific. Next you’ll be posting pictures of peasant maidens standing in rice fields.

          • Randy M says:

            Besides, rice is a social construct.

          • MNH says:

            @dndrnrsn, fantastic comment

          • Cerebral Paul Z. says:

            Chex your privilege.

          • Aapje says:

            Just read the Basmati Curve, white rice is better than brown rice, although Asian rice is even better.

          • dndnrsn says:

            We’ve been over it before: wheat nationalism is bad.

          • Protagoras says:

            I can’t agree about white rice being superior to brown rice. I originally started eating brown rice because I had a girlfriend who preferred it for nutrition reasons, but she’s long gone and I still buy only brown rice for myself these days because I find I like the texture better.

          • Aapje says:

            I agree, but the truth wouldn’t work for my bad joke.

          • Iain says:

            PSA: I moved recently, and found a bag of brown rice that I had forgotten about at the back of a cupboard. Turns out that, while white rice will last nigh-on forever, brown rice only has a shelf life of about six months. No matter how long I cooked it, my jambalaya never stopped being crunchy. In five years of feeding my girlfriend, it’s the only time we’ve ever thrown up our hands and gone out to a restaurant instead.

            Moral of the story: if your brown rice has an expiry date, believe it.

    • Brad says:

      It depends on what you mean by too controversial. Do you mean you will be banned if you discuss them or that you will get more heat than light? For the former the list of bans is a good place to get an idea of the de facto moderation policy (https://slatestarcodex.com/comments/). There’s some words that are banned by the software (if you try to use them your post will not post) but working around the filter is not considered an automatically bannable offense. For the latter you are just going to have to read and make your own judgments.

      As for the parameters of culture war free threads, it has been the subject of some back and forth in recent ones. Some try to police away any threads that involve any issue with any saliency in contemporary American politics. Myself, I think it should be limited to the specifically cultural hot button issues and that discussing e.g. the minimum wage should be fine. AFAIK there’s be no official statement on the subject.

      • Zodiac says:

        Huh, I would never have assumed there was a word filter here.
        I am mostly just trying to not get on peoples nerves. I think I understand now what to avoid and what is fair game.

    • Jaskologist says:

      Basically, don’t talk about US politics.

      • Anonymous says:

        That leaves only Chinese and Russian internal politics. In the realm of politics, anyway.

  19. Tibor says:

    So Turkey had a referendum yesterday about transitioning to a presidential system (where the president can even intervene in the judiciary branch). This gives Erdogan a lot of extra power and potentially another 10 years in the office (the tenure of the president is 5 years with one potential re-election).

    The outcome is not that surprising, what is a bit surprising to me is how close it was despite the Russian-like manipulation and persecution of the opposition – Erdogan won with slightly above 51%.

    What is perhaps a bit disturbing is that Turks living in Germany or Austria voted overwhelmingly (about 70%) for Erdogan’s constitutional changes (whereas American Turks voted with about 85% against the change). In Turkey, the three largest cities (Istanbul has a population of about 15 million, which means that about 20% of Turks live in Istanbul) were against expanding the presidential powers. I think the stark difference between German Turks and American (or even Turkish Turks) should be examined a bit. On one hand, it is sort of clear what’s going on – American Turks are probably predominantly from an educated middle class, whereas German Turks are overwhelmingly working class and with low education. And in Turkey there is obviously no bias in the population. But perhaps one should think about the immigration policy if what your country attracts is people who are in a sense “less western” than an average Turk in Turkey. I think the reason is that Germany offers quite a generous welfare state. This is not very attractive to Turks with good job prospects and opportunities who don’t want to live off welfare, but it is very attractive to those whose alternative is living in a village in Anatolia. And of course the “initial” Turkish population in Germany comes from the “Gastarbeiter” (guest workers) and their families who were invited in the 70s to do mostly manual labour the Germans didn’t want to do any more. A part of the problem might be that until the 90s or so, the idea was that these people eventually go back to Turkey and the policy was not taking into consideration that they won’t…which probably contributes to their rather bad integration in the society on average.

    • Aapje says:

      You have to keep in mind that:
      – Many of the Turkish mosques in the West are controlled by Diyanet (a Turkish state institution).
      – Turkish people living in the West get all of the propaganda, but see far less of the actual day to day problems that undermine the propaganda.
      – It’s an explicit policy of Turkey to keep control over people of Turkish descent.

      • pedrodegiovanni says:

        That doesn’t explain why american turks were against the referendum.

        • Aapje says:

          American Turks appear to have come mostly between 1820 and 1920, while Dutch Turks came between 1960 and now (and probably the same for Germany). So Dutch Turks have 40-140 fewer years of integration. Integration was probably easier back then, since there wasn’t satellite TV to keep people connected to the culture of their country of origin.

          Also, the republic of Turkey was established in 1920, so the Americans Turks presumably weren’t affected by the nationalist propaganda.

          Apparently, most of the American Turks actually came from the Balkans, so they may have had less connection to Turkey in the first place.

          AFAIK, most American Turks are now highly assimilated and the ones who aren’t, are a fairly small number of relatively recent migrants. These have a very different make-up compared to the European migrants, as it is mostly the highly educated that went to the US in the last decades.

          • Tibor says:

            The Turks who came to the US in the 19th century won’t have a voting right in Turkey today. Those who do must have been recent immigrants just like in Germany or the Netherlands. Also, if anything, Erdogan represents a break with the legacy of Atatürk and the republican Turkey.

          • Matt M says:

            Tibor,

            Going back to your original theory vis-a-vis the welfare state, I would imagine that America attracts primarily Turkish immigrants who are at the highest level of accomplishment. Those who have good job prospects in Turkey, but can do even better.

            Then consider that America is known as a ruthless capitalist society wherein people with talent and ability get rich, but the safety net is minimal. Sounds awful for those with low prospects or even middle class prospects. But for the truly elite, talented, and confident, the ceiling is much higher.

          • Tibor says:

            @Matt M: Perhaps. Does Switzerland also have this reputation? The results from Switzerland were more or less the same as those from the US (whereas pretty much all of the rest of Europe where there is a significant Turkish minority were strongly supportive of Erdogan -significantly more so than the Turks in Turkey themselves).

          • Aapje says:

            @Matt M

            It’s even worse, because the Turkish and Moroccan guest workers were specifically selected for being poorly educated, so they wouldn’t compete for the better jobs with Dutch natives.

            So it’s not necessarily the US that did much better, it was these European countries intentionally picking the least educated they could find (thinking that they would go back anyway – oops).

          • Matt M says:

            Not sure about Turkish perception of Switzerland, but IIRC it DOES regularly outperform the US on the Index of Economic Freedom. Also I think they generally have stricter immigration standards, do they not? (Recall the story about a local community being able to refuse someone permission to obtain citizenship on the grounds that they are “not sufficiently Swiss”)

          • Tibor says:

            @Aapje, Matt: Of course, I am not saying the US starts with the same people and just does the integration better. But my point is that a lot of the “undesirable” hard to integrate people don’t even come to your country if they have no chance to get access to social welfare (or only after a long period of time). And also the people who do come have more motivation to integrate themselves. For example, if they don’t speak the language, they won’t find a job or only a very bad one and since they cannot fall back to welfare, they have no other options.

            @Matt: Yes, Switzerland is economically freer than the US, but it probably has a more expansive welfare state than the US. Getting Swiss citizenship is extremely difficult, coming to Switzerland and getting a working visa is easier for highly qualified people or Schengen area nationals (that means almost all of EU except for Romania, Bulgaria and Croatia plus also non-EU countries Iceland and Norway).

            So yes, one way is basically filtering out poor people. But I don’t think it is necessary. I think that the problem are not poor people per se, but people living off welfare. If you filter out poor people you filter out people who are likely to live off welfare but if you cut the welfare state instead, you let the capable poor people come in, which I think is a better solution. The conservative immigration policy of Switzerland is stable, but I think it is also suboptimal.

          • Aapje says:

            @Tibor

            I think that welfare pull is greatly overrated. These people came to work hard and they did…until the economic crisis and a lot of industry started automating or moving to Japan/China/etc.

            They didn’t become unwilling to work after that, their labor became less and less needed.

            Basically, these were people with low education who were used to fill a worker gap that was very temporary…but the people turned out not to be temporary.

          • Tibor says:

            @Aapje: You could be right and I think you are very likely right with the Gastarbeiter in particular. However, welfare is what allows those people to stay without adapting to the new circumstances once the temporary work disappears. It might be due to cultural differences but there’s a parallel with Vietnamese in the Czech republic – they also came during the 80s and for similar reasons the Turks did to Germany (except that this was organized by the communist governments of the two countries). After the velvet revolution the jobs they came to do ceased to exist. But since the Vietnamese weren’t (as far as I know) eligible for social welfare (and it was not and still quite is not on the German level) they had to either come back to Vietnam (which was and still is communist, so not a very attractive prospect) or find a way to get by without it. And they seem to have integrated themselves (I specifically say that since I think the Czechs really did not do very much to help them) extremely well. It could be due to a different cultural make up, or due to a different socioecomical one (I guess the average Vietnamese Czech immigrant had a higher education than an average Turkish Gastarbeiter). Also, the circumstances were still a bit different. But I think it is also at least partly because of necessity.

            Welfare simply makes it easy for you to stay once you are “not needed”. In its absence you either have to go back home or come up with something new to do. Either way, it is much harder for parallel societies and what are sometimes essentially “ghettos” to form. Since you’d end up with immigrants who are all essentially capable people, you’d also see a lot less prejudice and xenophobia towards them from the native population. After all, nobody could say that they live off their money. The most Czechs say about the Vietnamese is that they avoid paying taxes (as if everybody doesn’t try to do that if there’s an opportunity). But nobody can say that they are lazy or that they are criminals, because the evidence is so overwhelmingly against that. And again, it might be due to some innately good things about the Vietnamese culture, but it could also be because only the capable people come and stay.

            At the same time, you do raise a fair point. Under the Czech law you get an access to the welfare payments and municipal voting rights after 5 years of having had a stable income and a permanent residency in the country, so I guess that if a technological change makes a large number of jobs suddenly obsolete and there’s been a large minority that has been in the country for a few years, this might still create problems. On the other hand maybe it’s a one time thing? I mean, the most simple labour has already largely been replaced by machines in Europe and so only qualified immigrants will find a job in the first place. since those people will have a higher education and will on average be smarter, there are going to be fewer problems with them if their jobs are replaced sometimes in the future. They will be more likely to adapt to the new conditions as well as integrate themselves into the society before that happens. But yeah, you raise a good point and it is at least not obviously true that cutting welfare for recent immigrants solves all your problems.

          • Aapje says:

            @Tibor

            Culture definitely plays a large role as well. The Netherlands also had substantial numbers of migrants from poor parts of Italy between 1949 and 1975, who worked in the mines and such. There even was a riot in one place (newspaper article in Dutch, about conflicts over girls and people yelling ‘macaroni’ and ‘spaghetti’ to Italians).

            However, they either went back or integrated fairly quickly (probably because they were OK with marrying Dutch women, which most of them did).

      • Tibor says:

        That is a plausible explanation only as long as they only do that in Europe. I think it is definitely true that they try to influence the German Turkish population more than any other outside of Turkey. I would not say they get all the propaganda though, it’s not like they only go to the mosques and the German media are not exactly pro-Erdogan. Also, they probably visit relatives in Turkey sometimes, so they do get someone​ feedback.

        The thing is that from the perspective of someone from a poor region in Eastern Turkey, Erdogan has really helped the country a lot. Not so much for some from Constantinople.

    • Paul Brinkley says:

      How likely is it that that German Turkish support is being fabricated by pro-Erdogan elements? Stuffing the ballot box? The way I’m imagining it: it’s easy to fake, and hard for anti-Erdogan elements to confirm, since such records are outside the country; they’d fabricate American Turkish records, too, but American Turks are likely enough to call this out that they don’t bother trying.

      OTOH, I’d honestly be less surprised if these figures are close enough to legit. I know next to nothing about Turkish politics. About the only thing making this believable to me is all the things I’ve heard Erdogan doing.

      • Matt M says:

        I dunno, I feel like if you wanted to keep this covert, you’d be more likely to fabricate it such that like, 60% of US and 60% of German Turks vote for Erdogan, rather than leave the US alone at 40% and make it 80% for Germany or whatever.

      • Tibor says:

        Why do you think US Turks are more likely to call this out? Btw, Swiss Turks were about 60% against the change of government system, Czech (very few in total and only about 550 people voted) ones were 85% against, whereas French or Dutch (I am less sure about the Dutch Turks) were about on the same levels with German Turks. Of course, Erdogan was mostly focusing on Germany and the Netherlands. But I don’t think it is that easy for him to do this. I am not sure how the counting works, whether it is done by Turkish citizens in Germany, or sent to Turkey first. I know that Czechs voting from abroad have to go to the embassy (I think), but there are no significant Czech minorities anywhere, so maybe the Turks have it a bit more developed in Germany.

        There seems to have been some shenanigans with the unstamped envelopes…and of course extremely one-sided state funded propaganda in which the opponents of the presidential system were likened to terrorists, many journalists were imprisoned (including a correspondent of “Die Welt”, who is still in prison despite German protests) and so on. That Erdogan won so closely given all this is actually rather surprising, at least to me.

        • AnonYEmous says:

          given that the results were replicated in france i’m not sure, but could this have something to do with german – turkish politics? Maybe a strong Erdogan can then pressure Germany to treat the German Turks better, or something. Which is pretty uncharitable to German Turks in its own way but provides a strong motive.

        • Paul Brinkley says:

          Why do you think US Turks are more likely to call this out?

          Shrug. Just a feeling that US Turks would have marginally better access to broadcast media.

    • Zodiac says:

      Somebody made the effort and got some further numbers for German Turks:
      Number of Turks that can vote in Germany: 1.430.000
      Turks that voted: 530.000
      Of those voted yes: 63%, in numbers 333.900
      Number of Turks in Germany overall (including all that can’t vote): 2.998.000
      Source (sadly in German)

      That doesn’t seem as bad as many make it out to be, especially considering that there seems to be evidence that potential no voters had their passes confiscated.

      • Tibor says:

        If their passports were confiscated, how would you tell who is pro or against Erdogan?

        I guess then that the main result from Germany is that most Turks there don’t care about Turkey very much (or they’ve given up hope), which is sort of understandable.

    • Murphy says:

      Migrant populations can be weird.

      At a time when groups like the IRA were becoming increasingly unpopular in Ireland itself they were still receiving support and funding from migrant populations in America and elsewhere.

      Kids growing up in american-irish communities would hear a view on the situation handed down from their parents and grandparents. Similarly you’ll sometimes come across people who speak a downright archaic dialect of Irish in American communities for pretty much the same reason.

      Sometimes migrant communities can become a sort of snapshot of social views when the bulk of them moved. The mainstream moves on while they remain frozen.

      • Tibor says:

        I am not sure about the populace, but Turkey in the 70s or 80s was probably more secular than it is today. On the other hand, these people might remember the military regime and while military was keeping the state secular whenever it looked like it could cease to be, they were hardly angels. Which makes one wonder whether Atatürk wasn’t too fast on the westernization. Sure, he managed it in the cities, but the entire post-Ottoman history of Turkey is one putsch after another to keep the country the way Atatürk envisioned…until Erdogan. Also, Atatürk was hardly a democrat himself, at best he was something like an enlightened dictator (which is not so strange though given the time period, just a few decades before him, most of Europe was ruled by absolutist monarchs). Also, there is little one can say against enlightened dictators provided that they really are and stay enlightened 🙂

  20. Aapje says:

    In my earlier post about Dutch education, I explained how it has early tracking, around age 12, albeit usually with a bridge class afterwards for the first year of secundary education that allows changing levels before too much is invested.

    Early tracking runs a greater risk of misplacement, but also allows for teaching to be more at the level of the student. The former problem can be solved in various ways. One way is to let students drop down or go up a level. The latter is more difficult to achieve than the former, since students will then need to catch up. Also, in the Dutch system, students at the highest level of secondary education take one more course than at the second highest. So they then need to take on an additional course and catch up if necessary. Going from VMBO to HAVO is even harder, since they don’t match up so well. So very few students go up a level and more, but still few, go down a level (the Inspectorate is bullish on schools over this, which has resulted in complaints that schools are being too conservative in their admissions).

    Traditionally, the Dutch education system has allowed for students to move up a level after getting a diploma. See this picture for example. At the bottom there is the primary education in yellow. Above it in green, the 4 branches of VMBO and next to it, HAVO and VWO. Then above those, you have higher education in blue and purple. People normally go from VMBO to MBO, from HAVO to HBO (basic college) and from VWO to HBO or the university (top right). The arrows show various paths, including the possibility to move up a level.

    At the secundary education level, 8% of students go up a level after getting a diploma, while at the MBO level, 40% do so. Some students do this multiple times, called ‘stapelen,’ which means piling up (as in: piling up diplomas). A big advantage of taking this route is that it works like a ‘save point’ in a game. If a student fails at a higher level, they still have a diploma at a lower level to fall back on. Furthermore, it seems to work very well for students with motivation problems, who have difficulty with Dutch or such. If they can solve their problems during their education, they can benefit by moving up. Taking such a route does take more time than the direct route and thus is also more costly for the government.

    The austerity measures of the last decades and fairly recent attempts to really increase educational efficiency have hampered the ability for students to do this, however. For example, there is ‘cascade funding’ where schools get less money if a student takes more time to get a diploma. So students who do MBO level 2, then 3, then 4 stay at the same school for 6-7 years (a single school tends to offer all levels). The school then gets less funding for the latter years and no funding for the 7th year. So a measure that is intended to encourage schools to get their students to a diploma quickly, can result in them getting a lesser education, quickly, rather than a better education, slowly.

    Something similar is true for the move from VMBO to HAVO, where relatively many students fail after taking that step (the former doesn’t prepare very well for the latter). This hurts the metrics of the HAVO school. A decade or so back, the Inspectorate starting publishing the metrics, so parents can see how schools rate. However, this also means that schools have started really optimizing these metrics. So they are making it hard to go from VMBO to HAVO.

    Another reason why the long route has become more difficult is because students get less funding today (just a few years back, we went to the American system with big college loans). So such a long route is much more costly today than in the past.

    Most likely/hopefully, the pendulum will swing back soon, as a lot of people are getting upset over these problems.

    • Aapje says:

      Note that there is an inherent conflict between punishing students for being lazy and/or schools for not educating students efficiently & enabling students to maximize their talents. The more bullish you are on efficiency, the more you encourage students and schools to take a safe route, with no risks.

      There is also an inherent conflict with designing education around the needs of certain groups of students and allowing people to switch between different types of education. If one curriculum is designed more around practical skills and another more around theoretical skills, it’s going to be hard for students to move up.

      • gbdub says:

        Is the “stapelen” a relatively recent phenomenon? I’m wondering if it maps to the American phenomenon where credentials are worth less (yesterday’s bachelors degeree is today’s masters), or the general “no jobs right now, guess I’ll get a graduate degree”.

        From your description it sounds like there are a lot of people trying to move up a level. Ideally I suppose that would mean that everyone ends up in the same place, with some just taking longer. But if a lot are failing, does that mean they shouldn’t have tried, or that the system is poorly designed for how people are using it? My gut sense is that in a well designed tracking system, the top 10% or maybe quartile of one level would do about as well, and have equally good outcomes, as the bottom group of the level above (since they are presumably pretty close in overall ability). How close does this system come to that?

        The chart is helpful – do you have any idea what percentages take each path?

        • Aapje says:

          Is the “stapelen” a relatively recent phenomenon?

          I’ve found one source that says that it was common in the 1970’s and 80’s, but was discouraged from the 90’s and on. Right now, a lot of people are arguing that it should be encouraged.

          But if a lot are failing, does that mean they shouldn’t have tried, or that the system is poorly designed for how people are using it?

          I would argue that if they couldn’t predict that they would fail, the system was poorly designed 🙂

          I think that the system is too much focused right now on a clean looking system, disregarding the actual needs of a substantial group. Ideally, there would be incentives to helping people move up (and perhaps down) more easily. For example, one could pay a school for the performance of the student in future schools (with more pay the better the kids do). Then they would have an incentive to prepare the student well, rather than just give them a diploma ASAP. That might also help against diploma inflation.

          The chart is helpful – do you have any idea what percentages take each path?

          VMBO theoretical/mixed -> HAVO was 18% in 2010 and 13% in 2013, with the rest going mostly to MBO level 3 and 4.

          HAVO -> VWO was a few percent in 2007, with the rest going to HBO (78%) or taking a gap year (~14%).

          VWO -> University was 71% in 2007, with the rest going to HBO (13%) or taking a gap year.

          MBO level 1 or 2 to a higher level of MBO was 50% in 2007. For students with a migration background this was 60%.

          MBO level 3 to MBO level 4 was 33% in 2007. For students with a migration background this was 50%.

          MBO 4 -> HBO was 40% in 2007. For students with a migration background this was over 50%*.

          HBO bachelor -> University was 7% in 2015.

          * These higher numbers suggest that students with a migration background greatly benefit from having these routes.

    • Edward Scizorhands says:

      Why does everything need to happen in one large block?

      Can’t we say that a student is at math level 5.4 and at reading level 4.0 and assign her to math and reading classes with others at that approximate level? And then after the semester, the teacher notes what their scores are and what level they should move on to next.

      That’s one way. Another might be to say that the student needs to learn Beginning Fractions, and gets assigned to a class with people from a similar range that also need to do Beginning Fractions.

      This is harder to do than our current system. To which I say: so what? The primary goal of an education system should not be “easy to run.” We have computers these days that can group people together.

      • cassander says:

        >To which I say: so what? The primary goal of an education system should not be “easy to run.” We have computers these days that can group people together.

        “easy to run” should not be a goal. “sufficiently easy to run” definitely should be. Students are not infinitely fungible. You only have so many teachers, so many classrooms, so many chairs per classroom, etc. Lesson plans have to be designed around some fixed period of time. At some point, you’re going to have to break people up into more or less arbitrary chunks to deal with real world friction.

      • Aapje says:

        @Edward Scizorhands

        One thing that I noticed during my final exams of my secondary education is that reading ability was earning me a much higher grade for many subjects that were not about reading. Reading the questions carefully and applying some logic lead me to a lot of answers, even if I hadn’t memorized something exactly.

        Later at the university, I was partnered with a student who was unable to write a paper (at even undergraduate level, not talking about a proper scientific paper). I was quite amazed at how he managed to get so far without that ability and I was also not happy at having to do all the work.

        Anyway, my point is that there are general skills that are generally useful, as well as crossover between subjects. Teaching and testing students merely based on one subject requires dumbing down the education and making our teaching even more mono-disciplinary.

        Real life is not mono-disciplinary.

        • Anonymous says:

          One thing that I noticed during my final exams of my secondary education is that reading ability was earning me a much higher grade for many subjects that were not about reading. Reading the questions carefully and applying some logic lead me to a lot of answers, even if I hadn’t memorized something exactly.

          Every test is an intelligence test.

  21. Well... says:

    A few Open Threads ago I listed the Neal Stephenson novels I’d read in order of how much I liked them and then asked for recommendations for more hard sci-fi.

    I tried out one of the recommendations: “Rainbows End” by Vernor Vinge. I’m not sure I hated it, but I disliked it enough to quit reading within the first hundred pages or so, which is something I almost never do. (I’m usually one of those people who stubbornly insists on finishing even the books I don’t like.) I also can’t put my finger on exactly why I disliked the book so much, and haven’t spent much time thinking about it. If other people have read both “Rainbows End” and some of the Neal Stephenson books I listed, we can discuss the differences and then maybe what I didn’t like about “Rainbows End” will pop out at me.

    “Snow Crash” was also recommended but it was already a book I had planned to read. (I was on a wait list for it at the library when I asked for recommendations last time.) I just finished “Snow Crash” tonight, so my updated list goes like this: (1 = liked most)

    1. Seveneves
    2. Anathem
    3. Snow Crash
    4. Zodiac
    5. Reamde (still liked it a lot and enjoyed reading it)

    More suggestions please! Repeat old ones if you want to, I doubt I’d go back and search for the original thread.

    Also, general discussion about Neal Stephenson’s writing is welcome.

    • hoghoghoghoghog says:

      I also kind of disliked Rainbow’s End, even though I loved the novella of which it is an expansion (Fast Times at Fairmont High). The novella is way more idea-driven (as a mathy person I loved the idea of humans discovering powerful general heuristics, and I’m amazed that particular fantasy never occured to me or anyone else I know before). But when Vinge tries to write detailed characters/plots it feels somehow exhausting.

      (This is why Charles Stross is more fun – the character that he worked hardest to make plausible is a sentient sex robot)

    • Nancy Lebovitz says:

      Rainbows End struck me as a real disappointment, though I can’t think specifically of why it seemed dulll. Irritating main character? And my feeling was that shredding books to make the permanent computer copies was just trolling people who like books.

      Anyway, if you want sense-of-wonder hard sf, I recommend Egan’s Diaspora.

    • 1soru1 says:

      Stephenson has won the Prometheus award for libertarian sci-fi twice, once for Seveneves, your top pick. So clearly the nominators share something of your taste. Of the several authors have won it three times, Ken Macleod is IMO the best writer.

      5 books by him matched in theme to the corresponding Stephenson books:

      1. Cosmonaut Keep
      2. The Sky Road
      3. Descent
      4. Intrusion
      5. The Execution Channel

      • MacLeod is an interesting thinker, but none of his books struck me as all that good as stories.

        • 1soru1 says:

          Traits he shares with Stephenson, IMO.

          Personally, I read SF for ideas first and foremost. Character, plot and and prose quality are secondary; if bad, they are flaws in a gemstone. Whereas bad ideas, or no ideas. means the whole thing is made out of plastic. Which disqualifies it from being of value, whatever i ts shape.

    • J Mann says:

      1) Did you not read The Diamond Age and Cryptonomicon or are they >5 on your list. If you haven’t read it, that’s I recommend them highly – they’re my favorite two Stephenson books, for different reasons. Also, you should dig up and read The Big U for grins.

      2) I recommend The Island of the Day Before, by Umberto Eco. It’s not science fiction, but if you like those books, I think you’ll like it a lot.

      3) You might like Dan Simmons Hyperion saga. It’s not as deliberately deep, but still pretty engaging and thought provoking.

      • Murphy says:

        I’m somewhat lukewarm on most Stephenson books but The Diamond Age is among my all time favorite books.

      • Well... says:

        All the books I’ve read by Stephenson are included in my list. As I read more, I add them and rank them. I’ll check out Diamond Age and Cryptonomicon next, as my library makes them available to me.

    • rlms says:

      I don’t think it’s especially Stephensonish, but Too Like The Lightning is very good. It has something to appeal to all SSC readers: replacement of nation states with voluntary government-like entities for the libertarians; engagement with Enlightenment ideas in a way I think Death Eaters would like; and being vaguely reminiscent of The Man Who Was Thursday for Chesterton fans.

      • Well... says:

        It has something to appeal to all SSC readers: replacement of nation states with voluntary government-like entities for the libertarians; engagement with Enlightenment ideas in a way I think Death Eaters would like; and being vaguely reminiscent of The Man Who Was Thursday for Chesterton fans.

        You’re making an assumption about SSC readers that might be true in general but isn’t true in my case.

        • rlms says:

          You might well like it even if you don’t fall into any of those categories: I do, despite not being a libertarian or a Death Eater and not having read any of Chesterton’s other work.

      • quanta413 says:

        Yeah, but if those groups of SSC readers are separate groups who think the other two groups are dumb although maybe allies at a given moment, isn’t that actually a net negative? They get one idea they like and two they hate.

        I say this with tongue firmly in cheek much how I read original post mind you.

    • cassander says:

      1. Cryptonomicon
      2. Baroque Cycle (I probably enjoyed this more than Cryptonomicon. but that has to do with my particular historical interest more than the pure merit of the work)
      3-4. Snow Crash and Diamond Age are about equal in my eyes.
      5+. The rest

    • JohnBuridan says:

      I love your ranking of Stephenson books! Same as mine. Seveneves was so good! For some reason I didn’t like Diamond Age and couldn’t finish Cryptonomicon…

      It seems like you like the same types of ideas that I enjoy:

      Dictionary of the Khazars by Pavic
      Name of the Rose by Eco
      Non-fiction:
      Empires of the Word by Ostler

      Got to go, can’t sit here thinking forever!

      • Well... says:

        I like realistic space stuff.

        I enjoyed the thrills I got from Seveneves–while I was reading it I’d glance up at the moon a few times each night, relieved it was still there.

        I loved the “take you to all these different crazy places and environments” aspects of Anathem (in addition to the space part), Reamde, and Snow Crash.

        I like the didactic element to Stephenson’s writing too–each book of his, I learn new things or at least get introduced to enough cool new things to go off and decide to learn more about them on my own.

        Stephenson’s earlier style–expressive, a little sarcastic, almost like Chuck Palahniuk–is fun but I don’t think his later writing’s any worse without it.

        And Stephenson keeps you turning the pages, which is at the end of the day pretty important in a novelist.

    • JohnBuridan says:

      I love your ranking of Stephenson books! Same as mine. Seveneves was so good! For some reason I didn’t like Diamond Age and couldn’t finish Cryptonomicon…

      It seems like you like the same types of ideas that I enjoy:

      Dictionary of the Khazars by Pavic
      Name of the Rose by Eco
      Non-fiction:
      Empires of the Word by Ostler

      Got to go, can’t sit here thinking forever!

    • Admin says:

      Update: picked up a copy of Diamond Age, will begin reading tonight.

  22. AnonYEmous says:

    here’s an offhand question for everyone here:

    How would you conduct a study to show the effectiveness of the jury in a jury trial, in a way that is transferable to jury trials generally? As in, how often they will get a case right.

    (note: I’m not in a position to use this in any way, it just interested me)

    • Brad says:

      You might be able to do something with DNA and exoneration rates. Like if a jurisdiction went back and tested all the physical evidence it had from the pre-DNA era and you found some percentage was the person in prison and some percentage wasn’t. It wouldn’t be any good though if it was a prisoner driven process, it would have to be something comprehensive or random.

      • Evan Þ says:

        Also, it would need to be compared against non-jury trials in some way that was carefully adjusted for confounders. I really have no idea how to do that adjustment, though, since lawyers have been carefully considering for decades what sort of defendant would do better with or without a jury trial.

        • Brad says:

          That’s a good point. Bench trials aren’t as much of an issue as plead outs, which have played a big part in the U.S. criminal justice system for a long, long time.

          Though the very fact that their are innocent people that plead guilty gives you some idea as to what those defendants thought of the reliability of the jury trial.

          • FacelessCraven says:

            @Brad – “Though the very fact that their are innocent people that plead guilty gives you some idea as to what those defendants thought of the reliability of the jury trial.”

            Isn’t this less about the reliability of a jury trial, and more about systemic overcharging by prosecutors as a tactic to force more pleas? If the difference between plea charges and trial charges are stark enough, any amount of uncertainty in the outcome becomes unacceptable. People will pay a lot more to go from one bullet to zero bullets in Russian Roulette than they will to go from two to one, right?

          • Brad says:

            Fair enough that it is more complicated than just one variable, but I think the same is true for the trial penalty as it is for the trial reliability.

            If, for example, the jury trial had reliability of 99% I don’t think there’d be much plea bargaining by innocents even if the trial penalty was 2 or 3x.

            If you wanted to model it you’d probably have to come up with some kind of function for the perceived disutility of prison sentences of various lengths. I’m not sure what the answer is there.

          • Matt M says:

            If, for example, the jury trial had reliability of 99% I don’t think there’d be much plea bargaining by innocents even if the trial penalty was 2 or 3x.

            The problem is, nobody involved knows these statistics, or has anything close to a legitimate guess. The choice they are given is “two years prison guaranteed OR enter this mysterious black box known as trial where 12 random people will spit out a result that could be anything from ‘you go free, sorry for the inconvenience’ to ’20 years prison'”

          • Brad says:

            I don’t think that’s accurate. ADAs and the defense attorneys, especially PDs, can make a legitimate guess in many cases. They bargain in the shadow of a shared understanding of the range and probability of likely outcomes if they went to trial (and less admirably their own incentives).

          • AnonYEmous says:

            If, for example, the jury trial had reliability of 99% I don’t think there’d be much plea bargaining by innocents even if the trial penalty was 2 or 3x.

            as an unrelated aside, if the jury had a 99% reliability would there be plea bargaining at all? Or would it just be a case of criminals saving the state time and in return receiving a shorter sentence to save them time?

          • Eltargrim says:

            @AnonYEmous

            That is frequently the case with pleas as they stand. Prosecutors generally want to clear cases quickly, criminals want to minimize the punishment they receive, so both benefit from avoiding a trial. That’s the point of plea bargains when used as intended.

            Plea bargains can be used effectively in a system where the jury is reliable. If they’re only 90% reliable, however, a 100% chance of 6 months and a record may look nicer than a 10% chance of 10 years.

          • Winter Shaker says:

            The whole ‘how to avoid perverse incentives in the criminal justice system’ is an area that seems to pit things that might actually work against the fact that the people who could actually implement such things generally don’t want to. But it seems to me that if you wanted to get some tabs on it, a rule that a certain randomly selected cases where the defendant has already accepted a plea bargain are chosen to go to trial anyway (with no adverse effect on the severity of the sentence that the defendant has already accepted, but with an onus now on the prosecution to prove their case or else the defendant goes free).

            Prosecutors can be graded according to how many cases they actually win, relative to how large a gap they were threatening the defendant with in severity of sentence if it goes to trial vs severity of sentence if they accept a plea bargain. That on its own might not be enough to embarrass overzealous prosecutors into toning it down, but it might at least allow society to get a handle on how many overzealous prosecutors there actually are.

            (Or, I guess, a fairly hard limit on how big a discount a prosecutor is allowed to offer in exchange for a plea bargain, but that offers less opportunity to Do Science)

          • Matt M says:

            randomly selected cases where the defendant has already accepted a plea bargain are chosen to go to trial anyway

            Not a bad idea, but definitely has some flaws.

            Is the defendant going to be fully motivated to spend the time, effort, expense, on a trial where they were fully willing to accept the plea? Is a public defender going to sufficiently prioritize such cases? Is the prosecution? Will witnesses still be compelled to testify? Won’t most people involved see this as a pointless waste of their time?

          • Winter Shaker says:

            Well, the defendant still goes free if they win the trial, so I expect that would be a reasonable incentive. And I’d *hope* that most people wouldn’t consider ‘participating in the process of calibrating the justice system to make sure it actually tends to produce justice’ to be any *more* a waste of their time than sitting on a normal jury – indeed, there’s no reason the jurors should even have to know that the particular case they are deciding is a plea-bargain calibration case. Though I agree that, realistically, if they *did* know, probably a lot of people would take it less seriously.

            For that matter, I’d *hope* that prosecutors would be incentivised to take them seriously too, because the cases where a defendant has accepted a plea are disproportionately the ones that ought to be a slam-dunk if they go before a jury, and a decent prosecutor ought to want to be seen to be good at securing pleas from jury convictions against the guilty while avoiding false positives of pleas from the innocent (obviously this system does not solve the problem of prosecutors who are skilled at securing jury convictions against the innocent) … but to the degree that the sort of people who care about securing convictions against anyone brought before them regardless of guilt are disproportionately likely to be the sort of people to become prosecutors in the first place, you are probably right.

          • I’ve been thinking about the problem of plea bargaining and overcharging in the context of the book I’m writing on legal systems, since some other systems had related problems and in some cases solutions to them.

            In Periclean Athens, where criminal cases were privately prosecuted and the successful prosecutor usually got a share of the fine, there was an obvious incentive to target people who were wealthy and unpopular. One way of dealing with it was that if the prosecutor failed to get at least twenty percent of a large jury to vote for conviction, the prosecutor was fined. The rule for their version of tort cases was that if the prosecutor lost, he owed damages to the prevailing defendant of one sixth the amount he had claimed the defendant owed him. We don’t know enough about the system to tell if that applied to all tort cases, only ones where the prosecutor got less than 20% of the jurors to support him, or some other subset, but it at least applied to some.

            One possible version in our system of state prosecutors would be a rule under which any time a prosecutor failed to get at least three jurors to vote for conviction on a charge, that was a point against him, and a certain number of points resulted in his being fired. That would be a sizable disincentive to deliberate overcharging. But it only works if the defendant goes to trial, and he still has a strong incentive not to.

            Another version would be a rule saying that if the defendant is acquitted on any charge, the sentence for charges on which he is convicted must be the lowest sentence permitted for that offense. The prosecutor could still try to scare the defendant by overcharging, but if the high charge failed the defendant would do at least as well, possibly better, than the plea bargain offer.

            I am not sure there is any U.S. jurisdiction where either of those could get implemented, but they at least suggest possible approaches for dealing with the incentives of both prosecutor and defendant that produce the current problem.

          • Aapje says:

            @Friedman

            In Periclean Athens, where criminal cases were privately prosecuted and the successful prosecutor usually got a share of the fine, there was an obvious incentive to target people who were wealthy and unpopular. One way of dealing with it was that if the prosecutor failed to get at least twenty percent of a large jury to vote for conviction, the prosecutor was fined.

            I don’t see how that solves much, because if you think that a person is unpopular enough that you can get 51% to vote to convict based on poor evidence, then you’d get 20+% of the votes in most cases, unless you are really bad at framing people. But then the entire business model was bad anyway. One can assume that people who run this scam are pretty good at framing people, or they wouldn’t do it.

            On the other hand, there is a definite risk to scare off actual victims when going after a popular person, if they think that it’s likely that there is strong prejudice in his/her favor.

            Both scenarios can present themselves 100% identically to a jury (a person testifying that another person did something wrong without much additional evidence). One would actually expect the ‘framers’ to have better cases than the worst cases of actual victims (as the latter have to make do with what happened, while the former can wait for the best opportunity/set up).

          • Jordan D. says:

            I’ve often thought that the criminal arena would be an interesting place to do a European-style loser-pays system for cases that get to trial. The reason attorney’s fees are so rare in civil cases is the fear that they would discourage less wealthy people from seeking redress, but prosecutors in America are well-funded and we don’t really want them to prosecute weak cases. That could also encourage Public Defender offices to take strong cases to trial, since they wouldn’t be stuck paying if they won.

            Heck, maybe it would even temper the public zeal for prosecution if overzealous prosecutors were seen as throwing tax dollars at unpopular groups.

          • Matt M says:

            Jordan,

            Is there a possibility that such a system would influence jurors?

            Like, if it’s clear the defendant is generally a scumbag probably up to no good, but the prosecution cannot specifically prove he is guilty of what he’s being charged with, the jury may be motivated to find guilty, because the prospect of having “our” tax dollars given to this fool is sufficiently unpalatable.

          • Brad says:

            @Jordan D.
            It’s an interesting idea, but I worry about adverse selection against public defenders of last resort as private lawyers skim off the perceived best cases.

            A perception that only guilty people have PDs could work significant unfairness.

          • Jordan D. says:

            @Matt:

            Maybe. I’ve never run a real jury trial, but my understanding from juror interviews and the litigators I’ve talked to is that juries as a whole usually do a reasonable job of adhering to the elements laid out. Perhaps in this case the judge could give an extra instruction to keep in mind that any reimbursement would not be retained by the defendant?

            @Brad:

            Eugh, I hadn’t thought of that. Public defenders have historically been prone to negative stereotyping and smear campaigns, though.

            On the other hand, would that ultimately be worse than the present situation, where PD offices are regularly overloaded far beyond capacity? Even granting that private offices would pick over the easiest cases they could, that might still give the public defenders more time to, e.g., work on cases that will probably turn out guilty but might merit a lower sentence.

          • Brad says:

            I hate to play the cynic, but what makes you think PD funding will be held steady if caseloads fall?

      • Douglas Knight says:

        Someone did that. For rape convictions that had useable DNA on ice, 15% were the wrong man.

        • Acedia says:

          Wow, 15% is disturbingly high.

        • It says between 8% and 15%.

          That’s still higher than the figures I have seen, but all of them had some problems. And it’s “sexual assault” which is broader than what is usually called rape.

          • Douglas Knight says:

            If you ask a stupid question, you get a stupid answer, 8%. If you ask the question I specified, the only question anyone would ever ask if they weren’t fishing for lowball answers, you get the answer 15%.

          • Matt M says:

            Isn’t rape the only instance of sexual assault in which DNA evidence might conclusively prove anything?

            DNA can neither prove nor disprove whether or not someone briefly groped a woman on a street corner.

          • DNA can neither prove nor disprove whether or not someone briefly groped a woman on a street corner.

            But it can prove whether he was the one who groped her and ejaculated onto her, even if he didn’t penetrate. Or bled on her.

          • Aapje says:

            @Matt M

            And then only cases where the accused/convicted person disputes having sex with the alleged victim.

            If it’s about consent, DNA evidence may be useless.

          • dndnrsn says:

            I would guess that cases where someone convicted of rape was exonerated by DNA evidence would be cases where the rapist was unknown to the victim and did so using force or explicit use of force, then the cops grabbed the wrong guy, and the victim mistakenly identified him as the rapist. Later, DNA evidence shows it was someone else. For example.

      • AnonYEmous says:

        that sounds like a winner t b h

        • AnonYEmous says:

          on reflection this response could be a bit more respectful

          Thanks to everyone who pointed out that old cases with newly relevant DNA evidence can be used to evaluate jury excellence. Now my curiosity is gone but I might look more into what was revealed as a result of those cases later.

          • Protagoras says:

            As I understand it, the main conclusions of those who have looked into this are that juries often believe eyewitnesses, and eyewitnesses are incredibly unreliable.

      • Nornagest says:

        I don’t think that’d give you a good estimate of the fraction of false convictions. It might put a floor or a ceiling on them, depending on whether cases with physical evidence that you can get DNA from are more or less likely to lead to a conviction (my money’s on “more”), but we have no reason to think that cases with that kind of evidence are typical.

    • doubleunplussed says:

      You could select a jury twice as large as normal, then split it into two groups randomly and let them view the same case without the two groups being able to communicate. You then measure what fraction of the time the two juries agree, and compare to random chance. That should tell you whether the juries are reliably measuring *something*, but unfortunately it won’t tell you whether that something is the truth or not. They could just be succumbing to the same bias or whatever. But it would be something.

      • Matt M says:

        I’d definitely be interested in something like that. My prediction is that the amount of times where “Jury A votes unanimous guilty” and “Jury B votes unanimous innocent” would be shocking. I’ve never served on a jury, but I have a pretty strong belief that they are likely dominated by a couple strong personalities, with everyone else just going along with the group, for various reasons.

        • James Miller says:

          Yes, but this isn’t as bad as it seems because most not-hard cases are settled before trial.

        • Edward Scizorhands says:

          Don’t people doing major lawsuits have mock juries which see all the trial information? How well do these do at predicting the actual outcomes?

          • Jordan D. says:

            IIRC, they have moderate predictive power in civil trials but aren’t used in criminal trials. Mostly because civil trials have depositions available to test with ahead of time, but they’re rare in criminal trials.

            In most civil cases, the question is less ‘did X do it’ and more ‘how liable is X for having done it in this situation’, so it wouldn’t surprise me to see juries broadly agree on the outcome but split harder on remedy.

    • J Mann says:

      That’s a tough problem because there are so many different factual situations – I don’t necessarily expect juries to have the same effectiveness in patent trials as in an assault case.

      Also, you can reasonably assume that most easy cases filter out by settlement: the ones that make it to trial are where both sides think they have enough of a shot that it doesn’t make sense to settle/plead/etc., or where at least one side is too emotionally invested to settle on the math.

      Lastly, a lot of trials turn on judgment calls that make it hard to determine the correct answer. When the parties signed a contract amendment saying that in return for avoiding change fees, it was ok to delay construction so long as it was finished by “the summer,” should they have understood that to mean the scientific definition or a common understanding, and if the second, then what was the completion date?

      doubleunplussed idea was pretty good. If you had unlimited money, you could try to run mock trials where you know the answer, but I’m not sure how representative that would be.

      Actually, for big cases, people do run mock pre-trials, so you could compare those to the actual verdict on cases that go to trial, but they’re abbreviated and their results affect the way the lawyers litigate the actual case, so there are problems.

    • Edward Scizorhands says:

      Once a year, put on a mock trial where everyone is actors but the jury, who think it’s a real trial.

      Have there be a lot of emotional and circumstantial evidence against the accused. However, have it crystal clear, to anyone paying attention, that there is a 100% alibi for some fact that is essential to the prosecution proving their case. See if the jury notices it, or even cares.

      • Civilis says:

        I think a better idea would be to recreate a trial using actors and the transcript from a real trial in front of a jury. Once you establish how reproducible the results of the original trial are, this would allow you to even change some variables, like the race or gender of some of the participants.

  23. Machina ex Deus says:

    So I have this friend who’s at least as smart as I am, and can work about three times as hard. I’m trying to get him to think beyond just working for someone else, since (a) that would limit him to working on things and in ways other people are smart enough to think up and (b) he no longer technically needs to work a steady job, having cashed out his options from a large startup he worked for (but was’t a founder or tenth employee of or anything).

    He kind of has to admit he’s smart, sine he’s had a successful technical career after getting a humanities Ph.D. And I think he’s sick of people who are less smart ordering him around and limiting his options. But he seems stuck on working for some existing organization.

    I think he’s capable of writing a great technical book: an intro to one of the functional languages he likes to use, or something about writing Domain-Specific Languages. (Apart from his thesis, the only other book-length things he’s done are translations.)

    I honestly believe he’d be much happier plugging away at a book or a niche project for six months than going around interviewing for jobs he doesn’t need and won’t like. And the world would end up a better place.

    How do you get someone to understand their potential?

    • Evan Þ says:

      I suppose you’re familiar with early retirement blogs like Mr. Money Mustache? They tend to assume that people will have projects to fill their early retirement with, but they’re largely what opened my mind to see that it was possible.

      What does he like to do aside from work? Would he like to do it all the time? (Those aren’t the same question; I would hate to write fiction all day long.) Is there some project (maybe a programming project; maybe work for some nonprofit) that he thinks he’d love to do if he just had time?

    • Matt M says:

      I honestly believe he’d be much happier

      This strikes me as a dangerous phrase/assumption. You should be very VERY sure about this before attempting to act. Some people (even very smart ones) legitimately like following others direction. It’s comforting. You can focus your energies on solving the problem rather than thinking about what problems to solve.

      Your assumption seems to rest on the premise that the ONLY reason someone would “take orders” is because they aren’t smart enough to do bigger and better things. But he may not want to do bigger and better things. Or he may not be ready for them just yet.

      • CatCube says:

        Yeah, I think I’m a pretty good engineer (though we’ll see with the PE coming up on Friday!) However, I do not do well with being 100% self-directed. When I’m part of a larger organization, I can accomplish things. When I’m on my own, I just run from one shiny thing to the next without actually finishing anything.

      • martinw says:

        Exactly. Some people want to be Captain Kirk, and others will be happier as Scotty or Mr. Spock. There’s no shame in preferring the latter, and it’s not a sign of lower intelligence or any other negative personality trait.

        (added) And it’s not just about preferring to follow orders rather than make your own decisions. A large corporation can achieve things that no individual can, and a person may want to be a part of that even if they get to be just a small cog in the machine.

        (added #2) To the grandparent: if this guy is at least as smart as you and financially successful, then why are you so sure that you know better than him what’s the best career choice for him? Would it not make more sense to approach him with this question, from an angle of “can you please explain this to me so that I may learn from it”?

      • Aapje says:

        Some people (even very smart ones) legitimately like following others direction.

        The people who give directions can also be very smart. An alternative strategy might be to look into jobs which are highly satisfying.

        For example, I don’t think that there is an upper limit to the ability of the NSA to tax even the smartest people.

      • Chalid says:

        It might very well be that he needs to just find a better company. Some companies are much better at identifying and rewarding exceptional talent than others.

    • Deiseach says:

      Not everyone is an entrepreneur or is a leader type. There’s no guarantee he’d be any happier being the boss of a team than he is working under a boss.

      If he has an idea himself for a small business or self-employment, he could try it. But pushing him to do something he’s not keen on won’t help and may damage your friendship.

      I suppose the compromise is ask him – if he really can afford six months or a year out of the workplace – to take a sabbatical and work on something like that technical book. If he does it and likes it, great. If he can’t do it, no harm done – at the end of the six months/year he goes back to job hunting. He’s not committing himself to being an early retiree or giving up work altogether.

    • Nancy Lebovitz says:

      Why can other-optimizing be such a very tempting project?

      • Deiseach says:

        Because we’re a nosy species that can’t help interfering in one another’s lives. Part of it is genuinely wanting to help others be happier and better-off and part of it is vicarious achievement (“if I were in his shoes, this is what I’d want to do”) and part of it is motes and beams; it’s easy to see the small flaws and ‘if you only did this’ in the lives of others when we can’t or won’t fix the big holes in our own lives.

        • Aapje says:

          It’s also easier to ridicule/ignore the irrational desires that other people have than to give up/ignore your own irrational desires:

          Me: That other person could just lose weight by not eating so much.
          Me: Oh look, tasty food. MUST EAT IT.

    • Machina ex Deus says:

      I’ll attempt to reply to a bunch of replies at once:

      In terms of what he enjoys doing, he’s kind of a Renaissance Man: reading history or poetry (some of it in English), running mildly long distances, learning a new martial art, messing around with a new programming language, listening to music, playing music (he’s had professional gigs as a keyboard player), and spending time with his wife and kids (sometimes in combination with one of the above).

      (The fact that I’m his friend, and not sick with envy of him, is kind of remarkable, now that I think about it.)

      I appreciate the words of caution about me saying, “I honestly believe he’d be much happier…”; the background is that every single job he’s ever had has made him unhappy, so him finding something that leaves him neutral would be a big improvement.

      He’s dealing with the lack of direction now by taking online courses. There’s probably no danger of him running out, but eventually he’s going to do something else, and absent some nudging, he’s going to drift into another standard technology job.

      If I could pay Alan Kay or David Gelernter to visit his house and send him on a quest, I would; he’s a finisher who could use a visionary who’s as at least as smart as he is.

      It’s possible that there are existing organizations that would not make him miserable, but I don’t know how to find them. He’s not really interested in the kind of stuff I do, or I’d have gotten him hired already.

      And I completely understand not wanting to be an entrepreneur or a manager: neither of us likes the business side of things, and neither of us wants to manage other people.

      Other-optimizing is driven by a lot of things; in this case, possibly it’s that my friend, in a professional context, is like a better version of me, so if he can’t find an awesome job (even now that he has little pressure to), what hope is there for me? Another part of it is that he just can’t see how awesome he is, from his inside view, and I suspect that if he had a more accurate idea, he’d chart a better course (defined as one that would make him happier during the day).

      • Aapje says:

        in this case, possibly it’s that my friend, in a professional context, is like a better version of me, so if he can’t find an awesome job (even now that he has little pressure to), what hope is there for me?

        Your mistake here is that your friend doesn’t seem to be limited by ability, but by the things that he considers satisfying to do.

        If you have lesser abilities, but also are easier to satisfy, it can be easier for you to find a satisfying job (or not). Your conclusion that if your friend can’t find happiness, you can’t find happiness, is thus mistaken, since you are focusing on the wrong variable in a multi-variate problem.

    • MoebiusStreet says:

      I’m author of a technical book. I can tell you that except in very rare cases, this is not a way to get rich, or even to get by. At the time I did it, I worked on the project virtually every hour outside my regular job, nights and weekends, for nearly six months. From this I made a modest 4-digit sum which dried up in under two years since the technology moves so fast. And this was about 20 years ago, when there wasn’t nearly so much competition from free information on the Internet.

      Don’t get me wrong: if I could go back in time, I’d still do the same things. It might not have directly made money, but it’s about the best resume you could show a prospective employer. The respect that it engenders is nice, and this also helped my wife’s family (many of whom are extremely highly regarded in their areas) accept me as more than just a street urchin. And I have to admit that it just made me feel good about myself, especially when I saw that the book had also been translated into Chinese.

      But being realistic, writing a technical book isn’t a reasonable way to make a living except for a few very rare people.

  24. keranih says:

    What is a day or event that you think should totally get world-wide attention by being featured on the Google homepage?

    My choice would be May the 3rd, the date of Eclipse’s first race in 1769.

    (Happy Easter to all people of good will!)

    • Evan Þ says:

      Christ is risen!

      My nominations for Google doodles:

      – Easter. At least, in years (like this year!) where Gregorian and Julian Easters coincide.

      – July 20, first manned moon landing. (Alternatively, July 21st, when Armstrong stepped out on the lunar surface.)

      – 24 October, Peace of Westphalia.

    • Matt M says:

      I’m a big supporter of garbage man day.

      Or more generally, that more honor and respect should go to the people who are responsible for the day to day running of the basic functions of society.

      • JohnBuridan says:

        Yes!

        I had a teacher in secondary school who said, being a garbage man is an honorable job, and if you go home at night and read the Aeneid, then you really are living a good life.

      • NIP says:

        I see that I’m in good company. Don’t forget janitors, as well! As you say, there are a lot of people who make sure society runs smoothly but who are beneath the notice of most.

        being a garbage man is an honorable job, and if you go home at night and read the Aeneid, then you really are living a good life

        Having a job where you can do an honest days’ labor without interacting with customers or managers and then return home to do whatever you want really is the good life. I’d probably kill for a job like that. Dunno why more people aren’t actually jealous of garbage men, janitors, landscapers, etc.

        • Matt M says:

          My dad was a janitor and seemed pretty pleased with his life overall. Most of his complaints were of the nature that he did his job TOO well and they kept trying to promote him into like, janitor management positions, which he wanted nothing to do with (and that it was virtually impossible to find employees who were at all reliable and hardworking).

          • NIP says:

            Your dad sounds like a pretty cool guy. I agree with him that a stable and peaceful life situation is better than trying to climb the ladder for more money and status.

            Most of his complaints were of the nature that he did his job TOO well

            Unions, amirite? I had a (very) short stint as a janitor years ago, and I also encountered this, although they didn’t offer to promote me, just bitched at me quite vehemently. So being a stupid kid, I quit. I’ve been looking for another job as a janitor for some time.

          • maybe_slytherin says:

            My dad was a janitor and seemed pretty pleased with his life overall.

            But was he pleased with his overalls?

          • Matt M says:

            Never had to wear them! Ended up being another area in which he annoyed his bosses. When he originally started, they were allowed to wear whatever they wanted. At some point, management basically came up with a “janitor uniform” for them all. He told them he didn’t have to wear it because it wasn’t in the union contract. He never wore it, but took a ton of grief from his bosses for his refusal. Got away with it because he was basically the Hawkeye Pierce of janitors – too good and too popular to fire.

        • The Nybbler says:

          Garbage men and landscape laborers get up early and do hours of hard labor each day. Nasty smelly labor in the case of garbagemen. The landscape laborers get paid peanuts, and often enough they don’t even have a job; they’re often day laborers so the next day they’ve got to be hanging out at the Home Depot at 0-dark-30 hoping for a job. The garbagemen get paid a decent wage by labor standards, but it won’t make anyone rich. In either case I doubt they have much energy left to live “the good life” when they get home.

          • NIP says:

            I can’t speak for garbagemen, but I have worked as a landscape laborer and I couldn’t disagree more. It was great! I got paid the same day every day I worked, in cash, at fifteen dollars an hour. Never once had to hang around a Home Depot and there was always plenty of work. You just need to find an affluent area and a boss who isn’t a skinflint. Perhaps easier said than done for some people, but that was my experience. Plus, if you’re smart, you can save up money for your own equipment and be your own boss. I imagine that’s quite difficult to do if you have a family or don’t like living a spartan lifestyle for a few years, but not at all impossible for the type of young single folks who are attracted to that type of work in the first place.

            In either case I doubt they have much energy left to live “the good life” when they get home.

            That depends on that person’s energy levels and their definition of “the good life”, I suppose. And do garbagemen work every day of the week? I thought they got some days off, surely.

          • Art Vandelay says:

            I also used to work as a landscape gardener, it was a great job, although not especially well paid. I experienced the exact opposite of not having enough energy for the good life: I was far more dynamic in making the most of my free time than I have been at any other point in my life.

          • NIP says:

            I was far more dynamic in making the most of my free time than I have been at any other point in my life

            Are you me? I’m clinically depressed, and before landing that job I was in a pretty bad spell. Afterwards, though, the sunshine and hard exercise, not to mention the sense of self-worth and satisfaction for being told “good job” everyday for once and being handed a stack of twenties, did wonders for my sense of drive. I probably got more accomplished every day during that time than at any other in my life.

          • Nancy Lebovitz says:

            In Philadelphia, the garage is in bags which the garbage men pick up and put in a nearby truck. They wear gloves.

            So, they’re on their feet. The garbage bags get smelly during the hot part of the year, though I don’t know how bad that gets generally speaking– the garbage frequently gets picked up at around 7AM on my street.

            There’s a weight limit on the garbage bags, but I don’t know how physically taxing it gets for the garbage men as a full time jobs.

            In short, it’s a physical job, but not as awful as the Nybbler says.

          • Matt M says:

            But everyone thinks it’s a horrible, smelly, low-class job. Which is precisely why it deserves recognition.

        • FacelessCraven says:

          @NIP – “Having a job where you can do an honest days’ labor without interacting with customers or managers and then return home to do whatever you want really is the good life.”

          I used to believe this. Then I got such a job working in a paint factory. I was good at it, but I’m much happier being an artist.

          • NIP says:

            Diff’rent strokes for different folks, fam. You gotta do what makes you happy…er than the alternatives.

        • Anonymous says:

          Having a job where you can do an honest days’ labor without interacting with customers or managers and then return home to do whatever you want really is the good life. I’d probably kill for a job like that. Dunno why more people aren’t actually jealous of garbage men, janitors, landscapers, etc.

          Yup. Having worked in a variety of capacities and professions, I can’t say I liked any line of work much more than the ones that:
          a) did not require very much long-term independent decision-making (stuff like deciding the “class architecture of a Java application”, rather than deciding “which side of the house I should paint today”),
          b) did not require dealing with people very much,
          c) did not involve “homework”.

          Doesn’t pay much, though, and being independently wealthy beats most if not all professional work.

          • NIP says:

            Doesn’t pay much, though, and being independently wealthy beats most if not all professional work.

            I’m sure being independently wealthy beats a lot of things, but then there are weirdos like me that are actually content to live cheaply. Unless by “independently wealthy” you just mean someone who doesn’t live extravagantly but can survive off their assets alone, in which case I agree. But I think even if you don’t get paid a ton, if you’re frugal and smart you can save enough over the years to live like that. There’s a lot of factors that could effect someone’s ability to do that, of course, and I’m not one of those people who think hard work and frugality will always win the day. I’ve been poor for too long to think that. Not an impossible dream, though.

      • Jaskologist says:

        I’ve long maintained that Mike Rowe’s Dirty Jobs is the conservative equivalent of some show with a gay deaf black midget in it. It’s not that I particularly enjoy watching it, but doing so makes me feel like a good person.

        • Randy M says:

          Because it is attempting to raise the status of people you feel should be raised in public esteem, relative to where they are?

          • Loquat says:

            Kinda, yeah: my (semi-rural, conservative) father-in-law once explicitly told me that he thinks shows like that are useful for reminding the upscale folk that the working class exists, and does useful things.

        • massivefocusedinaction says:

          I love Dirty Jobs because it’s How Stuff is Made with a little more human interest. I’m fascinated by production lines, steps, and details. It’s fun to think about the engineer who created each step in a process, and try to guess at what the process looked like before they came along.

    • Well... says:

      International Day of the Nacho.

      Because nachos rule.

  25. Siah Sargus says:

    DC Meetup Thread!

    So, the DC meetup was a total blast. Shouts out to Robi for organizing it, and shouts out to Cassander for letting us use the apartment’s lounge! I hope that someone will chime in with more exact numbers, but we had an outstanding turnout. I would totally be willing to do something like this once a month.