Open threads at the Open Thread tab every Sunday and Wednesday

Links for September

You Are Not So Smart raises our awareness of the menace of Placebo Buttons. I’d always known a lot of crosswalk buttons don’t do anything, but the information about elevator “close door” buttons really hurt.

I hate to link to something as lowbrow as this, but I have to admit this page made me laugh more than anything else has all year: 30 Hilarious Autocorrect Struggles. I don’t even care if it’s real.

A couple different places are now saying they can alter crops to automatically fixate nitrogen. If true, it would reduce dependence on synthetic fertilizers which is good for the environment and great for billions of Third World farmers. The terms “world-changing” and “second Green Revolution” are being used. High end of estimates of benefit to the world economy in the hundreds of billions to trillions of dollars. Everyone seems to be doing exactly the right thing and throwing money at this idea.

Poverty in rural India is declining at an unprecedented rate.

Speaking of India, did you know Indian districts that elect Muslim (as opposed to Hindu) representatives tend to increase health and education outcomes in the districts they represent, including for non-Muslim constituents? This is so awkwardly out-of-left-field I can’t even tell if it’s politically incorrect or not.

A link that doesn’t have that problem: 11.8% of gender income gap apparently explained by menstruation, bizarre study says. Now I’m starting to wonder whether all those sarcastic feminist jokes about how we can’t have a female president because she would menstruate all over important legislation might not have just been ahead of their time.

The Guardian very kindly linked to my essay on scientism, and I guess that means I owe them one. So here is the Guardian’s new psychology blog, the groan-inducingly-named Head Quarters. Best of all, it’s co-written by Molly Crockett, my old high school science bowl teammate. You should trust everything she says.

So the good news is that this picture of Vikings attacking a Mars rover was taken without the use of any government funds. The bad news is that the investigation into whether the picture was taken with government funds cost somewhere between $40,000 and $600,000 in government funds. Penny wise and pound foolish? I, for one, demand the government fund an investigation to find out!

Speaking of gender gaps, a study on sentencing behavior finds that women receive prison sentences on average only about half as long as men, and that about half of that advantage is due solely to gender bias (rather than to a different pattern of offenses or different criminal history).

Collection Of Unusual Weather-Related Photos. Even if, like me, you think you’ve seen it all in terms of weather-related photos (the standard everyday lenticular clouds and mammatus formations most of these links end up as), you probably haven’t seen anything like these.

Dinner Specials Created In Honor Of Game Of Thrones Characters. Trigger warning for terrible puns.

I’m obviously an introvert, but it was hard not to cheer on this 15 Unmistakable Outrageously Secret Signs You’re An Extrovert article.

Move over evolutionary psychology, we have a new Least Romantic Explanation Ever For Human Romance.

Speaking of evolutionary psychology moving over, if you want to make a rational, well-founded, non-hysterical critique of evo psych ideas, this is the way to do it: Chickens Prefer Beautiful Humans. If this is true, it seems hard to square with the idea that human sexual preferences are specific human adaptations, at least in the area of facial features that this study examined.

From Robin Hanson: people can determine who won music contests more accurately based on video-without-audio than on audio-without-video. This includes trained musicians and music contest judges. In other words, a whole lot of ability-to-win-music-contests seems to be looking good while you actually play whatever.

I want to make it clear that I only appreciate this story on an ironic level: Robot Programmed To Fall In Love With A Girl Goes Too Far.

Reddit comes through again with Scientists, What’s The Craziest Thing In Your Field That You Suspect Is True But Is Not Yet Fully Supported By Data? Especially interesting are the birth control, bacteriophage, Lorimer burst and sildenafil/obesity responses. Creepiest (and seriously, don’t read this if you’re easily creeped out) is the one on hemispherectomy.

You can write integers in base phi and get relatively short, terminating strings. You can even do math with them. It’s not obvious why you’d want to. But you can.

Good use of the Proving Too Much technique and some nice biting satire: Why We Need Minimum Price Laws Today.

Continuing on this month’s theme of extreme techno-optimism: nanocrystalline cellulose is the new wonder material, and it’s made of wood pulp. Imagine something that can do everything plastics do, but better, and it doesn’t require fossil fuels or do weird things to your childrens’ gender.

Normal saline, the mainstay of medical treatment everywhere, costs somewhere between 44 cents and $546 per bag.

1973: Girls are bad at math. 1993: Girls aren’t bad at math, they’re just culturally conditioned to get anxious about it. 2013: Girls aren’t culturally conditioned to get anxious about math, they’re just culturally conditioned to tell people that they’re culturally conditioned to get anxious about it. 2033: ???

The number of miles driven in cars in the US – total as well as per capita – has been declining since well before the start of the Great Recession and continues to decline.

The Patel Motel Cartel is the name given to the observation that 1/3 of motels in America are owned by someone with the last name “Patel”.

How hard can it be to make a pie chart? Apparently, pretty hard! From Matt Vana on Facebook: WTF Visualizations

Every year, a town in Sweden builds a giant wooden goat. Every year, it…no, just read Wikipedia’s Gavle Goat Timeline for yourself.

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

30 Responses to Links for September

  1. Qiaochu Yuan says:

    So probably a large chunk of the music-contest observation is restriction of range. Among people good enough to participate in such contests, it seems most of the variance can be explained by non-auditory cues, but there’s a strong selection bias in the sample: the people in such contests are already the top .1% or whatever of musicians. This is the same effect behind that quote from a Google interview awhile ago about SAT scores not predicting performance at Google after 2 years, and various other similar observations.

  2. Ben Southwood says:

    In a glorious first, I had actually seen two of these before I saw this!

  3. Jack says:

    I don’t quite get the satire here. I don’t understand all the reasons for and against a minimum wage, and I tend to be biased towards the side of the argument I’m already familiar with, but it seems like the basic point is that capitalism works great when there’s competition on both sides, but is a disaster when anyone (who is not a benign dictator) has a monopoly (or oligopoly). There are many more workers than companies, so companies have all the bargaining power, so wages are often too low. (Except where there’s a very strong union of all workers, when wages can be too high.) So to redress the balance we have minimum wage laws. But there are ALSO more consumers than companies, so we have maximum price laws, not minimum price laws?

    • Swimmy says:

      You simply have the theory wrong. You can spend a long time studying economics and I doubt you’ll see it anywhere suggested in any major textbook that you have a monopsony or oligopsony when sellers greatly outnumber buyers. You have a monopsony when you have precisely one buyer (and no potential competition), and an oligopsony when you have very few. Dozens of fast food chains, many of them franchises and so still competing with each other, combined with thousands of small restaurants, even within a single city, does not an oligopsony make. There are some more complicated theories to explain how a whole boatload of employers can act like an oligopsony or monopsony without collusion, but they don’t appeal to there simply being a whole lot more workers than employers. I know of no economist who has ever suggested you need to have an equal number of buyers and sellers for competition to work well, or anything close.

      The satire works because it maps to (and links) specific arguments for minimum wages that 1) frequently do not appeal to any strong market failure arguments, and 2) ignore important Bayesian evidence to the contrary.

  4. BenSix says:

    Move over evolutionary psychology, we have a new Least Romantic Explanation Ever For Human Romance.

    I wonder how many of its authors were married.

    “So, honey, what have you been working on?”


    • Jack says:

      It might not be a problem. I enjoy eating, running, conversation and sex, presumably because of evolutionary pressure to do so, why not romance? Um, maybe 🙂

      • Deiseach says:

        I can’t comment on the article (is the rate of infanticide really that great amongst humans?) but I always thought that the notion of romance was similar to the reasons birds build bowers or collect bits of blue plastic and shiny blue coloured things or strut about displaying their magnificent plumage or doing mating dances:

        Men want sex (because, as Jack points out, it is enjoyable)
        Women want sex, too, but they have more invested because of the risk of pregnancy (guy gets orgasm, woman gets orgasm plus little bundle of joy to raise)

        Therefore, in order to persuade a woman that the cost is worth letting you into her knickers, ROMANCE!

  5. fs says:

    This is probably obvious to most people, but the “Robot Programmed To Fall In Love With A Girl Goes Too Far” article has no basis in fact. Someone at Vice posted about the hoax, if you’re interested.

    • Scott Alexander says:

      I don’t really understand the deflation claim. In fact, the second one seems to be “People are making a lot of this study, but it actually just means [exactly what it says and what I would expect people to assume it says]” (which actually sounds like it could be a powerful albeit evil rhetorical technique)

      On the other hand, I don’t know if I buy the argument that at the top level of musical performance, everyone’s so equitalented at music that visuals become comparatively more important. If music and visual are both important (as they obviously are), why doesn’t the process of getting to the top screen for visual skill as well as musical skill? That is, why model it as “music, then visual” as opposed to a reductio ad absurdum “judges only look at visuals usually, but by the top echelons everyone is so equally good at looking snazzy that they’re forced to pay attention to the music” or a more reasonable “judges at all levels pay attention to both”?

      • Douglas Knight says:

        Re: second paragraph

        People don’t compete very often and thus don’t get a lot of feedback about what wins competitions. They get a lot more time with tutors who think their job is to give feedback on the sound, not on the looks. Even if the tutors’ effective judgement is exactly the same as competition judges, telling people “better” / “worse” is much less useful for optimization than describing what the performer is and should be doing. But the explicit feedback is about the sound, since that’s what they think is the point.

  6. Carl Shulman says:

    “A couple different places are now saying they can alter crops to automatically fixate nitrogen. If true, it would reduce dependence on synthetic fertilizers which is good for the environment and great for billions of Third World farmers.”

    This comes at some expense in productivity: the energy to fix that nitrogen has to come from somewhere.

    • Only if the plants are currently limited by energy, and not nutrition (because the farmers already buy fertilizer). If this is the case, the farmer is already spending some of the yield on fertilizer. If this is cheaper for some subset of farmers than fertilizer (eg. those where price/productivity is very low), then it’s a boost in (effective) yield.

  7. houseboatonstyx says:

    Re chickens and ‘beautiful’ humans. Without digging into the study or anything, istr that one standard for human beauty is bluntness/softness/infantilism of features. Which commonly marks infants of all species.

    No hawknosed humans need apply.

  8. Rachael says:

    Those autocorrects seem to be among the less funny ones from The current most recent ones are better IMO, or the compilations like

  9. Nick T says:

    The chicken study doesn’t actually analyze ‘beauty’, it just uses a continuum of faces from feminine to masculine. I think it can be fairly summarized as showing that:

    * if you train a system, e.g. a chicken, to discriminate between moderately masculine and moderately feminine faces, it’ll discriminate more strongly between strongly masculine and strongly feminine faces
    * men and women tend to be more attracted to opposite-sex faces

    • Scott Alexander says:

      This makes so much sense that now I can’t remember what I could possibly have been thinking by posting that study in the first place.

    • Sarah says:

      Thanks, I knew the conclusion of that study was fishy, now I understand where it came from.

  10. Mary says:

    One notes that a “different criminal history” can itself be the fruit of gender bias, because you can have situations where charges are dropped against the wife in return for the husband pleading guilty.

  11. Deiseach says:

    Perhaps the Swedes should emulate the town of Kilorglin – get a real wild goat down from the mountain, crown him king for three days, raise him on a platform to reign over Puck Fair.

    Four hundred years and no-one has burned it down yet!

  12. Sarah says:

    I worry about the message “it’s important to overcome gender-stereotype-based fears!” actually reinforcing said fears.

    From what I’ve seen, there is no math gender gap. Period. Median math test scores in the US are equal between boys and girls. The gender breakdown on international math exams varies widely from country to country.

    There’s a gender gap in performance at the very high end (no female Fields medalists), but when we’re talking about long tails of lifetime achievement, it becomes less clear that we’re still talking about mathematical ability rather than “competitiveness”, “careerism”, or just institutional support.

    There is no damn gender gap in math. Most of the “gaps” (in test performance, math BA’s/PhD’s/employment, etc) have either vanished already or are rapidly narrowing.

    Gender is an endlessly fascinating subject and makes good press. Gender differences are interesting. Genuine inequities do exist in many professions and areas of life. But women-and-math is *basically not a thing* and becoming less of a thing every day. While it flatters my vanity to belong to a category that’s constantly in the news, from what I’ve seen of the facts, it’s incredibly overhyped.

    The persistence of “stereotype threat” as an experimental finding implies that salient memes matter a lot. Here’s a beneficial meme that happens to be true: girls are not actually worse at math than boys, and at every level of education, the percent of people getting degrees in math who are female has been rising by about 1% a year for decades.

  13. Paul Torek says:

    Ann Arbor pedestrian crossing buttons are not placebos. They’re much worse. They don’t change the length of time you’ll wait, but they hold up traffic for a very long time when your turn comes around. By which time, you’ll probably already have crossed due to a gap in traffic (perfectly legal here). Useful if you’re elderly or disabled; otherwise pointlessly cruel to drivers.

    I always suspected elevator “close doors” buttons though. It’s been a very long time since I’ve seen one do anything.

    • Sniffnoy says:

      …wait, seriously? I’d noticed the “hold up traffic” effect; I had no idea they didn’t shorten the actual wait though. Oh my…

    • estelendur says:

      If you have a traffic-activated light in Ann Arbor where one of the axes never activates unless there is a car *or the pedestrian crossing button is pressed*, then they are necessary during, say, rush hours. This includes most of the traffic lights in the State/Stadium/Packard triangle. This also constitutes a reason to go up on the sidewalk while on a bike if you don’t think a car will be along before the cycle starts over…

  14. Fnord says:

    Regarding base phi, it’s hardly the only irrational number for which that’s possible. It’s really easy to convert binary to base squareroot of 2.

  15. Pingback: Women Friends are Better Friends - My blog