• Chess Robots Are Getting Kind of Sneaky

    PhotoXpress/ZUMApress

    From Garry Kasparov, trying to get everyone to calm down about increasingly intelligent AI:

    Imagine a chess program with the potential to trounce human opponents and explain its moves to us, revealing the patterns that turn knowledge into practical wisdom, like a father teaching chess to his daughter or son.

    That sounds so cozy. Just me and my robot teaching me how to play chess by the dancing light of the fireplace before tucking me in for bed at night.

    And then, when we’re all asleep, killing us all!

    No, wait. That’s not my schtick. And then, when we’re all asleep, taking all our jobs!

  • Wall Street Got Its Tax Cut, and Now It’s Depressed

    I don’t follow the stock market super closely, but I was curious about why I haven’t heard Donald Trump bragging about it lately. This is why:

    The market seems to have spiked in anticipation of the tax cut, but then it slumped once it finally got its tax cut. Kinda strange. I guess they’ve got nothing to look forward to now except three years of Trump doing God-knows-what to wreck the economy in service of whatever moronic ideas happen to flit across his brain or capture his attention from Fox & Friends. That’s tough luck, guys, but you know what they say about lying down with dogs.

  • One Space or Two? That Recent Study Won’t Tell You.

    A recently released study tested whether it was better to put one space or two spaces after a period. A commenter at Kieran Healy’s Twitter feed comments:

    I guess this confusion isn’t surprising, but the “rule” has never changed. It’s always been two spaces for monospaced fonts, like the ones on typewriters, and one space for proportional fonts, like the one you’re reading now. There was good reason for this:

    • Monospaced fonts like this one already have a lot of air.  Two spaces make it easier to quickly distinguish when a new sentence begins.
    • Proportional fonts like this one are denser and don’t have that problem. Spaces and capital letters jump out at you.

    More to the point, proportional text in magazines, books, and newspapers is usually justified on both right and left—like this paragraph. When text is justified, typesetting equipment ignores your spaces entirely, putting in just enough to make the line lengths come out right. So it doesn’t really matter what you do.

    Still, what about the study? As it turns out, the authors tested only monospaced fonts, which is odd since almost no one uses them anymore. Here’s an excerpted and colorized version of their main result:

    Among people who preferred one space, reading speed was about 5 wpm higher with one space. Among people who preferred two spaces, reading speed was about 9 wpm higher with two spaces. In other words, people preferred whatever they preferred, and the difference wasn’t very much anyway.

    Oddly, though, the folks who preferred two spaces were generally faster readers than those who preferred one space. For $39 I suppose I could find out if the authors had anything to say about this. However, I suspect it means that the difference between the two groups isn’t random.

    So the whole study is pretty useless.  What would be more interesting would be a study of proportional fonts.  Over the past couple of decades, it’s become far more common to use a ragged right margin, like the one you’re reading now.  The reason is readability: By keeping the spacing constant, instead of tweaking it to make the line lengths come out evenly, the text is easier to read.  But with even spacing, would two spaces after a period make the text even easier to read?  I don’t know!  Someone should do a study to find out.

  • Trump Team Hired Firm to “Get Dirt” on Obama’s Iran Negotiators

    Colin Kahl with President Obama aboard Air Force One in 2015. The rest of Obama's foreign policy team is now insanely jealous of Kahl, wondering why they didn't make Trump's enemies list.Pete Souza/The White House/ZUMAPRESS

    On Saturday the Guardian Observer ran a story saying that Team Trump had hired an Israeli private intelligence agency to orchestrate a “dirty ops” campaign against some of the Obama folks who helped negotiate the Iran deal:

    People in the Trump camp contacted private investigators in May last year to “get dirt” on Ben Rhodes, who had been one of Barack Obama’s top national security advisers, and Colin Kahl, deputy assistant to Obama, as part of an elaborate attempt to discredit the deal….Among other things they were looking at personal relationships, any involvement with Iran-friendly lobbyists, and if they had benefited personally or politically from the peace deal.

    The story was just vague enough that I put it on the back burner, waiting to see if anyone else corroborated it. And someone did. Here’s a Twitter thread that Colin Kahl posted last night:

    The worst part of this is yet to come. That will be when this gets more attention and all the usual slimeballs chime in to say that there was nothing wrong with this at all. Oppo research is a normal part of politics, and checking to see if the Iran negotiators had a personal stake in the deal is perfectly reasonable. Nothing to see here, folks.

    I can hardly wait.

    UPDATE: The latest development is that Trump probably hired the same firm that Harvey Weinstein used to dig up dirt on the women he assaulted. However, they deny it.

  • Mueller Probe To Go Dark for Elections?

    From the Wall Street Journal:

    Mueller Probe Might Have to Go Dark for Midterm Election

    With six months to go until November’s midterm elections, Mr. Mueller’s investigation into Russian meddling in the 2016 presidential campaign…will reach a point this summer when Justice Department habits dictate that he will have to either finish his inquiries or go dark and stretch out his work until past November so he doesn’t appear to be trying to sway voters’ decisions, which would be at odds with Justice Department guidelines for prosecutors.

    Hahahaha. He doesn’t want to appear to be trying to sway voters’ decisions. Of course not. That would be at odds with DOJ guidelines. So very much at odds. Totally at odds. We can’t have that, can we?

  • Alfie Evans Is the Latest Martyr of the Pro-Life Movement

    Omar Marques/SOPA Images via ZUMA

    Alfie Evans was a two-year-old British infant who suffered from an irreversible neuro-degenerative disease that destroyed his brain and caused persistent seizures. His doctors eventually concluded that Alfie had lost all brain activity and recommended he be taken off life support. His parents fought this, but lost in court. At the end, they asked for permission to transfer him to a Vatican hospital, but that was refused too. Alfie died two days after his respirator was unhooked.

    Megan McArdle wrote a column suggesting that in cases like this the wishes of the parents should be given greater weight. In a follow-up, she wrote this:

    If a court is going to overrule parents who are trying to keep their child alive, then for the reasons laid out in the column, I demand a high burden of proof….This, I think, was part of the vitriol I triggered. After years of writing about health care, I have a great respect for doctors. But I also respect the limitations of what they can know. And I think a grave mistake of late modern liberalism is to overestimate the powers of science and expertise.

    Those powers are great, mind you; through science and expertise, we have conquered all sorts of disease and hunger and general human misery. I admire these achievements as much as anyone. But it’s no good tearing down ignorance and superstition if we just erect an idol to science in its place — which some of my interlocutors seemed perilously close to doing. Imperiously informing me that the courts were very careful and serious about their decision does not go very far towards proving that those judges were right. Stating that all the experts agreed about the decision simply begs us to ask the question: “Experts in what?”

    Alfie’s case is not the first of this type. And who knows? Maybe British law should be changed. But Alfie was examined by nearly a dozen doctors. Consultants from outside were called in. Alfie’s parents retained their own doctors. All of them concluded the same thing, based on MRI scans, EEGs, and various other tests. The judge in the case summarized it like this:

    The scan of 2 February 2018 confirmed the progressive destruction of the white matter of the brain which Dr R interpreted as now appearing almost identical to water and cerebrospinal fluid (CSF). In addition, new areas of signal abnormality were demonstrated in the deep grey matter of the basal ganglia. The thalami, which I have been told fire the pathways within the white matter which generate sensory perception is, Dr R points out, effectively invisible in the scan. In simple terms the thalami, basal ganglia, the vast majority of the white matter of the brain and a significant degree of the cortex have been wiped out by this remorseless degenerative condition.

    One of Alfie’s primary doctors said this:

    I believe that it is unlikely that Alfie feels pain or has sensation of discomfort but I cannot be completely certain of this….The continuation of active intensive care treatment is futile and may well be causing him distress and suffering. It is therefore my opinion that it is not in Alfie’s best interests to further prolong the current invasive treatment.

    This is not a case in which “experts in what?” is a reasonable question. The burden of proof wasn’t just high, it was immense—and yet there still wasn’t the slightest dispute. Alfie Evans was not “disabled.” He was brain dead:

    Professor Cross was clear that there was little in the way of reactive response. The EEG she told me was “markedly attenuated” which she clarified as “essentially flat.

    Despite this, Alfie became a cause célèbre on the pro-life right. The general tenor of their criticism was that the abortion lobby was no longer satisfied with killing infants in the womb. Now they wanted to kill them after they were born based on nothing more than their opinion about Alfie’s quality of life.

    I find it puzzling that this is a hill they want to die on. It’s much like the Terry Schiavo case, and in both cases the pro-life folks were defending the notion that a person could still be alive even if their brain was all but destroyed. Why? There are always judgment calls to be made about when life ends, but lack of brain activity is pretty clear cut. In the case of Alfie Evans, his parents were effectively performing an experiment to see how long a brain-dead child’s heart can be kept beating.

    In the past week conspiracy theories have flourished. Alfie was killed by a nurse while the parents were out of the room. Alfie was given a cocktail of four mysterious drugs shortly before he was taken off the ventilator. Alfie was killed by aluminum from vaccines. This is pizzagate territory, and it’s put hospital staff in considerable danger. There’s a real cost to this stuff.

    Still and all, there’s one more thing: why weren’t Alfie’s parents allowed to transfer him to the Vatican hospital? It would have been paid for privately, after all. What’s the harm?

    I don’t know. If I were the judge, maybe I would have allowed it. But this case was heard by half a dozen different courts, all of which ruled the same way. The evidence was absolutely clear that nothing could help Alfie, not in Britain and not in the Vatican. And keep in mind that the law is designed to protect more than one interest. For example, the law is designed to protect patients from charlatans who make promises they can’t deliver. A judge will naturally consider whether the pro-Alfie forces actually had Alfie’s best interests at heart, or if they were cruelly prolonging his parents’ pain for little more than cynical ideological purposes. A judge will also bring a certain distance to the case. It’s natural that parents will never give up on a sick child, but when that child is effectively dead it might be time for someone to gently but firmly insist that they face the truth and move on.

    Is that the place of a judge? We can have that argument. But wouldn’t it be better to have that argument in a case that’s genuinely on the bubble rather than a case in which it’s absolutely indisputable that the child’s brain is gone? Why is the pro-life right so attached to cases like this? Keeping the heart beating in a brain-dead child really doesn’t seem to serve anyone’s purpose.

  • Behold the Great Rudy Giuliani Apology Tour

    Remember all that weird stuff Rudy Giuliani spouted on Wednesday? The following morning Donald Trump wholeheartedly agreed with it, but the two of them must have gotten a very stern talking-to later in the day. Today Giuliani took it all back.

    Trump knowing about the Stormy Daniels hush money? Inoperative.

    The hush money being related to the campaign? Not at all. It would have happened regardless.

    Trump firing James Comey because he wouldn’t publicly exonerate Trump? He misspoke.

    Trump himself, of course, is now pretending that he never concocted this idiot plan with Rudy and that Rudy just screwed up when he went on TV:

    What an incredible bunch of morons. I guess what I’m really curious about is why they didn’t just fess up about the hush money in the first place, declare it as a campaign expense, and then pay whatever fine the FEC levied? It’s a fine. Who cares? Instead they’ve gotten themselves mired in a fantastical string of lies that change every day and are now threatening to derail Trump’s presidency.

    I dunno. They’re morons. I guess that’s all the explanation we need.

  • Friday Cat Blogging – 4 May 2018

    Say hello to Moloko, Professor Marc’s new cat. Isn’t he gorgeous?

    Prof M. has had a tough time lately. Mrs. M is allergic to cats, so they’re limited to Siberians. That has turned into a horrible succession of Siberians with inherited disorders, all from allegedly reputable breeders. Only one has survived and recovered. However, this disorder shows up young, so they finally went looking for an older cat. Moloko is seven years old, and his owner wanted to give him up to another home after their other cat died and Moloko seemed lonely. So Prof M and Mrs. M introduced Moloko to Mocha and the M household now has two cats again, both having fun chasing each other around. Don’t you love it when a story finally has a happy ending?

  • Let’s Not Celebrate the 3.9% Unemployment Rate

    There are lots of headlines today that are focused on the headline unemployment rate falling to 3.9 percent. I guess that’s why they call it the headline unemployment rate.

    But this month it’s a mirage. Here’s an excerpt from the household survey that the BLS uses to calculate the unemployment rate:

    Take a look at those numbers. The number of unemployed is indeed down by 239,000, but where did they go? Not to the ranks of the employed, which rose by only 3,000. It turns out they left the labor force entirely, which is why the civilian labor force fell by 236,000 even though the total population grew.

    So, sure, the unemployment rate is down, but it’s because 236,000 people gave up and quit looking for work—which means they no longer get counted as unemployed. This is bad news, not good.

  • My Lost Relativity Masterpiece, Explained

    This common illustration of how gravity works in general relativity is possibly the worst scientific analogy ever invented. It should be banned forever.

    Yesterday I mentioned in passing that the product of my dex-fueled nights a couple of years ago was a tutorial on special and general relativity. Why? Beats me. Ask the dex.

    Several people have asked me to post this masterpiece of pop science, but I’m afraid I can’t do that. Before I explain that, though, here were my goals. I wanted to explain relativity without using a number of tropes that are almost universal in relativity books:

    • No coordinate systems
    • No inertial frames
    • No aether
    • No observers—only measuring devices
    • No trampolines to explain gravity
    • No math more complicated than a square root

    Beyond this, my overarching goal was to try to persuade readers that time and space really are inextricably linked together. That is, I wanted to do it in a way that threaded through the entire tutorial and eventually made sense emotionally, not just mathematically. This is, I think, not truly possible thanks to the wiring of our brains, but I wanted to get close.

    Now, some of these tropes, like the first two bullets, are just plain hard to understand if you don’t have any background. That’s why I wanted to avoid using them. Some, like the traditional “trampoline” to illustrate gravity in general relativity, are wildly misleading. And the last bullet is a two-edged sword: there’s no hard math, but there is math. I’m not convinced that you can do even a modestly decent job of explaining relativity with literally no math.

    Anyway, I mostly accomplished this, although I think I may have caved in and finally mentioned inertial frames toward the end. So why can’t I post it? Is it because I’m embarrassed to let people see my writing? Please. Is it because I’m afraid of criticism? Nah. My overarching goal, after all, was to talk about relativity very differently than it’s normally talked about, and this is bound to attract boatloads of criticism.¹ It’s all part of the job.

    No, the problem is that my tutorial is wrong. At least, I’m pretty sure it’s wrong. There were some parts at the beginning that I never did quite manage to pound into shape before my doctor stopped the dex and I lost interest. There are also one or two things later on that might be right, but might also be wrong, and I never quite figured them out. I don’t mind getting beat up for being wrong, but I’m not really willing to publish stuff that I already know is probably wrong.

    On the bright side, in a couple of weeks I’ll be taking the evil dex once a week, which means I’ll have one night a week that I’m wide awake and twiddling my thumbs looking for something to do. So maybe I’ll take another look and see if I have some new ideas for how to fix it up. If I do, you’ll be the first to know.

    ¹This is especially true among relativity mavens, who all seem to have an almost religious view of the “proper” interpretation of the math.