Tyler Cowen links today to a paper suggesting that knowledge of a person’s social network is more important than specific knowledge about the person herself. For some reason I clicked. It turns out the paper is written by a bunch of math and physics types and includes stuff like this:

The predictive information contained within a user’s text can be characterized by three related quantities, the entropy rate h, the perplexity 2^h, and the predictability Π….We applied a nonparametric entropy estimator that incorporates the full sequence structure of the text. This estimator has been proved to converge asymptotically to the true entropy rate for stationary processes and has been applied to human mobility data.

In other words, it’s pretty much incomprehensible. However, it’s incomprehensible in a way that appeals to me, so I read it. I’ve got an hour or two to kill right now, so why not?

The authors are looking at something very specific when they talk about “predictability.” They analyzed the Twitter feeds of about a thousand people, and for each tweet they tried to predict what the next word in the tweet would be given the person’s past writing. Their question is, if you look solely at the writing of the person’s friends on Twitter, how accurately can you predict what they’re going to say? It turns out that this depends on how many friends you look at, with predictability topping out at around 15 friends.

So how good is predictability when you look at 15 friends? Pretty damn good:

At around nine friends, the friends were better predictors than the person’s own past writings, represented by the black line labeled Π (ego). At 15 friends, the friends are considerably better, and adding the person’s own past writing increases predictability by only 3.2 percentage points. The authors point out what this means:

This may have distinct implications for privacy: if an individual forgoes using a social media platform or deletes her account, yet her social ties remain, then that platform owner potentially possesses 95.1% of the achievable predictive accuracy of the future activities of that individual.

In a nutshell, this means you can delete your Facebook account all you want, and it barely matters. If I already know who your friends are, I have a continuing stream of information about what things you’re likely to buy and which candidates you’re likely to vote for.

This paper, obviously, uses a very restricted meaning of “predictability.” Nonetheless, it’s suggestive that other kinds of predictability may be accessible through your friends too. If your friends like guns or beauty supplies, there’s a pretty good chance that you do too. Likewise, if some hot new fashion hits the stores and your friends all hate it, you probably hate it too. And if your friends seem to be undecided about a political candidate and are potentially persuadable with a specific message? Well, you probably are too, even if you insist that you are truly a unique snowflake uninfluenced by fads and peer pressure.

In other words, we are all doomed. Unless you can convince all your friends to abandon social media, it barely matters if you yourself do. Big Brother is still watching.

Sen. Dianne Feinstein in 2017, introducing legislation to ban the sale and possession of bump-stocks. Her bill went nowhere, of course, thanks to NRA opposition and Republican cowardice.Bill Clark/Congressional Quarterly/Newscom via ZUMA

Donald Trump this morning:

Obama Administration legalized bump stocks. BAD IDEA. As I promised, today the Department of Justice will issue the rule banning BUMP STOCKS with a mandated comment period. We will BAN all devices that turn legal weapons into illegal machine guns.

— Donald J. Trump (@realDonaldTrump) March 23, 2018

This is actually sort of true, more or less. That’s unusual for Donald Trump. But the reason it’s true is that Obama didn’t interfere when the ATF concluded in 2010 that the law gave them no authority to regulate bump stocks. The Kenyan socialist gun-hating dictator-in-chief then followed the law! Hard to imagine, isn’t it?

Needless to say, Sen. Dianne Feinstein and others have introduced legislation over the years to ban bump stocks as part of a very modest assault weapons ban, but these bills never saw the light of day thanks to NRA and Republican opposition. Last year, Feinstein introduced a bill limited just to banning bump stocks, but it also died thanks to NRA opposition and Republican timidity. Big surprise, eh? So now Trump is going to try to do it via executive order.

Will this work in the face of previous rulings that bump stocks are legal under current law? Will the NRA sue to overturn the EO? If they do, who will the courts side with? We could know the answers to all these questions next week if Republicans simply passed a bill to ban bump stocks. The original ATF ruling was based on statutory interpretation, not constitutional law, so a new law would clean up any confusion instantly. But Republicans are afraid to take even this tiny step for fear of crossing the NRA, so instead we probably won’t know for years.

I’ve been playing around some more with high-speed photography ever since I discovered that my camera’s shutter speed goes up to 1/32,000th of a second. Below you can see the results. I’ve got two hummingbird pictures, both full-frame photos taken within a few feet. I’ve got a honeybee flying around. And I’ve got a wasp (?) of some kind. All were taken at 1/16,000th of a second. They’re shockingly good.

But I need some help. I’d love to putter around with this some more, but what can I use it for? It needs to be something outdoors, since ultra-high shutter speeds require a lot of ambient light, and it needs to be something really fast. Cars and soccer balls might as well be molasses at these speeds. I need fast. Anyone got any good ideas?

UPDATE: In comments, Chris Hill suggests the “wasp” is really a hoverfly. I’ll buy that. But its shape is very distinctive, and so are its markings (two slanted white stripes followed by three horizontal white stripes). I can’t find one that looks anything like that.

UPDATE 2: It appears to be a four-spotted aphid fly.

Last night I mentioned something to a friend about the driverless car crash in Arizona:

The fact that it’s an Uber car doesn’t surprise me. They’re exactly the kind of company that would cut corners: cheap optics, lousy safety drivers, pushing to drive at night before the software is ready, etc. A corporate culture that’s built to expand their taxi service no matter who gets hurt or who complains might build a good taxi service, but it’s the worst possible culture for mission-critical software.

My point was that Uber famously has a corporate culture built around being a bull in a china shop, pushing the limits of expansion as far as they could and focusing obsessively on doing everything at light speed, before authorities could stop them or competitors could catch up. This is not the kind of company that’s likely to slow down to make sure their driverless cars work properly before pushing to the next stage. And sure enough, here’s the New York Times today:

Uber’s robotic vehicle project was not living up to expectations months before a self-driving car operated by the company struck and killed a woman in Tempe, Ariz. The cars were having trouble driving through construction zones and next to tall vehicles, like big rigs. And Uber’s human drivers had to intervene far more frequently than the drivers of competing autonomous car projects….As of March, Uber was struggling to meet its target of 13 miles per “intervention” in Arizona.

….There also was pressure to live up to a goal to offer a driverless car service by the end of the year and to impress top executives. Dara Khosrowshahi, Uber’s chief executive, was expected to visit Arizona in April, and leaders of the company’s development group in the Phoenix area wanted to give him a glitch-free ride in an autonomous car. Mr. Khosrowshahi’s trip was called “Milestone 1: Confidence” in the company documents.

….When Uber moved to a single operator, some employees expressed safety concerns to managers, according to the two people familiar with Uber’s operations. They were worried that going solo would make it harder to remain alert during hours of monotonous driving.

This kind of culture might be OK for, say, Facebook, which doesn’t kill people if there are glitches in its apps. But if you’re launching a satellite into space or putting a driverless car on the road, there’s a whole different development and testing ethos involved. Uber just doesn’t have that. I’d barely trust them to develop software for a driverless golf cart, let alone a driverless car.

The Times reports that Uber was planning to apply for a license to run a driverless car service in December. December! A year ago their cars couldn’t go 13 miles without a problem, but they still thought they’d be ready for full driverless operation in December? That’s crazy. Off the top of my head, I’d want to see something like an average of 10,000 miles between incidents before I thought I was within a year of applying for a limited license.

But not Uber. That’s just not their style.

You might wonder why I keep harping on lead and crime. After all, as much as I’d like to get rid of the last remaining traces of environmental lead, the truth is that it’s mostly gone these days. What’s more, the crime wave of the 70s and 80s is long past. It’s historically interesting that lead poisoning played a big role in skyrocketing crime rates 40 years ago, but not all that pressing as a current policy issue.¹

There are a couple of reasons for my continued harping, and today I want to mention just one of them. As my post earlier this morning made clear, teenagers these days are less violent and better adjusted than they used to be, and this is a permanent change. A few decades ago, people were chronically apprehensive around teens in public places, afraid they might be assaulted or even killed if they so much as looked at them funny. And there was something to that. Back then, teenage brains had all been damaged by a lifetime of lead poisoning, often making them impulsive and violent. But that’s long in the past, and there’s no longer any excuse for this apprehension. Without lead poisoning to wreck their brains, they’re just ordinary teenagers, like teenagers of every past era

This is something that I wish everyone could internalize. Teenagers just aren’t unusually dangerous these days. If you chase one into a backyard and you see a glint in his hand, you probably don’t need to unload 20 rounds into his body as if you were trying to bring down a PCP-crazed rhinoceros.

At the risk of being misunderstood, I want to add that this is very much a racial thing. The lead epidemic hit blacks harder than whites, and this meant that the violence level of black teenagers rose more than it did for white teenagers. In the early 90s, even Jesse Jackson was famously scared of black teenagers. Cops internalized this, mixed it up in a stew with lots of old-school racism, and ended up killing a lot of black teenagers.

And they still do, even though the violence level of black teens also dropped more than it did for white teens once we removed lead from gasoline. In the year 2018, there’s just no excuse for cops or anyone else to routinely treat black teenagers as scary hoodlums who might kill them at the drop of a hat. They’re back to being ordinary people these days, just like teenagers of every other color.

Now, teenagers are still teenagers. They’re full of hormones, their brains aren’t fully integrated, they like to rebel against authority, and some of them are dangerous. And unlike teens of previous eras, they have easy access to handguns of all types. For those reasons and more, police officers have to be alert and wary around all teenage crime suspects. But do they need to unleash a hail of bullets at the slightest hint of danger from a black teenager? No. They just don’t. Arguments based on social justice might or might not mean much to most cops, but I’m offering them another one: teenagers of all races, and especially black teenagers, are fundamentally, permanently, a lot less violent than they used to be. It’s time to recognize that and adjust our policing strategies accordingly.

¹No need to email me about this. Obviously I agree that lead is an important public policy issue. All I mean is that as far as changes in crime and violence are concerned, lead is no longer a major factor.

I was up in Seattle last weekend visiting friends, and on Sunday afternoon I had lunch with one of my old roommates from Caltech days. I hadn’t seen him for about 40 years. Unlike me, he stuck it out and graduated, and now runs a company that makes fancy optical equipment (he’s an optical physics guy).

More to the present point, this being a Friday, he’s married and it turns out that his wife, Charlotte, is very much a cat person. Their current cat is named Scooter, and I was warned that he was a bit “grabby.” This turned out to be true, but Charlotte says he’s gotten better after ten years of work. That was true too: in my case, at least, he wasn’t at all serious about it. Just a lazy little flick of the paw that didn’t even come close to making contact. He eventually curled up next to me and started purring.

Naturally I took pictures.

Hooray! Everybody is getting more money in the 2018 spending bill. Well, almost everybody:

This comes from Emily Horton of the Center for Budget and Policy Priorities, who says this:

Policymakers need to recognize the depth of the IRS’s budget and workforce depletion, as well as the multi-year and multi-dimensional nature of the response required to successfully implement and enforce the new tax law. Rather than continue squeezing the agency’s funding, policymakers should give the IRS sufficient resources to perform its core functions of collecting revenues and enforcing the nation’s tax laws.

Unfortunately, our current policymakers do recognize the depth of the IRS’s workforce depletion. In fact, cutting their budget is designed to deplete their workforce. Why? Because our current policymakers are Republicans, and Republicans don’t want the IRS to perform its core function of enforcing the nation’s tax laws. After all, that would mean more audits of rich people, and that’s not something they want.

This is neither new nor a secret. Republicans have conducted a jihad against the IRS for decades, primarily because they don’t want their rich donors to be pestered with audits. It’s the next best thing to just cutting their taxes outright.

President Trump has signed the 2018 spending bill, but he wants everyone to know he’s not happy about it. You see, Democrats hate America and keep demanding that we fund domestic programs like national parks and food stamps and clean water and other frills that we don’t need. This makes it hard to build border walls and additional aircraft carriers, especially after cutting taxes by $1.5 trillion. The answer, Trump says, is to eliminate the filibuster so that Republicans can just pass anything they want. Also, he wants a line-item veto, even though that’s unconstitutional.

So that’s that for the big press briefing.

UPDATE 1: I’m watching right now, and Trump is laboriously reading off all the various military toys that have been funded by this bill. Helicopters, ships, planes, tanks: you name it, Trump is going to build it.

UPDATE 2: He’s very happy about “the opioids.”

UPDATE 3: He’s not happy with the $1.6 billion for the wall.

UPDATE 4: DACA didn’t get funded because of Democrats. Republicans wanted to fund it, but Democrats just wouldn’t do it. There is not a single Hispanic in the entire country who is going to buy this.

UPDATE 5: And that’s a wrap. POTUS has left the stage.

The Economist discovers that teenagers in the West are a lot less worrisome than they used to be:

Young people are indeed behaving and thinking differently from previous cohorts at the same age. These shifts can be seen in almost every rich country, from America to the Netherlands to South Korea. Some have been under way for many years, but they have accelerated in the past few … [They] are getting drunk less often … Other drugs are also falling from favour … Fighting among 13- and 15-year-olds is down across Europe … Teenagers are also having less sex, especially of the procreative kind … In short, young people are less hedonistic and break fewer rules than in the past. They are “kind of boring”, says Shoko Yoneyama, an expert on Japanese teenagers at the University of Adelaide. What is going on?

Indeed. What is going on? The Economist provides a few options:

• One possible explanation is that family life has changed….Fathers have upped their child-care hours most in proportional terms….Those doted-upon children seem to have turned into amenable teenagers.
• Another possibility is that teenagers and young people are more focused on school and academic work.
• Today’s young people in Western countries are increasingly ethnically diverse….Many of those immigrants arrive with strong taboos against drinking, premarital sex and smoking—at least among girls.
• Social media allow teenagers’ craving for contact with peers to be squared with parents’ desire to keep their offspring safe and away from harmful substances….Teenagers who communicate largely online can exchange gossip, insults and nude pictures, but not bodily fluids, blows, or bottles of vodka.

Do I even need to bother pointing out that not a single one of these explanations makes any sense? Teenagers from single-parent homes are also better behaved. A focus on school is an effect, not a cause. Non-immigrants are better behaved too. All of this better behavior predates the invasion of social media. And it’s happening mostly in rich countries, not, say, in the Middle East, even though they also have taboos and cell phones. These explanations aren’t even worth tossing out just to knock them down. They’re completely ridiculous.

The lengths that psychologists, sociologists, and other academics will go to in order to protect their turf is remarkable. The obvious answer here is: Today’s teenagers are the first generation in more than half a century to grow up completely lead free.¹ This is, sadly, not a sociological explanation. It doesn’t provide much scope for grand theorizing.² It requires you to believe in an “essentialist” explanation.³ It will produce no new research papers.

On the other hand, it has the virtue of almost certainly being true. So there’s that.

¹Well, not completely. But pretty close.

²In fact, it’s worse than that: it demolishes a whole bunch of pet theories built up over the years by liberals, conservatives, academics, and guys on barstools.

³“Essentialist” is generally used to mean something that’s inherent rather than environmental or societal. It has a bad odor thanks to all the people who claim that blacks do worse than whites because they’re biologically inferior: that is, their problems are essential to their genetic heritage, rather than being the result of white racism. Unfortunately, this has made academics suspicious of any non-sociological argument for anything because they’re afraid it provides a slippery slope to claims that blacks are genetically inferior to whites. This is understandable, but considering that everyone accepts the effects of lead poisoning on IQ and schoolwork—not to mention that lead poisoning is environmental, not genetic—it’s a little hard to understand why so many people resist the possibility that it also has other effects.

Yesterday:

Today:

Ha ha ha ha … ha. What an idiot. The bill that was passed yesterday, needless to say, was the same bill as the day before and the day before that.

The good news, I suppose, is that this is probably based on nothing more than Trump’s love of drama. He wants Mitch McConnell and Paul Ryan to call and beg him to sign the bill. They’ll sigh, make the call, and I suppose Trump will then sign it.