A few days ago, NBC News quoted a former intelligence official about the fallout from Edward Snowden's NSA leaks. "The damage, on a scale of 1 to 10, is a 12," he said.

At the time, I thought it was an odd thing to say. Obviously Snowden's leaks have been damaging to the NSA, and just as obviously they've caused the NSA enormous PR problems. Still, we've known for years that they were collecting telephone metadata. We've known they were subpoenaing email and online documents from tech providers like Google and Microsoft. We've known they were monitoring switching equipment and fiber optic cables. We certainly know a lot more details about this stuff than we used to, but the basic outline of NSA's capabilities hasn't really come as much of a surprise.

So what was this former intelligence official talking about? I suspect it was this:

The agency has circumvented or cracked much of the encryption, or digital scrambling, that guards global commerce and banking systems, protects sensitive data like trade secrets and medical records, and automatically secures the e-mails, Web searches, Internet chats and phone calls of Americans and others around the world, the documents show.

....Some of the agency’s most intensive efforts have focused on the encryption in universal use in the United States, including Secure Sockets Layer, or SSL; virtual private networks, or VPNs; and the protection used on fourth-generation, or 4G, smartphones.

....By this year, the Sigint Enabling Project had found ways inside some of the encryption chips that scramble information for businesses and governments, either by working with chipmakers to insert back doors or by exploiting security flaws, according to the documents. The agency also expected to gain full unencrypted access to an unnamed major Internet phone call and text service; to a Middle Eastern Internet service; and to the communications of three foreign governments.

....[In 2010, a] briefing document claims that the agency had developed “groundbreaking capabilities” against encrypted Web chats and phone calls. Its successes against Secure Sockets Layer and virtual private networks were gaining momentum.

But the agency was concerned that it could lose the advantage it had worked so long to gain, if the mere “fact of” decryption became widely known. “These capabilities are among the Sigint community’s most fragile, and the inadvertent disclosure of the simple ‘fact of’ could alert the adversary and result in immediate loss of the capability,” a GCHQ document warned.

That's a 12 on a scale of 1 to 10. The Snowden documents don't make clear precisely what NSA's capabilities are, or exactly what kind of encryption it can break. Nor is it clear how many of its new capabilities are truly due to mathematical breakthroughs of some kind, and how many are more prosaic hacking exploits that have given them more encryption keys than in the past.

Nonetheless, this is truly information that plenty of bad guys probably didn't know, and probably didn't have much of an inkling about. It's likely that many or most of them figured that ordinary commercial crypto provided sufficient protection, which in turn meant that it wasn't worth the trouble to implement strong crypto, which is a bit of a pain in the ass. (Recall, for example, Glenn Greenwald's admission that he "almost lost one of the biggest leaks in national-security history" because Snowden initially insisted on communicating with strong crypto and Greenwald didn't want to be bothered to install it.)

But now that's all changed. Now every bad guy in the world knows for a fact that commercial crypto won't help them, and the ones with even modest smarts will switch to strong crypto techniques that remain unbreakable. It's still a pain in the ass, but it's not that big a pain in the ass.

For what it's worth, this is about the point where I get off the Snowden train. It's true that some of these disclosures are of clear public interest. In particular, I'm thinking about the details of NSA efforts to infiltrate and corrupt the standards setting groups that produce commercial crypto schemes.

But the rest of it is a lot more dubious. It's not clear to me how disclosing NSA's decryption breakthroughs benefits the public debate much, unlike previous disclosures that have raised serious questions about the scope and legality of NSA's surveillance of U.S. persons. Conversely, it's really easy to see how disclosing them harms U.S. efforts to keep up our surveillance on genuine bad guys. Unlike previous rounds of disclosures, I'm a lot less certain that this one should have seen the light of day.

The Economist writes about robots:

No matter how flexible, easy to program and safe they are, collaborative workers may not be welcomed by human workers to begin with. The experience of Alumotion, an Italian distributor of UR’s robots, is illustrative. Workers fear being replaced by robots, says co-owner Fabio Facchinetti, so his salespeople carry demonstration units in unmarked cases and initially only meet a potential client’s senior management behind closed doors.

Roger that. So how do we make humans more accepting of robots? Part of the answer, as near as I can tell, is the usual: other, higher ranking humans will tell lies about how the robots will never, ever take away your job. They'll just help you do your job better! But there's also this:

Workers generally warm to collaborative robots quickly....And because workers themselves do the programming, they tend to regard the robots as subordinate assistants. This is good for morale....To keep human workers at ease, collaborative robots should also have an appropriate size and appearance. Takayuki Kanda of the ATR Intelligent Robotics and Communication Laboratories in Kyoto says that collaborative, humanoid robots should generally be no larger than a six-year-old, a size most adults reckon they could overpower if necessary.

....It turns out, for example, that people are more trusting of robots that use metaphors rather than abstract language, says Bilge Mutlu....He has found that robots are more persuasive when they refer to the opinions of humans and limit pauses to about a third of a second to avoid appearing confused. Robots’ gazes must also be carefully programmed lest a stare make someone uncomfortable.

....When a person enters a room, robots inside should pause for a moment and acknowledge the newcomer, a sign of deference that puts people at ease....It is vital that a robot of this sort is not perceived as hostile, but as having its owner’s best interests at heart....One way to do this is to give robots a defining human trait—the ability to make mistakes. Maha Salem, a researcher under Dr Dautenhahn, programmed a humanoid Asimo robot, made by Honda, to make occasional harmless mistakes such as pointing to one drawer while talking about another. When it comes to household robots, test subjects prefer those that err over infallible ones, Dr Salem says.

So this is how robots will eventually become our overlords. They will keep themselves small and supposedly easy to overpower. They will traffic in charming metaphors. They will pretend to care about our opinions. They will avoid eye contact. They will feign deference. They will simulate charming clumsiness. And, of course, they will mount a massive PR campaign aimed at getting Hollywood to portray robots not as the relentless killing machines they are, but as harmless, friendly little eco-bots. They will do all this while Skynet takes over behind the scenes. You have been warned.

The American economy added 169,000 new jobs last month, but about 90,000 of those jobs were needed just to keep up with population growth, so net job growth clocked in at 79,000. That's about the same as last month—except for the fact that last month's numbers got revised sharply downward today. Net job growth in July was reduced from 72,000 to 14,000, barely the breakeven point, and just for good measure, June's numbers were revised down a bit too.

The headline unemployment number declined slightly to 7.3 percent, but for the worst possible reason. It's not that more people were at work in August. In fact, fewer people were employed than in July. Normally this would produce a lower employment rate (and therefore a higher unemployment rate), but because lots of people exited the labor force entirely, the size of the civilian labor force dropped by 312,000 people. Here's how the employment arithmetic works out:

July: 144,285 / 155,798 = 92.61%

August: 144,170 / 155,486 = 92.72%

So there are fewer people working, but because the size of the labor force dropped so much, the employment rate actually went up by a tenth of a point. Likewise, the unemployment rate went down by a tenth of a point.

The question is why the labor force shrank. As it turns out, it's not because there were more discouraged workers. There were fewer. Nor were there more people forced into part-time work because of the bad economy. There were fewer. There were more people who switched to part-time work for noneconomic reasons, but presumably that doesn't reflect one way or the other on the state of the economy.

It's a little bit mysterious, and maybe someone with better economic chops will explain it all later in the day. In the meantime, there's one takeaway from this that's simple: This is a really weak jobs report. It's crazy that we've all but given up on both monetary and fiscal policy designed to fight this weakness.

Hale Stewart provides some reason today to be moderately bullish on the economy: the massive deleveraging that started when the recession hit has now pretty much run its course:

What started this recession was a massive build-up of debt in the system, largely related to the housing bubble. When the bubble burst, people had to sell the asset(s) underlying their debt, usually at a loss....This meant they had less money to spend on other items, meaning there was overall less economic demand, culminating in weak overall growth.

....Total household and non-profit debt, according to the Federal Reserve's Flow of Funds report, has been dropping since a little bit before the recession began....Total household debt has been declining as a percentage of GDP since mid-way through the last recession....Household obligations as a percent of disposable personal income are at very low historical levels.

....Put in economic terms, the above charts show the "balance sheet" recession is close to ending.

There are still some underlying structural weaknesses in the economy that should keep anyone from getting too excited: employment is weak; wages are stagnant; automation is starting to become a real threat; deleveraging still has a ways to go overseas; and income inequality is back on the rise. Still, leverage was the core reason the 2008 recession was so severe and the subsequent recovery was so weak. The fact that household leverage is back to its historical average is good news for the future.

The Guardian and the New York Times have blockbuster stories today about the NSA's surprising success in cracking encryption standards that until now everyone has considered safe. More about this later. But here's a single tidbit from the Times story:

In one case, after the government learned that a foreign intelligence target had ordered new computer hardware, the American manufacturer agreed to insert a back door into the product before it was shipped, someone familiar with the request told The Times.

I guess the lesson here is not to buy network equipment from US companies. I don't imagine that US purveyors of network equipment are very happy about this revelation.

Awkward question: Was this story timed to coincide with the G20 meeting in Russia?

From the LA Times today:

In a major shift in how California's 6.2 million public school students are taught and tested, state officials plan to drop the standardized exams used since 1999 and replace them with a computerized system next spring.

The move would advance new learning goals, called the Common Core, which are less focused on memorizing facts. They are designed instead to develop critical thinking and writing skills that take formerly separate subjects — such as English and history or writing and chemistry — and link them. Forty-five states have adopted these standards.

Click the link to read more about the clusterfuckish nature of this whole thing. But regardless of how you feel about Common Core, why the switch to computerized tests? Can't you test Common Core knowledge using pencil and paper? Beats me. But it's apparently going to cost the LA school district some serious money. Here's a story from yesterday:

Los Angeles school officials are acknowledging a new looming cost in a $1-billion effort to provide iPads to every student: keyboards. Officials so far have not budgeted that expense, but they said the wireless keyboards are recommended for students when they take new state standardized tests.

When I read that, I wondered why they suddenly needed iPads to take standardized tests. I guess now I know. Sort of.

In any case, I'd like to open up this thread to teachers or anyone else who wants to weigh in on the benefit of giving every kid an iPad. I think this is just about the most colossally dumb use of money I've come across in a long time. But naturally I want to keep an open mind. So educate me. Someone tell me why I'm wrong.

Yet another Republican commissioner of insurance—this time in Wisconsin—has announced that Obamacare will bring eye-popping increases in insurance rates to the beleaguered residents of his state. Premiums will increase by 50 percent for most people and will double for the least lucky cheeseheads. It's gonna be a catastrophe.

I don't have the energy to figure out how the numbers are being cooked this time around, and the announcement rather carefully provides no detail about how the commissioner's office came up with its startling figures. But it's certainly remarkable that these skyrocketing rates only seem to affect states with Republican administrations, isn't it? Just remarkable.

In any case, while we wait for someone to figure out the precise nature of the gameplaying going on in Wisconsin, an actual report done by an organization that's actually trying to compile accurate information concludes that premiums under Obamacare "are generally lower than expected." That's from the Kaiser Family Foundation, which provides several handy charts with expected average rates. Here's a sample from a few of the states nearest to Wisconsin:

Those seem pretty reasonable. So how did Wisconsin supposedly end up with such high rates? My guess is that they lowballed the current rates; did their comparisons between different kinds of coverage; and cherry picked the buyers to get the worst possible results. But that's just a guess. I'm sure that eventually someone will dig into this and get us a real answer.

Judd Legum reports that the latest vote estimate in the House on Syria has about 200 leaning against and 50 leaning in favor. Only about a dozen Republicans are leaning in favor.

There's obviously a bit of hypocrisy on both sides in this affair, but I have to say that watching Republican pols and conservative pundits get on their high horses about Syria has been pretty nauseating. These are guys who mostly have never met a war they didn't like, and until a few months ago were practically baying at the moon to demand that that President Obama stop diddling around and get serious about aiding the rebels and taking out the monstrous Bashar al-Assad. But now? Butter wouldn't melt in their mouths as they talk piously about the value of multilateral support; the need to give diplomacy a chance; the perils of regional blowback; the lessons of Iraq; and the fear of escalation if Assad retaliates. You'd think they'd all just returned from a Save the Whales conference in Marin County.

There are some Republicans who are perfectly serious about their desire not to get entangled in yet another Middle Eastern conflict. But most of them couldn't care less. Obama is for it, so they're against it. It's pretty hard to take.

So why did Bashar al-Assad launch a chemical weapon attack in the first place? It's a bit of mystery. McClatchy rounds up the evidence, including a new report from Der Spiegel about a phone call intercepted by German intelligence:

According to Der Spiegel, one of the parties in the intercepted phone call was a “high-ranking member of Hezbollah,” the militant Lebanese movement that’s sent fighters to support the Assad government. That Hezbollah member told the Iranian that “Assad had lost his temper and committed a huge mistake by giving the order for the poison gas use," according to the magazine’s account.

The U.S. intelligence assessment reached a similar conclusion, finding that the alleged use of chemical weapons may have been in part because of “the regime’s frustration with its inability to secure large portions of Damascus.”

....The German account goes further than others that have been released recently in providing details of Assad’s state of mind that might have played a role in the motivation for launching a chemical attack, noting that Assad sees himself embroiled “in a crucial battle for Damascus.”

It also said Assad’s forces had used a highly diluted chemical agent in previous attacks on rebels and that the high death count Aug. 21 might have been the result of “errors made in the mixing of the gas” that made it “much more potent than anticipated.” That would be consistent with a suggestion from an Israeli official, cited by The New York Times, that the attack was “an operational mistake.”

So it was all one big FUBAR, launched by a sociopath who lost control of himself and then bungled by a military unit that was incompetent. And now we're deciding what we ought to do about it.

Today, Yale law professor Dan Kahan presents evidence that we Americans really suck at math. "Correctly interpreting the data was expected to be difficult," he says of the test subjects from his latest study, and he turned out to be right. But all he was asking them to do was calculate a simple percentage. If 200 out of 300 people in one group get better by taking a pill and 100 out of 125 people in a different group get better by doing nothing, which is better? Taking a pill or doing nothing? You'd pass Kahan's test if you understood that the raw numbers aren't enough to get the right answer. You have to calculate the percentage of each group that got better.

It's disheartening that most people couldn't figure that out, though hardly unexpected. But what came next is....well, not unexpected, maybe. But certainly discouraging. Kahan ran the exact same test with the exact same data, except this time the question was about gun bans and crime levels. Half of the time, he presented data suggesting that a gun ban increased crime, while the other half of the time the data suggested that a gun ban decreased crime. And guess what? Among the subset of test subjects who were very good at math, they suddenly got really stupid if they didn't like the answer they got. Here's the chart:

This comes via Chris Mooney, who describes the whole thing in much more detail here. However, there's one big caveat: If I'm interepreting the dataset correctly, the sample size of highly numerate subjects is very small. Roughly speaking, there were about 30 liberals and 30 conservatives who were highly numerate and were given the gun ban version of the test. That's not a lot.

On the other hand, the effect size is pretty stunning. There's a huge difference in the rate at which people did the math correctly depending on whether they liked the answer they got. I'd like to see some follow-ups with more subjects and different questions, but it sure looks as if we'd probably see the same dismal effect.

How big a deal is this? In one sense, it's even worse than it looks. Aside from being able to tell that one number is bigger than another, this is literally about the easiest possible data analysis problem you can pose. If ideologues actively turn off their minds even for something this simple, there's really no chance of changing their minds with anything even modestly more sophisticated. This is something that most of us pretty much knew already, but it's a little chilling to see it so glaringly confirmed.

But in another sense, it really doesn't matter at all. These days, even relatively simple public policy issues can only be properly analyzed using statistical techniques that are beyond the understanding of virtually all of us. So the fact that ideology destroys our personal ability to do math hardly matters. In practice, nearly all of us have to rely on the word of experts when it comes to this kind of stuff, and there's never any shortage of experts to crunch the numbers and produce whatever results our respective tribes demand.

We believe what we want to believe, and neither facts nor evidence ever changes that much. Welcome to planet Earth.