Congress has a terrible approval rating. Individual representatives and senators each have approval ratings above 50%, generally above 60%, and will all be safely re-elected whenever their term is up. This cognitive dissonance, where "my" representative is the only decent person in congress would be really fascinating, if only it weren't happening in my own country and fucking up my own government.

The party out of the White House always says this. The Democrats abhorred the outrageous spending on military actions and Medicare expansion during the Bush years. The Republicans abhorred the outrageous spending on social programs during the Clinton years. "We spend to much and are dooming our children to poverty and economic collapse," has been a rallying cry since at least Reagan.

Strangely, neither party, once in power, actually reduces spending. Neither party is especially interested in changing those programs that actually affect the budget. What we currently get from the GOP is "We need to cut $1000B from the $600B discretionary budget, so we can afford to reduce revenues by $200B."

This tactic, of claiming the party in-power is destroying the country, while continuing exactly those same behaviors when the tables turn, is the partisan rhetoric that polarizes the people and prevents any rational compromise to solve actual problems.

ENIAC was arguably the first general purpose electronic computer, and it was built for the US military (a wing of their government). Zuse's Z3 was arguably the first general purpose electric computer, and it was built for the German Air Ministry (a wing of their government). Of course, these devices are practical applications representing the culmination of centuries of prior basic research. The basics of electrical and electronic science was performed in the seventeenth and eighteenth centuries, largely by aristocrats who could be considered to be functionally the government of the day, many of whom supported their work with the labor of their tenants, which was pretty nearly taxation at the time. It wasn't until the nineteenth century that practical applications (notably the telegraph) began to appear. Even after the point where commercial interests began to involve themselves, taxation and government programs continued to support development of technologies like transistors, network communication technologies, and photolithography.

If ethical people refuse to work for the NSA, then only unethical people will work for the NSA. One might question whether that is already the case.

It is unlikely that NSA evidence will ever be reported as such in open court. Their whole argument for classifying these programs is that knowledge of them would eliminate their usefulness. Further, presenting that evidence in court would open it to legal challenge, and most of us hope that the only court willing to support these programs is the FISA kangaroo court. No, much better not to present the NSA's part of the evidence as anything more than 'an anonymous tip.' Especially if you can detain indefinitely without charges anyone you think might be convicted based on the secret evidence.

I think Pop Sci, and their explanation of this policy in particular, do a great job of stating that they do want to encourage real debate and discussion of science topics, for exactly the reason that we don't know everything yet. The problem is deciding what constitutes "real" debate and discussion.

If you've ever tried to debate gun control, abortion, or vivisection with someone, you know that facts and logic go right out the window, and every statistic you throw up in support of your position can be countered by a matching and completely incompatible statistic from the other side. Neither side will change their position, which is based more on emotion and personal ethics than reason. It is issues like these that have defined the way in which many people see and understand debate. There's little distinction between repeatably tested facts and weaker forms of evidence. Whoever yells louder or can more vehemently discredit the opposition 'wins.'

Historically, scientific debate has been a (sometimes only slightly) loftier process, largely restricted to experts (loosely defined) and objective evidence. It generally uses more formal language that excludes emotional phrases like "fucking moron." There are people in the general public who have the interest to really follow the arguments and raise excellent and interesting points. Or even just to raise relevant questions that help clarify the discussion for the less expert. PopSci should be lauded for having tried to allow the most open and inclusive discussion possible. Nor is it any surprise that when science is used to support one or another public policy, then the scientific discussion gets clouded by political discussion. People are a lot more passionate about their political positions than their scientific positions, so that side of the debate will quickly overwhelm the less passionate, more technical scientific debate.

I see this decision as PopSci's admission that they can't separate the political and scientific discussions fairly, and will have to revert the scientific discussion to the more formal forum of articles and letters-to-editor. I don't see that as a bad thing - maybe it will help people recognize the difference between scientific debate and political debate.

You may not have noticed, but the major change to surveillance in the past couple decades is that official interest is not longer required. Human attention is no longer required. You need not do anything to rise to "worthy" of NSA snooping: they're doing it already.

Analog phone taps are an excellent demonstration: to tap a phone, you used to have to have a lawyer draft a warrant, have a judge authorize said warrant, pay some guy to drive a over to the subject house and install a physical device on the identified wire, then pay some other guy to record and listen to any conversations. Major expenses that would only be taken if there was reasonable likelihood of getting actionable information. Today, some geek in the back room greps on a database they've already archived.

The reason they haven't come around knocking on your door isn't that you're "a good person," but just that your particular sins have not been grepped yet. You're no more than 3 steps from Aaron Alexis: know someone who knows someone; visited the same blog; bought the same brand of shoe. Enough such coincidences, and all of a sudden, you're worthy of human attention and intervention. Then, god forbid you own a pressure cooker.

It doesn't become surveillance when a human looks at the data, it's surveillance when they collect the data

Of course they will. "Obamacare" is not some kind of monolithic, one-size-fits all program that uniformly matches symptom A with pathology B and treatment C. It's a framework within which private companies are expected to compete, innovate, and distinguish themselves. There's no way a single payer system would ever get implemented, and this rich new market of people who have to buy insurance is a glorious opportunity for the invisible hand to optimize medical care, with some guidance to minimize the fraudulent offerings.

He's allowed to leave; he doesn't have to go down with the ship in the event of every accident. But he's also supposed to be the most capable and informed person on the boat and the most qualified to organize evacuation efforts. It's his responsibility and obligation to do everything he can to ensure the safety of people who have entrusted their lives to his judgement.

A captain abandoning his still-occupied ship is like a homeowner sneaking out the back without telling his guests that the kitchen is on fire.

Ships of this size are rarely dismantled in a drydock. Usually they're run up on the beach at Alang or Chittagong and cut apart, mostly by hand. You can actually see these operations in google maps. Check the satellite view of Alang, Gujarat, India, and you'll see dozens of ships in all stages of disassembly.

Like a terabyte HD filled with /dev/rand. OTP is obviously not a good solution for routine encryption, but is a reasonable option for even fairly large amounts of sensitive data. Even video conferencing. Although frankly, if your pad is a couple of terabytes, you can probably reuse it safely, at least once or twice. It does require a shift of paradigm - many of us have gotten used to the notion that nearly-unbreakable encryption is "easy." Wrap your data in a 2048-bit symmetric-key algorithm, and bing-bang-bop, you're safe. The revelation that this might not be true will encourage people to return to tiered protocols, presumably concentrating "real" secrets into a narrower-bandwidth, more identifiable, but also more secure channel. If being one packet among trillions no longer provides any meaningful sense of anonymity, then you might as well put all your secret data in a big red box and protect that box very well.

Remember that a lot of science is about coming up with actual evidence to support or refute the way you think the world works.

This is a little different, though. Wings, for example, have definitely evolved independently several times. This is why bird wings look nothing like cockroach wings. The genetic code that represents cockroach wings and bat wings are not very similar, but it's a great example of convergent evolution producing qualitatively similar function through very different processes. Likewise, it could happen that bats and dolphins solve the echolocation problem in completely different ways. What these people show is that, if you start with the same fundamental structures (ie, mammals), then 1) the modifications that allow for good echolocation are quite limited, and 2) happen frequently enough that they can be selected in multiple populations in a conserved pattern.

I think this is pretty exciting. Most mutations turn out to be non-functional or lethal; only a small percentage happen in the coding domains and produce a functional-but-different protein. This work is evidence that, on a large scale and across whole families of genes, those same mutations occur, provide an advantage, and get reinforced in the population. And that it happens over just a few million years. It suggests that, in order to develop echolocation, none of these species developed a whole new organ system, but just refined their existing structures. Gradual changes over time, without any wholesale de novo introduction of features.

Yeah, you might expect that all to be true beforehand, because it's completely consistent with the way we think genetics and evolution work, but now there's quantitative data to support it.

Underlying genetic evolution is the notion that genes pick up random mutations over time. Most of these have no effect on function, so you can estimate how long ago two species diverged by counting how many differences are in the genome. These guys had the clever idea of taking species that we think diverged a long time ago, but that have a similar trait (ie, echolocation), with the hypothesis that the genes controlling that might be more similar, even in these very different animals, than the genes for dissimilar traits.

Imagine a software project that forks and is maintained by separate groups. Over time, the two projects look more and more different. Now imagine that both of these forks end up with a new feature in common that didn't exist in the pre-fork code. The study hypothesis is essentially that code related to the new feature will be similar between the two projects, where code associated with other features that aren't the same between projects, will be more different

Genetically, this might happen either because the random mutations in hearing genes that facilitate echolocation facilitate echolocation in any environment, provide a survival advantage, and become conserved in multiple environments. As dolphins and echolocating bats diverged, acute hearing was favored in both species, so their hearing genes are more similar than those of echolocating and non-echolocating bats, even though the genes for "wings" are more similar between bats than dolphins. Because it seems unlikely that the large number of differences that separate bats as a group from dolphins might have come up separately, the study proposes that the first bat-ears were as different from dolphin-ears as bat wings are different from dolphin flippers, and that the specialization into echolocating bats brought those hearing genes closer together, following a convergent path.

This is a much more subtle form of convergent evolution that, say, wings. "Wings" as a feature provide a definite advantage, and wing structures have evolved multiple times in multiple forms. The genes that define insect wings are completely different from those that define bat wings. It's a dramatic demonstration of nature using multiple solutions to the same problem. The current study suggests that nature is also capable of finding the same solution from multiple starting points.

In the US, the median income is $40k. $1M tax requires an 'extra' 25 jobs beyond what people would take to feed and clothe themselves. The workplace fatality rate is 3.5 per 100,000 (source), or 1 per 28,600. This means you get $1.1B of revenue per fatality. US personal tax receipts are almost $2T, so you could argue that the federal government kills almost 1800 people per year through the tax burden.

This income include social security and medicare, and I'm quite certain that spending on those two programs alone saves more than 2000 people/year.

Thus demonstrating Scheier's point that we're really not very good at estimating risk

Per capita spending on students is about $10k/year, summed across all levels of government. For comparison, per capita spending is $16k/year for medicare recipients. Government places a 60% higher priority on keeping retired people healthy than on educating the next generation. In my school district, I pay three times as much for trash collection as for schools. The total spending on education may be a large number, but it is a shockingly small fraction of government spending.

You know, there was a time when education was supposed to produce well-rounded, generally capable human beings. Physical fitness is part of that. Competition is part of that. Very often, the people who compete well in reading don't compete well in baseball, and vice-versa, so the presence of an athletic program helps people find their own areas of strength.

Sport has definitely taken excessive dominance in some districts, but that's not a good reason to extract those programs from all districts. This notion that school is essentially a job training program is not healthy. School is a largely consequence-free opportunity to expose (or force) young people into diverse activities to help them find their own aptitudes and interests.

-Robert A. Heinlein