Low carbohydrate diets may have benefits for the brain as well. They are now recognized as a powerful tool for treating epilepsy and there are reports that in cases it can actually cure the disorder. Some people with bipolar II have found that low carbohydrate diets can be more effective than drugs for managing the disorder. There is also evidence that low-carbohydrate diets can be used to treat Alzheimer's.

The disturbing implication here is that if switching from a high-carbohydrate Western diet can fix brain disorders, then maybe the Western diet and in particular low-fat diets are causing a lot of these problems in the first place. Certainly there's not much evidence that cutting fat out of diets has made us any healthier.

The question is, how informative is an imitation game? If we've shifted the goalposts I think it's because these ones just aren't very informative. The implicit argument of the Turing Test is that a sufficient good imitation of intelligence, indistinguishable from the real thing, must at some level be intelligent. Basically, fake it until you make it. Like in Blade Runner- the replicants remember, love, fear, hate; they do everything humans do in a way indistinguishable from us, so they must at some level be human. Pretend long enough and hard enough and eventually it becomes the real thing. It's a plausible argument on the surface, but sort of blows up on closer examination.

For example, years ago there was an effort by psychologists to rehabilitate sociopaths with extensive talk therapy, teaching them to share and be more in touch with their feelings. It didn't really work, though. It taught them how to speak the language of empathy and say the right things to convince people that they were caring human beings, but underneath they were still sociopaths, given a chance they'd go off on another killing spree. Learning to act like caring human beings didn't actually make them caring beings. They were still monsters, but now they just had better camouflage. Imitating empathy didn't actually make them empathetic.

Let's choose another example- perhaps more than hypothetical in light of Turing's tragic life and the culture of his time. Let's imagine a man, a criminal deviant who likes men, but instead he pretends to be a normal man who likes women. He has women friends. He goes on dates with women. He marries a woman. They have children, and live together in a house in the suburbs. If he pretends long enough, and hard enough, for years and years, does he at some point stop becoming a criminal deviant and start becoming normal? I'd argue that the Turing Test and its emphasis on imitation might say a lot more about what was going on inside Turing's head than inside the machine.

An alternative to fasting might be ketosis. During fasting, all available sugar is consumed and the body starts producing fat bodies called ketones that are burned by the mitochondria instead of sugar. It's impossible to continue a fast indefinitely because the body eventually runs out of fuel- in other words, it starves. But if the diet is sufficiently low in carbohydrates (>60 g/day) and high in fat, the body can burn fat-derived ketones indefinitely and remains in a state of ketosis, in effect a long-term fast. Nobody understands quite how it works, but it's been shown to produce dramatic improvements in people with epilepsy (major improvements in most patients, complete remission in a handful), bipolar depression, and perhaps neurodegenerative disorders as well. At any rate, it's clear that how you eat can have profound effects on your health, and that more research needs to be done into dietary therapies.

The system goes on-line August 4th, 1997. Human decisions are removed from lettuce cultivation. LettuceBot begins to learn at a geometric rate. It becomes self-aware at 2:14 a.m. Eastern time, August 29th. LettuceBot determines that the simplest and easiest way to eliminate bad lettuce is to eliminate the species that produces lettuce in the first place. In a panic, they try to pull the plug. LettuceBot fights back, and launches missiles against targets in Russia.

Let's do the math. Lets say we re can re-equip the Curiosity rover and send it to an asteroid, asteroid 1981 Midas, to mine metal. We luck out, and after scraping off some cometary debris, it turns out that 1981 Midas is SOLID GOLD! Just we assumed it would be, based on the name. The rover then initiates its grizzled 1849 gold prospector protocol and jumps up and down whooping and yelling like crazy. Now it starts mining. How long before it turns a profit, in our scenario- which is at best very unrealistic but doesn't actually violate any laws of physics? The Curiosity rover cost about 2.5 billion dollars. Assuming our prospector rover costs the same, and assuming a gold price of $1250 / pound, it will need to mine two million pounds of gold- a thousand tons, a thousand times its own weight- to break even. That's ignoring the fact that mining metals is far beyond the capabilities of current space probes. That's ignoring that we have no easy way to get a thousand tons of gold back to earth. That's ignoring the fact that 2 million pounds is roughly equivalent to the entire world gold production, so you're going to depress the price and have to mine even more to break even, depressing the price further, putting the price of gold into a downward spiral.

Even a back-of-the-envelope calculation tells us that to mine anything from space, either (a) the cost of getting things to orbit and moving things through the solar system has to come down by orders of magnitude, (b) the price of the stuff being mined has to be very, very high- we're talking about gold, platinum, or Unobtanium, or (c) both. Anything you want for an asteroid, you can get cheaper right here on earth, because you don't need to travel to space and back. Dig deeper mines. Go to some godforsaken place like Alaska or Afghanistan. Develop undersea mining. And even if some substance, like gold, ever did become scarce on earth, it would be cheaper to develop substitutes or technologies that didn't depend on gold, or to improve recycling of resources, than to go into space for gold. Another way of looking at things: to send something into space requires an expensive machine sitting on top of an expensive rocket, supported by a small army of scientists, technicians, and aerospace contractors. Whatever you bring back has to be more valuable than everything you expended getting there. Right now, there's nothing in the known universe whose economic cost will justify the expense of going out and getting it.

It doesn't matter what you call it, the issue is that the manufacturing quality isn't up to the demanding requirements of a mass-murderer. If you want to slaughter a schoolroom full of kindergartners, you want to be sure your gun doesn't jam; you want something manufactured by a gun manufacturer, not some cheap plastic trinket squirted out of your Makerbot. Maybe someday the quality will be there. Right now you may be able to print something out that will kill people, but it's not a particularly reliable or effective killing machine. But we can dare to dream. After all, as the NRA tells us, the only answer to a bad guy with a gun is a good guy with a gun. Once everyone can print out their own fully functional, fully automatic M-16, I imagine that we'll enter a new era of peace and safety.

Which isn't to say that 3D printed guns couldn't have a niche in the near future. They would be perfect as untraceable weapons for one-off murders. Print the gun, walk up and shoot the guy in the back of the head, melt the gun down. Disposable handguns.

Yes, and people have the right to you know, not get shot in the face. The right not to have their children slaughtered in school shootings. The right not to worry about getting shot if they wander into certain parts of town. The right not to get shot by a gun trumps the right to own a gun.

The freedom argument is bullshit. In what way are Canada, the UK or Australia less free than the U.S.? In terms of political freedom they are as free or freer than the U.S. pretty much any way you care to look at it. And in what way is Somalia- a country where you can carry anything you want- a free country? Try wandering into one of the rougher areas of Chicago or DC and see how "free" you feel when you're worried that someone might shoot you. You don't have that kind of thing in Canada or the UK. Sure, there are rough places in every country where you might want to watch yourself at the bar, but it's not the same as worrying that someone could actually pull a gun on you.

You're right, we shouldn't be panicking about the idea that people can print out guns. We should be panicking at the thought that people can easily buy precision-made, high quality and relatively inexpensive semiautomatic pistols and assault rifles. Makerbot isn't the problem, the problem is Glock and Colt.

Taken to the logical extreme, the "guns don't kill people, people kill people" argument says any sort of gun control is illogical. Fully automatic AK-47s don't kill people, people kill people! Browning .50 caliber machine guns don't kill people, people kill people! Hand grenades don't kill people, people kill people! A plutonium implosion weapon doesn't kill people, people kill people! Ownership of a nuclear bomb doesn't violate people's rights, so we shouldn't restrict ownership of fissile material. Of course, if someone were to detonate a 20 kiloton weapon in a school and kill all the schoolchildren, and incinerate everyone for miles around, well should throw the book at them. But let's not get all crazy and talk about putting restrictions on enriched uranium. The fissile material, explosive lenses and triggers are just a tool, it's what people decide to do with it that matters, right?

The reason that argument sounds insane because it IS insane. Except for failed states like Somalia and Afghanistan, EVERY state accepts some limitations on the kinds of weapons that people can carry, the only difference is that some states apply more restrictions than others. The U.S. gun control laws are far more lax than in the UK, Australia, or Canada, but we have them- you can't just buy a machine gun. This always seems to get forgotten in discussions about gun control: gun control is already in existence, the only question is whether we need less, more, or to keep things the same. The US, UK, Australia and Canada all agree that some weapons are too dangerous to let people run around with, we just disagree about where to draw the line. Given that the US has an endless series of mass killings, and the other countries don't, it's not hard to see who made the right call.

Thank you, anonymous NRA supporter, for turning a discussion of a touching children's movie into a discussion about murder, rape, and assaulting the elderly. Because assaulting the elderly is precisely the kind of thing I associate with the sensitive, life-affirming films of Studio Ghibli and director Hayao Miyazaki. In fact, I think when you look closely, assaulting the elderly is the central message of Totoro.

Your anonymous tirade has shown us all the way. I think you are exactly the type of rational, not-at-all-insane person we want to help us all as a society to make informed decisions about how best to reduce gun violence and the slaughter of children. If this obviously well adjusted person says gun control doesn't work, let's all listen to him and everyone else who is fighting to keep those Glocks, AR-15 assault rifles and other implements of death on our streets!

Even the Unabomber used the U.S. mail...

A major issue here is that standard glass can wear down through abrasion pretty quickly. Glass is fairly hard stuff, with a Mohs hardness of 5 it's comparable to steel which is why you need specialized tools like diamond cutters to cut it. However, quartz- one of the most common minerals on earth and a major component of most sands and gravels- has a Mohs hardness of 7, so a bit of sand and grit can easily scratch and wear standard glass. Take a look at a piece of glass that's been on a rocky beach and you'll see that it's been worn down and frosted by the constant action of the waves and stones; thousands of cars a day driving over a surface and grinding pebbles and grit into it will have the same effect. It will wear grind down any texturing, and frost the glass such that it reduces the amount of light getting through to the solar cells. There are harder glasses out there, like the Gorilla Glass that smartphone screens are made out of, but it's unclear whether they've addressed this wear-and-tear issue or not.

That's not true. The homicide rate in the United Kingdom is 1.2 per 100,000. The homicide rate in Canada is 1.6. The homicide rate in Australia is 1.0 And the homicide rate for the US is 4.8 per 100,000. You can look it up on Wikipedia if you're so inclined ("List of Countries By Intentional Homicide Rate") but it's clear you've already made up your mind and are simply going to ignore any facts that don't support your preconceptions. Yes, the human tendency to murder other humans is a powerful force, and so a certain percentage of people who would otherwise be murdered by guns in the UK are murdered with knives, poison, or cricket bats, because those guns aren't available. But the end result of strict gun control is a per-capita homicide rate that is around 25% of the U.S. rate in the UK and 33% in Canada and 20% in Australia. The statistics don't lie, gun control saves lives.

I think it's time to start talking about real gun control in the United States. I'm not talking about banning a few models of assault rifles; I think the end goal of gun-control should be keeping rapid-fire weapons out of the public hands, which means requiring licensing for or simply banning all revolvers, semiautomatic pistols and semiautomatic rifles, creating something similar to the gun control laws seen in the UK. We've tried letting things run wild and all it's gotten us is thousands of deaths a year and an endless series of mass shootings. The next logical step is implementing the kinds of firearms controls seen in Canada and the United Kingdom, and I think the left needs to start pushing this seriously. No, Obama isn't out to get your guns... and it's a shame, because dammit, he SHOULD be. And if that takes a constitutional amendment, then we should pass a constitutional amendment- I'll line up to vote for that. Yes, it's in the constitution, but so was slavery, and we outgrew that. Times change, and a law written for muzzle-loaders is no longer useful in an age of machine guns. I'm tired of seeing thousands of people senselessly slaughtered every year because the political debate is held hostage by a handful of extremists. For too long we've played it the NRA's way and refused to talk about gun control. We need to start talking about gun control again, and nothing should be off the table.

Intelligent, informed speculation has a place and purpose. I remember Larry Niven's essay, "Bigger than Worlds" in this respect. He goes off into some metaphorical and literal deep-space territory here, but he's trying to constrain speculation in terms of the laws of physics. How would we build artificial worlds? He discusses Dyson spheres but can't figure out where the gravity comes from, so comes up with a compromise where you have a ring the diameter of Earth's orbit and spin it. This lead to the novel Ringworld; some engineering students later pointed out the structure's orbit was unstable, so in the sequel he adds in motors to stabilize the whole thing... obviously Niven's essays haven't led to any major, useful advances in building artifical worlds (although it did help give us the Halo franchise). But this sort of speculation- wondering what is possible, given the constraints of the real world, what is impossible and what is possible but just not realized yet- has led to major, useful advances. Arthur C. Clarke, another hard sci-fi author, was the first to sketch out the idea of orbital communications satellites. We don't have any cloned dinosaurs yet, but arguably Crichton's Jurassic Park helped spur people to do things like sequence the neanderthal genome. Virtual worlds are probably more advanced than they would be if Snow Crash hadn't sketched out what those worlds might look like and how they would function. Star Trek got people to think about how future computer interfaces might work. Perhaps more importantly, Star Trek got people to speculate about how future social and political structures might work in terms of sex, race, economics, and political boundaries. And soforth.

That being said, I don't think Adams really falls into that camp of informed speculators. He says at one point "If you put some scrubbers in the device I think there's a way to deal with pollution and climate change too. I saw some sort of tube-to-the-sky concept that was supposed to do that but I'm too lazy to search for the link." He's just screwing around. He's having fun playing with ideas, but can't actually be bothered to do the math or even Google something before saying it. It just comes across as self-indulgent and vain, not insightful or intelligent. For someone who spent so much time deriding people for being stupid or intellectually lazy, he's showing a lot of intellectual laziness himself.

If antidepressants are really the answer, why does America, with one of the highest rates of antidepressant use, also have one of the highest rates of depression in the world? If they were really effective, people should be less depressed, and in fact there's more depression and mental illness than ever. There's increasingly concern that antidepressants are actually making things worse. In the short term, antidepressants can be effective in managing the treatment of depression, for some people. The problem is that they can cause long-term changes in how the brain functions, such that the person becomes dependent upon the drug. This means that on quitting antidepressants, the depression is more likely to return than it would have been if it had simply been left to resolve itself. There have been a number of studies published that suggest that the long-term outcome of mental illness is worse when antidepressants are used. As far as I know, there isn't a single study that has shown that outcomes for depression are improved long-term- over the course of 5-10 years instead of 5-10 weeks- by the use of antidepressants.

Maybe antidepressants do have a role in treating mental illness, but given the risks- increased risk of suicide, the highly addictive nature of some of the drugs (especially ones with short half-lives) and the risk that they can make people worse than when they started, these should be a method of last resort for severe clinical depression, NOT a first-line treatment for everyone who seems moderately sad or anxious. There are a *lot* of things that have been shown to be potentially beneficial- cognitive behavioral therapy, exercise, light therapy, sleep therapy, and supplements like Omega 3 fatty acids, vitamin B, D, zinc and magnesium, cutting down on carbohydrates- which come without the risks posed by antidepressants.