I'd be interested in reading the source to see what the argument is. Off the top of my head, the Irish Potato Famine strikes me as a pretty real famine. It was certainly exacerbated by political pressures, and they were growing monocultures in the first place because of the pressure for productivity. But it was a real crop failure, and they learned to reduce their dependence on a single crop.

Certainly it could have been handled better, and far fewer people would have died. But I still think the death toll would have counted as a famine, or at best a famine barely averted by aid. I'd put it in a different category from starvation caused by war or corruption. Even the Great Chinese Famine could be chalked up to politics without too much of a stretch, but there are still crop failures due to drought and disease.

Since the agricultural revolutions of the past few centuries and especially the last few decades, we're so awash in food that aid will always be stymied by people rather than lack of calories. But I'd put the tipping close closer to 40 years than 400.

The terminator gene solves the gene-spreading problem, but it introduces the problem of leaving farmers permanently at the hands of Monsanto. They are forced to buy new seeds every year.

They can, of course, opt out, but then they miss out on Monsanto's improvements. So we've got a conflict of expectations not entirely unlike Slashdot's frequent outrage about EULAs that effectively mean you don't own your own software, or even hardware.

As I understand it, most farmers buy seeds anyway, because the plants don't breed true to type. But there was particular worry about poor nations, where the farmers are closer to being completely broke, and this looked suspiciously like indentured servitude.

I'm not taking a position on the argument here, just clarifying what it's about.

They had to push that back. This is the year of Linux on the Flying Car. We'll get to Linux on the Desktop right after that.

Oh, I certainly don't: there's a permanent speed trap there.

It's conceivable that there's a reason for it. The road as a whole should be a major arterial, but it's got an awful lot of stop lights. (This is just outside of Washington, DC, which has practically no proper arterials.) At rush hour, allowing people to go faster on this section than the overall speed of the road would be worse for traffic.

What's really needed is to substantially restrict access to that road and make it a highway, though I'm sure that the businesses and residences along that road would hate it. The problem is systemic: there are no arteries and nobody wants to turn their stretch of road into one. There are zero interstates, so the roads are under a variety of local jurisdictions. I'm sure plenty of people complained to the county and state about that segment of road, but it's just a disaster for the whole region to deal with. And so it isn't.

Er, anyway, that's kinda beside the point, which is really that what's needed is for the traffic engineers to design for steady flow and for people to follow it, even if they'd be more comfortable at some other speed, especially when lanes are limited. But it's easier said than done in a metropolitan area.

It's best if people all move at more or less the same speed. It keeps them better spaced. People driving much slower than that can cause as many difficulties as people driving much faster.

We recognize the dangers of driving too fast, and most people try to keep it to near the speed limit, at least as long as the limit is set properly. Some are set very badly, and that's hazardous. You get a mix of people traveling at a safe but illegal speed with people obeying the law.

Fortunately, I've found that most speed limits aren't too badly off. I'm sure there are jurisdictions where they're deliberately mis-setting them as revenue generators, but I don't encounter many of them. (I can name one not too far from my house, where a four-lane divided road with minimal access has a 30 MPH speed limit... and a speed camera on a big downhill leg. That's going to get people killed, because everybody who knows about the speed limit jams on their brakes and goes 25. And the road is a major arterial, or it could be, if they didn't deliberately limit the flow rate so badly. The road is, of course, a nightmare at rush hour and a speed-trap revenue source the rest of the time.)

How the f*** does Inhofe get to vote "yes" on this, when he's said "it's a hoax" loudly and repeatedly in the past? He's still the chair of the Environment committee. Is there any chance that this change of heart at least going to keep him from railroading scientists?

We don't really get to ask that question. It's subsumed by the existing question about whether we're contributing significantly. Since the answer turned out to be "no", there's no point in asking the next question.

Mind you, "no" is a stupid answer, but that's the point. There's no way to discuss the right question, because we're still too busy being stupid about the wrong question.

I hadn't realized it had gotten as high as 17. The number is arbitrary, of course, at this point it's tricky to remember that decade between the START treaty and 9/11 where we genuinely didn't expect the world to come crashing down on our heads.

Interesting. I'm surprised it's so low, since it's considerably less than the likely liability from a single accident.

Perhaps it's because, as the AC sibling post says, you don't get the money back at the end, and they pay claims out of the pot of money they've collected. Making them effectively your insurer, with one massive up-front payment. (That web page does distinguish it from self-insurance, which they do only for fleets.)

I'm not sure how many people would be helped by that. You'd probably need to guarantee at least half a million dollars in your retirement account; even those who take their retirement seriously (a depressingly small fraction, according to polls) don't get that until well advanced in their careers.

And those who take their retirement seriously should not be risking their entire retirement account on this risk. There's a decent chance that if they're in an accident, they too will need medical care and lose work, and they'd be drawing down precisely the account they'd be depending on.

There are certainly some people who can afford to self-insure, but I suspect that they're about the same people who are in the SEC's class of accredited investors, who are capable of taking big risks with their money without becoming destitute if they fail. Not quite the 1%, but perhaps the 5%. And for them, insurance is already a very small percentage of their expenses.

Thanks for that. That really helps improve the perspective.

It sounds as if the real win would be to build functionality into the device. Many refrigerators use the freezer as a cold store. The objects inside are frozen, and there's more latitude to lower the temperature still more without further damage as long as it remains frozen.

So you could lower the setpoint when electricity is cheap, then use that to drive the refrigerator when electricity is expensive.

I don't know if this is feasible or cost-effective; it would require more electronics (my fridge is dumb) and more engineering.

Is there really that much room in the deadband on a refrigerator that we can save significant amounts of electricity? We're talking about food spoilage here; letting the food get above 40F can be potentially lethal.

I'm not an engineer, but I am a cook, and we are extremely careful about the amount of time food spends in the danger zones. We're cautious, to be sure, but we have to consider the case of the most-susceptible people. I don't know how much room there is to slacken the parameters, and I'm sure there's some, but I'd need to see some numbers to know if the risks we're running are worth the savings we'd get. A fridge costs something like $150 per year to run, which is significant, but you'd need to demonstrate that we can save a hefty percentage of that to make it worth messing with.

That graph is plotting months, rather than years. Those are monthly spikes rather than yearly ones. (Spikes in the differential over the long-term average for that month, so you don't end up seeing the seasonal swings.) The number being discussed in the article is the global mean temperature for the entire year. Other years had higher spikes but this past one had the highest yearly mean.

Thanks for the analysis, but why would it be lighter than a conventional train?

You compare it to the monorail, but at the least it's going to have to support the tube plus the train cars itself. The tube is static rather than dynamic load, which will make things a bit easier, but it still seems like the pillars will have to be a good deal stronger than the ones in the picture. Is it simply that new materials and not having to share tracks with existing trains allow for different, lighter construction? Or the fact that it's passengers and not cargo?

I'm all for Elon Musk, and he's succeeded so often that I'm going to assume he hasn't overlooked the obvious. (I even own a bit of Tesla stock.) But I'm concerned that it might be too optimistic, and that when reality kicks in it will lose the large advantage that's needed to overcome the entrenched resistance.