Or you could just leave the manual controls in the self-driving car. There was some story about Google proposing to not put steering wheels in its cars, but I don't think anybody seriously things the first self-driving cars aren't going to have manual controls.
There's quite a bit of tire dust off roads. I'd be surprised if brake dust contributed much, but might as well throw it in. The particulate emissions in modern gasoline car exhaust is extremely low, so it's not terribly surprising that tire dust is responsible for the majority.
When you're changing speed or direction there's more wear on the tires because they have to transfer more force. Its hard to see that on the tires themselves, but you can see it on the roads. Gravel roads usually have washboards at intersections and pavement gets more potholes.
You don't need a fancy graphics card to stack images. It's done in regular MRI all the time too. The problem is, signal to noise ratio goes up approximately linearly with field strength, but recovering that SNR by averaging takes N^2 images, which take N^2 time to acquire.
I agree. I think basic income is a good solution to the problem, and also a way out of our current mess. Societies work better when the gap between rich and poor isn't too big. Basic income also provides a safety net, including for people who want to take risks in order to do something completely new, while still providing an incentive to do something productive.
I disagree with you on one bit: the jobs are going away. They always have. People used to work an actual eight or ten hours a day and the jobs they were working at were critical to the survival of both individuals and civilization. Most of the jobs we have today are more make-work. Some of the highest profile ones are literally random number generators, and most white collar jobs only involve a few hours of productive work per day. Yet as a society we've bought into the idea that we must submit to the whims of a corporate master for at least eight hours a day, five days a week, even if we spend most of it refreshing Slashdot and Facebook.
I hope when the machines come for the white collar jobs we'll consider doing something different. That we'll look hard at equitable ways of sharing our ever increasing wealth without the unnecessary pageantry.
Many times we've had automation kill off labour jobs. Our economies used to be virtually all primary production. Now they're 70%+ service. Whenever technology obsoleted labour jobs, people switched into white collar work. Now technology is coming for the white collar jobs. We can continue to play the "must have work" game as we've done in the past by doubling down on paying people to do useless things, or we can actually embrace the freedom that having machines do our work gives us.
That's a much more believable scenario than the implied "Uber wants to lower surge pricing for you their customer!"
A proper free market would also provide everyone with good information about what the price is going to be. Predicting demand would let Uber do that, although they probably don't want to because it would further cut into demand.
Depends on the treatment. Generic therapy is intrinsically pretty cheap, and is likely to get much cheaper with foreseeable technological developments. Organ transplantation is inherently much more expensive, but also likely to get cheaper with near term technology.
Tying a piece of mouldy bread to your arm and designing a new antibiotic from scratch are a wee bit different. Only one of them is engineering. We made all manner of poultices and potions, sure, but that doesn't mean they worked, or that any of them worked any better than tying a piece of mouldy bread to your arm.
It is possible to build simple things without understanding the principles behind them. More complex things require some theoretical knowledge but not necessarily a complete understanding. But having a workable mathematical theory lets you rapidly refine your design (no more slow trial and error), and it's usually at that point that the concept really becomes widely feasible.
Kinetic energy isn't that simple in a relativistic universe. I have to transfer energy to accelerate, right? But then it takes MORE energy to "decelerate" back to my starting velocity. Why? Where did the energy go?
The GP is pointing out that in a relativistic universe kinetic energy makes more sense the other way around. It's not some property the object gains when it accelerates, it's the energy required to accelerate the object. The fact that it requires energy to produce an acceleration is inertia. It's like relativistic friction.
I think the idea is that there's a mathematical link between Unruh radiation and the cosmic event horizon, which you can sort of understand as the wavelength of the Unruh radiation having to fit inside the universe. Because that event horizon changes when you accelerate and energy is quantized, if Unruh radiation is responsible for inertia then inertia must be quantized too.
The explanation for the em drive appears to be that if you confine the system to a small space then the quantization of inertia becomes much bigger, to the point where it can create a Casimir-like effect that's observable as thrust.
As an actual scientist, I disagree. Some scientists do have an ego problem, but there are others who are willing to look into crazy ideas. Eventually, if the idea is a good one, the evidence will mount to the point where it becomes accepted. It's good to have high standards of evidence. I argue that the danger of ego is that it lowers the standard of evidence for some things. "That makes sense" is the dangerous statement.
The safe bet, which is what the majority of physicists should be going with, is still that the EM drive is a mistake. But it's looking promising enough now that putting a few more resources into it is probably a good idea. Which is exactly what's happening.
We didn't make antibiotics. We found them. A little later we learned how to purify them.
Today we can make synthetic versions of found antibiotics, which requires a lot of theory. We'd like to be able to design antibiotics too, and there's a lot of interest in developing the theory that would allow it.
The control experiment DID show a zero measurement. They certainly had a control with no power going to the device, and IIRC they had another control where something critical was removed.
They also had a test option that removed certain geometric features (strakes inside the cavity I think) which the American inventor said were critical and the UK inventor said were useless. The guy from the UK turned out to be correct.
Building an ion propulsion engine big enough to be used on a manned mission is a pretty big boost. Rocket engines have been used for a couple thousand years, but building one big enough to put humans into space was kind of a big development.
Slashdot Mirror
And unless you live in some barbaric place, there are a lot of laws about fencing and other safety features around pools.
Plus it's more dangerous to have a pool than unvaccinated kids *now*, when most people are vaccinated.
Or you could just leave the manual controls in the self-driving car. There was some story about Google proposing to not put steering wheels in its cars, but I don't think anybody seriously things the first self-driving cars aren't going to have manual controls.
There's quite a bit of tire dust off roads. I'd be surprised if brake dust contributed much, but might as well throw it in. The particulate emissions in modern gasoline car exhaust is extremely low, so it's not terribly surprising that tire dust is responsible for the majority.
When you're changing speed or direction there's more wear on the tires because they have to transfer more force. Its hard to see that on the tires themselves, but you can see it on the roads. Gravel roads usually have washboards at intersections and pavement gets more potholes.
You don't need a fancy graphics card to stack images. It's done in regular MRI all the time too. The problem is, signal to noise ratio goes up approximately linearly with field strength, but recovering that SNR by averaging takes N^2 images, which take N^2 time to acquire.
I agree. I think basic income is a good solution to the problem, and also a way out of our current mess. Societies work better when the gap between rich and poor isn't too big. Basic income also provides a safety net, including for people who want to take risks in order to do something completely new, while still providing an incentive to do something productive.
I disagree with you on one bit: the jobs are going away. They always have. People used to work an actual eight or ten hours a day and the jobs they were working at were critical to the survival of both individuals and civilization. Most of the jobs we have today are more make-work. Some of the highest profile ones are literally random number generators, and most white collar jobs only involve a few hours of productive work per day. Yet as a society we've bought into the idea that we must submit to the whims of a corporate master for at least eight hours a day, five days a week, even if we spend most of it refreshing Slashdot and Facebook.
I hope when the machines come for the white collar jobs we'll consider doing something different. That we'll look hard at equitable ways of sharing our ever increasing wealth without the unnecessary pageantry.
Many times we've had automation kill off labour jobs. Our economies used to be virtually all primary production. Now they're 70%+ service. Whenever technology obsoleted labour jobs, people switched into white collar work. Now technology is coming for the white collar jobs. We can continue to play the "must have work" game as we've done in the past by doubling down on paying people to do useless things, or we can actually embrace the freedom that having machines do our work gives us.
That's a much more believable scenario than the implied "Uber wants to lower surge pricing for you their customer!"
A proper free market would also provide everyone with good information about what the price is going to be. Predicting demand would let Uber do that, although they probably don't want to because it would further cut into demand.
Kind of like that, except without the video out.
Sure. All the deep learning libraries that support GPU computation on Linux beg to differ with you. TensorFlow doesn't run on Windows at all.
It's a mobile GPU in a USB stick. Kinda handy if you've got some device with a USB port but no GPU.
No, I couldn't think of any examples either.
It's just a mobile gpu chip with a USB interface.
A gas plant in Australia apparently. It's in the article.
Slashdot got trolled by someone in the UK who thinks their new nuke plant is too expensive.
Yes. Simple statistical concepts are within reach of everyone. One very simple one will take you a long way:
1) Is the conclusion based on a single analysis, or multiple analyses of a single dataset? If so it's interesting, but not conclusive.
You know the population growth rate is in decline, and is projected to go negative in the future? It's already negative in most of the western world.
People "going forth and populating the land" isn't something that's going to be happening much longer.
Depends on the treatment. Generic therapy is intrinsically pretty cheap, and is likely to get much cheaper with foreseeable technological developments. Organ transplantation is inherently much more expensive, but also likely to get cheaper with near term technology.
Tying a piece of mouldy bread to your arm and designing a new antibiotic from scratch are a wee bit different. Only one of them is engineering. We made all manner of poultices and potions, sure, but that doesn't mean they worked, or that any of them worked any better than tying a piece of mouldy bread to your arm.
It is possible to build simple things without understanding the principles behind them. More complex things require some theoretical knowledge but not necessarily a complete understanding. But having a workable mathematical theory lets you rapidly refine your design (no more slow trial and error), and it's usually at that point that the concept really becomes widely feasible.
Kinetic energy isn't that simple in a relativistic universe. I have to transfer energy to accelerate, right? But then it takes MORE energy to "decelerate" back to my starting velocity. Why? Where did the energy go?
The GP is pointing out that in a relativistic universe kinetic energy makes more sense the other way around. It's not some property the object gains when it accelerates, it's the energy required to accelerate the object. The fact that it requires energy to produce an acceleration is inertia. It's like relativistic friction.
I think the idea is that there's a mathematical link between Unruh radiation and the cosmic event horizon, which you can sort of understand as the wavelength of the Unruh radiation having to fit inside the universe. Because that event horizon changes when you accelerate and energy is quantized, if Unruh radiation is responsible for inertia then inertia must be quantized too.
The explanation for the em drive appears to be that if you confine the system to a small space then the quantization of inertia becomes much bigger, to the point where it can create a Casimir-like effect that's observable as thrust.
As an actual scientist, I disagree. Some scientists do have an ego problem, but there are others who are willing to look into crazy ideas. Eventually, if the idea is a good one, the evidence will mount to the point where it becomes accepted. It's good to have high standards of evidence. I argue that the danger of ego is that it lowers the standard of evidence for some things. "That makes sense" is the dangerous statement.
The safe bet, which is what the majority of physicists should be going with, is still that the EM drive is a mistake. But it's looking promising enough now that putting a few more resources into it is probably a good idea. Which is exactly what's happening.
They also make mistakes, change their minds, and experience regret. So yeah.
We didn't make antibiotics. We found them. A little later we learned how to purify them.
Today we can make synthetic versions of found antibiotics, which requires a lot of theory. We'd like to be able to design antibiotics too, and there's a lot of interest in developing the theory that would allow it.
The control experiment DID show a zero measurement. They certainly had a control with no power going to the device, and IIRC they had another control where something critical was removed.
They also had a test option that removed certain geometric features (strakes inside the cavity I think) which the American inventor said were critical and the UK inventor said were useless. The guy from the UK turned out to be correct.
Building an ion propulsion engine big enough to be used on a manned mission is a pretty big boost. Rocket engines have been used for a couple thousand years, but building one big enough to put humans into space was kind of a big development.