Well, first, real-world price is the price at which people are willing to buy and sell the good. So, if there are trades happening over the course of milliseconds, then you should expect that price to change over the course of milliseconds. There isn't anything unusual about that. For example, if you want to buy oranges, and I say that I will sell you oranges for $1.00, the price is $1.00 to you if you want to buy oranges. If somebody else says they'll sell oranges to you at $0.80 instead, then you'll buy from them instead of me, and the price just fell $0.20. How long did it take the price to fall $0.20? However long it took the other guy to make the $0.80 sell offer. That could have been a month after I made my $1.00 offer, or it could have been 0.2 ms after I made my $1.00 offer. Clearly, if he wants people to buy his oranges instead of mine, it's in his benefit to make his offer as quickly as possible, because after you buy oranges from me, you won't want to buy any more oranges. So if he waits a month, he may have nobody buying at that price.

Conversely, if I offer to sell you 5 oranges at $1.00 and they immediately sell at that price, I'm going to offer my next 5 oranges at $1.20. How fast did the price rise $1.20? How fast did I make my next offer? I could continue selling oranges at $1.00 for a month, but if people are buying a ton of them, and I think I can sell all my oranges for more money, it's in my advantage to up the price as quickly as possible. So, milliseconds after your order went through, I could decide to sell the next batch at $1.20.

There's absolutely nothing nefarious about millisecond trades and price changes, if that's all that's going on. The only difference from "real world phenomena" is that the brokers have algorithms to increase or decrease the share price automatically based on the supply and demand it sees. In a very high trade volume situation, that time matters. If you're faster than your competitor, people are buying and selling *from you* because your prices are always better, closer to the optimal given the supply and demand for the stocks. That's how you make money being faster.

Second, HFT helps you get the "real world value" because the way you get a "real world value" is through iteration. When I decide to sell you oranges at $1.00, that's not the real-world value of oranges. That's a guess I made at the price, assuming there would be exactly enough demand for oranges at $1.00 as I have the ability to supply it. If people are willing to buy it at a higher price, I'll find that higher price faster the quicker I can perform trades and vary my price, and the more trades that I can make. Same if people are only willing to buy it at a lower price. It's no different than, say, if I want to find the square root of a number via the Babylonian Method. If I have a computer running at a low clock frequency, each iteration might take a second. If I have a computer running at a high clock frequency, each iteration might take a microsecond. They both get to the same answer, but a higher clock frequency gets you that answer faster. Again, nothing nefarious about that, and it means that at any one point a human looks at the price of stocks, it's a value that most accurately reflects that equilibrium price between buyers and sellers, because all the iterations are happening very fast.

What *is* somewhat nefarious is that apparently some trading houses are noticing you just bought all the oranges they were selling at a particular price. Then they assumed that you're likely trying to buy oranges from your competitors as well, for a similar price, at the same time. So, because they have a faster connection to the other trading house, they start buying oranges from competitors before your request to buy gets there. When your request to buy arrives, they tell you, "we're no longer selling at any oranges at that price." So the original place just bought up YOUR cheap oranges, and they get to sell it at a slight profit margin. It's not exactly front running, because that would be if your broker, once you've placed an order to buy oranges, instead of buying at your behest, goes out and buys himself a bunch of oranges, then sells them to you at a higher price. He knew what the demand was going to be, because you told him, and he caused the price to go up as a result. In this case, the trading house is buying up more oranges, but they don't *know* that you've placed on order elsewhere. They're making an assumption and taking a risk, based on some algorithm that predicts that type of stuff with some probability. That said, I will agree that's ethically iffy, because they are acting on knowledge nobody else will have for the next few milliseconds, and trading while they have that advantage.

Also, give this a try.

It's a good question. Try this video

You don't have the right to have someone not point a camera at you. You can leave, and cover your face, but you can't really force them to stop. You can politely tell them that it makes you uncomfortable, but if they want to be assholes about it, there's no law against being an asshole. There's definitely a law against you assaulting said asshole and/or stealing their property.

Oh? What exactly are you doing to other people when you film them? Stealing their soul?

If you're not touching somebody, you're not assaulting them. If you're not following them as they try to leave you, you're not harassing them. Filming somebody is not doing anything to them anymore than loudly talking about them to somebody else, so that they can hear. They're peripherally involved, they might be annoyed by it, but they don't have any right to stop you.

I don't know what bar you go to, but I've never seen that, ever. In fact, if the bar has live music, I've never been to one without at least 20% of people recording.

I don't think so, but let's say you're right. The trick is that recognizing such an expectation of privacy cannot mean that you just ban cameras you're uncomfortable with. You have to ban security cameras. You have to ban reporters who are covering a story. You have to ban people taking selfies at bars and other locations where it's not possible to ensure someone who didn't consent will show up in the background.

If the majority of people in a jurisdiction are willing to go with that, then yeah, the law should be changed. I think they're not. I think the first time you take your phone out to take a picture of something cool you've seen and other people tell you that you can't do that, you're going to throw a fit. People don't really think they have an expectation of privacy at those locations, they just feel uncomfortable when they see a camera next to them because it makes it obvious that they have no expectation of privacy, and they don't like to be reminded of that. They like to pretend they're not ending up on the background of tons of pictures, or being laughed at by the police who is reviewing security tapes because someone's wallet was stolen at the same time you were at the bar getting slapped for the stupid line you tried to use.

No. In a public location, bringing a high quality video camera out, setting up on a tripod, and pointing it at straight at you is perfectly acceptable. It's a public location, you have no privacy. You are within your rights to leave, to cover your face, to turn your back to the camera. You can't attack the owner of said camera, or take the camera away.

At the bar you own, or at your house, or at any private property in which the owner doesn't want the device on, you absolutely have the right to kick anybody out who doesn't follow the rules. But somebody is perfectly within their rights to stand in the public street and point a camera at your house window. The only legal recourse you have is to close the blinds.

This obviously depends on the jurisdiction in your area, and whatnot. However, for a lot of the US, that happens to be true. And it's the way it should be.

Exactly. However, that's not a difficult problem to solve. What the Nature article fails to address is the real problem: it's not easy to publish papers that do nothing but confirm the findings of another paper.

The article talks about how a researcher had his dreams of being published dashed once he failed to achieve a similar p-value upon attempting to reproduce his own research. This is bullshit. Journals should be selective, yes. They should be selective in terms of whether experiments have been run with proper methodology, and whether the study supports the conclusions made by the author. They shouldn't be selective based on p-value. For proper science to occur, not only should said researcher have been able to publish his first low p-value study, he should have been able to publish his second high p-value study. Other researchers at different institutions should attempt to reproduce the work and publish their positive or negative findings. Only once dozens of said studies are performed can we actually start to draw a conclusion: if only 1 in 20 experiments show a p-value below 0.05, well that doesn't actually disprove the null-hypothesis, it's evidence for it.

Even if the 5% chance of a paper being bullshit was an upper bound, that would still be a really plausible scenario. Replicating experiments are a fundamental part of science, but journals are only interested in unique experiments yielding positive results with low p-values. Either that or negative results replicating a particularly important paper that everyone takes for granted. Ideally, while grad students are still new and learning the ropes, their research should consist of replicating others' research and publishing the result, whatever it may be. It's the perfect job to get them started before they've had the chance to do significant research of their own, and it's incredibly valuable to the community at large.

Is that right? I was under the impression GR is supposed to work fine orbiting a black hole, and that you only ran into problems once you got close to the singularity, because at that point it's dealing with small scales. Simultaneously, quantum mechanics is unlikely to work there either, because it's dealing with very large gravity in small scales. Basically, the singularity is the point where you run outside the scope where either theory works correctly: GR works well with gravity at large scales, but works poorly at small scales, and quantum mechanics works well at small scales, but doesn't work well when gravity is a significant factor (therefore the need for the development of a quantum gravity theory).

I'm not a physicist though, so I'd be glad to be corrected if my understanding is incorrect.

I never understood that argument. You guys argue that people who have insurance should pay their premiums in proportion to how likely they are to use it. You consider that the fairest possible payment system. However, if you take that to its logical conclusion, you should only charge people who actually end up using it. So you should go ahead and eliminate insurance altogether, and you have the fairest model possible: only people who get into car accidents pay the costs, only people who get sick pay medical costs, only people who get robbed suffer their losses.

The entire point of insurance is to make the payment unfair in order to diminish the payment by spreading the risk among everyone. You agree to pay something, even though you hope to never have to cash in on the insurance, so that if you do have to cash in, everybody else who doesn't need to cash in subsidizes you, and you pay less. You do this for peace of mind. What you should want isn't to pay commensurate to your risk, you should want everybody to pay equal rates, which will result in the lowest possible premium for everyone. If you determine that premium is too high for your risk level, that should mean you think your risk level is low enough to go without insurance.

That's going to inconvenience him about as much as spam mail in the spam folder, considering all the e-mail people send is going to be automatically filed away to the social folder. Your post ended up proving his point that people don't actually understand how the feature work.

I was somewhat pissed off that Google made accepting those e-mails the default in the google+ settings, but I can see why some people would turn it on. In any case, anyone can turn it off.

Right. Which is why 1kW / hr doesn't make any sense. Power is a rate, watts is equivalent to Joules per second. So 1 kW / hour is an acceleration in the draw of energy, or 0.278 J / s^2.

They also didn't mean 1 kW * hr for their energy, because the rest of the sentence specified that "1 kW / hr" was $0.10 and he parked for less than 30 minutes. So yeah, clearly they meant just 1kW, and the time period is 30 minutes. Or 0.5 kW * hr for the energy.

I live in the South. The consequences are going to have people talk to you about how your position is unwise. Which I believe would happen quite often. You may increase your risk of getting keyed, but the only people who would do that are the people who randomly key cars anyway, they just found what they're deeming a justification this time.

The GP was making a joke. Of the same type you'd hear from a friend who is a fan of a different football team you are saying that you won't be safe at his house during the game.

Your quote is from the original article from March. In the next link he talks about the latest November update, which reintroduced sleep mode.

That said, he's wrong that the latest update doesn't fix the problem. I own a Model S, and I went from losing about 5 miles off my rated range in 8 hours to losing about 1 mile per 14 hours. So, what's the difference between my car and his? Well, based on the pictures he posted, which has snow on the ground, he lives somewhere far colder than South Carolina, where I live. So his car is using more power for thermal management of the batteries.

But wait, you say. The article says, "It's a popular myth among Model S owners that much of the vampire power goes to keep the battery warm during cold nights. This is simply not true. According to Tesla, there is no thermal management of the Model S battery when the car is turned off and not charging--no matter how cold it gets."

True, guy. However, let's examine your testing methodology: "For each test, I charged the car up in the evening to its usual selected level (In my case, about 80 percent). Then I removed the charge plug. I allowed the car to sit unplugged overnight and on into the next day, until I needed to drive it. (Typically a span of 12 to 24 hours.) Before driving it, I plugged it back in to top off the vampire-depleted battery back to its original level. Then I checked the kWh-meter."

And...when you plug it in to charge it, the pumps come on, and they start heating up your battery for safe charging. There's your so-called vampire load. My car, in a warmer environment, doesn't have to spend as much energy doing that.

Furthermore, he says: "The three tests showed vampire losses of 2.3 kWh in 17 hours, 1.9 kWh in 23 hours, and 4.2 kWh in 18 hours...I can't explain the wide variation in the vampire draw over the three tests."

Maybe he should try correlating it with temperature.