This just in: Guy accusing company of misconduct emphasizes how ethical everybody else is! Film at 11.
Really? I spent time in Africa (volunteering in Ghana), and nearly every business, individual, and official I dealt with was corrupt to some degree. I was actually surprised when a taxi driver reminded us of the agreed-upon rate, and it was the same as what we'd agreed.
Now, we're not talking Western white-collar corruption, where an executive will duck out to play golf and write it off as a business expense. We're talking about a construction company who won't start work until you start paying them so they can get supplies, then pay a little more to fix the tire on their truck, then more for gas, then more to hire a new worker, then more to rush the job that's now running behind. Contracts mean nothing, because the court officials will decide in favor of whoever offers them the biggest bribe without getting caught by the few people in government who think that corruption is wrong, despite its widespread cultural approval.
If you want to get things done in Africa, the best way is to hire a local who can navigate the local politics (because he grew up in the village, is a member of the local tribe, and has been in good standing all his life), and bribe the local chief to pressure the village to help you. Give that local a budget for bribes, and just accept that they happen. It's certainly not the Western standard of business, but it works. Things get done, the bribes are just a cost of doing business, and the locals appreciate that you're playing their game, instead of forcing your silly ideals onto their ancient traditions.
IP addresses can be easily hijacked, and it's still no indication of actual wrongdoing within US jurisdiction. Perhaps Google gives IP addresses out to any company who wanted to pay them $10/month, as so many ISPs do. According to the post currently below this one, that's exactly what Google's done.
Assuming that Google is ultimately at faulty, the only way I see where Google would be liable under US law, would be if each of the involved branches were first found to be liable. Then there's a reasonable chance that the branches were coordinated by the parent corporation, and it can be pursued.
Misunderstood instructions make things even more difficult, as the subordinate was "just following orders" from a superior who didn't think he was giving them!
The people who think we're able to accurately model the world economy are the same ones who think we're able to model world climate. Maybe, one day, with enough understanding of how all the major underlying principles work - both independently and interacting with one another - we'll be able to model either of them.
The climate has been behaving roughly the same for tens of thousands of years. The global economy has been around for about 30.
For a few thousand years, we've been able to model the climate with absolute accuracy, if little precision. Once each year, the northern hemisphere gets colder, and the southern hemisphere is warmer. Six months later, those roles are reversed.
Similarly, we've been able to model simple events within the economy with remarkable accuracy. Company stocks almost always go up after announcing a new product, and down during a big (really big, not just "important to Slashdotters") lawsuit. Of course, the chaotic events with subtle causes (like exactly when an economic bubble will burst or where a tornado will strike) are much more difficult to predict. All that we can do now is try our best.
Of course, we can take the same asinine comparison to ridiculous conclusions. Neither mathematics nor psychology perfectly predicts the behavior of drivers, so perhaps we shouldn't install any traffic lights until we can perfectly predict where collisions will occur? We can't predict software vulnerabilities, so we shouldn't install any current antivirus software. We can't predict how a person's life will go, so we should stop having children.
As the population increases, individual psychology gives way to statistics, in terms of usefulness. Even "extremely complex" behavior follows probability distributions, and can be expected to remain within those distributions.
Let's consider that a captain of a sinking ship may or may not stay by his post to drown. In fact, let's make a wager on it. Now, before you decide what position of the gamble to take, go ask a nearby ship captain (assuming you're not in Utah) if he'll go down with his ship. Let's assume he says he will. If our wager is based on that one specific captain's choice, then you'll have the best chance of winning, of course. If, however, we wager based on whether "the next" or other suitable random captain drowns, then your brief interview (and the resulting psychological insight) is nearly useless, and a better predictor of the outcome would be the historic statistics of drowning captains.
No mathematical model is ever perfectly accurate. Rather, the models are refined as more data is available, accounting for the influences of old models. What is the currently accepted standard is merely the latest iteration of refinement.
Psychology (as a foundation for economic theory) is no more accurate, but merely produces models without the presence of descriptive numbers. On a playground merry-go-round, where a mathematical model may describe a 90% chance of a child falling off within the next 2 minutes, a psychological model predicts only that a child will fall off soon.
Despite the summary, the patent describes a particular specific application, where every aspect of the system is based around minute-granularity timestamps, rather than mere dates. The application also follows a particular architecture, which is all nicely explained in the actual patent itself. The half-day example was just mentioned for clarity, and does not affect the patent itself.
If your data's integrity is absolutely necessary, Hadoop (or more specifically HBase, which is the part most closely analogous to a database) is probably not for you. On the other hand, if you're working with statistics or any other application where an error affects your product trivially, you may find the speed is worth it, bearing in mind that changes are broadcast across the cluster "eventually". The strengths and weaknesses of Hadoop are different from a traditional database enough that I'd caution your architect against migrating an existing application just for the sake of speed. There are numerous pitfalls and headaches down that path.
New Madrid is also primarily under Missouri. Its effects in Ohio consist of influencing other faults and (in the 1812 earthquake you mentioned) causing residents in a town to leave their homes. For comparison, note that California has had significant earthquakes roughly every year along the San Andreas fault. What's actually in Ohio is far smaller and even less active than the main New Madrid area.
The fracking debate currently has little to do with logic. There's precious little actual evidence of causing harm (or evidence of causing no harm), and the regulators know this. That's why they resist making regulations, even though it's pissing off the general public who just want to see anything happen. Meanwhile, the industry just continues operating as normal. So far, the vast majority of studies (from both sides) conclude nothing one way or the other. Those that do (again, in both directions) are deeply flawed, lacking silly things like control groups.
By my understanding, it would be trivially simple for the energy industry to run a causative study, by testing the output of several wells near a fracking site before and after any fracking operation. The company knows where they're building their new well, and could spend a few hundred thousand dollars of their propaganda budget to drill a well and run some tests. Of course, they never will out of fear that something bad might be discovered, and even if there was no evidence of contamination, the study would never be believed by the anti-fracking protesters.
Within the last month, the EPA announced that it found the first case of a water supply being contaminated by fracking, in a community near a shallow fracking well. The EPA itself has stated that the study needs still more review and reconsideration before it should be used as a basis for any regulations. Of course, both sides have already fired up their most imaginative writers, either condemning the study or overstating its significance.
I can't tell if you intend to include environmentalists in your classification of "everyone with an agenda", but it would certainly be justified. There's enough propaganda from both sides of the science-free debate to disrupt honest attempts at science. It's sickening.
Ohio is on many small faults, the largest of which is the New Madrid fault. There are a few dozen significant earthquakes each year, the vast majority of which cannot be felt.
Despite what you "would think", data is easy to find.
Having an opinion doesn't make one a shill or a troll, especially when there's as much evidence supporting his opinion as your own. The problem is there's absurdly little research from both the pro- and anti-fracking camps. On the one hand, Ohio's seismic activity has increased lately. On the other hand, it has been very inactive since the 1930's, and still remains relatively stable today. Then, of course, there's the possibility (mentioned many times already in this discussion) that releasing pressure could reduce the risk of a larger earthquake.
Comparing Ohio's seismic activity to California is ridiculous. In Ohio, the last big earthquake in 1937 toppled a few weak chimneys. In California, an equivalent earthquake (magnitude 5.4) happened in July of 2010. The faults in Ohio, even when active, pale in comparison to California's eternal fear of the next "big one".
There's no consensus among relevant experts about fracking's effects, but there's plenty of people willing to protest vehemently one way or the other. GP is right to call this out as fear-mongering.
I'm going to assume you're not trolling at this point, though the utter rejection of match makes it unlikely you'r being serious. Oh well. It must be dinnertime.
there is absolutely no way that your brain does calculus in order to walk around an obstacle.
Actually, yes, the brain does some basic physics computations (including calculus) naturally. The answers aren't exact, and there aren't numbers given for output, but the process is straightforward: To predict the path of an obstacle (or several), first watch for a moment to determine the rates of change (velocity v and acceleration a), then apply that to a "black box" formula or neural network (x=x[0] + vt + at^2), then continue to check the result as more observations are available. Physicists have been observing and verifying for the past few thousand years to ensure that the formula we use now is as accurate as our current data allows. Human brains also observe, but they don't have the same accuracy. Instead, brains rely on having lots of repetitive observations, to hopefully train their neural networks to be a closer approximation of accurate. This is why baseball players practice: to improve their aim by refining the neurons they use.
The same concept applies to many operations that humans are good at, like facial and environmental recognition, quantitative estimation, and memory. The brain has neural networks for these important functions, and when someone's brain is damaged, these functions are often impaired. It's important to note that the neural networks are usually far faster than current digital technology, partly due to being analog in nature, partly due to parallelism, and partly (probably) due to using algorithms we don't know or understand yet.
If we did somehow have a magic digital circuit that could do visual recognition and compute the path of objects as fast as the brain can, difficult tasks like robotic driving would be far simpler (index all visible objects, figure out their paths, then avoid them). We're not there yet, of course, but that's no reason to throw our collective hands up in despair and eschew science forever, just because our current technology is imperfect.
it's not about probability.
Yes, it is. Any pilot must understand the risks of flying, and how to mitigate them. If the pilot's instruments tell him that the plane's speed has slowed, and there is no ground reference (such as when flying through fog, or at night, or even just at cruising altitude), he must understand that there is a chance the plane has actually slowed or hit less dense air, and there's a chance the instrument's malfunctioning. As you mention, the pilot could ask the tower - but the tower's observations are less accurate than even a malfunctioning sensor, can't monitor air conditions as well, and by the time the response comes back, the plane is likely to have stalled. A good pilot knows the probabilities of such problems happening, and knows the probabilities and severities of damage associated with each response.
A human pilot has a few options: Ignore the sensor, call the tower, or angle the nose down immediately. Ignoring the sensor or calling the tower both lead to a chance of a stall, which means the plane will drop and damage will be done anyway (possibly much more severely, up to and including a complete crash landing with no power). Angling the nose down slightly may reduce the risk, but may cause some discomfort to a few passengers. Angling down heavily will remove almost all chance of stalling, but is very likely to cause discomfort to several passengers. Given the options, it's perfectly reasonable (and safest in the long run) to opt for the steep dive, and not kill passengers.
Consider the alternative headline, if the math of risk assessment were ignored in favor of the more human (and more risky) behavior were followed: Computer considered sensor warning as malfunction, crash killed 335
That's assuming that the computer knows what a "nose-dive" even is, or why it's (usually) a bad thing. It would have to know every problem, every tactic, and every risk, and nothing would actually be safer, though the program would be far more complex..
Instead, the "psychological" program thinks "We're going a lot slower than we should for this altitude. Oh no! We're going to stall, and it's only by sheer luck that we haven't already! Why are we this high, anyway? The pilot told me to go this high, but maybe he entered the flight plan wrong. Maybe there was a fight in the cockpit, and that last change wasn't really supposed to happen. Quick! Let's go down to denser air as fast as we can! It's either that or stall, crash, and kill everyone on board!". It still dives, because the basic problem hasn't changed: The sensor failed, and had no redundancy. There's still a bit of self-doubt, but without enough relevant information about what's going on, the program still takes the option that causes the least damage, in terms of probability: the nose-dive.
AI involves large amounts of both math and psychology, and that's not going to change. A major aspect of psychology is to effectively reverse-engineer human actions to determine their underlying mental algorithms. Current AI programs simulate the algorithms, and we compare the emergent behavior to what's observed in humans (or other, simpler, animals). They are effectively the same problem, working in opposite directions.
I like the general concept here, but it isn't much more sustainable than sending up supplies.
you still need to send up the raw material.
Which could be included on the regularly-scheduled crew launches, like food. Having a stock of material on board means that if some part breaks, it's likely fixable without an extra unscheduled launch, which is currently a very expensive option.
A state is any politically-distinct entity, which can be as small as a single town, or as large as the whole EU. The word is much more versatile than the particular usage in the name "United States of America".
The State of the Union address states the state of the state of states.
I've recently become enamored with Tiny Core Linux. It seems ideal for this situation, because it includes so little by default (which is why I'm using it for a VM hosting a single application). It's very easy to make your own custom LiveCD version, including only, for example, a JVM, a testing application, and X.
1. Is it a correct thing to allow interpretation of Constitution?
Yes. The societal context in which the Constitution is viewed changes, as do the very meanings of words (such an "insure" in the preamble to the Constitution, which now relates almost exclusively to financial matters). As an example, consider the curtailing of "free speech" to exclude speech which causes "imminent lawless action". Falsely warning about a bomb in a crowded building is extremely likely to cause assault, theft, and vandalism as people try to escape. Merely advocating illegal behavior at an indeterminate time in the future is not imminent, and is thus not prohibited. There are, of course, other laws that can affect how speech may be presented. You can not abuse or harass others with your ideas, for example. In my opinion, nobody should have a Constitutionally-protected right to be a jackass.
2. Is it a correct thing to allow the government live on debt?
Yes, more or less. First, a large portion of the government's debt is long-term obligations that are not yet fully funded, nor expected to be. If the government has said it will pay several million dollars for a new fighter jet over the next 20 years, that full several-million dollar figure is counted as debt, even though only a small part of it is actually due now. Planning for future expenses is a reasonable thing to do, no? Another large portion of debt is a financial device to free up quantities of money for other uses. More on that shortly.
3. Is it a correct thing to allow the government control money supply and cost?
Yes, when necessary. The government acts (financially) as a large single entity, so if anyone's going to control the money supply, it's going to be the government. Is that control really necessary, though?
According to modern monetary theory, the answer is again "yes". A strong economy is one where money moves freely and quickly, and everyone gets what they want. In other words, "to each according to his need". When the government adds money to the economy through the Federal Reserve Bank, it also adds an equal amount of debt. Increasing the monetary supply allows the public to have more money to spend immediately, with the knowledge that said money will disappear again shortly. It enables a strong economy to be built (or rebuilt), and when the economy is running again, the money supply can be reduced gradually to improve efficiency and reduce the effects of inflation.
There is a riddle about a man dying, leaving his 19 horses to be divided among his three sons, with the eldest receiving half, the middle receiving one fourth, and the youngest receiving one fifth. After trying for several days to figure out how to divide a horse, the local wise man came and brought his own horse, adding it to the pool. The eldest received 10, the middle received 5, and the youngest received 4. The wise man then took his horse and left.
The economy works similarly. With the temporary addition of money, transactions can be processed faster and easier, and operations can go more smoothly. The rapid response of the Federal Reserve Bank reduces the effect of recessions, and speeds recovery.
4. Is it a correct thing to give the government power to insure people in any way (from deposit insurance to health and retirement)?
Yes, when it's a matter that will "insure domestic Tranquility" and "promote the general Welfare".
Deposit insurance reduces the effect of bank runs. Before 1933, when a bank was in danger of closing, customers would rush to get their money out, before it became lost in the disappearing bank. Since banks can't keep all of their holdings on hand at once, some people would inevitably lose their savings. Now, there is no need to make that rush to withdraw, because even if banks are closed, the money is protected. Less damage from bank runs means the banks keep more money available, are are less likely to
Well, of course, if you "perform" professionally at bachelor parties, then perhaps your Facebook page is a marketing tool for your entertainment business. In that case, it should present an image suitable to your profession. If that means insulting your boss to help potential customers identify with you, then so be it.
Slashdot Mirror
Shh... rational thought isn't allowed here. This is Slashdot.
This just in: Guy accusing company of misconduct emphasizes how ethical everybody else is! Film at 11.
Really? I spent time in Africa (volunteering in Ghana), and nearly every business, individual, and official I dealt with was corrupt to some degree. I was actually surprised when a taxi driver reminded us of the agreed-upon rate, and it was the same as what we'd agreed.
Now, we're not talking Western white-collar corruption, where an executive will duck out to play golf and write it off as a business expense. We're talking about a construction company who won't start work until you start paying them so they can get supplies, then pay a little more to fix the tire on their truck, then more for gas, then more to hire a new worker, then more to rush the job that's now running behind. Contracts mean nothing, because the court officials will decide in favor of whoever offers them the biggest bribe without getting caught by the few people in government who think that corruption is wrong, despite its widespread cultural approval.
If you want to get things done in Africa, the best way is to hire a local who can navigate the local politics (because he grew up in the village, is a member of the local tribe, and has been in good standing all his life), and bribe the local chief to pressure the village to help you. Give that local a budget for bribes, and just accept that they happen. It's certainly not the Western standard of business, but it works. Things get done, the bribes are just a cost of doing business, and the locals appreciate that you're playing their game, instead of forcing your silly ideals onto their ancient traditions.
4 is "Add a large markup by charging extra for cheaper/free services, when all you really do is forward everything to Google"
IP addresses can be easily hijacked, and it's still no indication of actual wrongdoing within US jurisdiction. Perhaps Google gives IP addresses out to any company who wanted to pay them $10/month, as so many ISPs do. According to the post currently below this one, that's exactly what Google's done.
Assuming that Google is ultimately at faulty, the only way I see where Google would be liable under US law, would be if each of the involved branches were first found to be liable. Then there's a reasonable chance that the branches were coordinated by the parent corporation, and it can be pursued.
Misunderstood instructions make things even more difficult, as the subordinate was "just following orders" from a superior who didn't think he was giving them!
The people who think we're able to accurately model the world economy are the same ones who think we're able to model world climate. Maybe, one day, with enough understanding of how all the major underlying principles work - both independently and interacting with one another - we'll be able to model either of them.
The climate has been behaving roughly the same for tens of thousands of years. The global economy has been around for about 30.
For a few thousand years, we've been able to model the climate with absolute accuracy, if little precision. Once each year, the northern hemisphere gets colder, and the southern hemisphere is warmer. Six months later, those roles are reversed.
Similarly, we've been able to model simple events within the economy with remarkable accuracy. Company stocks almost always go up after announcing a new product, and down during a big (really big, not just "important to Slashdotters") lawsuit. Of course, the chaotic events with subtle causes (like exactly when an economic bubble will burst or where a tornado will strike) are much more difficult to predict. All that we can do now is try our best.
Of course, we can take the same asinine comparison to ridiculous conclusions. Neither mathematics nor psychology perfectly predicts the behavior of drivers, so perhaps we shouldn't install any traffic lights until we can perfectly predict where collisions will occur? We can't predict software vulnerabilities, so we shouldn't install any current antivirus software. We can't predict how a person's life will go, so we should stop having children.
As the population increases, individual psychology gives way to statistics, in terms of usefulness. Even "extremely complex" behavior follows probability distributions, and can be expected to remain within those distributions.
Let's consider that a captain of a sinking ship may or may not stay by his post to drown. In fact, let's make a wager on it. Now, before you decide what position of the gamble to take, go ask a nearby ship captain (assuming you're not in Utah) if he'll go down with his ship. Let's assume he says he will. If our wager is based on that one specific captain's choice, then you'll have the best chance of winning, of course. If, however, we wager based on whether "the next" or other suitable random captain drowns, then your brief interview (and the resulting psychological insight) is nearly useless, and a better predictor of the outcome would be the historic statistics of drowning captains.
No mathematical model is ever perfectly accurate. Rather, the models are refined as more data is available, accounting for the influences of old models. What is the currently accepted standard is merely the latest iteration of refinement.
Psychology (as a foundation for economic theory) is no more accurate, but merely produces models without the presence of descriptive numbers. On a playground merry-go-round, where a mathematical model may describe a 90% chance of a child falling off within the next 2 minutes, a psychological model predicts only that a child will fall off soon.
Despite the summary, the patent describes a particular specific application, where every aspect of the system is based around minute-granularity timestamps, rather than mere dates. The application also follows a particular architecture, which is all nicely explained in the actual patent itself. The half-day example was just mentioned for clarity, and does not affect the patent itself.
Probably not Ron Paul any time.
His preferred position on economics is to ignore that silly mathematics stuff and go with a mix of psychology and gut instinct.
A corporation is a joint effort of one or more people. Are you saying that individuals should be stripped of rights when they start to cooperate?
As Microsoft will be quick to tell you, there are many different forms of "open", and only a few of them exempt you from getting screwed.
If your data's integrity is absolutely necessary, Hadoop (or more specifically HBase, which is the part most closely analogous to a database) is probably not for you. On the other hand, if you're working with statistics or any other application where an error affects your product trivially, you may find the speed is worth it, bearing in mind that changes are broadcast across the cluster "eventually". The strengths and weaknesses of Hadoop are different from a traditional database enough that I'd caution your architect against migrating an existing application just for the sake of speed. There are numerous pitfalls and headaches down that path.
New Madrid is also primarily under Missouri. Its effects in Ohio consist of influencing other faults and (in the 1812 earthquake you mentioned) causing residents in a town to leave their homes. For comparison, note that California has had significant earthquakes roughly every year along the San Andreas fault. What's actually in Ohio is far smaller and even less active than the main New Madrid area.
The fracking debate currently has little to do with logic. There's precious little actual evidence of causing harm (or evidence of causing no harm), and the regulators know this. That's why they resist making regulations, even though it's pissing off the general public who just want to see anything happen. Meanwhile, the industry just continues operating as normal. So far, the vast majority of studies (from both sides) conclude nothing one way or the other. Those that do (again, in both directions) are deeply flawed, lacking silly things like control groups.
By my understanding, it would be trivially simple for the energy industry to run a causative study, by testing the output of several wells near a fracking site before and after any fracking operation. The company knows where they're building their new well, and could spend a few hundred thousand dollars of their propaganda budget to drill a well and run some tests. Of course, they never will out of fear that something bad might be discovered, and even if there was no evidence of contamination, the study would never be believed by the anti-fracking protesters.
Within the last month, the EPA announced that it found the first case of a water supply being contaminated by fracking, in a community near a shallow fracking well. The EPA itself has stated that the study needs still more review and reconsideration before it should be used as a basis for any regulations. Of course, both sides have already fired up their most imaginative writers, either condemning the study or overstating its significance.
I can't tell if you intend to include environmentalists in your classification of "everyone with an agenda", but it would certainly be justified. There's enough propaganda from both sides of the science-free debate to disrupt honest attempts at science. It's sickening.
Ohio is on many small faults, the largest of which is the New Madrid fault. There are a few dozen significant earthquakes each year, the vast majority of which cannot be felt.
Despite what you "would think", data is easy to find.
Having an opinion doesn't make one a shill or a troll, especially when there's as much evidence supporting his opinion as your own. The problem is there's absurdly little research from both the pro- and anti-fracking camps. On the one hand, Ohio's seismic activity has increased lately. On the other hand, it has been very inactive since the 1930's, and still remains relatively stable today. Then, of course, there's the possibility (mentioned many times already in this discussion) that releasing pressure could reduce the risk of a larger earthquake.
Comparing Ohio's seismic activity to California is ridiculous. In Ohio, the last big earthquake in 1937 toppled a few weak chimneys. In California, an equivalent earthquake (magnitude 5.4) happened in July of 2010. The faults in Ohio, even when active, pale in comparison to California's eternal fear of the next "big one".
There's no consensus among relevant experts about fracking's effects, but there's plenty of people willing to protest vehemently one way or the other. GP is right to call this out as fear-mongering.
interactive autonomous systems
So you've rediscovered closed feedback loops?
I'm going to assume you're not trolling at this point, though the utter rejection of match makes it unlikely you'r being serious. Oh well. It must be dinnertime.
there is absolutely no way that your brain does calculus in order to walk around an obstacle.
Actually, yes, the brain does some basic physics computations (including calculus) naturally. The answers aren't exact, and there aren't numbers given for output, but the process is straightforward: To predict the path of an obstacle (or several), first watch for a moment to determine the rates of change (velocity v and acceleration a), then apply that to a "black box" formula or neural network (x=x[0] + vt + at^2), then continue to check the result as more observations are available. Physicists have been observing and verifying for the past few thousand years to ensure that the formula we use now is as accurate as our current data allows. Human brains also observe, but they don't have the same accuracy. Instead, brains rely on having lots of repetitive observations, to hopefully train their neural networks to be a closer approximation of accurate. This is why baseball players practice: to improve their aim by refining the neurons they use.
The same concept applies to many operations that humans are good at, like facial and environmental recognition, quantitative estimation, and memory. The brain has neural networks for these important functions, and when someone's brain is damaged, these functions are often impaired. It's important to note that the neural networks are usually far faster than current digital technology, partly due to being analog in nature, partly due to parallelism, and partly (probably) due to using algorithms we don't know or understand yet.
If we did somehow have a magic digital circuit that could do visual recognition and compute the path of objects as fast as the brain can, difficult tasks like robotic driving would be far simpler (index all visible objects, figure out their paths, then avoid them). We're not there yet, of course, but that's no reason to throw our collective hands up in despair and eschew science forever, just because our current technology is imperfect.
it's not about probability.
Yes, it is. Any pilot must understand the risks of flying, and how to mitigate them. If the pilot's instruments tell him that the plane's speed has slowed, and there is no ground reference (such as when flying through fog, or at night, or even just at cruising altitude), he must understand that there is a chance the plane has actually slowed or hit less dense air, and there's a chance the instrument's malfunctioning. As you mention, the pilot could ask the tower - but the tower's observations are less accurate than even a malfunctioning sensor, can't monitor air conditions as well, and by the time the response comes back, the plane is likely to have stalled. A good pilot knows the probabilities of such problems happening, and knows the probabilities and severities of damage associated with each response.
A human pilot has a few options: Ignore the sensor, call the tower, or angle the nose down immediately. Ignoring the sensor or calling the tower both lead to a chance of a stall, which means the plane will drop and damage will be done anyway (possibly much more severely, up to and including a complete crash landing with no power). Angling the nose down slightly may reduce the risk, but may cause some discomfort to a few passengers. Angling down heavily will remove almost all chance of stalling, but is very likely to cause discomfort to several passengers. Given the options, it's perfectly reasonable (and safest in the long run) to opt for the steep dive, and not kill passengers.
Consider the alternative headline, if the math of risk assessment were ignored in favor of the more human (and more risky) behavior were followed: Computer considered sensor warning as malfunction, crash killed 335
you don't grasp a glass by dete
That's assuming that the computer knows what a "nose-dive" even is, or why it's (usually) a bad thing. It would have to know every problem, every tactic, and every risk, and nothing would actually be safer, though the program would be far more complex..
Instead, the "psychological" program thinks "We're going a lot slower than we should for this altitude. Oh no! We're going to stall, and it's only by sheer luck that we haven't already! Why are we this high, anyway? The pilot told me to go this high, but maybe he entered the flight plan wrong. Maybe there was a fight in the cockpit, and that last change wasn't really supposed to happen. Quick! Let's go down to denser air as fast as we can! It's either that or stall, crash, and kill everyone on board!". It still dives, because the basic problem hasn't changed: The sensor failed, and had no redundancy. There's still a bit of self-doubt, but without enough relevant information about what's going on, the program still takes the option that causes the least damage, in terms of probability: the nose-dive.
AI involves large amounts of both math and psychology, and that's not going to change. A major aspect of psychology is to effectively reverse-engineer human actions to determine their underlying mental algorithms. Current AI programs simulate the algorithms, and we compare the emergent behavior to what's observed in humans (or other, simpler, animals). They are effectively the same problem, working in opposite directions.
I like the general concept here, but it isn't much more sustainable than sending up supplies. you still need to send up the raw material.
Which could be included on the regularly-scheduled crew launches, like food. Having a stock of material on board means that if some part breaks, it's likely fixable without an extra unscheduled launch, which is currently a very expensive option.
I once saw a saw saw a saw...
A state is any politically-distinct entity, which can be as small as a single town, or as large as the whole EU. The word is much more versatile than the particular usage in the name "United States of America".
The State of the Union address states the state of the state of states.
I've recently become enamored with Tiny Core Linux. It seems ideal for this situation, because it includes so little by default (which is why I'm using it for a VM hosting a single application). It's very easy to make your own custom LiveCD version, including only, for example, a JVM, a testing application, and X.
1. Is it a correct thing to allow interpretation of Constitution?
Yes. The societal context in which the Constitution is viewed changes, as do the very meanings of words (such an "insure" in the preamble to the Constitution, which now relates almost exclusively to financial matters). As an example, consider the curtailing of "free speech" to exclude speech which causes "imminent lawless action". Falsely warning about a bomb in a crowded building is extremely likely to cause assault, theft, and vandalism as people try to escape. Merely advocating illegal behavior at an indeterminate time in the future is not imminent, and is thus not prohibited. There are, of course, other laws that can affect how speech may be presented. You can not abuse or harass others with your ideas, for example. In my opinion, nobody should have a Constitutionally-protected right to be a jackass.
2. Is it a correct thing to allow the government live on debt?
Yes, more or less. First, a large portion of the government's debt is long-term obligations that are not yet fully funded, nor expected to be. If the government has said it will pay several million dollars for a new fighter jet over the next 20 years, that full several-million dollar figure is counted as debt, even though only a small part of it is actually due now. Planning for future expenses is a reasonable thing to do, no? Another large portion of debt is a financial device to free up quantities of money for other uses. More on that shortly.
3. Is it a correct thing to allow the government control money supply and cost?
Yes, when necessary. The government acts (financially) as a large single entity, so if anyone's going to control the money supply, it's going to be the government. Is that control really necessary, though?
According to modern monetary theory, the answer is again "yes". A strong economy is one where money moves freely and quickly, and everyone gets what they want. In other words, "to each according to his need". When the government adds money to the economy through the Federal Reserve Bank, it also adds an equal amount of debt. Increasing the monetary supply allows the public to have more money to spend immediately, with the knowledge that said money will disappear again shortly. It enables a strong economy to be built (or rebuilt), and when the economy is running again, the money supply can be reduced gradually to improve efficiency and reduce the effects of inflation.
There is a riddle about a man dying, leaving his 19 horses to be divided among his three sons, with the eldest receiving half, the middle receiving one fourth, and the youngest receiving one fifth. After trying for several days to figure out how to divide a horse, the local wise man came and brought his own horse, adding it to the pool. The eldest received 10, the middle received 5, and the youngest received 4. The wise man then took his horse and left.
The economy works similarly. With the temporary addition of money, transactions can be processed faster and easier, and operations can go more smoothly. The rapid response of the Federal Reserve Bank reduces the effect of recessions, and speeds recovery.
4. Is it a correct thing to give the government power to insure people in any way (from deposit insurance to health and retirement)?
Yes, when it's a matter that will "insure domestic Tranquility" and "promote the general Welfare".
Deposit insurance reduces the effect of bank runs. Before 1933, when a bank was in danger of closing, customers would rush to get their money out, before it became lost in the disappearing bank. Since banks can't keep all of their holdings on hand at once, some people would inevitably lose their savings. Now, there is no need to make that rush to withdraw, because even if banks are closed, the money is protected. Less damage from bank runs means the banks keep more money available, are are less likely to
Well, of course, if you "perform" professionally at bachelor parties, then perhaps your Facebook page is a marketing tool for your entertainment business. In that case, it should present an image suitable to your profession. If that means insulting your boss to help potential customers identify with you, then so be it.