You don't deter navies with ICBMs. The USSR sure didn't. Long-range supersonic anti-ship missiles are the way to go.
Both the US and the USSR had the capability of conducting nuclear war against naval task forces, potentially with ICBMs. That doesn't mean that they didn't also develop the means to fight conventionally against them as well. Nuclear war is basically a last resort that nobody wants to use, and it is politically useful to have an option between tolerate anything your enemy does and kill everybody on the planet.
The potential for nuclear warfare has a big impact on naval tactics. In a conventional war you bunch your ships up to provide mutual support (if your anti-air ships are scattered over a 100 mile radius then an attack from any direction would only be engaged by a single ship, but if they're all in a tight ring around the capital ships then potentially all of them can defend against any attack). In a nuclear war you want to spread ships out so that a single warhead won't take out the entire task force. That said, anyone with a reasonable number of nuclear weapons and half-decent intel on target location is almost certain to wipe out a naval group. The US doesn't have that many carriers, and if you MIRV a large area with a few ICBMs you're likely to wipe out any you target. But, the chances of a tactical nuclear war staying tactical doesn't doesn't seem good. One guy nukes a task force at sea, the next guy nukes a task force in harbor, then you're nuking bases, and many of those are right next to cities. Many believe that as soon as one nuke flies, they basically all fly (at least as far as the superpowers are concerned - if NK fired one off and the US retaliated against them only either conventionally or in a limited nuclear strike I don't think that would necessarily start WWIII).
Nukes really aren't much of a deterrent in any case. What would happen if the Russians invaded Alaska en masse? Would the US nuke them and thus kill virtually every person on the planet? No, we'd get into a big conventional war. Nukes in the possession of stable governments really are just a deterrent against full-scale nuclear war, since that is about the only condition they'd be used under.
They're more of a deterrent against conventional attack if they're in the hands of a nutcase. I could see NK taking out Tokyo over an invasion - even though it would have no impact on the result of the war. For a dictatorship the loss of even a fair portion of the population is acceptable but the loss of the dictator is absolutely unacceptable.
A lot would have to change before India would really be concerned about the US. However, for those missiles to be a credible threat they'd have to reach the US in numbers that could thwart a defensive system. Even SLBMs are only a threat if they are on subs capable of evading US attack subs. I imagine that the US routinely trails any ballistic missile sub in the ocean - in the event of war they could be sunk before they fired a shot (assuming the US subs get the message fast enough).
There is always the risk that a bunch of Indian subs all leave with orders to just launch without warning at a pre-arranged time as a first strike, and unless the US subs have orders to open fire as soon as they make launch preparations they'd get off many if not all of their missiles. It would be unlikely that US subs would have such orders unless a state of near-war existed, and if things were that bad they'd probably just sink them as soon as they saw them. If subs routinely had such orders you'd see shots fired anytime somebody conducted a test launch.
I use it in the same way. It still goes into my inbox so I don't have most of the pain associated with filtering into a bazillion folders, but I can at least get a notification when somebody RSVPs to the event that I'm hosting in 15 minutes, and not when Aunt Tizzy sends out cat pictures.
Seriously! That's just over 1% of the entire population. Chances are that if you live in the US one of your neighbors has VPN access to the thing at that rate.
Actually, if they create a fighter with the performance of the F-35, it wouldn't be a problem at all... as the F-35 is massively expensive
When a fighter jet costs $150M/plane it usually means that the plane takes $10M in materials and labor to build, and $140M goes towards paying off the costs of designing the thing in the first place. It is really a $10M plane with a $1T design phase (or whatever the figure is).
Somebody copying the plane only needs to pay the $10M/plane - they don't have to redesign the whole thing from scratch. I'm sure it won't cost them nothing to start from the US blueprints, but overall it will be WAY cheaper.
taking years longer to develop
Not an issue for China. They'll just wait until we're done, and then roll out the copies after a year or two of reverse-engineering. In the meantime nobody is flying the thing.
and still can barely get off the ground.
Also not an issue for China. They'll just wait until we figure out all the problems and then copy the design that actually works.
Copying is WAY cheaper than inventing. Even if all they had as a photo of the thing it would be much cheaper. How many overall designs were tossed because using thrust vectoring vs a lift fan was an unclear design decision? The US has to spend hundreds of millions on prototypes and testing to figure out which design is better. The Chinese just have to see what we picked. If the whole VTOL design turns out to be impractical and gets canceled then they get the same data point that we get but for zero cost.
Today it is easy to point out what the design of the space shuttle was bad, even without the blueprints. Anybody who is interested in submarines knows that a 7-blade propeller is much quieter than a 4-blade one, but for many years this was a closely guarded secret that just a glance at a propeller would have leaked.
When you're doing something that has never been done before most of the cost is only incurred by the first person to have to figure it out. That's the price of innovation. Followers can always do it much cheaper.
I'm running Steam on linux right now - its binaries are neither SUID root, nor do they have any capabilities assigned. At least, not the version I'm running on my distro.
However, there is no inherent flaw in using hardware-based random number generation with a non-deterministic source of entropy.
There isn't any flaw in using ANY method of generating random numbers if it involves a non-deterministic source of entropy. The problem is that nobody can prove that such a thing exists. You can only say that something looks random and that nobody has publicly disclosed that they've found a pattern in it.
I agree that most physicists think that the timing of individual decay events is non-deterministic, along with many other quantum events. However, you're talking about an area of science that is hardly understood from first principles. We can't even describe how gravity works at the quantum scale, and yet we KNOW that gravity is real. Who is to say that somebody won't figure out that we're wrong about things being non-deterministic?
The only time you can know anything with certainty is in the world of mathematical proofs, and you can only say that something in math is certain in the sense that it doesn't contradict any axioms that you have accepted. The problem is that this is all an abstraction - there is no way to prove that anything in the real world behaves in any particular way to mathematical certainty. You can only obtain increasing levels of confidence through science.
It's really just finally getting to the point where it is getting easier to buy bitcoins with THINGS instead of MONEY if you want to maintain your pseudonymity.
Sure, assuming you can find another party to sell your things to who doesn't log your identity. However, chances are the money you buy in this way is traceable to the person who sold it to you, which means that they'll probably get questioned by the police when you use that money to do something illegal. If that becomes a trend, expect to find few people willing to trade things for bitcoins without ID.
That case was novel because it was the first time a court DIDN'T toss somebody in jail indefinitely for failing to disclose an encryption key. The only reason they ruled this way was because the prosecution couldn't give evidence that it was likely that the defendant knew the key, and divulging the key would have therefore provided evidence that he did, and was therefore associated with the encrypted content. The defendant argued that he had just received the hard drive and did not know anything about the encrypted content on it. If it were a computer he had bought new the court would have ruled differently.
I agree that the way the courts handle these cases is completely contrary to the 5th amendment, but you can argue that all you want while you rot in jail. Just google for contempt of court for not disclosing encryption keys for many examples.
Presumably someone as paranoid about anonymity as this would have their wallet file encrypted.
I agree with the rest of your post about reasonable suspicion, but encryption won't get you anything. In the US they can keep you in jail for life for failing to divulge an encryption key, even if you were only charged with a crime that carried a six month sentence.
I thought using a new address for every transaction was the advised method of using bitcoin. I don't use it myself, but I remember reading something along the lines of "never reuse your addresses" in the documentation.
How do you get bitcoins into the new address? You either have to mine it, buy it, or transfer it from another address. Most likely you transfer it, and what account you transferred it from is public record. So, all your accounts are effectively linked, even if you use a new one for every transaction.
Sure, you can have multiple completely-independent accounts, but you can't move money between them unless you pass it out through cash, or you mine it. The ability to pass it out through cash is limited due to the developments in TFA - banks are asking for ID. So, unless you mine ALL the cash you use in discrete unrelated bundles, you're traceable. Mining is pretty expensive and a hassle, though it can be done (though over time it will become less and less practical).
And if the contract stipulates that AT&T can add whatever surcharge they like, why are customers complaining?
Such a contract would likely be unenforceable as a matter of law, or at least it should be. By their very nature the terms of contracts cannot be modified except by mutual consent. Contracts can include variable rates, but usually they need to be very well-defined. A contract that says you'll pay whatever expenses we happen to incur without any choice in the matter wouldn't qualify.
Yeah, this. What the AC may be confused about is that faster than light travel is (as far as we know) not possible in space, but the distance between two points can increase faster than light could travel because there's nothing stopping space itself from expanding that fast.
That's relativity, but the nice thing about this is that you can't actually observe it happening. If you put two golf balls in a room and then expanded the space between them so that they move a light year apart in a second, you'd never see either golf ball moving faster than the speed of light away from you, whether you are viewing from the perspective of either ball, or an observer in the room. You'd just see them move apart at a speed very close to the speed of light.
Objects further away than the cosmic horizon are in theory moving away from us faster than the speed of light. However, if we look at them all we see is a background of objects close to the horizon that are so red-shifted that they're barely detectable at all moving away at almost the speed of light. We can't see beyond the horizon (hence the name), and we can only see the horizon itself as a wash of particles coming from every direction (otherwise known as the cosmic microwave background).
I'm sure he was talking about an orbit with a low inclination in general, not one that was exactly zero. In any case, it does tend to launch to the east. If you wanted to launch from the east coast into a polar orbit it would probably be cheaper to do it at a more northern facility like Wallops Island.
However, if you were to use a mortgage product as a short-term loan facility you are benefiting from a lower rate. In order to prevent abuse of the system there is an early repayment penalty. It's fairly straight forward.
This really only makes sense if the bank had incorporated some of its up-front costs into the interest rate, instead of the closing costs. If the bank is covering all of its transactional costs via the up-front fees then you could open and pay-off a mortgage three times a day and they'd only profit from it.
As long as the market continues to offer a broad range of terms I don't really have a problem with this practice. Choices generally are good. I'd consider a ban on the practice if everybody started shifting towards only offering these kinds of loans, because they tend to create lock-in which isn't good for the market.
Why? It is because they are truly random. Each single outcome is just as likely as every other, which means that in the long run, the outcomes will occur pretty much equally often (give or take).
Each outcome cannot be equally likely. An atom of uranium at any moment has to be more likely to decay than an atom of carbon, since the former decays and the latter does not.
Sure, radioactive decay certainly seems random on short terms. However, it is entirely possible that some day we'll come up with a way of describing its behavior and it will turn out not to be random.
A good post, but I'm not sure you understand hardware based random number generation. At least one way to do it is have a small amount of radiactive material. Although it decays predictably in the long term (half life) it is random in the short term.
How do you KNOW that it is random in the short term? The fact that you cannot currently predict its behavior does NOT necessarily mean that it is random. If it were truly random, then why is there a long-term predictable behavior?
I'm not sure that this really helps clarify whether this is actually a quantum computer.
The question is whether the D-Wave is a quantum computer, not whether it is faster than a classical computer. A quantum computer that is slower than a desktop calculator is still useful for research purposes.
Agreed. If anything antibiotics and vaccines have completely opposite mechanisms of action.
An antibiotic taken as a medication kills bacteria directly, assisting the immune system and making its job easier. In the case of bleaching every surface in your house, it means that the immune system never sees the bacteria in the first place. The same is true of other external use of antibiotics (killing of bacteria before it gets into your body).
A vaccine provokes your immune response against a pathogen without exposing you to the risk of developing the disease (or a greatly reduced risk). Your immune system does all the work, and as a result it is able to do the job entirely on its own much more effectively at a later time.
Comparing the approaches, the disinfectant approach is like bleaching your house 3x/day, and the vaccine approach is like rolling around in the mud and not washing before dinner. I'd be very hesitant to associate the problems of the one with the other.
Let me ask you a question – is your objection that felons who have served their time can't vote or that the standard for felonies – those major crimes against society – has been watered down? Because it sounds to me that it is the watering down of felonies that is your issues – and I would agree with you there.
I would object to both.
Somebody convicted of a crime is either a danger to society or they aren't. If they are dangerous to be allowed in public, then they shouldn't be allowed in public - full stop. If they aren't, then quite badgering about it for the rest of their life. Frankly our criminal justice system needs to be a lot less punitive and a lot more rehabilitative. I'm fine with deterring crime, but clearly that on its own doesn't work. If a criminal can't be rehabilitated then they should get a life sentence, even if all they did was beat somebody up. If they can be rehabilitated, then they should be released once they're able to function normally in society, even if they killed 35 people. As a citizen my concern is not whether the guy across the street was appropriately punished like a 12 year old, but rather whether they're capable of not acting like a 12 year old now.
Sure, the system will always be imperfect, but I don't really see much value in permanent sanctions. By all means use parole (and by that I mean an invasive probation where you actually help the parolee re-integrate over years with heavy contact), but once they're just an ordinary citizen, let them be an ordinary citizen (heaven forbid that criminals that rehabilitate have something to look forward to).
When you turn people into second-class citizens they'll start acting like second-class citizens.
Banning gun ownership by felons is also doubly silly. If you think they're not dangerous then why ban gun ownership? If you do think they have no regard for the law, then why do you think outlawing gun ownership will stop them when whatever law they previously broke failed to do so?
Imagine you step into a teleporter, in an atomic instant you disappear and an exact replica of you appears on the other side. In this hypothetical example the position and composition of every particle that composes you is perfectly replicated, momentum and all (no messy Uncertainty Principle issues).
In this highly idealized scenario, you could debate whether your consciousness really does teleport, or whether the guy who steps out the other side merely thinks that it does.
Of course, you could probably make the same argument with motion in general. When you move every particle in your body moves along, and at one point in time they are at one place, and at the next moment they are at another. I'm not sure how teleportation is any different beyond the distance being larger.
Of course, to the degree that teleportation is imperfect, then the matter is compounded.
I think another huge challenge is just going to be wiring them up correctly initially.
I used to think a brain was just a box full of neurons that had inputs on one side and outputs on the other and they were trained like any other simple neural network, but with far many nodes. As I've read up on things (and lived with somebody with anomic aphasia that has improved over time) I realized that nothing could be further from the truth.
There is very clearly some kind of basic architecture to the brain that involves incredibly intricate connections that are basically pre-wired but which evolve over time. There are specialized regions of the brain, and most operations involve many regions of the brain working together in various ways. There are both short- and long-distance connections all over the place.
My personal theory is that there are many building-blocks of functionality that are suited to particular purposes, and that as brains evolved these got stitched together in ways that led to all kinds of emergent properties. You have areas like the cerebellum which seem (to my poorly educated brain) to be a bit like the simplistic neural network model, and I suspect the rest of the brain uses it like a programmer might use a computer to automate certain types of operations (like feedback loops for balance and who knows what else). I suspect that different regions of the brain "use" each other in similar ways, almost symbiotically.
If I had the time to really study this stuff I'd probably start by trying to understand fundamental building blocks of neural networks. I figure that developmentally this stuff has to form in some kind of almost fractal pattern since embryonic cells really only can keep track of their immediate surroundings and cell division counts. What kinds of patterns lead to what kinds of processing abilities?
Really fascinating stuff - it will be truly amazing if we ever figure it out.
Yup. To the point where quantum effects impact a brain, it might be impossible to replicate a human mind. You could probably make a decent facsimile, but it could differ in subtle ways.
When you decide where to eat lunch, for all we know the decision was influenced by a cosmic ray that originated in a supernova halfway across the universe.
Slashdot Mirror
You don't deter navies with ICBMs. The USSR sure didn't. Long-range supersonic anti-ship missiles are the way to go.
Both the US and the USSR had the capability of conducting nuclear war against naval task forces, potentially with ICBMs. That doesn't mean that they didn't also develop the means to fight conventionally against them as well. Nuclear war is basically a last resort that nobody wants to use, and it is politically useful to have an option between tolerate anything your enemy does and kill everybody on the planet.
The potential for nuclear warfare has a big impact on naval tactics. In a conventional war you bunch your ships up to provide mutual support (if your anti-air ships are scattered over a 100 mile radius then an attack from any direction would only be engaged by a single ship, but if they're all in a tight ring around the capital ships then potentially all of them can defend against any attack). In a nuclear war you want to spread ships out so that a single warhead won't take out the entire task force. That said, anyone with a reasonable number of nuclear weapons and half-decent intel on target location is almost certain to wipe out a naval group. The US doesn't have that many carriers, and if you MIRV a large area with a few ICBMs you're likely to wipe out any you target. But, the chances of a tactical nuclear war staying tactical doesn't doesn't seem good. One guy nukes a task force at sea, the next guy nukes a task force in harbor, then you're nuking bases, and many of those are right next to cities. Many believe that as soon as one nuke flies, they basically all fly (at least as far as the superpowers are concerned - if NK fired one off and the US retaliated against them only either conventionally or in a limited nuclear strike I don't think that would necessarily start WWIII).
Nukes really aren't much of a deterrent in any case. What would happen if the Russians invaded Alaska en masse? Would the US nuke them and thus kill virtually every person on the planet? No, we'd get into a big conventional war. Nukes in the possession of stable governments really are just a deterrent against full-scale nuclear war, since that is about the only condition they'd be used under.
They're more of a deterrent against conventional attack if they're in the hands of a nutcase. I could see NK taking out Tokyo over an invasion - even though it would have no impact on the result of the war. For a dictatorship the loss of even a fair portion of the population is acceptable but the loss of the dictator is absolutely unacceptable.
A lot would have to change before India would really be concerned about the US. However, for those missiles to be a credible threat they'd have to reach the US in numbers that could thwart a defensive system. Even SLBMs are only a threat if they are on subs capable of evading US attack subs. I imagine that the US routinely trails any ballistic missile sub in the ocean - in the event of war they could be sunk before they fired a shot (assuming the US subs get the message fast enough).
There is always the risk that a bunch of Indian subs all leave with orders to just launch without warning at a pre-arranged time as a first strike, and unless the US subs have orders to open fire as soon as they make launch preparations they'd get off many if not all of their missiles. It would be unlikely that US subs would have such orders unless a state of near-war existed, and if things were that bad they'd probably just sink them as soon as they saw them. If subs routinely had such orders you'd see shots fired anytime somebody conducted a test launch.
I don't mind getting Aunt Tizzy's emails. I just don't need to be alerted anytime they come in.
I use it in the same way. It still goes into my inbox so I don't have most of the pain associated with filtering into a bazillion folders, but I can at least get a notification when somebody RSVPs to the event that I'm hosting in 15 minutes, and not when Aunt Tizzy sends out cat pictures.
No argument there. Even if the plane only costs $10M to build, it STILL costs $10M to build and trying to build it for $8M will probably not end well.
It also has 4.2M users, which is far too many.
Seriously! That's just over 1% of the entire population. Chances are that if you live in the US one of your neighbors has VPN access to the thing at that rate.
Actually, if they create a fighter with the performance of the F-35, it wouldn't be a problem at all... as the F-35 is massively expensive
When a fighter jet costs $150M/plane it usually means that the plane takes $10M in materials and labor to build, and $140M goes towards paying off the costs of designing the thing in the first place. It is really a $10M plane with a $1T design phase (or whatever the figure is).
Somebody copying the plane only needs to pay the $10M/plane - they don't have to redesign the whole thing from scratch. I'm sure it won't cost them nothing to start from the US blueprints, but overall it will be WAY cheaper.
taking years longer to develop
Not an issue for China. They'll just wait until we're done, and then roll out the copies after a year or two of reverse-engineering. In the meantime nobody is flying the thing.
and still can barely get off the ground.
Also not an issue for China. They'll just wait until we figure out all the problems and then copy the design that actually works.
Copying is WAY cheaper than inventing. Even if all they had as a photo of the thing it would be much cheaper. How many overall designs were tossed because using thrust vectoring vs a lift fan was an unclear design decision? The US has to spend hundreds of millions on prototypes and testing to figure out which design is better. The Chinese just have to see what we picked. If the whole VTOL design turns out to be impractical and gets canceled then they get the same data point that we get but for zero cost.
Today it is easy to point out what the design of the space shuttle was bad, even without the blueprints. Anybody who is interested in submarines knows that a 7-blade propeller is much quieter than a 4-blade one, but for many years this was a closely guarded secret that just a glance at a propeller would have leaked.
When you're doing something that has never been done before most of the cost is only incurred by the first person to have to figure it out. That's the price of innovation. Followers can always do it much cheaper.
I'm running Steam on linux right now - its binaries are neither SUID root, nor do they have any capabilities assigned. At least, not the version I'm running on my distro.
However, there is no inherent flaw in using hardware-based random number generation with a non-deterministic source of entropy.
There isn't any flaw in using ANY method of generating random numbers if it involves a non-deterministic source of entropy. The problem is that nobody can prove that such a thing exists. You can only say that something looks random and that nobody has publicly disclosed that they've found a pattern in it.
I agree that most physicists think that the timing of individual decay events is non-deterministic, along with many other quantum events. However, you're talking about an area of science that is hardly understood from first principles. We can't even describe how gravity works at the quantum scale, and yet we KNOW that gravity is real. Who is to say that somebody won't figure out that we're wrong about things being non-deterministic?
The only time you can know anything with certainty is in the world of mathematical proofs, and you can only say that something in math is certain in the sense that it doesn't contradict any axioms that you have accepted. The problem is that this is all an abstraction - there is no way to prove that anything in the real world behaves in any particular way to mathematical certainty. You can only obtain increasing levels of confidence through science.
It's really just finally getting to the point where it is getting easier to buy bitcoins with THINGS instead of MONEY if you want to maintain your pseudonymity.
Sure, assuming you can find another party to sell your things to who doesn't log your identity. However, chances are the money you buy in this way is traceable to the person who sold it to you, which means that they'll probably get questioned by the police when you use that money to do something illegal. If that becomes a trend, expect to find few people willing to trade things for bitcoins without ID.
That case was novel because it was the first time a court DIDN'T toss somebody in jail indefinitely for failing to disclose an encryption key. The only reason they ruled this way was because the prosecution couldn't give evidence that it was likely that the defendant knew the key, and divulging the key would have therefore provided evidence that he did, and was therefore associated with the encrypted content. The defendant argued that he had just received the hard drive and did not know anything about the encrypted content on it. If it were a computer he had bought new the court would have ruled differently.
I agree that the way the courts handle these cases is completely contrary to the 5th amendment, but you can argue that all you want while you rot in jail. Just google for contempt of court for not disclosing encryption keys for many examples.
Presumably someone as paranoid about anonymity as this would have their wallet file encrypted.
I agree with the rest of your post about reasonable suspicion, but encryption won't get you anything. In the US they can keep you in jail for life for failing to divulge an encryption key, even if you were only charged with a crime that carried a six month sentence.
I thought using a new address for every transaction was the advised method of using bitcoin.
I don't use it myself, but I remember reading something along the lines of "never reuse your addresses" in the documentation.
How do you get bitcoins into the new address? You either have to mine it, buy it, or transfer it from another address. Most likely you transfer it, and what account you transferred it from is public record. So, all your accounts are effectively linked, even if you use a new one for every transaction.
Sure, you can have multiple completely-independent accounts, but you can't move money between them unless you pass it out through cash, or you mine it. The ability to pass it out through cash is limited due to the developments in TFA - banks are asking for ID. So, unless you mine ALL the cash you use in discrete unrelated bundles, you're traceable. Mining is pretty expensive and a hassle, though it can be done (though over time it will become less and less practical).
And if the contract stipulates that AT&T can add whatever surcharge they like, why are customers complaining?
Such a contract would likely be unenforceable as a matter of law, or at least it should be. By their very nature the terms of contracts cannot be modified except by mutual consent. Contracts can include variable rates, but usually they need to be very well-defined. A contract that says you'll pay whatever expenses we happen to incur without any choice in the matter wouldn't qualify.
Yeah, this. What the AC may be confused about is that faster than light travel is (as far as we know) not possible in space, but the distance between two points can increase faster than light could travel because there's nothing stopping space itself from expanding that fast.
That's relativity, but the nice thing about this is that you can't actually observe it happening. If you put two golf balls in a room and then expanded the space between them so that they move a light year apart in a second, you'd never see either golf ball moving faster than the speed of light away from you, whether you are viewing from the perspective of either ball, or an observer in the room. You'd just see them move apart at a speed very close to the speed of light.
Objects further away than the cosmic horizon are in theory moving away from us faster than the speed of light. However, if we look at them all we see is a background of objects close to the horizon that are so red-shifted that they're barely detectable at all moving away at almost the speed of light. We can't see beyond the horizon (hence the name), and we can only see the horizon itself as a wash of particles coming from every direction (otherwise known as the cosmic microwave background).
I'm sure he was talking about an orbit with a low inclination in general, not one that was exactly zero. In any case, it does tend to launch to the east. If you wanted to launch from the east coast into a polar orbit it would probably be cheaper to do it at a more northern facility like Wallops Island.
However, if you were to use a mortgage product as a short-term loan facility you are benefiting from a lower rate. In order to prevent abuse of the system there is an early repayment penalty. It's fairly straight forward.
This really only makes sense if the bank had incorporated some of its up-front costs into the interest rate, instead of the closing costs. If the bank is covering all of its transactional costs via the up-front fees then you could open and pay-off a mortgage three times a day and they'd only profit from it.
As long as the market continues to offer a broad range of terms I don't really have a problem with this practice. Choices generally are good. I'd consider a ban on the practice if everybody started shifting towards only offering these kinds of loans, because they tend to create lock-in which isn't good for the market.
Why? It is because they are truly random. Each single outcome is just as likely as every other, which means that in the long run, the outcomes will occur pretty much equally often (give or take).
Each outcome cannot be equally likely. An atom of uranium at any moment has to be more likely to decay than an atom of carbon, since the former decays and the latter does not.
Sure, radioactive decay certainly seems random on short terms. However, it is entirely possible that some day we'll come up with a way of describing its behavior and it will turn out not to be random.
A good post, but I'm not sure you understand hardware based random number generation. At least one way to do it is have a small amount of radiactive material. Although it decays predictably in the long term (half life) it is random in the short term.
How do you KNOW that it is random in the short term? The fact that you cannot currently predict its behavior does NOT necessarily mean that it is random. If it were truly random, then why is there a long-term predictable behavior?
I'm not sure that this really helps clarify whether this is actually a quantum computer.
The question is whether the D-Wave is a quantum computer, not whether it is faster than a classical computer. A quantum computer that is slower than a desktop calculator is still useful for research purposes.
Agreed. If anything antibiotics and vaccines have completely opposite mechanisms of action.
An antibiotic taken as a medication kills bacteria directly, assisting the immune system and making its job easier. In the case of bleaching every surface in your house, it means that the immune system never sees the bacteria in the first place. The same is true of other external use of antibiotics (killing of bacteria before it gets into your body).
A vaccine provokes your immune response against a pathogen without exposing you to the risk of developing the disease (or a greatly reduced risk). Your immune system does all the work, and as a result it is able to do the job entirely on its own much more effectively at a later time.
Comparing the approaches, the disinfectant approach is like bleaching your house 3x/day, and the vaccine approach is like rolling around in the mud and not washing before dinner. I'd be very hesitant to associate the problems of the one with the other.
Let me ask you a question – is your objection that felons who have served their time can't vote or that the standard for felonies – those major crimes against society – has been watered down? Because it sounds to me that it is the watering down of felonies that is your issues – and I would agree with you there.
I would object to both.
Somebody convicted of a crime is either a danger to society or they aren't. If they are dangerous to be allowed in public, then they shouldn't be allowed in public - full stop. If they aren't, then quite badgering about it for the rest of their life. Frankly our criminal justice system needs to be a lot less punitive and a lot more rehabilitative. I'm fine with deterring crime, but clearly that on its own doesn't work. If a criminal can't be rehabilitated then they should get a life sentence, even if all they did was beat somebody up. If they can be rehabilitated, then they should be released once they're able to function normally in society, even if they killed 35 people. As a citizen my concern is not whether the guy across the street was appropriately punished like a 12 year old, but rather whether they're capable of not acting like a 12 year old now.
Sure, the system will always be imperfect, but I don't really see much value in permanent sanctions. By all means use parole (and by that I mean an invasive probation where you actually help the parolee re-integrate over years with heavy contact), but once they're just an ordinary citizen, let them be an ordinary citizen (heaven forbid that criminals that rehabilitate have something to look forward to).
When you turn people into second-class citizens they'll start acting like second-class citizens.
Banning gun ownership by felons is also doubly silly. If you think they're not dangerous then why ban gun ownership? If you do think they have no regard for the law, then why do you think outlawing gun ownership will stop them when whatever law they previously broke failed to do so?
Imagine you step into a teleporter, in an atomic instant you disappear and an exact replica of you appears on the other side. In this hypothetical example the position and composition of every particle that composes you is perfectly replicated, momentum and all (no messy Uncertainty Principle issues).
In this highly idealized scenario, you could debate whether your consciousness really does teleport, or whether the guy who steps out the other side merely thinks that it does.
Of course, you could probably make the same argument with motion in general. When you move every particle in your body moves along, and at one point in time they are at one place, and at the next moment they are at another. I'm not sure how teleportation is any different beyond the distance being larger.
Of course, to the degree that teleportation is imperfect, then the matter is compounded.
I think another huge challenge is just going to be wiring them up correctly initially.
I used to think a brain was just a box full of neurons that had inputs on one side and outputs on the other and they were trained like any other simple neural network, but with far many nodes. As I've read up on things (and lived with somebody with anomic aphasia that has improved over time) I realized that nothing could be further from the truth.
There is very clearly some kind of basic architecture to the brain that involves incredibly intricate connections that are basically pre-wired but which evolve over time. There are specialized regions of the brain, and most operations involve many regions of the brain working together in various ways. There are both short- and long-distance connections all over the place.
My personal theory is that there are many building-blocks of functionality that are suited to particular purposes, and that as brains evolved these got stitched together in ways that led to all kinds of emergent properties. You have areas like the cerebellum which seem (to my poorly educated brain) to be a bit like the simplistic neural network model, and I suspect the rest of the brain uses it like a programmer might use a computer to automate certain types of operations (like feedback loops for balance and who knows what else). I suspect that different regions of the brain "use" each other in similar ways, almost symbiotically.
If I had the time to really study this stuff I'd probably start by trying to understand fundamental building blocks of neural networks. I figure that developmentally this stuff has to form in some kind of almost fractal pattern since embryonic cells really only can keep track of their immediate surroundings and cell division counts. What kinds of patterns lead to what kinds of processing abilities?
Really fascinating stuff - it will be truly amazing if we ever figure it out.
Yup. To the point where quantum effects impact a brain, it might be impossible to replicate a human mind. You could probably make a decent facsimile, but it could differ in subtle ways.
When you decide where to eat lunch, for all we know the decision was influenced by a cosmic ray that originated in a supernova halfway across the universe.