Not necessarily. If there needs to be specific infrastructure at the arriving time, then you'll not see time travelers before that specific infrastructure was installed, which likely isn't before setting up that infrastructure.
For example, imagine a special stabilizer field which must be at your destination if you don't want to disintegrate as soon as you re-enter spacetime. Then any time before invention and deployment of those stabilizer fields won't be possible destinations for time travelers (to see what happens if you try anyway, look to Tunguska:-)).
While the concept of a paradox is indeed about our language and logic, the point is that theories about the universe are also just in our language and logic. The point is that it's possible to say nonsense, and sometimes such nonsense shows up by the statement (or collection of statements) being paradox. So basically time travel paradoxes mean that any description of the universe which involve classical causality and time travel are internally inconsistent, and therefore cannot be used to describe the real world. The question if it is possible to have a theory with time travel, but without paradoxes is the question if any theory which includes time travel (and causality) can be internally consistent and therefore a possible description of the real universe. If there are no consistent theories with time travel, then time travel doesn't exist - not because the universe cares about our language and logic, but because time travel, which itself is a concept which exists only inside our language and logic, and would have to be mapped to our observations of the outer world, would be a meaningless concept. Describing the real world with meaningless concepts is futile, therefore we have to demand that our concepts we use to describe the world are free of internal inconsistencies, not because the world would care about those inconsistencies, but because or description of the world with those concepts would not work.
Or in short: A useful description of reality is free of paradoxes not because nature cares about our descriptions, but because descriptions with paradoxes can't actually describe anything, and therefore especially cannot describe reality.
If you use quantum mechanics to combine a nondeterministic Turing machine with a teleportation device, you get a time machine. However you still can't correct the past.
Neither. Assuming he's right (and the Technology Review article correctly reproduced his claims; I haven't actually read the arXive article yet), he proved that you cannot have non-linear quantum mechanics without time travel. Given that some people try to resolve the measurement problem by adding nonlinearities, that's certainly an interesting result.
The concept of time travel itself shows misunderstanding of what time is. It basically posits an extra fifth dimension that's the "real" time, and treats our time as a dimension just like one of the three spatial dimensions.
No. That may be a common misconception in SF time travel, but it's not the essence of the idea of time travel. The misconception which leads to this "hidden meta-time" is the idea of an universal, absolute time. But for our real universe, we already know since Einstein that there's no absolute time, and two different observers may disagree about how much time was between two events, and may even disagree about the temporal order of causally unrelated events. Time travel is nothing than a logical extension of this concept, where observers also disagree about the temporal order of causally related events. There's absolutely no meta-time needed.
The spin of a photon is better known as its polarization. To get a heads/tails result, you had to choose a direction to measure the spin on. According to your claim, the photon is still there, so you managed a demolition-free polarization measurement, so the photon now actually has that polarization (i.e. if you repeat the same measurement again, you get a 100% probability, while measuring the opposite polarization you get a 0% probability). Ok, here's one way how you get a 50% probability:
If you measured a circular polarization, then now measure a linear polarization (any linear polarization will do).
If you measured linear polarization, measure linear polarization at an angle of 45 degrees to your original linear polarization measurement.
If you measured elliptical polarization, choose linear polarization at 45 degrees to the main axes of the corresponding ellipse.
I don't think the authors expect to see post-selection time machines any time soon. However such work may enable us to understand better the structure of quantum mechanics. By exploring what is and isn't possible in theory, we get guidelines in which directions it makes sense to look for adaptions of quantum theory (e.g. for quantum gravity), and which directions are better avoided.
There's no red. Probably you've seen a summary from the mysterious future. Post-selected time travel has been used successfully on Slashdot for quite some time now.:-)
One of these quantum universes has to have every quantum event probability = 0%, and one p=100%.
No. That would violate the laws of quantum mechanics. Show me a measurement event which has 0% probability and another one having 100% probability where the two are not separated by a superselection rule, and I readily show you a measurement event on the same system which has 50% probability.
In 10 years of professional programming, I don't think I've ever seen RTTI not disabled. Its little to no gain for a high cost. As for the X_cast functions- no point to them. If I want to cast, I'll just use (type). If I'm explicitly casting, I know that I want it cast and I don't want it to throw an exception. Why would I cast something that isn't a valid cast?
Sure, and you never make mistakes. Being less error-prone is the primary reason those *_casts exist. And those *_cast functions which are substitutes for C style casts don't throw exceptions. They behave exactly like the corresponding C style casts, except that they don't compile if the type of cast isn't possible (unlike C style casts, which silently switch to the meaning of reinterpret_cast).
However, dynamic_cast (the only one which depends on RTTI) is different in that it does something you cannot do with C style casts. Now, it's also something you seldom want to do, but in those rare cases where you want it, it's really the best tool for the job. (And BTW, even dynamic_cast doesn't throw exceptions if you use it with pointers; only the reference version throws, as there's nothing meaningful it could do in that case; note that reference dynamic_casts are easily avoided by just using the address-of operator and doing a pointer dynamic_cast).
But for the high cost: As long as you don't actually use it, the only cost is a few extra static tables, which most likely won't even be loaded into memory unless used. So unless you are programming embedded systems, I don't see any value in disabling RTTI.
Flamebait? Moderators should learn the difference between "Flamebait" and "I do not agree!"
I'm actually undecided in this case. An important fact is that there are possibilities for Firefox to support H.264 without getting the license, by just delegating it to the OS (Windows will most probably have the codec installed; Linux users are usually aware of the problem and can decide themselves whether to install the codec). What they can't do is to support it directly in the browser. But I don't think users would actually notice the difference. Therefore it's not a question of necessity, but one of policy.
Sometimes supporting non-free things from free software is important. For example, GNU software generally also runs on non-free operating systems (and if there's one person you cannot accuse of not caring enough about free, it's Stallman). It's generally extra work to support those operating systems, so it's clearly a deliberate decision. On the other hand, sometimes it can make sense to not support something, by principle.
The question here is what's more dangerous: Firefox supporting H.264 through the operating system increases the danger of getting stuck with a proprietary (patent encumbered) video codec. Firefox not supporting H.264 increases the danger of one of the most prominent and successful OSS product declining in usage, and may harm the general impression and thus adoption of open source software.
Slashdot Mirror
Well, the spyware that was silently installed from that DVD accurately recorded your action :-)
The post I answered to asked for a citation. And a citation is exactly what I provided. I don't know what you pretend that I pretended.
Not necessarily. If there needs to be specific infrastructure at the arriving time, then you'll not see time travelers before that specific infrastructure was installed, which likely isn't before setting up that infrastructure.
For example, imagine a special stabilizer field which must be at your destination if you don't want to disintegrate as soon as you re-enter spacetime. Then any time before invention and deployment of those stabilizer fields won't be possible destinations for time travelers (to see what happens if you try anyway, look to Tunguska :-)).
Of course this question is meaningless because there are no trees. :-)
Citation needed.
David Deutsch, Quantum mechanics near closed timelike lines, Phys. Rev. D 44, 3197–3217 (1991)
Only if time travel allows you to go back into the past arbitrarily far.
Creating a time travel from two FTL travels is a standard exercise in special relativity.
See http://en.wikipedia.org/wiki/Relativity_of_simultaneity
While the concept of a paradox is indeed about our language and logic, the point is that theories about the universe are also just in our language and logic. The point is that it's possible to say nonsense, and sometimes such nonsense shows up by the statement (or collection of statements) being paradox. So basically time travel paradoxes mean that any description of the universe which involve classical causality and time travel are internally inconsistent, and therefore cannot be used to describe the real world. The question if it is possible to have a theory with time travel, but without paradoxes is the question if any theory which includes time travel (and causality) can be internally consistent and therefore a possible description of the real universe. If there are no consistent theories with time travel, then time travel doesn't exist - not because the universe cares about our language and logic, but because time travel, which itself is a concept which exists only inside our language and logic, and would have to be mapped to our observations of the outer world, would be a meaningless concept. Describing the real world with meaningless concepts is futile, therefore we have to demand that our concepts we use to describe the world are free of internal inconsistencies, not because the world would care about those inconsistencies, but because or description of the world with those concepts would not work.
Or in short: A useful description of reality is free of paradoxes not because nature cares about our descriptions, but because descriptions with paradoxes can't actually describe anything, and therefore especially cannot describe reality.
You can't have one without the other.
If you use quantum mechanics to combine a nondeterministic Turing machine with a teleportation device, you get a time machine. However you still can't correct the past.
Neither. Assuming he's right (and the Technology Review article correctly reproduced his claims; I haven't actually read the arXive article yet), he proved that you cannot have non-linear quantum mechanics without time travel. Given that some people try to resolve the measurement problem by adding nonlinearities, that's certainly an interesting result.
More like: "If results are B, we know that any possible nonlinearity in QM will be less than (small value)".
No. That may be a common misconception in SF time travel, but it's not the essence of the idea of time travel. The misconception which leads to this "hidden meta-time" is the idea of an universal, absolute time. But for our real universe, we already know since Einstein that there's no absolute time, and two different observers may disagree about how much time was between two events, and may even disagree about the temporal order of causally unrelated events. Time travel is nothing than a logical extension of this concept, where observers also disagree about the temporal order of causally related events. There's absolutely no meta-time needed.
First post!!
Also, last post. Try to figure that one out!
Simple: Only post.
The spin of a photon is better known as its polarization. To get a heads/tails result, you had to choose a direction to measure the spin on. According to your claim, the photon is still there, so you managed a demolition-free polarization measurement, so the photon now actually has that polarization (i.e. if you repeat the same measurement again, you get a 100% probability, while measuring the opposite polarization you get a 0% probability). Ok, here's one way how you get a 50% probability:
Note that these cases cover all possibilties.
They got eaten by dinosaurs.
I don't think the authors expect to see post-selection time machines any time soon. However such work may enable us to understand better the structure of quantum mechanics. By exploring what is and isn't possible in theory, we get guidelines in which directions it makes sense to look for adaptions of quantum theory (e.g. for quantum gravity), and which directions are better avoided.
Yes, but the keys will be postselected to be in the worst case possible.
There's no red. Probably you've seen a summary from the mysterious future. Post-selected time travel has been used successfully on Slashdot for quite some time now. :-)
Has nobody anything snarky to say?
They are too busy trying to build postselection time machines. Expect to lose your first post status as soon as one of them succeeds.
No. That would violate the laws of quantum mechanics. Show me a measurement event which has 0% probability and another one having 100% probability where the two are not separated by a superselection rule, and I readily show you a measurement event on the same system which has 50% probability.
And what about the non-fucking internet?
In 10 years of professional programming, I don't think I've ever seen RTTI not disabled. Its little to no gain for a high cost. As for the X_cast functions- no point to them. If I want to cast, I'll just use (type). If I'm explicitly casting, I know that I want it cast and I don't want it to throw an exception. Why would I cast something that isn't a valid cast?
Sure, and you never make mistakes. Being less error-prone is the primary reason those *_casts exist.
And those *_cast functions which are substitutes for C style casts don't throw exceptions. They behave exactly like the corresponding C style casts, except that they don't compile if the type of cast isn't possible (unlike C style casts, which silently switch to the meaning of reinterpret_cast).
However, dynamic_cast (the only one which depends on RTTI) is different in that it does something you cannot do with C style casts. Now, it's also something you seldom want to do, but in those rare cases where you want it, it's really the best tool for the job. (And BTW, even dynamic_cast doesn't throw exceptions if you use it with pointers; only the reference version throws, as there's nothing meaningful it could do in that case; note that reference dynamic_casts are easily avoided by just using the address-of operator and doing a pointer dynamic_cast).
But for the high cost: As long as you don't actually use it, the only cost is a few extra static tables, which most likely won't even be loaded into memory unless used. So unless you are programming embedded systems, I don't see any value in disabling RTTI.
Flamebait? Moderators should learn the difference between "Flamebait" and "I do not agree!"
I'm actually undecided in this case. An important fact is that there are possibilities for Firefox to support H.264 without getting the license, by just delegating it to the OS (Windows will most probably have the codec installed; Linux users are usually aware of the problem and can decide themselves whether to install the codec). What they can't do is to support it directly in the browser. But I don't think users would actually notice the difference. Therefore it's not a question of necessity, but one of policy.
Sometimes supporting non-free things from free software is important. For example, GNU software generally also runs on non-free operating systems (and if there's one person you cannot accuse of not caring enough about free, it's Stallman). It's generally extra work to support those operating systems, so it's clearly a deliberate decision. On the other hand, sometimes it can make sense to not support something, by principle.
The question here is what's more dangerous: Firefox supporting H.264 through the operating system increases the danger of getting stuck with a proprietary (patent encumbered) video codec. Firefox not supporting H.264 increases the danger of one of the most prominent and successful OSS product declining in usage, and may harm the general impression and thus adoption of open source software.