Most developers are at risk of losing their jobs if anybody realizes how bad they actually are. Automation would just be the thing that shows that, not the thing that replaces them. Good developers will not get replaced until we have working strong AI, which is not happening any time soon. (A senior member of the IBM Watson team put it as "certainly not in the next 50 years" to me.)
In ordinary patch submission, there are two instances with actual intelligence and understanding: The patch creator and the maintainer. Here, there is only one: The maintainer. This violated the 4-eye principle. If the maintainer makes a mistake, the most stupid (in a non-obvious way) code makes it into the software.
Automated tools should never be used to decide anything. They should always only provide input to a human expert that knows exactly how the input was created and that there is no intelligence in that mechanism.
Unfortunately, a lot of incompetent coders (and there are a lot of them, not only because this is PHP) do use sample code frequently. I agree that for a competent coder, the whole thing is probably a non-issue, but that is not the reality of things.
You have no clue what you are talking about. Due to fundamental physical limitations, QCs will never scale the way digital computers did for a long time.
And you just outed yourself as utterly clueless. There are no "rounds" in RSA. You are thinking of a Feistel-construction or the like, which is something completely different. Incidentally, I am in the know but you would not even understand what that means.
So? If it is high-order polynomial, things stay secure. You can do public-key crypto with, say , effort n in in one direction and n^4 in the other. Requiring NP is just convenient and if you can get it, go for it. But it is not required at all.
Indeed. This tech has scaled sub-linear for 4 decades now. It is very likely it will only get worse at larger sizes. It may well hit a wall at sizes far below what is needed to threaten modern encryption and it will certainly not get there anytime soon. These are not classical computing scaling factors were you got a factor of 16 in just 8 years for a long time.
I disagree. After 40 years of failure, the probability("real soon") is at worst "low" but realistically "very low". And the impact is not "ginormous", but rather "moderate". That makes risk = low... very low.
Even most encryption is not threatened. A working, scaling QC is nowhere near as magic as people believe. These things are useless except for a few tasks and even for them (factorization) they may have huge constants in their run-times.
Indeed. And that is just my point. QC is a crapshot at this time. It may at some time be valuable, it is not today and will not be for a long time. That does not mean stop all research, but that does certainly mean do not prioritize it and do not put major emphasis in decision making on what it may or may not eventually deliver. Now, it is possible that at some future time some other tech becomes available that makes higher-intensity research into QCs a good idea, but at the moment this is not the case and the whole thing is a large bubble of hot air.
Exactly. Incidentally, the slide-rule example is limited pretty much by noise and measurement precision. The same is true for classical digital computers (at some scale and speed you are losing bits and digital computations become infeasible) and the huge success for classical computers comes from them having dealt very effectively with noise. It looks now like noise is the bane of QCs as well, but at a scale where they have not yet scaled to any useful size as classical computers hang that bar very high.
Nice example! Technologies do plateau, the question is where. For classical computing we are pretty much there now. But we had a fed decades of rapid progress before and these things are very powerful and useful. For Quantum Computers, it looks like they pretty much plateaued as well or are about too, bit at a scale were they are pretty useless and a modern pocket calculator can beat them easily.
You are comparing apples and oranges. Nuclear fusion has at least two observable instances where it works large-scale: 1. The sun 2. Hydrogen bombs. Nothing like that exists for QC.
What nonsense is that? We are talking about mathematical proofs here, not physical ones. A mathematical proof requires a consistent ("completely true") theory as a basis, period. If that is not the case, you can prove any statement that can be made in the theory to be both false and true and your theory is completely worthless.
Now, if you want a physical observation with high confidence, that is something else. But such an observation cannot be used as a starting point of mathematical deduction. Mathematical proofs deals with absolute truth. That is both their strength and their weakness.
Quantum computing has failed to perform for something like 40 years now. Any other technology this abysmally bad has just been scrapped. But somehow there are a lot of really clueless people that think this is magic and will suddenly scale and whatnot. There is absolutely no indication for that and a ton of indications to the contrary.
Theoretical concepts based on a consistent theoretical axiom system. Only have approximations in reality. Anybody that really studied them knows that. All the theory here can give is hints what may or may not work in practice, it cannot do any absolute predictions.
For that matter, since we're talking about Quantum Mechanics, I doubt anyone would be shocked if two inconsistent answers are sometimes both true.
That would be incredible cool! It would mean something like observer-dependent laws of physics though and all physics would suddenly only be an approximation. That would require some major re-thinking of all theoretical physics. But given what we know about Quantum Mechanics, I would not be surprised.
Oh? Somebody solved quantum-gravity while I was not looking? Now? Then the standard model is still known to be wrong, but nobody knows were exactly it is wrong and it is even possible that the whole thing has to be scrapped And hence you cannot prove anything "mathematically" based on it, since any mathematical proof critically requires a consistent (i.e. completely true) theory as basis. Seriously, this is proof-theory 101. Do you know nothing?
Wow. You really _are_ an idiot. Did anybody solve quantum-gravity while I was not looking? No? Then the standard model is still known to be wrong and you cannot actually "mathematically" prove anything about reality using it.
Incidentally, even working quantum computers are useful for very little. Almost all classical computing is not threatened, because a QC would simply be useless for it. There are a few special algorithms where a QC would do very well, but that is it. Hence there is no feeling threatened on my part at all.
Slashdot Mirror
Most developers are at risk of losing their jobs if anybody realizes how bad they actually are. Automation would just be the thing that shows that, not the thing that replaces them. Good developers will not get replaced until we have working strong AI, which is not happening any time soon. (A senior member of the IBM Watson team put it as "certainly not in the next 50 years" to me.)
In ordinary patch submission, there are two instances with actual intelligence and understanding: The patch creator and the maintainer. Here, there is only one: The maintainer. This violated the 4-eye principle. If the maintainer makes a mistake, the most stupid (in a non-obvious way) code makes it into the software.
Automated tools should never be used to decide anything. They should always only provide input to a human expert that knows exactly how the input was created and that there is no intelligence in that mechanism.
Unfortunately, a lot of incompetent coders (and there are a lot of them, not only because this is PHP) do use sample code frequently. I agree that for a competent coder, the whole thing is probably a non-issue, but that is not the reality of things.
You wish. I just point out the problem with the story. You did summarize that nicely though.
This is a wholly artificial panic. Tells you some people have something rather bad to hide and hence they are hyping this.
You have no clue what you are talking about. Due to fundamental physical limitations, QCs will never scale the way digital computers did for a long time.
IBM is desperate for relevancy these days. They are on their knees.
And you just outed yourself as utterly clueless. There are no "rounds" in RSA. You are thinking of a Feistel-construction or the like, which is something completely different. Incidentally, I am in the know but you would not even understand what that means.
So? If it is high-order polynomial, things stay secure. You can do public-key crypto with, say , effort n in in one direction and n^4 in the other. Requiring NP is just convenient and if you can get it, go for it. But it is not required at all.
Indeed. This tech has scaled sub-linear for 4 decades now. It is very likely it will only get worse at larger sizes. It may well hit a wall at sizes far below what is needed to threaten modern encryption and it will certainly not get there anytime soon. These are not classical computing scaling factors were you got a factor of 16 in just 8 years for a long time.
I disagree. After 40 years of failure, the probability("real soon") is at worst "low" but realistically "very low". And the impact is not "ginormous", but rather "moderate". That makes risk = low ... very low.
Even most encryption is not threatened. A working, scaling QC is nowhere near as magic as people believe. These things are useless except for a few tasks and even for them (factorization) they may have huge constants in their run-times.
Indeed. And that is just my point. QC is a crapshot at this time. It may at some time be valuable, it is not today and will not be for a long time. That does not mean stop all research, but that does certainly mean do not prioritize it and do not put major emphasis in decision making on what it may or may not eventually deliver. Now, it is possible that at some future time some other tech becomes available that makes higher-intensity research into QCs a good idea, but at the moment this is not the case and the whole thing is a large bubble of hot air.
Exactly. Incidentally, the slide-rule example is limited pretty much by noise and measurement precision. The same is true for classical digital computers (at some scale and speed you are losing bits and digital computations become infeasible) and the huge success for classical computers comes from them having dealt very effectively with noise. It looks now like noise is the bane of QCs as well, but at a scale where they have not yet scaled to any useful size as classical computers hang that bar very high.
Nice example! Technologies do plateau, the question is where. For classical computing we are pretty much there now. But we had a fed decades of rapid progress before and these things are very powerful and useful. For Quantum Computers, it looks like they pretty much plateaued as well or are about too, bit at a scale were they are pretty useless and a modern pocket calculator can beat them easily.
You are comparing apples and oranges. Nuclear fusion has at least two observable instances where it works large-scale: 1. The sun 2. Hydrogen bombs. Nothing like that exists for QC.
What nonsense is that? We are talking about mathematical proofs here, not physical ones. A mathematical proof requires a consistent ("completely true") theory as a basis, period. If that is not the case, you can prove any statement that can be made in the theory to be both false and true and your theory is completely worthless.
Now, if you want a physical observation with high confidence, that is something else. But such an observation cannot be used as a starting point of mathematical deduction. Mathematical proofs deals with absolute truth. That is both their strength and their weakness.
Quantum computing has failed to perform for something like 40 years now. Any other technology this abysmally bad has just been scrapped. But somehow there are a lot of really clueless people that think this is magic and will suddenly scale and whatnot. There is absolutely no indication for that and a ton of indications to the contrary.
I should be concerned over a QC that can factor 100 bit numbers? My RSA key is 2048 bit. No reason for concern at all.
Theoretical concepts based on a consistent theoretical axiom system. Only have approximations in reality. Anybody that really studied them knows that. All the theory here can give is hints what may or may not work in practice, it cannot do any absolute predictions.
So, one now gets modded "troll" for accurately describing the scientific state-of-the-art? How pathetic is that?
For that matter, since we're talking about Quantum Mechanics, I doubt anyone would be shocked if two inconsistent answers are sometimes both true.
That would be incredible cool! It would mean something like observer-dependent laws of physics though and all physics would suddenly only be an approximation. That would require some major re-thinking of all theoretical physics. But given what we know about Quantum Mechanics, I would not be surprised.
Oh? Somebody solved quantum-gravity while I was not looking? Now? Then the standard model is still known to be wrong, but nobody knows were exactly it is wrong and it is even possible that the whole thing has to be scrapped And hence you cannot prove anything "mathematically" based on it, since any mathematical proof critically requires a consistent (i.e. completely true) theory as basis. Seriously, this is proof-theory 101. Do you know nothing?
Indeed. But all these mindless cheerleaders do not even know what a QC can do. It is utterly pathetic.
Wow. You really _are_ an idiot. Did anybody solve quantum-gravity while I was not looking? No? Then the standard model is still known to be wrong and you cannot actually "mathematically" prove anything about reality using it.
Incidentally, even working quantum computers are useful for very little. Almost all classical computing is not threatened, because a QC would simply be useless for it. There are a few special algorithms where a QC would do very well, but that is it. Hence there is no feeling threatened on my part at all.
Indeed. And there is the little problem that QCs may never work and we may actually find out how to fix Physics instead.