... then this cannot be any good. Time to terminate it. We really do not need more cretins with huge visions and zero understanding of how things actually work.
Unless it has to happen in the real universe. In fantasy-land, this is certainly true, you just need a fairy to wave her wand over it.
Seriously, for short-enough messages, even the venerable Enigma is completely secure. (I remember something like 4000 bits of encrypted message per key-setting being required in order for a break becoming feasible, given unlimited computing power.)
But nil-whits will repeat any such bullshit, because they mistake science for being about opinions and beliefs.
Also commercially viable flying cars, actually intelligent machines, a home-helper robot for everybody, a viable, self-sufficient Lunar or Martian colony, faster-than-light travel, and other complete drivel those that misuse science and technology as a surrogate for religion are willing to believe in.
I completely agree. I know some people that worked in that area 20 years ago, and they knew back then that Quantum Computing is extremely unlikely to ever scale without exceptionally huge penalties and hence is useless as actual computing machinery. The physics makes larger machines exponentially more difficult to build. Digital computers can sub-divide problems into smaller problems. That makes them scale well, but not even they scale linearly with effort, just look at the absence of increased CPU-speeds over the last 5 years or so. Quantum computers require you to put in increasingly higher effort for every bit you add as you always need to process the whole problem in each step. Most people do not even begin to understand why that is so and what it means.
The people doing this back then justified this by saying it was interesting theoretical research, and it is. It has no known practical application at this time though. The claims in the direction of crypto solely serve to keep the people funding this research happy (and ignorant).
Quantum Computing does _not_ scale, as it cannot subdivide problems. You argument is completely bogus and in fact shows the opposite of what you think it shows.
No. This cardinal rule actual is "attacks never get worse". It usually gets stated as "attacks always get better", because that is easier to understand, but it is not quite correct.
My take is that such a machine may well be impossible in any practical sense. Even if it it possible, from looking at the speed (or rather extreme slowness) of the progress made in Quantum Computing, it may well take several hundred years to get there. And when we are there, we can simply make the numbers larger. Currently, RSA is recommended to use 4096 bits for new designs, that is a factor of 256 larger than what can be "cracked" by quantum computers. Remember count that people have been working on this for 25 years or longer.
As to symmetric encryption, Quantum Computing (if it ever can do it at all) halves the number of bits in the key-strength. That means for example that AES-256 will be secure against Quantum Computing forever, as around 80-100 bits there is a fundamental limit to brute-forcing things in this universe. Only if the key strength can be massively reduced by other means does Quantum Computing even factor here. And then you could do things like triple-AES-256, making Quantum Computing completely useless again.
But seriously, it took them 25 years getting from factoring 12 to factoring a 16 bit number. That strongly suggests logarithmic speed-up over time at best, i.e. some future generation may be seeing a 32 bit number factored, and 64 bit may already be infeasible. That also means no modern block-cipher will ever be in reach.
As soon as this becomes a factor in arrests, somebody will sell a tool to do it. This is a stunt, nothing more. It has no lasting positive effects. Well, maybe it can be used fro even more surveillance.
There is absolutely no problem with them having a 10GbE Card in there and, say 64 CPUs or so (and Linux or FreeBSD to handle that load;-). There is also absolutely no problem with the IP address actually being that of load-balancers. So, yes, there are no technical reasons why the data from hundreds of millions of users cannot be sent to just 2 IP addresses.
Ordinary engineering and typical engineers assume a friendly environment, i.e. the absence of intentional sabotage and hacking. This state of affairs is not true with globally networked infrastructure and sensors operating outside of protected spaces. What these people lack is what Bruce Schneier calls "the security mind-set". It involves not only thinking about how things can be made to work, but also how they can be intentionally broken and subverted. Having it is critical. That most people designing software and software-driven systems these days do not have it the main reason why IT security is in such an abysmally bad state these days.
And there is you problem: You are scratching a lot of surfaces of applications of mathematics and that is it. As such you are never doing anything real and no, you are not giving people "a foundation", you are misrepresenting a whole subject area. School "mathematics" is basically worthless for any STEM studies as it never covers how actual mathematics works. Ever had them create a theory? (Can be a simple one...) Ever had them develop a proof technique and them have them proof that it works? (Again, can be simple.) Ever had them formalize something?
So, no, you are not teaching mathematics. You are teaching applications of mathematics and not a lot of it. The whole exercise is a worthless waste of time.
Oh, and incidentally that you think talent is not special just means that you lack the experience of having it and that you either never have seen it at work or are deeply in denial. A very common thing in people with huge egos but only mediocre skills.
There is no known connection between talents in the mental space and genetics. That is a red herring. We have at this time absolutely no clue where talent comes from. We can describe its properties though: It cannot be created by education, it is separate from intelligence (but may need it as tool) and it cannot be replaced by anything else and it is critical for good-to-excellent performance in some areas.
Incidentally, we have no clue what intelligence is and how it is created. We can only observe its effects. What we do know is that it is only observable in connection with consciousness. All those neuro-"scientists" that claim the understand these things are full of it and their finding either do not explain anything or are unsound. They basically try to deduce how a computer works from staring at the screen.
You do not think it is worthwhile to invest a few hours into a tool that you are going to use literally for thousands of hours? You are right, that is hilarious. Human stupidity is really boundless.
You should have looked a tiny bit longer. Then you would have noticed that you can run one of a few hundred different window managers on Linux. What the distribution installer gives you as default is merely an example.
I never noticed. Of course, I never used KDE or Gnome or some other crappy "modern" window-manager. fvwm has all I need and excellent customization possibilities in addition.
You seem to have missed that on Linux you can completely ignore whatever crappy (e.g. KDE, Gnome) window-manager the distribution has standardized on. I have been running Debian with my own fvwm configuration for ages and that configuration I first created on sun-os. I had to do one rewrite when fvwm2 came out, but that is it.
Ah, so there is a propaganda guideline. Fits. It nicely explains why the same fallacies and emotional bullshit is vomited by the pro-systemd sect every time. They seem to have fallen for "it it is written, then it is truth" and completely have stopped to actually think about anything.
I guess it depends on how you define it; Any person of average or better intelligence is capable of becoming a programmer or mathematician intellectually.
And there you are fundamentally wrong. Unless you include very bad programmers and very bad mathematicians. It requires specific talent, raw intelligence alone does not cut it at all. Of course, a way above average "raw intelligence" is a required but not sufficient condition for ever becoming any good in those fields. And that is why most people cannot do it.
Specifically for mathematicians, economic factors are secondary. If you have a high level of talent and aptitude, you will get a stipend or other support. Mathematicians have long ago realized that there is absolute no substitute for specific talent and that those few that have it need to be found and supported. Admittedly, coder quality is currently on a race to the bottom, because they all want "cheap" and do not realize how extremely expensive that is in the long run.
Example: I recently had a case where a large customer needed some advanced piece of C code. They were unable to find anybody in their 10'000+ people IT organization that could do it. I ended up doing it at full consulting rates, and there they had to finesse it, as they could pay less than halve that per hour for a "top expert level" external coder. It is absolutely no surprise that so much software sucks under these economic border-conditions. A good coder is not a technician, but a highly talented, experienced and skilled engineer. Paying peanuts only gets you monkeys. We do not need more monkeys in coding, we need to stop paying only peanuts.
And the other thing I have consistently noticed: "Software engineering" teams are almost universally far, far too large. People try to do with 10-20 semi-to-incompetent coders what would require one or two really good ones. It is no surprise these projects fail so often. Nobody has an overview over the project. There is no "chief engineer" that knows how all parts fit and what approaches do and do not make sense. Instead people try to muddle through somehow.
If I look at Brook's classical "surgical team", it has one chief engineer, one assistant, and one tool-maker and that is it on the coder side. This is still the most efficient team possible (unless one expert can do it alone), you can only increase efficiency by using better people. Making the team larger universally _decreases_ efficiency. Yet what they do is having 10 barely capable people work on one piece of software. These insights were well-documented 40 years ago, and still the same amateur-level mistakes are being made. Managers do not even understand the basics. Yes, you may get 4 cheap coders for the price of one really good one, but you get a maximum of like 15% the productivity from the 4-bad-coders team that you get from the good one, if you are lucky. More often than not, the cheap coders will completely mess it up and end up having _negative_ productivity.
I had that too, some 40 years ago. Came in very handy when I went to university (gave me a head-start in both propositional logic and elementary set theory), never really needed it before that in school. They stopped it a few years later, because it was "too difficult".
And then these people come to university, and we need to deprogram them first from all the bad crap they have been learning. It is, incidentally, not about the latest stuff, but about the advanced stuff. Example: If you cannot clearly identify whether OO is suitable or unsuitable for for a certain project and justify this in detail, then you have no business teaching OO. As a large part of the industry is only finding out now that OO has (rather severe) disadvantages, you do need to follow what is going on. Mathematics and Literature fundamentals are moving extremely slow, CS (and coding) fundamentals are still very much in flux.
I disagree on the importance of quality of the teaching for coding. The critical, core competence any good coder has is aptitude and talent. That cannot be created or taught. Sure, those few that have it should get good teachers, but for most people trying to learn to code is completely futile and will never have good results, regardless of teacher quality. I have seen this time and again in students.
That said, good teachers are still critical, as the serve as multiplicators. Even if they only improve some skills in some pupils, this is very much worthwhile. Unfortunately, good teachers also teach independent thinking, and that is something those in power do not desire. Hence the teaching profession gets starved and the dumbest ones just good enough get selected for it.
I think that is wrong. The problem is that if you are good at "coding" in school, you are still very likely to be a bad coder. And I personally know several people that went into a CS program without coding skills and did well. School does not give you any realistic picture of what a career in a field would be like. Teach the basic things well, instead of stuffing even more specialty stuff in there and doing it badly.
Slashdot Mirror
... then this cannot be any good. Time to terminate it. We really do not need more cretins with huge visions and zero understanding of how things actually work.
Unless it has to happen in the real universe. In fantasy-land, this is certainly true, you just need a fairy to wave her wand over it.
Seriously, for short-enough messages, even the venerable Enigma is completely secure. (I remember something like 4000 bits of encrypted message per key-setting being required in order for a break becoming feasible, given unlimited computing power.)
But nil-whits will repeat any such bullshit, because they mistake science for being about opinions and beliefs.
Also commercially viable flying cars, actually intelligent machines, a home-helper robot for everybody, a viable, self-sufficient Lunar or Martian colony, faster-than-light travel, and other complete drivel those that misuse science and technology as a surrogate for religion are willing to believe in.
I completely agree. I know some people that worked in that area 20 years ago, and they knew back then that Quantum Computing is extremely unlikely to ever scale without exceptionally huge penalties and hence is useless as actual computing machinery. The physics makes larger machines exponentially more difficult to build. Digital computers can sub-divide problems into smaller problems. That makes them scale well, but not even they scale linearly with effort, just look at the absence of increased CPU-speeds over the last 5 years or so. Quantum computers require you to put in increasingly higher effort for every bit you add as you always need to process the whole problem in each step. Most people do not even begin to understand why that is so and what it means.
The people doing this back then justified this by saying it was interesting theoretical research, and it is. It has no known practical application at this time though. The claims in the direction of crypto solely serve to keep the people funding this research happy (and ignorant).
Quantum Computing does _not_ scale, as it cannot subdivide problems. You argument is completely bogus and in fact shows the opposite of what you think it shows.
No. This cardinal rule actual is "attacks never get worse". It usually gets stated as "attacks always get better", because that is easier to understand, but it is not quite correct.
My take is that such a machine may well be impossible in any practical sense. Even if it it possible, from looking at the speed (or rather extreme slowness) of the progress made in Quantum Computing, it may well take several hundred years to get there. And when we are there, we can simply make the numbers larger. Currently, RSA is recommended to use 4096 bits for new designs, that is a factor of 256 larger than what can be "cracked" by quantum computers. Remember count that people have been working on this for 25 years or longer.
As to symmetric encryption, Quantum Computing (if it ever can do it at all) halves the number of bits in the key-strength. That means for example that AES-256 will be secure against Quantum Computing forever, as around 80-100 bits there is a fundamental limit to brute-forcing things in this universe. Only if the key strength can be massively reduced by other means does Quantum Computing even factor here. And then you could do things like triple-AES-256, making Quantum Computing completely useless again.
But seriously, it took them 25 years getting from factoring 12 to factoring a 16 bit number. That strongly suggests logarithmic speed-up over time at best, i.e. some future generation may be seeing a 32 bit number factored, and 64 bit may already be infeasible. That also means no modern block-cipher will ever be in reach.
As soon as this becomes a factor in arrests, somebody will sell a tool to do it. This is a stunt, nothing more. It has no lasting positive effects. Well, maybe it can be used fro even more surveillance.
Something is wrong on your system. What you are seeing is a failure of DNS to resolve the host names, not the effects of a DDoS attack.
There is absolutely no problem with them having a 10GbE Card in there and, say 64 CPUs or so (and Linux or FreeBSD to handle that load ;-). There is also absolutely no problem with the IP address actually being that of load-balancers. So, yes, there are no technical reasons why the data from hundreds of millions of users cannot be sent to just 2 IP addresses.
Ordinary engineering and typical engineers assume a friendly environment, i.e. the absence of intentional sabotage and hacking. This state of affairs is not true with globally networked infrastructure and sensors operating outside of protected spaces. What these people lack is what Bruce Schneier calls "the security mind-set". It involves not only thinking about how things can be made to work, but also how they can be intentionally broken and subverted. Having it is critical. That most people designing software and software-driven systems these days do not have it the main reason why IT security is in such an abysmally bad state these days.
And there is you problem: You are scratching a lot of surfaces of applications of mathematics and that is it. As such you are never doing anything real and no, you are not giving people "a foundation", you are misrepresenting a whole subject area. School "mathematics" is basically worthless for any STEM studies as it never covers how actual mathematics works. Ever had them create a theory? (Can be a simple one...) Ever had them develop a proof technique and them have them proof that it works? (Again, can be simple.) Ever had them formalize something?
So, no, you are not teaching mathematics. You are teaching applications of mathematics and not a lot of it. The whole exercise is a worthless waste of time.
Oh, and incidentally that you think talent is not special just means that you lack the experience of having it and that you either never have seen it at work or are deeply in denial. A very common thing in people with huge egos but only mediocre skills.
There is no known connection between talents in the mental space and genetics. That is a red herring. We have at this time absolutely no clue where talent comes from. We can describe its properties though: It cannot be created by education, it is separate from intelligence (but may need it as tool) and it cannot be replaced by anything else and it is critical for good-to-excellent performance in some areas.
Incidentally, we have no clue what intelligence is and how it is created. We can only observe its effects. What we do know is that it is only observable in connection with consciousness. All those neuro-"scientists" that claim the understand these things are full of it and their finding either do not explain anything or are unsound. They basically try to deduce how a computer works from staring at the screen.
You do not think it is worthwhile to invest a few hours into a tool that you are going to use literally for thousands of hours?
You are right, that is hilarious. Human stupidity is really boundless.
You should have looked a tiny bit longer. Then you would have noticed that you can run one of a few hundred different window managers on Linux. What the distribution installer gives you as default is merely an example.
I never noticed. Of course, I never used KDE or Gnome or some other crappy "modern" window-manager. fvwm has all I need and excellent customization possibilities in addition.
You seem to have missed that on Linux you can completely ignore whatever crappy (e.g. KDE, Gnome) window-manager the distribution has standardized on. I have been running Debian with my own fvwm configuration for ages and that configuration I first created on sun-os. I had to do one rewrite when fvwm2 came out, but that is it.
Ah, so there is a propaganda guideline. Fits. It nicely explains why the same fallacies and emotional bullshit is vomited by the pro-systemd sect every time. They seem to have fallen for "it it is written, then it is truth" and completely have stopped to actually think about anything.
I guess it depends on how you define it; Any person of average or better intelligence is capable of becoming a programmer or mathematician intellectually.
And there you are fundamentally wrong. Unless you include very bad programmers and very bad mathematicians. It requires specific talent, raw intelligence alone does not cut it at all. Of course, a way above average "raw intelligence" is a required but not sufficient condition for ever becoming any good in those fields. And that is why most people cannot do it.
Specifically for mathematicians, economic factors are secondary. If you have a high level of talent and aptitude, you will get a stipend or other support. Mathematicians have long ago realized that there is absolute no substitute for specific talent and that those few that have it need to be found and supported. Admittedly, coder quality is currently on a race to the bottom, because they all want "cheap" and do not realize how extremely expensive that is in the long run.
Example: I recently had a case where a large customer needed some advanced piece of C code. They were unable to find anybody in their 10'000+ people IT organization that could do it. I ended up doing it at full consulting rates, and there they had to finesse it, as they could pay less than halve that per hour for a "top expert level" external coder. It is absolutely no surprise that so much software sucks under these economic border-conditions. A good coder is not a technician, but a highly talented, experienced and skilled engineer. Paying peanuts only gets you monkeys. We do not need more monkeys in coding, we need to stop paying only peanuts.
And the other thing I have consistently noticed: "Software engineering" teams are almost universally far, far too large. People try to do with 10-20 semi-to-incompetent coders what would require one or two really good ones. It is no surprise these projects fail so often. Nobody has an overview over the project. There is no "chief engineer" that knows how all parts fit and what approaches do and do not make sense. Instead people try to muddle through somehow.
If I look at Brook's classical "surgical team", it has one chief engineer, one assistant, and one tool-maker and that is it on the coder side. This is still the most efficient team possible (unless one expert can do it alone), you can only increase efficiency by using better people. Making the team larger universally _decreases_ efficiency. Yet what they do is having 10 barely capable people work on one piece of software. These insights were well-documented 40 years ago, and still the same amateur-level mistakes are being made. Managers do not even understand the basics. Yes, you may get 4 cheap coders for the price of one really good one, but you get a maximum of like 15% the productivity from the 4-bad-coders team that you get from the good one, if you are lucky. More often than not, the cheap coders will completely mess it up and end up having _negative_ productivity.
I had that too, some 40 years ago. Came in very handy when I went to university (gave me a head-start in both propositional logic and elementary set theory), never really needed it before that in school. They stopped it a few years later, because it was "too difficult".
And then these people come to university, and we need to deprogram them first from all the bad crap they have been learning. It is, incidentally, not about the latest stuff, but about the advanced stuff. Example: If you cannot clearly identify whether OO is suitable or unsuitable for for a certain project and justify this in detail, then you have no business teaching OO. As a large part of the industry is only finding out now that OO has (rather severe) disadvantages, you do need to follow what is going on. Mathematics and Literature fundamentals are moving extremely slow, CS (and coding) fundamentals are still very much in flux.
I disagree on the importance of quality of the teaching for coding. The critical, core competence any good coder has is aptitude and talent. That cannot be created or taught. Sure, those few that have it should get good teachers, but for most people trying to learn to code is completely futile and will never have good results, regardless of teacher quality. I have seen this time and again in students.
That said, good teachers are still critical, as the serve as multiplicators. Even if they only improve some skills in some pupils, this is very much worthwhile. Unfortunately, good teachers also teach independent thinking, and that is something those in power do not desire. Hence the teaching profession gets starved and the dumbest ones just good enough get selected for it.
I think that is wrong. The problem is that if you are good at "coding" in school, you are still very likely to be a bad coder. And I personally know several people that went into a CS program without coding skills and did well. School does not give you any realistic picture of what a career in a field would be like. Teach the basic things well, instead of stuffing even more specialty stuff in there and doing it badly.
Oooooh, an Ad Hominem! The last argument of utter incompetence.