I haven't read about TOR for a while, and yes doing a bruteforce MitM attack would be impractical with current technology, but all encryption can be broken, that's all I was implying:p
But that's okay because everyone can also live forever through nanotechnology, so why worry about going to jail for a few years?
The main benefit of the GPL over the BSD license is that it requires improvements in software to be given back to the community. I think that a hybrid model that allows for commercial exploitation might have a better chance, basically a BSD style license with time limits on proprietary releases. After 5 or 10 years, any software modified under the license would revert to the original license, allowing improvements back into the free software community. It's just an end run around the insane copyright terms that are the norm now.
Probably the ideal situation would be a dual license under both the GPL and the time limited BSD style license so that after the time period has expired for proprietary incorporation of the software, the modifications are relicensed under the dual license, so that neither the GPL or BSD camps would ever lose any code. An obvious practical benefit is that GPL and BSD projects could begin sharing their code under the dual license as well.
Such a dual license would improve developers' freedom to use free software, and end users would eventually obtain full rights to any proprietary modifications made by the developers. Terms could also be included that would automatically terminate the proprietary license if the availability of support or patches was removed, or if the company went bankrupt.
Watch the laptops until they reveal enough information about themselves to drop them a friendly call (make sure to state their name and address) to inform them that they have stolen property, and that if they'd like to avoid being arrested they should perform a blind drop of all the stolen equipment at a location you're familiar with. Maybe just have them come in to your business (if it's large enough and public) and drop them in an empty conference room or turn it in to lost and found, saving you a lot of trouble. If they never reveal their phone number, an email would probably suffice as well.
Not "finished" in terms of "completed", you dolt. "Finished" in terms of "polished", like the finish on a piece of furniture.
I agree, shiny things are far more valuable than functional or innovative things. I think the phrase you were really looking for was "a polished turd."
If what you are saying was the case, nothing would ever reach version 2.0, or 1.0.1 for that matter. One Point Oh is just a way of saying "this product does everything we wanted it to do, to our satisfaction". The point being that with 18,000 people, they should be able to get a product to do everything that they want it to do; if not, then their managers need to be sacked for presenting the coders and testers with a moving target.
In other words, it's an arbitrary selection of features that were picked for an arbitrary number to be applied to them when they are put together into one packaged result. Yay.
In reality, projects are never finished. People just stop working on them. If projects were perfectly tied to specifications, then specs would never be completely finished, just stopped prematurely. By keeping a core set of features stable while adding features continually development goes faster, smoother, and obviously has the benefit of more testers.
Just to be clear, I think it's stupid that google has so many things in beta still. I mean, seriously, froogle and gmail are still beta? It's been what, 5 years?
Well, they're still adding new features. Do you want them to stop? Do you want to set up a new email account at gmail2.com to try out new features instead?
I take Google's "beta" label as warning that they don't want to guarantee that their software will work, that it won't eat your data, and that they won't suddenly remove a bunch of features you like.
Read any EULA for consumer software and it says exactly the same things.
You'd think an 18,000 person company would be able to release a finished project once in a while.
How do you "finish" a web based project? "Well, that's done, no one will ever think of anything new to do with this software, or any way to make it easier or better!" "Gee, we've indexed the entire Internet this month, so I guess we're done!"
Re:Hate to Jones in on the thread...
on
What NAS To Buy?
·
· Score: 1
Why pay for an interface to S3 when there are library bindings for several languages that make it pretty easy to do it on your own? gpg + perl + Net::Amazon::S3 makes synchronizing encrypted versions of files and directories pretty straightforward.
False. Stochastic theories do not exactly predict the outcome of every experiment, they only predict probabilities (a tautology). So, one is free to posit another mechanism (call it souls, or free wills, or whatever you like), which does determine the outcome of some experiments. It is entirely plausible that this mechanism does so in a way that does not affect the distribution of experimental results, and is thus empirically indistinguishable from the same stochastic theory without such a supposition. Deterministic theories leave no such room. Once the state is determined at any time t, it is determined for all other times, come hell, high water, or free wills external to the universe.
The question then is whether the mechanism for free will in this universe has free will in the larger universe containing it, reductio ad absurdum. Essentially, either free will must be able to exist entirely within a universe or not at all.
A definition of free will that could exist in our universe without question is this: Free will is the lack of perfect foresight. By divorcing the models and decisions that we make from the actual laws of the universe, our choices are freed from perfect isomorphism with the nature of their outcomes. In other words, our imperfect nature grants us free will in the form of not knowing with certainty what the outcome of our actions will be. Our actions may be deterministic and we may retain free will by the observation that our concept of self is not mathematically isomorphic to the bits of the universe that compose our body and brain. Essentially, our model of ourself can possess free will because it is an imperfect model. We clearly don't mean that the atoms composing our bodies have their own free will that determines our choices, we mean that the abstract concept of a person (which we believe ourselves to be) has an abstract property called free will. Even as our mental models of ourselves are influenced by advancing physics, we will never be able to formulate a perfect model of ourselves, and therefore never quite able to be rid of our free will, no matter how hard we try.
If that seems like a cop-out, consider that humans can only experience the universe subjectively (through an imperfect model) to begin with, so defining properties of our imperfect model is simply the best that we can do. If we were perfectly objective we would have to possess a perfect model of the universe, and it would render free will nonexistent. We would simply know the universe in full all at once, including ourselves and all our intentions, future choices and actions, and their ultimate outcomes.
It is worth noting that my "mechanism" would constitute a hidden variable theory, sort of. The sort of is because "hidden variable theory" is usually understood to mean a way to make quantum mechanics deterministic or complete, that is, it introduces realism. There are of course sharp theoretical limits on such theories. But one which does not quite make quantum mechanics complete, especially if it is also nonlocal, is not so limited.
Bhom's interpretation of quantum mechanics (guiding waves for deterministic particles) works as a deterministic model, but it doesn't add any explanations or make new predictions. Nevertheless, it allows a fully deterministic universe and avoids the many worlds interpretation, essentially making that issue orthogonal to physics for all practical purposes (I don't think quantum immortality is practical until it's actually tested by dying in the current universe. It's certainly not a repeatably testable hypothesis within a single universe or an entire branch of universes). Free will also appears to be orthogonal to physics as we know it for the same reason, which implies that it's truly a philosophical problem based on the model of ourselves that we want to use. We could define a model of ourselves where we do not have free will, but I posit that such a model could explain nothing that a model with free will could not also explain. My definition of free wi
Good job there, you're really kicking the shit out of your straw man. Hint 'domination, self-deception, naive arrogant independence, hating homos, xenophobia, punishment, and pre-emptive war, and fake moral authority' are not what conservatives believe in.
Honestly, I think this whole notion that a bunch of average Joe Americans with handguns and rifles is going to stand up to a trained professional army in any meaningful way is utterly ridiculous.
Yeah, forget what we learned in Korea, Vietnam, Afghanistan, and Iraq. Modern armies have *NO* trouble putting down a little insurrection by common people with small arms and improvised weapons.
There are two reasons you're wrong. One is entropy, and the other is one way functions.
Entropy forces causality to appear in physical systems. A boiled egg is highly correlated with a heated raw egg, but I challenge you to explain away the causation from one state to the other.
One way functions are quite similar, and probably a result of the same physical properties of matter. When a key is used to encrypt data, there is a high correlation between the original data, the key, and the encrypted data, but causation clearly flows from encrypting data with the key to the encrypted data state, and not from the encrypted state to a derived key and the original data. It's just a limitation of human (and our machines) abilities, but it nevertheless presents very strong evidence for the practical existence of causation.
Yep, but what you won't have is the possibility of hit for all the legal copies you made that the license covered. Similarly, You can't violate the GPL and act as if it was never valid. You have to determine a breach or terms and then revoke the license in which case, any distribution from there on out would be a copyright violation.
There are no distributable copies of GPL software that do not contain the GPL itself and either the entire source code or a notice that the source is available. One violates the GPL at the derivative work stage by not including the GPL and source offer, and hence there is no distribution license by the time one gets around to pressing CDs or burning ROMs. Fair use allows everyone to make all the derivative works they want (at least in the opinion of the FSF lawyers), but at that point they have give up all distribution rights for those derivative works by not including the source or notice.
Now, you're right that it would be a completely civil matter if a company published GPL software with a notice that they would provide source, and then they never did. At that point they would be in violation of the GPL for not complying with the source distribution license, but they would not be responsible for all the software distributed before it became apparent that they were not going to give out the source. Any further distribution would presumably be illegal after 3 years (the minimum period after an initial release during which the source must be available) had passed since the first distribution, at which point it could be considered obvious that the license had not been followed, thus invalidating it. A Real Lawyer could probably establish a shorter "reasonable" time-frame outside of which the license could be deemed broken by failing to provide source code.
In this case, the lack of a prominent notice of source availability and lack of actual source would be sufficient to terminate the GPL for the derivative work that the company created, removing their right to distribute before they even shipped the first box.
Sorry, but exactly when did Title 17 begin including or referencing the GPL?!?!?
Copyright law does not say anything about providing source code. It merely says that you must have a license. It does not dictate what the license says, nor does it say you must follow the terms of the license.
Quite simply, if you are not following the terms of a license when distributing a copyrigthed work, you are in violation of copyright statutes and can be held civilly and criminally liable.
If you buy a dvd or cd, you are granted a license to perform the work non-publically, and not make and sell copies. If you violate the license by publicly performing the work or copying it and putting it on a P2P network, you've violated both copyright law and the terms of the license. Criminal penalties apply to the violation of the letter of the law (no distribution or performance without license) and civil responsibility results from the breach of license terms.
More to the point, if you buy a volume licensed Windows CD, you can make as many copies as you need to within your business to support installation activities. You will still get busted by the FBI for uploading the CD ISO to a P2P network. Just because you are licensed for SOME redistribution or copying rights does not mean you have ALL rights and no criminal responsibility if you violate the license.
Their research could go into "humane" forms of weapons or perhaps even robot soldiers. A government would probably be more reluctant to send flesh-and-blood soldiers against robots, which are cheap compared to human lives.
Conversely, governments would be more likely to send their own robots to war; whether it was against humans or other robots wouldn't matter. The goal of many people is to reduce overall human suffering, not just suffering within their own country.
I'm glad you mentioned Turing because this is rather a propos. IMO, Turing and his ideas are the worse things to have happened to computer science (mod me down as a troll, if you hate Kuhn and/or the free flow of ideas). I am sure you'll disagree on what I have to say but hey, nobody has a monopoly on opinions. Consider that the computer industry is faced with three major crises: software unreliability, low productivity and the parallel programming problem. Guess what? Not one of Turing's supposedly brilliant ideas is of any help. Not a single one! You know why? It's because Turing's ideas are the cause of those crises.
Stupid people are the reason for software unreliability, low productivity, and the lack of parallel programs. Synchronization primitives for parallel processing are *old*, as in 40 to 50 years old. Software unreliability was also solved around that time them with the introduction of formal proofs for algorithms. Technically, if a piece of computer code lacks a proof of correctness, it can't even be called an algorithm to begin with; it's just a heuristic.
You were probably expecting The Next Big Thing(TM) to come out as a library with bindings for your favorite programming language that would magically solve your synchronization and security problems, right? heh.
Regarding your ideas about science, how can massive paradigm shifts occur if they aren't based on existing trusted scientific experiments? You could claim that you have a great new model of gravity, but unless it contains Newton's inverse square law for objects at non relativistic speeds, it fails. If it doesn't explain general relativity (and specifically points out how it is better than GR), then it fails. Same with your brand new theories of computer science. If you can't prove that whatever new methods you come up with are a) turing complete, and b) formally correct (using existing mathematical tools), then they fail. If you can't understand why scientists might like to have new theories proven by way of trusted ones, perhaps you should go sell magic crystals to people or predict their futures.
Why do I care if some big oil CEO if it will save me and everyone else $0.25 a gallon? I still end up ahead!
Because if the big oil CEO didn't get rich (say the oil was nationalized) it would probably save everyone $0.50 a gallon. That's the real tradeoff. People realize that a resource in the ground can be exploited differently and benefit them more than any currently proposed scheme. The problem is that one word of nationalizing oil causes everyone to run around screaming "communism!", despite the fact that the government already nationalized all the roads and rails you can use gas on.
It's slow because you have to aptget unfree-leachware-closed-source to get an accelerated driver which will probably fuck the machine up.
As opposed to a closed source driver from ATI that fucks your machine up on Windows? Or even better one of those noname brand cards that have better support on Linux than the crappy drivers on the CD that come with the card? 2D acceleration is enabled for all major brands of video cards (ATI, Nvidia, Via, and Intel [and probably Matrox too, but I don't have a card to test it on]) in the default free Xorg drivers. Sometimes the X configurator in the distro decides to use a generic framebuffer for no good reason. It's easy to fix.
I can Ctrl+C and Ctrl+V to cut'n'paste between applications.
The Windows clipboard beats any clipboard KDE or Gnome has. This is primarily an application level failure, since X doesn't even know what applications put on the clipboard (and none of the applications know what another application has put on, either). Just throwing a mime-type on the clipboard blob would probably do wonders for KDE/Gnome clipboard functionality. The content negotiation X supports is just not specific enough (and it's old formats).
Now it used to be that X Windows had network transparency and Windows didn't. But now there's Remote Desktop and VNC to handle that. And it was a far better idea to optimize for the common case where the application is running locally than across the network.
Remote Desktop is not truly network transparency; it's just remote desktop. Same with VNC. X lets you run applications anywhere and view them anywhere. The usefulness of such a thing is waning, but it's still possible to do GUI work on a server without taking up its own resources for graphics, instead letting your fancy new workstation do it. What would be a truly killer application is the ability to migrate applications from one X server to another without ending them. I haven't seen that put into common use, though. X basically has all the abstraction necessary to make truly distributed computing work seamlessly. I have to think pretty hard to find an example of a Win16 application running "well" for any reasonable definition of "well." The Win16 subsystem was almost worse than DOS in terms of design and implementation.
These days a 32 or 64 bit Windows app in C or C++ is seriously simple to write and it very efficient since you build the UI out of built in Windows classes. You subclass 'em and override the 1% of behaviour you want to change. And since you're only overriding 1%, everyone knows how to use it. You can make small exe files too, e.g. uTorrent.
I've done Win32 programming. It's not *that* wonderful. Neither is X11. Neither was that abomination of C++ classes Visual Studio used to use. The number of times I wished I could avoid the stupid class abstractions and just use Win32 function calls was amazing. But why bother with that low a level when there's Gtk and Qt? Small executables are usually a function of the libraries included, essentially. You can bloat your code in many ways, but the GUI toolkits are probably not a majority of the size. Shared libraries have a big impact on actual executable size, too.
That's where many weaknesses in cryptographic software have been found. With any luck the virus writers just borrowed the encryption code out of OpenSSL in the Debian tree.
Slashdot Mirror
I haven't read about TOR for a while, and yes doing a bruteforce MitM attack would be impractical with current technology, but all encryption can be broken, that's all I was implying :p
But that's okay because everyone can also live forever through nanotechnology, so why worry about going to jail for a few years?
Wouldn't the existence of such a header break SPF? Spam could just come "forwarded" from the spoofed sender.
Simply stating "I believe it can" is the realm of religion. What evidence would it take to convince you that it isn't possible after all?
Death?
The main benefit of the GPL over the BSD license is that it requires improvements in software to be given back to the community. I think that a hybrid model that allows for commercial exploitation might have a better chance, basically a BSD style license with time limits on proprietary releases. After 5 or 10 years, any software modified under the license would revert to the original license, allowing improvements back into the free software community. It's just an end run around the insane copyright terms that are the norm now.
Probably the ideal situation would be a dual license under both the GPL and the time limited BSD style license so that after the time period has expired for proprietary incorporation of the software, the modifications are relicensed under the dual license, so that neither the GPL or BSD camps would ever lose any code. An obvious practical benefit is that GPL and BSD projects could begin sharing their code under the dual license as well.
Such a dual license would improve developers' freedom to use free software, and end users would eventually obtain full rights to any proprietary modifications made by the developers. Terms could also be included that would automatically terminate the proprietary license if the availability of support or patches was removed, or if the company went bankrupt.
Watch the laptops until they reveal enough information about themselves to drop them a friendly call (make sure to state their name and address) to inform them that they have stolen property, and that if they'd like to avoid being arrested they should perform a blind drop of all the stolen equipment at a location you're familiar with. Maybe just have them come in to your business (if it's large enough and public) and drop them in an empty conference room or turn it in to lost and found, saving you a lot of trouble. If they never reveal their phone number, an email would probably suffice as well.
Not "finished" in terms of "completed", you dolt. "Finished" in terms of "polished", like the finish on a piece of furniture.
I agree, shiny things are far more valuable than functional or innovative things. I think the phrase you were really looking for was "a polished turd."
If what you are saying was the case, nothing would ever reach version 2.0, or 1.0.1 for that matter. One Point Oh is just a way of saying "this product does everything we wanted it to do, to our satisfaction". The point being that with 18,000 people, they should be able to get a product to do everything that they want it to do; if not, then their managers need to be sacked for presenting the coders and testers with a moving target.
In other words, it's an arbitrary selection of features that were picked for an arbitrary number to be applied to them when they are put together into one packaged result. Yay.
In reality, projects are never finished. People just stop working on them. If projects were perfectly tied to specifications, then specs would never be completely finished, just stopped prematurely. By keeping a core set of features stable while adding features continually development goes faster, smoother, and obviously has the benefit of more testers.
Just to be clear, I think it's stupid that google has so many things in beta still. I mean, seriously, froogle and gmail are still beta? It's been what, 5 years?
Well, they're still adding new features. Do you want them to stop? Do you want to set up a new email account at gmail2.com to try out new features instead?
I take Google's "beta" label as warning that they don't want to guarantee that their software will work, that it won't eat your data, and that they won't suddenly remove a bunch of features you like.
Read any EULA for consumer software and it says exactly the same things.
You'd think an 18,000 person company would be able to release a finished project once in a while.
How do you "finish" a web based project? "Well, that's done, no one will ever think of anything new to do with this software, or any way to make it easier or better!" "Gee, we've indexed the entire Internet this month, so I guess we're done!"
Why pay for an interface to S3 when there are library bindings for several languages that make it pretty easy to do it on your own? gpg + perl + Net::Amazon::S3 makes synchronizing encrypted versions of files and directories pretty straightforward.
False. Stochastic theories do not exactly predict the outcome of every experiment, they only predict probabilities (a tautology). So, one is free to posit another mechanism (call it souls, or free wills, or whatever you like), which does determine the outcome of some experiments. It is entirely plausible that this mechanism does so in a way that does not affect the distribution of experimental results, and is thus empirically indistinguishable from the same stochastic theory without such a supposition. Deterministic theories leave no such room. Once the state is determined at any time t, it is determined for all other times, come hell, high water, or free wills external to the universe.
The question then is whether the mechanism for free will in this universe has free will in the larger universe containing it, reductio ad absurdum. Essentially, either free will must be able to exist entirely within a universe or not at all.
A definition of free will that could exist in our universe without question is this: Free will is the lack of perfect foresight. By divorcing the models and decisions that we make from the actual laws of the universe, our choices are freed from perfect isomorphism with the nature of their outcomes. In other words, our imperfect nature grants us free will in the form of not knowing with certainty what the outcome of our actions will be. Our actions may be deterministic and we may retain free will by the observation that our concept of self is not mathematically isomorphic to the bits of the universe that compose our body and brain. Essentially, our model of ourself can possess free will because it is an imperfect model. We clearly don't mean that the atoms composing our bodies have their own free will that determines our choices, we mean that the abstract concept of a person (which we believe ourselves to be) has an abstract property called free will. Even as our mental models of ourselves are influenced by advancing physics, we will never be able to formulate a perfect model of ourselves, and therefore never quite able to be rid of our free will, no matter how hard we try.
If that seems like a cop-out, consider that humans can only experience the universe subjectively (through an imperfect model) to begin with, so defining properties of our imperfect model is simply the best that we can do. If we were perfectly objective we would have to possess a perfect model of the universe, and it would render free will nonexistent. We would simply know the universe in full all at once, including ourselves and all our intentions, future choices and actions, and their ultimate outcomes.
It is worth noting that my "mechanism" would constitute a hidden variable theory, sort of. The sort of is because "hidden variable theory" is usually understood to mean a way to make quantum mechanics deterministic or complete, that is, it introduces realism. There are of course sharp theoretical limits on such theories. But one which does not quite make quantum mechanics complete, especially if it is also nonlocal, is not so limited.
Bhom's interpretation of quantum mechanics (guiding waves for deterministic particles) works as a deterministic model, but it doesn't add any explanations or make new predictions. Nevertheless, it allows a fully deterministic universe and avoids the many worlds interpretation, essentially making that issue orthogonal to physics for all practical purposes (I don't think quantum immortality is practical until it's actually tested by dying in the current universe. It's certainly not a repeatably testable hypothesis within a single universe or an entire branch of universes). Free will also appears to be orthogonal to physics as we know it for the same reason, which implies that it's truly a philosophical problem based on the model of ourselves that we want to use. We could define a model of ourselves where we do not have free will, but I posit that such a model could explain nothing that a model with free will could not also explain. My definition of free wi
...with at least one hot chick who is there of her own free will.
But she'd be watching Fox News.
Good job there, you're really kicking the shit out of your straw man. Hint 'domination, self-deception, naive arrogant independence, hating homos, xenophobia, punishment, and pre-emptive war, and fake moral authority' are not what conservatives believe in.
So why did they vote for Bush twice?
1 per 10,000 isn't too bad. Are your laws laxer than Alaska? We don't require concealed carry permits anymore, and can shoot people committing arson.
Honestly, I think this whole notion that a bunch of average Joe Americans with handguns and rifles is going to stand up to a trained professional army in any meaningful way is utterly ridiculous.
Yeah, forget what we learned in Korea, Vietnam, Afghanistan, and Iraq. Modern armies have *NO* trouble putting down a little insurrection by common people with small arms and improvised weapons.
So the blindingly obvious solution is for liberals (read: democrats in the U.S.) to buy guns. Duh.
There are two reasons you're wrong. One is entropy, and the other is one way functions.
Entropy forces causality to appear in physical systems. A boiled egg is highly correlated with a heated raw egg, but I challenge you to explain away the causation from one state to the other.
One way functions are quite similar, and probably a result of the same physical properties of matter. When a key is used to encrypt data, there is a high correlation between the original data, the key, and the encrypted data, but causation clearly flows from encrypting data with the key to the encrypted data state, and not from the encrypted state to a derived key and the original data. It's just a limitation of human (and our machines) abilities, but it nevertheless presents very strong evidence for the practical existence of causation.
Yep, but what you won't have is the possibility of hit for all the legal copies you made that the license covered. Similarly, You can't violate the GPL and act as if it was never valid. You have to determine a breach or terms and then revoke the license in which case, any distribution from there on out would be a copyright violation.
There are no distributable copies of GPL software that do not contain the GPL itself and either the entire source code or a notice that the source is available. One violates the GPL at the derivative work stage by not including the GPL and source offer, and hence there is no distribution license by the time one gets around to pressing CDs or burning ROMs. Fair use allows everyone to make all the derivative works they want (at least in the opinion of the FSF lawyers), but at that point they have give up all distribution rights for those derivative works by not including the source or notice.
Now, you're right that it would be a completely civil matter if a company published GPL software with a notice that they would provide source, and then they never did. At that point they would be in violation of the GPL for not complying with the source distribution license, but they would not be responsible for all the software distributed before it became apparent that they were not going to give out the source. Any further distribution would presumably be illegal after 3 years (the minimum period after an initial release during which the source must be available) had passed since the first distribution, at which point it could be considered obvious that the license had not been followed, thus invalidating it. A Real Lawyer could probably establish a shorter "reasonable" time-frame outside of which the license could be deemed broken by failing to provide source code.
In this case, the lack of a prominent notice of source availability and lack of actual source would be sufficient to terminate the GPL for the derivative work that the company created, removing their right to distribute before they even shipped the first box.
Sorry, but exactly when did Title 17 begin including or referencing the GPL?!?!? Copyright law does not say anything about providing source code. It merely says that you must have a license. It does not dictate what the license says, nor does it say you must follow the terms of the license.
Quite simply, if you are not following the terms of a license when distributing a copyrigthed work, you are in violation of copyright statutes and can be held civilly and criminally liable.
If you buy a dvd or cd, you are granted a license to perform the work non-publically, and not make and sell copies. If you violate the license by publicly performing the work or copying it and putting it on a P2P network, you've violated both copyright law and the terms of the license. Criminal penalties apply to the violation of the letter of the law (no distribution or performance without license) and civil responsibility results from the breach of license terms.
More to the point, if you buy a volume licensed Windows CD, you can make as many copies as you need to within your business to support installation activities. You will still get busted by the FBI for uploading the CD ISO to a P2P network. Just because you are licensed for SOME redistribution or copying rights does not mean you have ALL rights and no criminal responsibility if you violate the license.
Just not the same without Manny's broken English.
Their research could go into "humane" forms of weapons or perhaps even robot soldiers. A government would probably be more reluctant to send flesh-and-blood soldiers against robots, which are cheap compared to human lives.
Conversely, governments would be more likely to send their own robots to war; whether it was against humans or other robots wouldn't matter. The goal of many people is to reduce overall human suffering, not just suffering within their own country.
I'm glad you mentioned Turing because this is rather a propos. IMO, Turing and his ideas are the worse things to have happened to computer science (mod me down as a troll, if you hate Kuhn and/or the free flow of ideas). I am sure you'll disagree on what I have to say but hey, nobody has a monopoly on opinions. Consider that the computer industry is faced with three major crises: software unreliability, low productivity and the parallel programming problem. Guess what? Not one of Turing's supposedly brilliant ideas is of any help. Not a single one! You know why? It's because Turing's ideas are the cause of those crises.
Stupid people are the reason for software unreliability, low productivity, and the lack of parallel programs. Synchronization primitives for parallel processing are *old*, as in 40 to 50 years old. Software unreliability was also solved around that time them with the introduction of formal proofs for algorithms. Technically, if a piece of computer code lacks a proof of correctness, it can't even be called an algorithm to begin with; it's just a heuristic.
You were probably expecting The Next Big Thing(TM) to come out as a library with bindings for your favorite programming language that would magically solve your synchronization and security problems, right? heh.
Regarding your ideas about science, how can massive paradigm shifts occur if they aren't based on existing trusted scientific experiments? You could claim that you have a great new model of gravity, but unless it contains Newton's inverse square law for objects at non relativistic speeds, it fails. If it doesn't explain general relativity (and specifically points out how it is better than GR), then it fails. Same with your brand new theories of computer science. If you can't prove that whatever new methods you come up with are a) turing complete, and b) formally correct (using existing mathematical tools), then they fail. If you can't understand why scientists might like to have new theories proven by way of trusted ones, perhaps you should go sell magic crystals to people or predict their futures.
So, when gas jumped from $2.00 to $4.00, who makes an extra $292,000,000,000 this year? Could they have afforded to make $73,000,000,000 less?
Why do I care if some big oil CEO if it will save me and everyone else $0.25 a gallon? I still end up ahead!
Because if the big oil CEO didn't get rich (say the oil was nationalized) it would probably save everyone $0.50 a gallon. That's the real tradeoff. People realize that a resource in the ground can be exploited differently and benefit them more than any currently proposed scheme. The problem is that one word of nationalizing oil causes everyone to run around screaming "communism!", despite the fact that the government already nationalized all the roads and rails you can use gas on.
It's slow because you have to aptget unfree-leachware-closed-source to get an accelerated driver which will probably fuck the machine up.
As opposed to a closed source driver from ATI that fucks your machine up on Windows? Or even better one of those noname brand cards that have better support on Linux than the crappy drivers on the CD that come with the card? 2D acceleration is enabled for all major brands of video cards (ATI, Nvidia, Via, and Intel [and probably Matrox too, but I don't have a card to test it on]) in the default free Xorg drivers. Sometimes the X configurator in the distro decides to use a generic framebuffer for no good reason. It's easy to fix.
I can Ctrl+C and Ctrl+V to cut'n'paste between applications.
The Windows clipboard beats any clipboard KDE or Gnome has. This is primarily an application level failure, since X doesn't even know what applications put on the clipboard (and none of the applications know what another application has put on, either). Just throwing a mime-type on the clipboard blob would probably do wonders for KDE/Gnome clipboard functionality. The content negotiation X supports is just not specific enough (and it's old formats).
Now it used to be that X Windows had network transparency and Windows didn't. But now there's Remote Desktop and VNC to handle that. And it was a far better idea to optimize for the common case where the application is running locally than across the network.
Remote Desktop is not truly network transparency; it's just remote desktop. Same with VNC. X lets you run applications anywhere and view them anywhere. The usefulness of such a thing is waning, but it's still possible to do GUI work on a server without taking up its own resources for graphics, instead letting your fancy new workstation do it. What would be a truly killer application is the ability to migrate applications from one X server to another without ending them. I haven't seen that put into common use, though. X basically has all the abstraction necessary to make truly distributed computing work seamlessly. I have to think pretty hard to find an example of a Win16 application running "well" for any reasonable definition of "well." The Win16 subsystem was almost worse than DOS in terms of design and implementation.
These days a 32 or 64 bit Windows app in C or C++ is seriously simple to write and it very efficient since you build the UI out of built in Windows classes. You subclass 'em and override the 1% of behaviour you want to change. And since you're only overriding 1%, everyone knows how to use it. You can make small exe files too, e.g. uTorrent.
I've done Win32 programming. It's not *that* wonderful. Neither is X11. Neither was that abomination of C++ classes Visual Studio used to use. The number of times I wished I could avoid the stupid class abstractions and just use Win32 function calls was amazing. But why bother with that low a level when there's Gtk and Qt? Small executables are usually a function of the libraries included, essentially. You can bloat your code in many ways, but the GUI toolkits are probably not a majority of the size. Shared libraries have a big impact on actual executable size, too.
That's where many weaknesses in cryptographic software have been found. With any luck the virus writers just borrowed the encryption code out of OpenSSL in the Debian tree.