... computer malware copy the lifecycle of their organic counterparts; some viruses and parasites, in order to remain in the host, evolved to grant some advantage to the host in question. The analogy in the computer realm is this "greyware" - the advantage being some valid function (legitimate program), and the parasitic or viral aspect being the malicious part of the greyware in question.
Not all viruses evolve in this manner, however. Some just entangle themselves so deeply it's impossible to remove, or to remove it would cause great harm. Fortunately, there's not many computer examples of the latter, but see cryptovirology for ideas how they might in the future.
In the end it matters not, if Negroponte wants Sugar on Windows all he has to do is ask that wealthy corporation to invest some of their ill gotten gains in porting the open source code themselves.
If Microsoft ports Sugar to Windows, perhaps they could call it.. Chigurh.
In order to make it economically unsound for the spammers, you'd have to make it economically annoying for the rest of humanity. More annoying than simply putting up with the spam.
Not necessarily. If you have a trust network or database telling you which sources are more likely to spam (like RBL but with degrees instead of "either you're a spammer or you're not"), mail servers could demand more of sources that are likely to spam. Just connect this thing to another network of cryptographic time stamp servers (who, themselves, don't permit a single address to get more than a single token in a given interval), and demand that legitimate users send no more than say, 10 mails per minute and spammy users send no more than 0.1 mail messages per minute. Boom, spam zombies are slowed down by 100x.
That's an economic solution to the degree that the cryptographic timestamp servers print money and the RBL-alikes lets one adjust supply and demand. If you can't trust the timestamp servers, a poor man's approximation could be proof of work (like Hashcash, but use something memory bound since memory speed doubles more slowly than CPU power). See this paper about that strategy.
You talk about "computing power" as if it plateauing would prevent the singularity, and Kurzweil et al. make diagrams like these that pretty much link Moore's law and the Singularity together.
As for brute force not leading to scalable intelligence, just take a quick look at tree search. That's exponential; it's just that, as an example, the game tree for chess is narrow enough that computers can beat grandmasters. And it's not just chess. Many other puzzles, when generalized, are NP-complete, and many games PSPACE-complete (at least). To make headway into the harder areas here is going to require real effort and ingenuity, not just traveling up the computing power graph.
Therefore it doesn't matter where computing power will plateau, if it will. As long as the algorithms are exponential for most real cases, the horizon will always catch up with you.
So when, mighty imaginer, will computing power plateau?
Who knows, but it's not important. You can't make a fire build a house, no matter how powerful fuel you're pouring on it. In other words, it's going to take more than just brute force to scale up intelligence.
Capabilities would help a lot (and get rid of "buffer overflows let you run anything as the user"), but they wouldn't be a cure-all unless you could somehow solve the trust problem. Otherwise, the.xyz.exe the spam carries around will simply say "To show the Dancing Pigs Gallery, you must give access to the internet for ports 1024 so it can search the web for all new dancing pigs!". The user grants capabilities, and boom.
As opposed to raytracing, which is so extremely parallel that it scales nearly linearly. In other words, your 3-way real-time raytracer graphics card (if/when such a beast is ever made) would perform at about 2.8x the one-GPU variant. And unless the Rapture^Wsingularity keeps you from getting a 192-GPU card, it'd render at 170x the reference or so.
(Of course, there's the question of global illumination. I don't know if those can be parallelized as easily, but there was a story about distributed photon mapping here some time back, where they used Blue Gene.)
In other words, the corporations are paying people to go looking for exploits in our brains. Full disclosure and all that, right? The problem is that once they've found the exploit, you can't just go and get a patch from the vendor. They're not full disclosure either; a better analogy would be those zero-day trading scenes where crackers sell exploits to the krasnaya mafiya.
Just why should this be legal?
(If you want to be picky about it, it's more like privilege escalation than rooting.. but I'm straining the analogy. And note that the market only works if we're "rational actors" - totally bug-free.)
That can't be the reason in itself. It shows that having a plurality but not a majority would be a real mess, but not that it'd be impossible. Even in plurality voting, it'd be perfectly possible (in a theoretical manner) for say, each state to elect a different candidate, which would cause the winner to have something way short of a majority.
More likely is Duverger's law, that plurality favors a two-party system. The reason isn't hard to see, since plurality requires a binary choice, and you can't code a preference for more than (this guy above all the others) with a binary choice. Therefore you get vote splitting, and from that in turn you get two-party rule.
Now, you may say that the states, knowing about Duverger, would consciously pick plurality in order to avoid the scheming of a winner short of a majority in the electoral college. But I think that's unlikely; they're not that sophisticated.
In simpler words, there would be nothing stopping an individual state from saying, for instance, "we're going to elect the candidate for our state by Condorcet, and then have all our electors vote for whoever was the Condorcet winner". The state in question might have to rewrite its faithless elector laws, but since it'd be altering the electoral system, it would already have to rewrite a lot of other laws. Such a modification wouldn't be unconstitutional, but it could cause a return to the scheming you mention (if state Condorcet winners differ too greatly).
Or perturb the logic. The easy way is just to look at how polymorphic viruses did it. The hard way is to get out your disassembler and change
cmp eax, edx
jle offset
to cmp edx, eax
jae offset (insert your own variation here). Have a program read all cmp eax, edx (or cmp edx, eax) opcodes and output 0 for the first and 1 for the second.
List two software innovations (i.e. something not copied) done by the linux/hobbyist community please.
How about that Apache - the web server, you know, where it's Microsoft doing the copying? Unless they both count as "copies" of whoever made the first web server. Or if you want something tangible, I'm sure there's some kernel trick that's unique; like the O(1) scheduler (especially the Really Fair - or whatever it's called now - modifications and subsequent flamewar) or some of the more obscure file systems. The PaX break-on-execute-for-x86 trick may also count, but I'm not sure if it's been used before on a non-linux platform.
Without patents, the result will be predictable: most people will keep their algorithms a closely guarded secret. The result will be that academia will suffer as algorithms go from publicly disclosed patents to trade secrets.
And then a clever hacker will reverse engineer the algorithm and leak it to the world. Short of DMCA-type problems (which is an entirely different mess), there's nothing the companies can do since there are no more software patents, and if the prevalence of cracks show anything, it's that any program can be reverse-engineered.
I haven't used Tor in a while, but can't you blacklist nodes explicitly based on their public keys or location? Blacklist all the internal China nodes and you're good... or is there more to it?
Yes. The important part, however, is that the limit is one of economics, and not of physics; the economic situation may change, so it's reasonable to do research into how to store hydrogen (as in this case) or use it (as with fuel cells and so on) even if hydrogen extraction is too expensive at the moment, or the technology hasn't gone from research to engineering yet.
There are many well known ways to extract hydrogen without creating carbon dioxide (electrolysis from renewable source of your choice, molten salt or gaseous generation IV reactors with thermocracking, etc); they're just not viable.
And only with a passive MITM attack. There's nothing to stop the man in the middle from having his own receiver and transmitter, and acting like A to B and B to A, sending photons of his own in reply to each (an active MITM attack); at least not unless the parties have exchanged a short password beforehand.
Sorry but I loose hours a week to deleting spam even with filters.
How about codifying this? Let's say that deleting a spam takes a second, and the guy sends a hundred million spam messages. Put him in jail for a hundred million seconds; that's 27777 hours or 1157 days, a little over three years. And what kind of spam king would send only a hundred million spam messages?
Slashdot Mirror
Does the crime of Slashdot first-posting get you on that list?
... computer malware copy the lifecycle of their organic counterparts; some viruses and parasites, in order to remain in the host, evolved to grant some advantage to the host in question. The analogy in the computer realm is this "greyware" - the advantage being some valid function (legitimate program), and the parasitic or viral aspect being the malicious part of the greyware in question.
Not all viruses evolve in this manner, however. Some just entangle themselves so deeply it's impossible to remove, or to remove it would cause great harm. Fortunately, there's not many computer examples of the latter, but see cryptovirology for ideas how they might in the future.
In the end it matters not, if Negroponte wants Sugar on Windows all he has to do is ask that wealthy corporation to invest some of their ill gotten gains in porting the open source code themselves.
If Microsoft ports Sugar to Windows, perhaps they could call it.. Chigurh.
The mind boogles.
In order to make it economically unsound for the spammers, you'd have to make it economically annoying for the rest of humanity. More annoying than simply putting up with the spam.
Not necessarily. If you have a trust network or database telling you which sources are more likely to spam (like RBL but with degrees instead of "either you're a spammer or you're not"), mail servers could demand more of sources that are likely to spam. Just connect this thing to another network of cryptographic time stamp servers (who, themselves, don't permit a single address to get more than a single token in a given interval), and demand that legitimate users send no more than say, 10 mails per minute and spammy users send no more than 0.1 mail messages per minute. Boom, spam zombies are slowed down by 100x.
That's an economic solution to the degree that the cryptographic timestamp servers print money and the RBL-alikes lets one adjust supply and demand. If you can't trust the timestamp servers, a poor man's approximation could be proof of work (like Hashcash, but use something memory bound since memory speed doubles more slowly than CPU power). See this paper about that strategy.
On a Gaussian, they're the same.
You talk about "computing power" as if it plateauing would prevent the singularity, and Kurzweil et al. make diagrams like these that pretty much link Moore's law and the Singularity together.
As for brute force not leading to scalable intelligence, just take a quick look at tree search. That's exponential; it's just that, as an example, the game tree for chess is narrow enough that computers can beat grandmasters. And it's not just chess. Many other puzzles, when generalized, are NP-complete, and many games PSPACE-complete (at least). To make headway into the harder areas here is going to require real effort and ingenuity, not just traveling up the computing power graph.
Therefore it doesn't matter where computing power will plateau, if it will. As long as the algorithms are exponential for most real cases, the horizon will always catch up with you.
So when, mighty imaginer, will computing power plateau?
Who knows, but it's not important. You can't make a fire build a house, no matter how powerful fuel you're pouring on it. In other words, it's going to take more than just brute force to scale up intelligence.
You know, that's a pretty strange way of writing "alyuminii".
Capabilities would help a lot (and get rid of "buffer overflows let you run anything as the user"), but they wouldn't be a cure-all unless you could somehow solve the trust problem. Otherwise, the .xyz.exe the spam carries around will simply say "To show the Dancing Pigs Gallery, you must give access to the internet for ports 1024 so it can search the web for all new dancing pigs!". The user grants capabilities, and boom.
No, you've got it wrong. Dollars are now dog Euros.
As opposed to raytracing, which is so extremely parallel that it scales nearly linearly. In other words, your 3-way real-time raytracer graphics card (if/when such a beast is ever made) would perform at about 2.8x the one-GPU variant. And unless the Rapture^Wsingularity keeps you from getting a 192-GPU card, it'd render at 170x the reference or so.
(Of course, there's the question of global illumination. I don't know if those can be parallelized as easily, but there was a story about distributed photon mapping here some time back, where they used Blue Gene.)
Now I'm going to be moderated a troll by someone who writes parodies of Gilbert and Sullivan operettas. Sigh.
I am the very model of a modern Slashdot forum troll
and when I post and moderate I always will be very droll...
In other words, the corporations are paying people to go looking for exploits in our brains. Full disclosure and all that, right? The problem is that once they've found the exploit, you can't just go and get a patch from the vendor. They're not full disclosure either; a better analogy would be those zero-day trading scenes where crackers sell exploits to the krasnaya mafiya.
Just why should this be legal?
(If you want to be picky about it, it's more like privilege escalation than rooting.. but I'm straining the analogy. And note that the market only works if we're "rational actors" - totally bug-free.)
That can't be the reason in itself. It shows that having a plurality but not a majority would be a real mess, but not that it'd be impossible. Even in plurality voting, it'd be perfectly possible (in a theoretical manner) for say, each state to elect a different candidate, which would cause the winner to have something way short of a majority.
More likely is Duverger's law, that plurality favors a two-party system. The reason isn't hard to see, since plurality requires a binary choice, and you can't code a preference for more than (this guy above all the others) with a binary choice. Therefore you get vote splitting, and from that in turn you get two-party rule.
Now, you may say that the states, knowing about Duverger, would consciously pick plurality in order to avoid the scheming of a winner short of a majority in the electoral college. But I think that's unlikely; they're not that sophisticated.
In simpler words, there would be nothing stopping an individual state from saying, for instance, "we're going to elect the candidate for our state by Condorcet, and then have all our electors vote for whoever was the Condorcet winner". The state in question might have to rewrite its faithless elector laws, but since it'd be altering the electoral system, it would already have to rewrite a lot of other laws. Such a modification wouldn't be unconstitutional, but it could cause a return to the scheming you mention (if state Condorcet winners differ too greatly).
Or perturb the logic. The easy way is just to look at how polymorphic viruses did it. The hard way is to get out your disassembler and change
cmp eax, edx
jle offset
to
cmp edx, eax
jae offset
(insert your own variation here). Have a program read all cmp eax, edx (or cmp edx, eax) opcodes and output 0 for the first and 1 for the second.
Would that make Bill the Tomator?
List two software innovations (i.e. something not copied) done by the linux/hobbyist community please.
How about that Apache - the web server, you know, where it's Microsoft doing the copying? Unless they both count as "copies" of whoever made the first web server. Or if you want something tangible, I'm sure there's some kernel trick that's unique; like the O(1) scheduler (especially the Really Fair - or whatever it's called now - modifications and subsequent flamewar) or some of the more obscure file systems. The PaX break-on-execute-for-x86 trick may also count, but I'm not sure if it's been used before on a non-linux platform.
Without patents, the result will be predictable: most people will keep their algorithms a closely guarded secret. The result will be that academia will suffer as algorithms go from publicly disclosed patents to trade secrets.
And then a clever hacker will reverse engineer the algorithm and leak it to the world. Short of DMCA-type problems (which is an entirely different mess), there's nothing the companies can do since there are no more software patents, and if the prevalence of cracks show anything, it's that any program can be reverse-engineered.
I haven't used Tor in a while, but can't you blacklist nodes explicitly based on their public keys or location? Blacklist all the internal China nodes and you're good... or is there more to it?
I'll tell you what's wrong with the IOC: All the committee-members in their rooms, dancing and singing...
Money, money, money
always sunny
in the rich man's world...
Yes. The important part, however, is that the limit is one of economics, and not of physics; the economic situation may change, so it's reasonable to do research into how to store hydrogen (as in this case) or use it (as with fuel cells and so on) even if hydrogen extraction is too expensive at the moment, or the technology hasn't gone from research to engineering yet.
There are many well known ways to extract hydrogen without creating carbon dioxide (electrolysis from renewable source of your choice, molten salt or gaseous generation IV reactors with thermocracking, etc); they're just not viable.
All theoretically of course.
And only with a passive MITM attack. There's nothing to stop the man in the middle from having his own receiver and transmitter, and acting like A to B and B to A, sending photons of his own in reply to each (an active MITM attack); at least not unless the parties have exchanged a short password beforehand.
Sorry but I loose hours a week to deleting spam even with filters.
How about codifying this? Let's say that deleting a spam takes a second, and the guy sends a hundred million spam messages. Put him in jail for a hundred million seconds; that's 27777 hours or 1157 days, a little over three years. And what kind of spam king would send only a hundred million spam messages?