People told you this about million times already, and I will repeat it: There _is_ demand for Linux version. But ASUS won't ship it, and if it does, it's very inferior to Windows offering.
Look at http://asus.as/ - official ASUS distributor for Czech Republic. There is a poll on the front page, and Linux/Windows have the same amount of votes. ASUS released EEE 900 (!) with Linux only after 4 months it released EEE 901 here and lots of people explicitly wanting the Linux version in the discussions. They won't release 901 version as of yet. There are other strange things - for example, if you bought EEE notebook in other EU country with Linux, you wouldn't have it repaired during warranty for free in Czechia; which differs from official ASUS policy on other notebooks. Also, you cannot get refund for Windows XP on EEE, which again differs from official ASUS policy on normal notebooks.
Maybe it will work differently. You will get cut off, but you still have to pay. Then, there is no problem for ISPs, in fact, more bandwidth can be overselled to others.
Thanks for the links, I will try to read them if I have time (I knew such proposal existed, I just didn't had the papers).
And no, I don't feel silly assuming hypothetical weaknesses in an unknown cryptosystem. Considering how cryptanalysis advanced in the last 20 years, I think it's quite safe to assume that some attacks to these systems will appear.
Actually, the Slysoft guy didn't give out that much of a hint. It was more psychological help than anything else - along the lines "it's not as difficult as it seems to be".
Yes, banking works on trust, but the point is, you will find out that it failed. Not so with election - you won't find out they have been tampered with.
Can you give me a source for that 1% figure? I find it very hard to believe; from my experience, there is at most one (out of several thousands) polling places where the things are contested, but either a recount or another voting will fix this.
Also, I have my doubts about the methods from the paper, though I haven't read it. For example, how resistant is the technique with respect to attacks on votes that have been cast by people who, for some reason (such as low technical skills), cannot check the correctness of the result? If someone else is going to do that for them, won't this compromise anonymity of their vote?
You know, having a theoretically good crypto algorithm is nice, but the correct implementation is in another ballpark.
But in banking system, you don't have to exactly trust the ATMs and stuff, because you can always check your account balance to see if correct amounts have been transferred.
Ballot boxes can be manipulated on small scale, but it's hard to do it on large scale. I don't know how many people are involved in vote counting in U.S., but here in Czech Republic, I would estimate we have one voting booth and ballot box per 1000-2000 people (there are no queues during the vote), and this place is manned by 3-6 people from different political parties, who also do the manual counting together. So it would mean to manipulate ten thousands of people to alter the results significantly (which I think is impossible to keep in secret). We also have more scalable system than U.S. in that there are no "contested regions" which are enough to manipulate in order to alter the results significantly.
I agree with that there exist (though still in theory) such methods. But: Most people will have trouble understanding them. How can you believe results of elections if you don't understand that the result is correct? And also, how can you be sure that politicians implemented the correct procedure, if you don't understand it? Compared to these methods, the paper ballot method is very simple and not much less secure.
So, until there are nanoreplicators that can change existing ballots in a closed box, I don't think these methods will be necessary. I still find them intriguing, though.
Actually, it's very reasonable. Soviets also tried to separate scientists into academies (which were research-oriented) and universities (which were teaching-oriented). And this system is not good, because the real good scientists cannot capture young people soon enough. We have remnants of such a system in Czech Republic, and the people from academies are competing to teach, because that way they can get fresh minds. It's not a good idea at all to separate teaching and research.
I believe it is not inevitable, but it's generally true that newer software is slower and no one I think knows what to do about it. The problem is that the modern pieces of software are large systems, and we don't know who to predict complexity of and optimize such large systems.
I think most of the bloat is accumulation of very small problems, for example in libraries. These problems may seem fine, but can be amplified in another layer of indirection. One would need to optimize through all the layers, and it would make impossible to write a complex program (or at least to write it quickly).
Maybe in the future, there will be compilers capable of that. But I don't see them now.
That's also the most idiotic solution. Sure, yes, reward those who got lucky and got large blocks in early days, and make pay those who actually use them.
Ok. Then it's just a misunderstanding, as I explain above. If this dissolution is either due to technology or bad management (and I agree both of them can be causes), it's outside of scope of economic theory (it has to be, otherwise the complexity would go through the roof), so from the economists' point of view, it's unnatural cause.
It's not natural way, because "natural" means it can happen in economic equilibrium (on macro scale, mind you).
If there is a technological progress, it means that there is no equilibrium. However, we could still approximate the areas where the technological progress has been very slow by equilibrium behavior (relative to the time scale chosen).
If dissolution cannot happen during the equilibrium, it is a good argument for regulation during such periods. But you are saying - if you wait long enough - which is basically a way to use time scale large enough that the economy cannot be considered in equilibrium anymore.
Not only that, it's even worse (though I agree completely with what you said). Even if you assume perfect information, Steve Keen has proven that the equilibrium for finite number of companies is always a cartel.
The classical argument about competition on free market that economists make is wrong, because it at one point relies the assumption that they affect each other so little that it can be neglected, but then they are trying to add up those zeros together. Another possible refutation is that if this argument was true, then the iterated prisoner dilemma would never have a cooperative stable solution.
You can only pack so much power into one processor, so from certain perspective, there is no other way than to go distributed. But I would still argue that systems that tend to pack as much as power as possible into as small space as possible (like IBM mainframes or large supercomputers) are more energy efficient than systems that use networks of smaller servers, like Google or Amazon. Which kinda proves my point. The purchasing cost, however, is a different matter, as I mentioned already (however, from 100 or so servers it's not so clear anymore).
I am actually making two points about advantage of centralization:
1. Less energy needed to move things around. 2. Resource sharing enabled by centralization.
You argue that the second point is crucial and not the first one, but in fact the second follows from the first. For example, you could in principle distribute the work you need to do, say, over the network of desktop machines, but it's much more efficient to distribute it around processors on a single large server. So you can share resources better on centralized system than distributed system.
And yes, it may also follow from my argument that for example web apps, which do a lot of network I/O (and not much actual computation on the server) between client and server, are in fact less efficient than desktop apps.
From physics, it's obvious that centralized computing is more energy efficient than distributed one. The longer distance you have to move energy (that encodes the information) to compute the results, the more energy you need. Also, centralization allows for better resource sharing.
The only issue is who pays for the costs. Mass production of computers allowed to decrease their costs to the point that distributed systems were cheaper than centralized ones. However, as the demand for computer power grows, energy spent on computing itself enters the equation, and the times will change again.
there are financial firm stockholders who took your advice who probably put a gun to their mouth and pulled the trigger.
I would say such people are mentally ill by any reasonable definition.
If there is reasonable chance that if Jobs suddenly left the company it would be bad for them, then it is reasonable for stock to decrease. It only decreased now because people suddenly realized that it could be a problem.
Note, I am not saying that speculations or rumors like this are any good. But if they get corrected quickly, they seem to be harmless for real long-term investors.
Slashdot Mirror
That would make sense if the service they're delivering is the refusal to give you access, I suppose.
That seems like an excellent business proposal.
I was actually joking.
People told you this about million times already, and I will repeat it: There _is_ demand for Linux version. But ASUS won't ship it, and if it does, it's very inferior to Windows offering.
Look at http://asus.as/ - official ASUS distributor for Czech Republic. There is a poll on the front page, and Linux/Windows have the same amount of votes. ASUS released EEE 900 (!) with Linux only after 4 months it released EEE 901 here and lots of people explicitly wanting the Linux version in the discussions. They won't release 901 version as of yet. There are other strange things - for example, if you bought EEE notebook in other EU country with Linux, you wouldn't have it repaired during warranty for free in Czechia; which differs from official ASUS policy on other notebooks. Also, you cannot get refund for Windows XP on EEE, which again differs from official ASUS policy on normal notebooks.
If this is not a cartel, I don't know what is.
Maybe it will work differently. You will get cut off, but you still have to pay. Then, there is no problem for ISPs, in fact, more bandwidth can be overselled to others.
Thanks for the links, I will try to read them if I have time (I knew such proposal existed, I just didn't had the papers).
And no, I don't feel silly assuming hypothetical weaknesses in an unknown cryptosystem. Considering how cryptanalysis advanced in the last 20 years, I think it's quite safe to assume that some attacks to these systems will appear.
Actually, the Slysoft guy didn't give out that much of a hint. It was more psychological help than anything else - along the lines "it's not as difficult as it seems to be".
Yes, banking works on trust, but the point is, you will find out that it failed. Not so with election - you won't find out they have been tampered with.
Can you give me a source for that 1% figure? I find it very hard to believe; from my experience, there is at most one (out of several thousands) polling places where the things are contested, but either a recount or another voting will fix this.
Also, I have my doubts about the methods from the paper, though I haven't read it. For example, how resistant is the technique with respect to attacks on votes that have been cast by people who, for some reason (such as low technical skills), cannot check the correctness of the result? If someone else is going to do that for them, won't this compromise anonymity of their vote?
You know, having a theoretically good crypto algorithm is nice, but the correct implementation is in another ballpark.
I would like to read that link, but it's paid-for paper. Maybe they should work on that first, if they want to get common people to trust this. :-)
But in banking system, you don't have to exactly trust the ATMs and stuff, because you can always check your account balance to see if correct amounts have been transferred.
Ballot boxes can be manipulated on small scale, but it's hard to do it on large scale. I don't know how many people are involved in vote counting in U.S., but here in Czech Republic, I would estimate we have one voting booth and ballot box per 1000-2000 people (there are no queues during the vote), and this place is manned by 3-6 people from different political parties, who also do the manual counting together. So it would mean to manipulate ten thousands of people to alter the results significantly (which I think is impossible to keep in secret). We also have more scalable system than U.S. in that there are no "contested regions" which are enough to manipulate in order to alter the results significantly.
I agree with that there exist (though still in theory) such methods. But: Most people will have trouble understanding them. How can you believe results of elections if you don't understand that the result is correct? And also, how can you be sure that politicians implemented the correct procedure, if you don't understand it? Compared to these methods, the paper ballot method is very simple and not much less secure.
So, until there are nanoreplicators that can change existing ballots in a closed box, I don't think these methods will be necessary. I still find them intriguing, though.
Actually, it's very reasonable. Soviets also tried to separate scientists into academies (which were research-oriented) and universities (which were teaching-oriented). And this system is not good, because the real good scientists cannot capture young people soon enough. We have remnants of such a system in Czech Republic, and the people from academies are competing to teach, because that way they can get fresh minds. It's not a good idea at all to separate teaching and research.
I believe it is not inevitable, but it's generally true that newer software is slower and no one I think knows what to do about it. The problem is that the modern pieces of software are large systems, and we don't know who to predict complexity of and optimize such large systems.
I think most of the bloat is accumulation of very small problems, for example in libraries. These problems may seem fine, but can be amplified in another layer of indirection. One would need to optimize through all the layers, and it would make impossible to write a complex program (or at least to write it quickly).
Maybe in the future, there will be compilers capable of that. But I don't see them now.
Is that wise? What about the configuration files? You will configure every program again?
Why not just hide their message in slashdot troll posts? Not like anyone reads them anyways...unless you know what you're looking for...
Maybe that's the real reason behind all those twitter conversations.
That's also the most idiotic solution. Sure, yes, reward those who got lucky and got large blocks in early days, and make pay those who actually use them.
No, because you also use them. Those people who have lots of IP addresses are not using them.
Ok. Then it's just a misunderstanding, as I explain above. If this dissolution is either due to technology or bad management (and I agree both of them can be causes), it's outside of scope of economic theory (it has to be, otherwise the complexity would go through the roof), so from the economists' point of view, it's unnatural cause.
It's not natural way, because "natural" means it can happen in economic equilibrium (on macro scale, mind you).
If there is a technological progress, it means that there is no equilibrium. However, we could still approximate the areas where the technological progress has been very slow by equilibrium behavior (relative to the time scale chosen).
If dissolution cannot happen during the equilibrium, it is a good argument for regulation during such periods. But you are saying - if you wait long enough - which is basically a way to use time scale large enough that the economy cannot be considered in equilibrium anymore.
Would you care to provide some examples where cartels dissolved naturally, and not due to evolutionary pressure from new technologies?
Not only that, it's even worse (though I agree completely with what you said). Even if you assume perfect information, Steve Keen has proven that the equilibrium for finite number of companies is always a cartel.
The classical argument about competition on free market that economists make is wrong, because it at one point relies the assumption that they affect each other so little that it can be neglected, but then they are trying to add up those zeros together. Another possible refutation is that if this argument was true, then the iterated prisoner dilemma would never have a cooperative stable solution.
You can only pack so much power into one processor, so from certain perspective, there is no other way than to go distributed. But I would still argue that systems that tend to pack as much as power as possible into as small space as possible (like IBM mainframes or large supercomputers) are more energy efficient than systems that use networks of smaller servers, like Google or Amazon. Which kinda proves my point. The purchasing cost, however, is a different matter, as I mentioned already (however, from 100 or so servers it's not so clear anymore).
I am actually making two points about advantage of centralization:
1. Less energy needed to move things around.
2. Resource sharing enabled by centralization.
You argue that the second point is crucial and not the first one, but in fact the second follows from the first. For example, you could in principle distribute the work you need to do, say, over the network of desktop machines, but it's much more efficient to distribute it around processors on a single large server. So you can share resources better on centralized system than distributed system.
And yes, it may also follow from my argument that for example web apps, which do a lot of network I/O (and not much actual computation on the server) between client and server, are in fact less efficient than desktop apps.
From physics, it's obvious that centralized computing is more energy efficient than distributed one. The longer distance you have to move energy (that encodes the information) to compute the results, the more energy you need. Also, centralization allows for better resource sharing.
The only issue is who pays for the costs. Mass production of computers allowed to decrease their costs to the point that distributed systems were cheaper than centralized ones. However, as the demand for computer power grows, energy spent on computing itself enters the equation, and the times will change again.
My daughter is 13 months old. She would not pass the Turing Test, yet is undeniably intelligent.
Wait until she is 13 years old, and hits puberty ... your opinion will change.
So she will then be able to pass Turing test, while being undeniably unintelligent?
there are financial firm stockholders who took your advice who probably put a gun to their mouth and pulled the trigger.
I would say such people are mentally ill by any reasonable definition.
If there is reasonable chance that if Jobs suddenly left the company it would be bad for them, then it is reasonable for stock to decrease. It only decreased now because people suddenly realized that it could be a problem.
Note, I am not saying that speculations or rumors like this are any good. But if they get corrected quickly, they seem to be harmless for real long-term investors.
That's quite funny response, actually, because I am not associated with twitter in any way. I am just wondering.