Interestingly, solar power actually increases the amount of heat delivered to the surrounding area. Ordinary wildlife, and sunny deserts in particular, reflect a significant percentage of energy back into space without absorbing it.
Photovoltaics would reduce the heat delivered if they weren't so inefficient - more heat is left behind than would have been absorbed by the ground in the first place. Power plants that concentrate sunlight to heat pipes have the same problem. The immediate area actually gets hotter than it was before.
Of course, this has little to do with aquathermal power...
Given the rates at which prices for storage, processing, and networking were plummeting, a problem that would take eight years to solve today could be solvable in four years two years out, and in two years two years from now. So by putting it off for two years, you'd shave two years off the project.
This may be true in the everyday world of cut-throat competition, but if we call this "optimal" and everybody does it, everybody waits two years, and nobody puts forth the effort to realize the gain in productivity. Leeching off of technology that hasn't been invented yet reaps the benefits of work paid for by whomever goes first. I call it a technologically-oriented game of chicken.
Looking more closely at NASA's past projects, you will find that NASA takes precisely that role - the government puts up huge sums of money on an unproven technology, and the world reaps the benefits years (or decades) later. From the taxpayer's perspective, the only important criterion is whether the indirect reward will pay back the taxpayer for the up-front costs.
A P233 is old hardware. I just (last week) disposed of a P200 in constant use since 1997. The thing never broke down over 8 years, but I couldn't put more than 64MB of RAM in it, so it wasn't running reasonably anymore. That is, the OS with minimal services and a firewall was acceptable so long as no time-sensitive applications were running. When I dismantled it, it was running Win2k as an Internet gateway, but it originally came with Win95.
IANAAIE (AI expert), but my intuition says that maintaining control of a large vehicle at 20+ MPH on rugged terrain requires one hell of a real-time sensor system. The system doesn't understand patterns as well as humans do just from stereoscopic images, so it has to compensate by gathering many other kinds of information in real-time. Also consider that the algorithms we normally use to process such information accurately can take a long time - longer than your average human reaction time.
In theory, you're right, a good driver wouldn't necessarily be able to do this task, which means anyone who wins this challenge might have created a car that drives better than we do.:)
At forward viewing angles, yes, the images would be blue-shifted, but this doesn't mean everything goes dark. Visible becomes UV, and infrared becomes visible. But this is angle-dependent. Light arriving from behind you is actually red-shifted.
And yes, pushing several hundred watts per square meter of visible light into the UV range would result in a terrible sunburn.
I can see a $499 Mac Mini filling the needs of a lot of government agencies. And it's immunity to viruses and spyware is a big bonus.
I'm not sure whether to consider this flamebait or just RDF. Mac hardware is not immune to malicious software, nor is OS X, the software. Making blanket statements like this despite a clear track record of vulnerabilities in this platform (and others) just distorts and polarizes opinions to the point that everyone loses sight of what causes vulnerabilities in the first place: poor software design.
I, personally, am primarily a Windows user, and I know Windows is vulnerable. But so are Linux and OS X. Don't lose sight of that.
So what happens when one of the human players on the other team does a slide tackle on the robot? Does the robot fall down? Does the robot get damaged? Does the human get injured? IANASP, but it seems like physical contact between opposing players is so common that replacing man with machine is either clumsy, scary, or both.
It's actually quite simple. Hash collisions are everywhere, no matter what hash function you use. They are all vulnerable to the birthday attack because if you use too many documents, you will have a non-trivial probability that two documents will have the same hash.
However, this isn't the real problem. Hashes can be secure so long as the collisions are random and sparse. For a secure 128-bit hash function, this is the case. You can't just keep trying documents until you find a random collision. That would take longer than the age of the universe given today's computers.
Actually, even if someone does find a random collision, it's not a real problem because software can be retrofit to detect constructions that use (or abuse) the known collision. Or the algorithm can be trivially changed so that the attacker has to start searching all over again.
Hash functions really become insecure when someone finds a formula to generate two documents with the same hash quickly. Or, even worse, when someone finds a formula that, given two documents, specifies the modifications for one of the documents to collide with the other!
I consider myself independent, so I'm not trying to be partisan here. Calling someone a traitor for exposing war crimes committed by other soldiers is absurd. The fact that yet more other soldiers may have been imprisoned and tortured to the point of false confessions is depressing, but completely irrelevant. If war crimes were committed, we need to accept that and punish the perpetrators according to the merits.
Whistle-blowers are not traitors. People who commit war crimes and then try to deflect blame onto the accusers are the traitors! Shame on those who are unwilling or incapable of seeing them for what they are.
This will dissipate any investors' fears pertaining to the validity of the lawsuit.
...to the validity of what lawsuit? SCO has never been part of a lawsuit. The real purpose of today's press conference is a surprise launch of our new product! It's called Linnux, and it's only $699 per seat!
Forgive me for replying to a troll, but I just can't resist.
It's not Apple that started this. Tom's Hardware started this. The tin-foil hats would say that Intel is ultimately behind it, but who knows. This all started with Tom's now-infamous video of an AMD processor going up in smoke after the heatsink "falls off".
I would argue that this increasing focus on heat dissipation in desktop PCs would likely not have come up if AMD's Athlon chips hadn't been smeared as defective space-heaters. What we're seeing now is just the backlash. Now people are noticing that, hey, look at how hot Intel's chips are getting! (Never mind that a heatsink coming off during operation of a properly maintained computer is practically impossible. Nobody seemed to care at the time.)
Well, yes, of course I realized that, but many schools (including Stanford) have plenty of financial aid available for those that need it. Level of financial need isn't even a factor in admissions - but I'm sure there are a few people who get shafted. And there really aren't many people on campus who pinch pennies to get by. The aid packages are pretty good.
One of my friends is only paying $400 per quarter in tuition, and most of his aid is grants. For someone in his position, textbooks are a pretty big chunk of his expenses, so this topic still applies.
Extortion (dictionary.com): 1.... the act or practice of wresting anything from a person by force, by threats, or by any undue exercise of power; undue exaction; overcharge.
IANAL. No, it's not textbook extortion. Threatening to sue someone for a bona fide violation of law is not extortion. It's deterrence. Also, five dollars a month per person cannot be considered "overcharge" even if the student can't avoid the charge. It's a legitimate service that the school has decided to pay for. We can question the merits of such a service but it's not illegal.
My main objection to "deals" like this is the effective imposition of an "RIAA Tax". I don't like monopolies charging for things someone might not even use just because they can. That said, for people who do like downloads, it's a pretty small price to pay for being legal. I wonder if it's still worth it after all the DRM restrictions (if any) are factored in.
Actually, I don't think this is a problem with America. It's a problem with each and every College and University that does it. The fact that most schools here do it simply indicates that most schools have been sliding down the slippery slope towards abandoning their first and foremost priority: education.
And yes, IMHO, professors selling their textbooks to their own students for profit is an inherent conflict of interest that undermines instructive value.
My personal experience is pretty telling. I have had to buy expensive books with no resale value, but many of those are (to me) worth keeping. Most of the core undergraduate engineering classes here have no required textbooks, just optional reference books. All of the course curriculum in most of my classes is contained in slides and handouts that stay relatively constant over time and teach the material better than any textbook could.
The professors always pass on these notes to whomever teaches the class next, presumably without any royalties other than their existing salaries, which are pretty high. The professors here are, by far, concerned about maintaining the best curriculum, and, for them, that means sidestepping the textbook industry entirely.
And.. what school am I talking about? Stanford University.
I realize that everybody knows increasing file sizes only buys you so much time, but there's another much more important reason why this is a terrible idea.
CPU processing "power", which could be represented for the purposes of this situation as instructions per second, is increasing exponentially over time at a rate of approximately 25-50% per year. Note that this is not as fast as Moore's Law, because that applies to the number of transistors per chip, not instructions per second.
Telecommunications, on the other hand, has been increasing in bandwidth at a much faster rate. Very roughly, we're talking about an increase in consumer download capability from 14 kbps to 1Mbps or more for the target market, all in the last 12 years. That comes out to an average exponential growth of 71% per year. And core routing capabilities are growing probably even faster than that, but let's not go there.
Now, if we assume that the amount of computation for video is roughly proportional to the data size, your ability to "bloat" the video stream with more pixels is bounded NOT by your Internet bandwidth, but by your computational limits.
Thus, the point of this argument is, even if you stress the user's computer to its limit when playing video, the stress level on the user's Internet connection will keep falling at 20-45% per year. The user will find it progressively easier to get videos through downloading, and there is NOTHING companies can do to stop it.
Discouraging, boycotting, or flat-out disallowing companies from hocking their products through spam would be great. This would probably cover a minority of spam-financed revenues, however.
It seems that the real money comes from spam campaigns that are already illegal. They're from shady or non-existent companies. They're selling counterfeit products that sometimes even have recklessly dangerous ingredients added. And, of course, there are also the get-rich-quick scams.
Going after Viagra makes no sense here unless it can be shown that Pfizer is actually contributing to the spam campaign in some way. AFAIK, all the Viagra emails you see are fraudulent ads not sent by Pfizer. Viagra is a prescription drug, remember - how is it even possible for legit online vendors to sell Viagra without verifying the prescription?
Slashdot Mirror
Error on line 1: mismatched metaphor. Go around the barn at high noon to avoid colloquialisms.
Interestingly, solar power actually increases the amount of heat delivered to the surrounding area. Ordinary wildlife, and sunny deserts in particular, reflect a significant percentage of energy back into space without absorbing it.
Photovoltaics would reduce the heat delivered if they weren't so inefficient - more heat is left behind than would have been absorbed by the ground in the first place. Power plants that concentrate sunlight to heat pipes have the same problem. The immediate area actually gets hotter than it was before.
Of course, this has little to do with aquathermal power...
Given the rates at which prices for storage, processing, and networking were plummeting, a problem that would take eight years to solve today could be solvable in four years two years out, and in two years two years from now. So by putting it off for two years, you'd shave two years off the project.
This may be true in the everyday world of cut-throat competition, but if we call this "optimal" and everybody does it, everybody waits two years, and nobody puts forth the effort to realize the gain in productivity. Leeching off of technology that hasn't been invented yet reaps the benefits of work paid for by whomever goes first. I call it a technologically-oriented game of chicken.
Looking more closely at NASA's past projects, you will find that NASA takes precisely that role - the government puts up huge sums of money on an unproven technology, and the world reaps the benefits years (or decades) later. From the taxpayer's perspective, the only important criterion is whether the indirect reward will pay back the taxpayer for the up-front costs.
A P233 is old hardware. I just (last week) disposed of a P200 in constant use since 1997. The thing never broke down over 8 years, but I couldn't put more than 64MB of RAM in it, so it wasn't running reasonably anymore. That is, the OS with minimal services and a firewall was acceptable so long as no time-sensitive applications were running. When I dismantled it, it was running Win2k as an Internet gateway, but it originally came with Win95.
IANAAIE (AI expert), but my intuition says that maintaining control of a large vehicle at 20+ MPH on rugged terrain requires one hell of a real-time sensor system. The system doesn't understand patterns as well as humans do just from stereoscopic images, so it has to compensate by gathering many other kinds of information in real-time. Also consider that the algorithms we normally use to process such information accurately can take a long time - longer than your average human reaction time.
:)
In theory, you're right, a good driver wouldn't necessarily be able to do this task, which means anyone who wins this challenge might have created a car that drives better than we do.
At forward viewing angles, yes, the images would be blue-shifted, but this doesn't mean everything goes dark. Visible becomes UV, and infrared becomes visible. But this is angle-dependent. Light arriving from behind you is actually red-shifted.
And yes, pushing several hundred watts per square meter of visible light into the UV range would result in a terrible sunburn.
I can see a $499 Mac Mini filling the needs of a lot of government agencies. And it's immunity to viruses and spyware is a big bonus.
I'm not sure whether to consider this flamebait or just RDF. Mac hardware is not immune to malicious software, nor is OS X, the software. Making blanket statements like this despite a clear track record of vulnerabilities in this platform (and others) just distorts and polarizes opinions to the point that everyone loses sight of what causes vulnerabilities in the first place: poor software design.
I, personally, am primarily a Windows user, and I know Windows is vulnerable. But so are Linux and OS X. Don't lose sight of that.
Actually, being based on C is not what makes verification problematic. Verification of programs in any Turing-complete language is undecidable!
As a simple example, try to verify whether the following program obeys certain "security" properties:
if (system.verify(program)) {do evil deed}
else {be nice}
This is the same classic style of counterexample used to prove undecidability of the halting problem, of virus detection, and so on.
IANASP = I am not a soccer player.
So what happens when one of the human players on the other team does a slide tackle on the robot? Does the robot fall down? Does the robot get damaged? Does the human get injured? IANASP, but it seems like physical contact between opposing players is so common that replacing man with machine is either clumsy, scary, or both.
It's actually quite simple. Hash collisions are everywhere, no matter what hash function you use. They are all vulnerable to the birthday attack because if you use too many documents, you will have a non-trivial probability that two documents will have the same hash.
However, this isn't the real problem. Hashes can be secure so long as the collisions are random and sparse. For a secure 128-bit hash function, this is the case. You can't just keep trying documents until you find a random collision. That would take longer than the age of the universe given today's computers.
Actually, even if someone does find a random collision, it's not a real problem because software can be retrofit to detect constructions that use (or abuse) the known collision. Or the algorithm can be trivially changed so that the attacker has to start searching all over again.
Hash functions really become insecure when someone finds a formula to generate two documents with the same hash quickly. Or, even worse, when someone finds a formula that, given two documents, specifies the modifications for one of the documents to collide with the other!
I consider myself independent, so I'm not trying to be partisan here. Calling someone a traitor for exposing war crimes committed by other soldiers is absurd. The fact that yet more other soldiers may have been imprisoned and tortured to the point of false confessions is depressing, but completely irrelevant. If war crimes were committed, we need to accept that and punish the perpetrators according to the merits.
Whistle-blowers are not traitors. People who commit war crimes and then try to deflect blame onto the accusers are the traitors! Shame on those who are unwilling or incapable of seeing them for what they are.
This will dissipate any investors' fears pertaining to the validity of the lawsuit.
...to the validity of what lawsuit? SCO has never been part of a lawsuit. The real purpose of today's press conference is a surprise launch of our new product! It's called Linnux, and it's only $699 per seat!
Forgive me for replying to a troll, but I just can't resist.
It's not Apple that started this. Tom's Hardware started this. The tin-foil hats would say that Intel is ultimately behind it, but who knows. This all started with Tom's now-infamous video of an AMD processor going up in smoke after the heatsink "falls off".
I would argue that this increasing focus on heat dissipation in desktop PCs would likely not have come up if AMD's Athlon chips hadn't been smeared as defective space-heaters. What we're seeing now is just the backlash. Now people are noticing that, hey, look at how hot Intel's chips are getting! (Never mind that a heatsink coming off during operation of a properly maintained computer is practically impossible. Nobody seemed to care at the time.)
Well, yes, of course I realized that, but many schools (including Stanford) have plenty of financial aid available for those that need it. Level of financial need isn't even a factor in admissions - but I'm sure there are a few people who get shafted. And there really aren't many people on campus who pinch pennies to get by. The aid packages are pretty good.
One of my friends is only paying $400 per quarter in tuition, and most of his aid is grants. For someone in his position, textbooks are a pretty big chunk of his expenses, so this topic still applies.
Go around the barn at high noon to avoid colloquialisms.
Extortion (dictionary.com): 1. ... the act or practice of wresting anything from a person by force, by threats, or by any undue exercise of power; undue exaction; overcharge.
IANAL. No, it's not textbook extortion. Threatening to sue someone for a bona fide violation of law is not extortion. It's deterrence. Also, five dollars a month per person cannot be considered "overcharge" even if the student can't avoid the charge. It's a legitimate service that the school has decided to pay for. We can question the merits of such a service but it's not illegal.
My main objection to "deals" like this is the effective imposition of an "RIAA Tax". I don't like monopolies charging for things someone might not even use just because they can. That said, for people who do like downloads, it's a pretty small price to pay for being legal. I wonder if it's still worth it after all the DRM restrictions (if any) are factored in.
Actually, I don't think this is a problem with America. It's a problem with each and every College and University that does it. The fact that most schools here do it simply indicates that most schools have been sliding down the slippery slope towards abandoning their first and foremost priority: education.
And yes, IMHO, professors selling their textbooks to their own students for profit is an inherent conflict of interest that undermines instructive value.
My personal experience is pretty telling. I have had to buy expensive books with no resale value, but many of those are (to me) worth keeping. Most of the core undergraduate engineering classes here have no required textbooks, just optional reference books. All of the course curriculum in most of my classes is contained in slides and handouts that stay relatively constant over time and teach the material better than any textbook could.
The professors always pass on these notes to whomever teaches the class next, presumably without any royalties other than their existing salaries, which are pretty high. The professors here are, by far, concerned about maintaining the best curriculum, and, for them, that means sidestepping the textbook industry entirely. And.. what school am I talking about? Stanford University.
Whoops, sorry, I goofed the exponential math. Revised figures:
:)
CPU power increases: probably 25-35% or less
Bandwidth increases: probably 40-50% or more
You get the idea
I realize that everybody knows increasing file sizes only buys you so much time, but there's another much more important reason why this is a terrible idea.
CPU processing "power", which could be represented for the purposes of this situation as instructions per second, is increasing exponentially over time at a rate of approximately 25-50% per year. Note that this is not as fast as Moore's Law, because that applies to the number of transistors per chip, not instructions per second.
Telecommunications, on the other hand, has been increasing in bandwidth at a much faster rate. Very roughly, we're talking about an increase in consumer download capability from 14 kbps to 1Mbps or more for the target market, all in the last 12 years. That comes out to an average exponential growth of 71% per year. And core routing capabilities are growing probably even faster than that, but let's not go there.
Now, if we assume that the amount of computation for video is roughly proportional to the data size, your ability to "bloat" the video stream with more pixels is bounded NOT by your Internet bandwidth, but by your computational limits.
Thus, the point of this argument is, even if you stress the user's computer to its limit when playing video, the stress level on the user's Internet connection will keep falling at 20-45% per year. The user will find it progressively easier to get videos through downloading, and there is NOTHING companies can do to stop it.
Discouraging, boycotting, or flat-out disallowing companies from hocking their products through spam would be great. This would probably cover a minority of spam-financed revenues, however. It seems that the real money comes from spam campaigns that are already illegal. They're from shady or non-existent companies. They're selling counterfeit products that sometimes even have recklessly dangerous ingredients added. And, of course, there are also the get-rich-quick scams. Going after Viagra makes no sense here unless it can be shown that Pfizer is actually contributing to the spam campaign in some way. AFAIK, all the Viagra emails you see are fraudulent ads not sent by Pfizer. Viagra is a prescription drug, remember - how is it even possible for legit online vendors to sell Viagra without verifying the prescription?