One problem with a demo of this sort is that it isn't possible to know how much was actually done by the quantum computer and how much was done by the classical front end computer. Typical Sudoku puzzles can
be solved essentially instantaneously by classical computers. 16 qubits isn't a heck of a lot of
quantum parellelism, so I do wonder how much of the real work they were doing.
They make it sound like they have built a device that will solve an arbitrary instance of a particular
NP-complete problem (the two dimensional Ising model in a magnetic field). Indeed, they argue that they
will be able to solve any problem in NP by using a conventional computer to convert that problem to the one they can solve. It would be a major breakthrough to show that any NP-complete problem can be solved in polynomial time on a quantum computer, but I don't think they've done it. (Factoring can be done in polynomial time on a quantum computer, but is not generally believed to be NP-complete.)
From their paper, "The third, and perhaps most interesting, time-complexity result--
the possibility of exponential speedup on almost all difficult instances
of NP-hard problems--comes from explicit numerical integrations of
Schroedinger's equation to simulate adiabatic computation on sets of
small instances of these problems..."
What they have is not a mathematical proof but some hand waving extrapolation based on small instances.
Most likely what they have is a faster way to do simulated annealing. If so, it will give good, but
not necessarily optimal, approximations to certain NP-complete problems. But approximations don't in general transfer across the polynomial time reductions used to convert one NP-complete problem to another.
So their idea of having a front end computer that transforms an arbitrary problem to their Ising model
doesn't look likely to work. Consider how magical it would be if you really could solve NP-complete problems in polynomial time. Almost any question in mathematics could be solved in minutes. Do you want a proof of the Riemann hypothesis? Chances are that if a proof exists it's reasonably small -- say, less than
1000 pages. (If it's much bigger than that no human will be able to understand it anyway.) Any such proof can be expressed in a formal language and verified in polynomial time by a conventional computer. So, just have your NP machine guess the proof and verify it.
Finally, keep in mind that this machine has only 16 qubits. A problem with just 16 qubits can be solved by
brute force on your laptop computer. Any dreams about solving monster traveling saleman problems will have to be defered until they have at least a few thousand qubits.
In short, what they have may be a useful next step in quantum computing but it sounds seriously over hyped.
Nvidia provides Linux binary drivers for their graphics chips. So Nvidia is already incurring the substantial costs of writing Linux device drivers. What that company doesn't do is provide a hardware specification of their devices that whould enable other people to write open source drivers. Common sense tells me this would be the cheaper route for Nvidia to take, but then I am not privy to whatever restrictive licensing arrangements Nvidia may have made with third parties.
Surely ATI and Nvidia have been aware for years that Linux hackers just want the specs to their cards, and are quite willing to do the programming themselves. For whatever reasons they have, they don't want to
release that information, and I don't think this announcement is going to change that.
With scientific journals the authors don't get paid (often they are asked to cough up fees to the journal for publishing it). The referees don't get paid either. The only one who gets paid is the publisher. So there's no reason to believe that moving on to a free, online publication would cause any loss in the quality of the submissions or of the referee process.
And that bit about how free publications will be censored by the government -- is anyone stupid enough to fall for that? If the government wants to censor something (say, for national security reasons, or
because Bush, Cheney, & Michael Crichton disagree with the results) it will be censored just as easily in a for profit journal.
First of all, that's "up to 101 years". The perp is still scheduled for his sentencing.
The actual sentence may be substantially less than that.
The perp used a phishing scheme, so he didn't merely annoy people with ads for V1@gr@ but
stole credit cards and robbed people.
Finally, phishers and spammers take hours out of the lives of millions of people, and
cost them many millions of dollars, so this is a theft of major proportions. The fact
that the perp only got to spend a tiny fraction of that himself is beside the point.
If you are robbed of $10,000, and then the thief burns $9900 of that, is it better because
he only gets to use $100?
Me, I'm delighted to see a phisher get the kind of sentence that will make other spammers and
phishers feel genuine fear. They should not think that they can make tons of money, hide it
somewhere, do 4 years in prison, and then come out and live a life of luxury.
Wrong. Radians per second, not cycles per second, is the appropriate measure.
To put it another way, the acceleration is v^2 / r, where v is the velocity at the circumference.
But v = circumferance * cycles per second = 2 pi r * cycles per second = r * radians per second.
So acceleration = r * omega^2 where omega is radians per second.
Actually, that review is dated October 15, 2004 and is for the 10k Savvio.
So I don't think it tells us anything about the noise level of the 15k Savvio.
At the rim of the platter there's nearly 8,000 g's of centripedal acceleration.
I wonder if the engineers have to take into account the gradual stretching out of the
disk.
The math: 1.25" radius =.03175 m.
15,000 rpm = 250 cps = 1,570.796 radians / sec.
acceleration = r * omega^2 =.03175 * (1570.796)^2 = 78,339.98 m / s^2.
1 g = 9.8 m/s^2, so acceleration = 7994 g's.
I tend to deplore DRM. But I also agree that GPLv3 won't stop it. The value of the GPL codebase above BSD and above the cost of proprietary code just isn't that great: neoTivo would just go BSD if not MS-proprietary.
I don't know where you get the strange idea that it is the intent of GPLv3 to stop all DRM.
The purpose of GPLv3 to prevent the use of DRM that subverts the intention of GPLv2.
For example, you won't be able to use DRM to prevent people from modifying the version of
gcc that you gave them.
GPLv3 has nothing directly to do with DRMed music. However, if music DRM was enforced in GPLv3ed
software you would have both the right and the ability to modify the code so as to subvert the DRM.
For that reason we can reasonably expect that companies that want to enforce DRM on music and videos
won't use GPLv3ed code to do it. And yes, neoTivo would use BSD. So? The writers of BSD don't
care whether recipients of the OS have the right to modify the code (otherwise they would have used
a license like the GPL) but by the writers of GPLed software do care (otherwise they would
have used a different license). Linus strikes me as a somewhat inconsistent person -- he released
his kernel under the GPL but doesn't really care if people subvert its intent. I wonder why he bothered.
Every provider here in the Netherlands has a support page that tells you how to set your GPRS and other data settings incase you didn't get your phone from them. What kind of retarded world are the providers they mention in TFA from?!
Try Exaile!!! It's a clone of Amarok that runs under Gnome. It's brand new, but it runs GREAT already. Very stable, nice UI, pretty fast. I wholeheartedly recommend it... I'd say it's the answer to the prayers of the Gnome users who kept the KDE libs around just to run amaroK.
Of course this just proves the parent poster's point. You either have to accept incompatible look and
feel across apps, and install multiple libraries that do pretty much the same thing, or you have to duplicate every app
across both Gnome and KDE. Linux is a mess, and that's a direct consequence of having no entity like Apple with its
Mac OS X in a position to lay down the law and set some real standards.
Relativity is "just a theory" as well, but your GPS system would fail totally if both the special
and general theories of relativity were not taken into account. Too many laymen think that
"theory" means "tentative hypothesis" when in fact many theories are about as well established
as any claim about the physical world could possibly be.
Just a wild guess - a complex number with floating points? (a+b*i where a and b are non-integer)
You guessed wrong. D has both complex reals and imaginary reals, the latter being complex numbers in which the real part is 0. They are distinct types. So far as I can see, this was so that operations like
z * 3i could be done with guaranteed two multiplications, rather than four multiplications and two additions as might happen when a dumb compiler sees z * Complex(0, 3). IMHO, it would have been better
to have just provided some guarantees that the compiler would be smart enough to detect the obvious cases where you have a pure imaginary number. I consider it to be extremely rude to have a numeric type that is not closed under multiplication.
As others have pointed out, there are no precious metals in our pennies and nickels. There are zinc,
copper, and nickel in them, all base metals. But the commodities boom caused by massive new demand in Asia combined with the plummeting dollar now means that even base metals are starting to look valuable. Of course the real causes of the declining dollar are
our massive trade and budget deficits, about which Congress will do nothing at all. No politician gets reelected by saying he's going to raise taxes and cut your benefits. It is always much more expedient to rob people's savings and earnings by allowing inflation to happen. After all, if oil costs more in dollars you can always blame greedy oil companies and Middle Easterners. So Congress will "solve" this problem by reintroducing steel pennies.
Yes, I recall that back in my youth I tried generating hydrogen via electrolysis of water with some table salt thrown in so the stuff would conduct. I got hydrogen, all right. But at the other end I got chlorine gas. I don't recommend that this process be scaled up, unless you want to re-enact World War I.
Makes me feel bad about the tuna sandwiches I had for dinner last night.
While many ocean dolphins do get killed by tuna nets, the species that went extinct was a river dolphin,
unique to the Yangtze. They were done in by the increasing pollution of that river. So instead of feeling bad about the tuna sandwiches you had you should feel bad about the cheap DVD player you bought -- not only did the people who put it together get paid slave wages, but the company that employed them didn't "waste" any money on pollution control.
The article notes that Mr. Phipps' comments are somewhat surprising, given that the recent open-sourcing of Java went forward with GPLv2
There is nothing surprising about this. GPL v3 in final, legally binding form doesn't exist yet, so
of course any GPLed software released now will use GPL v2. It will only be surprising if future releases
of Java don't use GPL v3 after it is finalized.
It's to the point where any left leaning pot shot gets +5 interesting?
Well, my parent post got +1 interesting, +1 insightful, and -2 flame bait. So now I can die
happy.
Just for the record, I am a pro-environment paleocon. I see nothing conservative about starting an unnecessary war. Foolish, yes. Criminal, yes. Conservative, no. I am puzzled by those who think that George Bush is a conservative, although I can fully understand why the liberals don't want to claim him.
Heck, just replace the President and Vice President with an abacus and there will be a marked improvement in performance. When was the last time an abacus started a war just for fun?
Slashdot Mirror
One problem with a demo of this sort is that it isn't possible to know how much was actually done by the quantum computer and how much was done by the classical front end computer. Typical Sudoku puzzles can be solved essentially instantaneously by classical computers. 16 qubits isn't a heck of a lot of quantum parellelism, so I do wonder how much of the real work they were doing.
They make it sound like they have built a device that will solve an arbitrary instance of a particular NP-complete problem (the two dimensional Ising model in a magnetic field). Indeed, they argue that they will be able to solve any problem in NP by using a conventional computer to convert that problem to the one they can solve. It would be a major breakthrough to show that any NP-complete problem can be solved in polynomial time on a quantum computer, but I don't think they've done it. (Factoring can be done in polynomial time on a quantum computer, but is not generally believed to be NP-complete.)
From their paper, "The third, and perhaps most interesting, time-complexity result-- the possibility of exponential speedup on almost all difficult instances of NP-hard problems--comes from explicit numerical integrations of Schroedinger's equation to simulate adiabatic computation on sets of small instances of these problems..."
What they have is not a mathematical proof but some hand waving extrapolation based on small instances. Most likely what they have is a faster way to do simulated annealing. If so, it will give good, but not necessarily optimal, approximations to certain NP-complete problems. But approximations don't in general transfer across the polynomial time reductions used to convert one NP-complete problem to another. So their idea of having a front end computer that transforms an arbitrary problem to their Ising model doesn't look likely to work. Consider how magical it would be if you really could solve NP-complete problems in polynomial time. Almost any question in mathematics could be solved in minutes. Do you want a proof of the Riemann hypothesis? Chances are that if a proof exists it's reasonably small -- say, less than 1000 pages. (If it's much bigger than that no human will be able to understand it anyway.) Any such proof can be expressed in a formal language and verified in polynomial time by a conventional computer. So, just have your NP machine guess the proof and verify it.
Finally, keep in mind that this machine has only 16 qubits. A problem with just 16 qubits can be solved by brute force on your laptop computer. Any dreams about solving monster traveling saleman problems will have to be defered until they have at least a few thousand qubits.
In short, what they have may be a useful next step in quantum computing but it sounds seriously over hyped.
Nvidia provides Linux binary drivers for their graphics chips. So Nvidia is already incurring the substantial costs of writing Linux device drivers. What that company doesn't do is provide a hardware specification of their devices that whould enable other people to write open source drivers. Common sense tells me this would be the cheaper route for Nvidia to take, but then I am not privy to whatever restrictive licensing arrangements Nvidia may have made with third parties.
Surely ATI and Nvidia have been aware for years that Linux hackers just want the specs to their cards, and are quite willing to do the programming themselves. For whatever reasons they have, they don't want to release that information, and I don't think this announcement is going to change that.
If you care about privacy, your Constitutional rights, getting out of Iraq, and fiscal responsibility you have only one choice: Ron Paul
As for Hillary, she'll take a different position next week, depending upon which way the wind is blowing.
With scientific journals the authors don't get paid (often they are asked to cough up fees to the journal for publishing it). The referees don't get paid either. The only one who gets paid is the publisher. So there's no reason to believe that moving on to a free, online publication would cause any loss in the quality of the submissions or of the referee process.
And that bit about how free publications will be censored by the government -- is anyone stupid enough to fall for that? If the government wants to censor something (say, for national security reasons, or because Bush, Cheney, & Michael Crichton disagree with the results) it will be censored just as easily in a for profit journal.
From the article: You lose the normal use of the Caps Lock key.
That is easily the best feature of the entire program.
First of all, that's "up to 101 years". The perp is still scheduled for his sentencing. The actual sentence may be substantially less than that.
The perp used a phishing scheme, so he didn't merely annoy people with ads for V1@gr@ but stole credit cards and robbed people.
Finally, phishers and spammers take hours out of the lives of millions of people, and cost them many millions of dollars, so this is a theft of major proportions. The fact that the perp only got to spend a tiny fraction of that himself is beside the point. If you are robbed of $10,000, and then the thief burns $9900 of that, is it better because he only gets to use $100?
Me, I'm delighted to see a phisher get the kind of sentence that will make other spammers and phishers feel genuine fear. They should not think that they can make tons of money, hide it somewhere, do 4 years in prison, and then come out and live a life of luxury.
Wrong. Radians per second, not cycles per second, is the appropriate measure.
To put it another way, the acceleration is v^2 / r, where v is the velocity at the circumference. But v = circumferance * cycles per second = 2 pi r * cycles per second = r * radians per second. So acceleration = r * omega^2 where omega is radians per second.
Actually, that review is dated October 15, 2004 and is for the 10k Savvio. So I don't think it tells us anything about the noise level of the 15k Savvio.
At the rim of the platter there's nearly 8,000 g's of centripedal acceleration. I wonder if the engineers have to take into account the gradual stretching out of the disk.
The math: 1.25" radius = .03175 m. .03175 * (1570.796)^2 = 78,339.98 m / s^2.
15,000 rpm = 250 cps = 1,570.796 radians / sec.
acceleration = r * omega^2 =
1 g = 9.8 m/s^2, so acceleration = 7994 g's.
I don't know where you get the strange idea that it is the intent of GPLv3 to stop all DRM. The purpose of GPLv3 to prevent the use of DRM that subverts the intention of GPLv2. For example, you won't be able to use DRM to prevent people from modifying the version of gcc that you gave them.
GPLv3 has nothing directly to do with DRMed music. However, if music DRM was enforced in GPLv3ed software you would have both the right and the ability to modify the code so as to subvert the DRM. For that reason we can reasonably expect that companies that want to enforce DRM on music and videos won't use GPLv3ed code to do it. And yes, neoTivo would use BSD. So? The writers of BSD don't care whether recipients of the OS have the right to modify the code (otherwise they would have used a license like the GPL) but by the writers of GPLed software do care (otherwise they would have used a different license). Linus strikes me as a somewhat inconsistent person -- he released his kernel under the GPL but doesn't really care if people subvert its intent. I wonder why he bothered.
No troll or flamebait points. So give me my $5.
The United States.
Of course this just proves the parent poster's point. You either have to accept incompatible look and feel across apps, and install multiple libraries that do pretty much the same thing, or you have to duplicate every app across both Gnome and KDE. Linux is a mess, and that's a direct consequence of having no entity like Apple with its Mac OS X in a position to lay down the law and set some real standards.
Relativity is "just a theory" as well, but your GPS system would fail totally if both the special and general theories of relativity were not taken into account. Too many laymen think that "theory" means "tentative hypothesis" when in fact many theories are about as well established as any claim about the physical world could possibly be.
I'm the guy with 6 trillion dollars and the rest of you are my serfs and thralls.
MWA HA HA HA!!
You guessed wrong. D has both complex reals and imaginary reals, the latter being complex numbers in which the real part is 0. They are distinct types. So far as I can see, this was so that operations like z * 3i could be done with guaranteed two multiplications, rather than four multiplications and two additions as might happen when a dumb compiler sees z * Complex(0, 3). IMHO, it would have been better to have just provided some guarantees that the compiler would be smart enough to detect the obvious cases where you have a pure imaginary number. I consider it to be extremely rude to have a numeric type that is not closed under multiplication.
ban myspace?
As others have pointed out, there are no precious metals in our pennies and nickels. There are zinc, copper, and nickel in them, all base metals. But the commodities boom caused by massive new demand in Asia combined with the plummeting dollar now means that even base metals are starting to look valuable. Of course the real causes of the declining dollar are our massive trade and budget deficits, about which Congress will do nothing at all. No politician gets reelected by saying he's going to raise taxes and cut your benefits. It is always much more expedient to rob people's savings and earnings by allowing inflation to happen. After all, if oil costs more in dollars you can always blame greedy oil companies and Middle Easterners. So Congress will "solve" this problem by reintroducing steel pennies.
Yes, I recall that back in my youth I tried generating hydrogen via electrolysis of water with some table salt thrown in so the stuff would conduct. I got hydrogen, all right. But at the other end I got chlorine gas. I don't recommend that this process be scaled up, unless you want to re-enact World War I.
Makes me feel bad about the tuna sandwiches I had for dinner last night.
While many ocean dolphins do get killed by tuna nets, the species that went extinct was a river dolphin, unique to the Yangtze. They were done in by the increasing pollution of that river. So instead of feeling bad about the tuna sandwiches you had you should feel bad about the cheap DVD player you bought -- not only did the people who put it together get paid slave wages, but the company that employed them didn't "waste" any money on pollution control.
The article notes that Mr. Phipps' comments are somewhat surprising, given that the recent open-sourcing of Java went forward with GPLv2
There is nothing surprising about this. GPL v3 in final, legally binding form doesn't exist yet, so of course any GPLed software released now will use GPL v2. It will only be surprising if future releases of Java don't use GPL v3 after it is finalized.
It's to the point where any left leaning pot shot gets +5 interesting?
Well, my parent post got +1 interesting, +1 insightful, and -2 flame bait. So now I can die happy.
Just for the record, I am a pro-environment paleocon. I see nothing conservative about starting an unnecessary war. Foolish, yes. Criminal, yes. Conservative, no. I am puzzled by those who think that George Bush is a conservative, although I can fully understand why the liberals don't want to claim him.
Heck, just replace the President and Vice President with an abacus and there will be a marked improvement in performance. When was the last time an abacus started a war just for fun?