A 1% pregnancy rate over two and a half years actually sounds very effective. I don't know the rates for other protection methods, or even unprotected, but I know they're not as good as 99% (in practice) over 2.5 years.
But 1/3 of the sample dropping out is not very promising. Side effects? Cherry picking? Guess we'll find out later.
Even if they use physical properties of the chip you can still clone it. Your clone will simply contain a digital copy of the physical information, instead of actually measuring it every time like the original chip.
That's actually not true, although many cons will no longer work. Lots of people with cancer spend thousands of dollars on cures that don't work [the seller doesn't always know this].
They showed, given very few assumptions [read the paper], that if humans are unpredictable then particles are unpredictable. A pretty basic logical equivalence: A->B === ~B->~A Therefore if particles are entirely predictable, humans are entirely predictable
Right, because no physicist could have ever though of that. "It's there but we just can't measure it" theories, aka "Hidden Variable Theories" aren't compatible with the predictions of QM, unless you throw in some other weird things like non-locality.
There's a demand for credit cards, people aren't going to "just use cash". Not to mention cash doesn't work nearly as well as credit over the internet. The idea is to fix the system, not throw it out.
I'm going to go out on a limb and say the core of the problem isn't the security of the computers, it's the fact that in order to use a credit card number you have to reveal it. There will always be some retailer or customer without a secure system. _We can't change this, it's too hard_.
I think the solution is a small device with an embedded secret key. All it has to do is sign data [secondary: show text, wireless, usb, etc].
For example, to complete a transaction, a store asks you to sign this: [
VISA Credit Transfer
"here's a one-line ad because we just can't help it!"
amount: 12.34$us
buyer: John Doe
seller: Matt's Grocery Store
date: August 7, 2008
buyer public key: 09 f9 11 02 9d 74 e3 5b d8 41 56 c5 63 56 88 c0
seller public key: 4B 3D BA 71 3B D8 56 43 2B A7 E8 F4 69 CA C5 5A
seller transaction id: 594864purplebunnies
protocol version: 1 ] Then the store also signs it, and sends it and the signatures to VISA, or whoever.
The beauty here is that the security is now entirely encapsulated in a) the signing device, and b) the plaintext format for requesting credit.
In the example I have given the buyer only has to check that the amount is correct because all other modifications give them free groceries. The store only needs to ensure they match the format specified by VISA, and that the buyer's signature is valid. VISA takes most of the work, checking that the format is correct, the signatures are valid, the transaction id is unique for the seller, the buyer has enough credit, etc.
I'm sure there are holes, but it's a hell of a lot better than what we have now.
A secure voting system, in the sense that every person can keep their vote a secret AND check that their vote has not been tampered with, requires every person to have a public/private key pair. It will be years, maybe decades, maybe never, before we start getting those kind of IDs.
While you're at it, complain about university students not making their books less expensive. Your beef is with publishers. They aren't the entire scientific community.
That's actually not true. 'Optimal' play means you are at least as good as every player, it doesn't mean you are the best against every player. You can sometimes do better against poor players by making moves that good players would counter.
A computer with a perfect model of their (non-optimal) opponent will do better in the long run than a computer just playing the probabilities.
The same reason insulation must display a resistivity value: so the customer knows what the hell they're buying. It might not be illegal to sell mature games to young kids, but it's illegal to put a teen rating on those games (or it should be!).
Spontaneous generation is a type of abiogenesis, but abiogenesis is not spontaneous generation. Abiogenesis refers to any theory about how life arose from non-life. Lookup abiogenesis on wikipedia for more.
It's becoming more commonplace for "creationist" to by default mean "young earth creationist", and that is what the GP was most likely referring to (definitely seems that way based on the context).
Strictly speaking, you're correct because "God created the universe" and "the theory of evolution is true" can both be true.
Math does not go together with multiple choice tests. Many math problems are hard to solve but easy to check, meaning it can be much faster (and easier) to just run each answer backwards than solving the original problem. As a simple example: suppose you're asked to integrate sin(x)*cos(x). Since differentiation is easy, you just differentiate all the answers and see that sin(x)^2/2 differentiates to sin(x)*cos(x). You could do an entire test on indefinite integration and *never even perform integration*. (Another example: prove X vs which proof proves X)
Of course some math problems don't have that 'easy to check' property, and they are more appropriate for multiple choice tests. Definite integration is an example of such a problem. But even then multiple choice tests are easier, because you can catch your own errors by comparing against the possible answers (not to mention the non-negligible chance of just guessing the correct answer).
Well, assuming you can't extract the private key from the chip, and assuming you can't get one of the keys used to make authorized private keys from the manufacturer: you can still crack the games.
At some point either the game or the hardware needs to check the TPM. Guess what: we can modify the game to not perform the check, and we can emulate the hardware except without the check. Emulating the TPM isn't even necessary.
Slashdot Mirror
Excellent post. Well written, insightful, and even a bit funny.
A 1% pregnancy rate over two and a half years actually sounds very effective. I don't know the rates for other protection methods, or even unprotected, but I know they're not as good as 99% (in practice) over 2.5 years.
But 1/3 of the sample dropping out is not very promising. Side effects? Cherry picking? Guess we'll find out later.
If computers get better at composing music (and they will), we should eventually see websites that stream newly composed music 24/7.
Instead of selling songs you sell composition.
Experts exchange puts all the responses at the bottom of the page, so they still come up in searches, but you're not likely to see them.
Get the aardvark and remove-it-permanently extensions for firefox and RIP all the clutter in the middle and it'll look just like any other forum.
You're assuming you're not going to have to share the prize.
Even if they use physical properties of the chip you can still clone it. Your clone will simply contain a digital copy of the physical information, instead of actually measuring it every time like the original chip.
We already have an algorithm that solves NP problems in P time, but only if P = NP.
http://en.wikipedia.org/wiki/P_%3D_NP#Polynomial-time_algorithms
That's actually not true, although many cons will no longer work. Lots of people with cancer spend thousands of dollars on cures that don't work [the seller doesn't always know this].
How is it unjustifiable?
They showed, given very few assumptions [read the paper], that if humans are unpredictable then particles are unpredictable.
A pretty basic logical equivalence: A->B === ~B->~A
Therefore if particles are entirely predictable, humans are entirely predictable
Right, because no physicist could have ever though of that. "It's there but we just can't measure it" theories, aka "Hidden Variable Theories" aren't compatible with the predictions of QM, unless you throw in some other weird things like non-locality.
http://en.wikipedia.org/wiki/Hidden_variable_theory
There's a demand for credit cards, people aren't going to "just use cash". Not to mention cash doesn't work nearly as well as credit over the internet. The idea is to fix the system, not throw it out.
I'm going to go out on a limb and say the core of the problem isn't the security of the computers, it's the fact that in order to use a credit card number you have to reveal it. There will always be some retailer or customer without a secure system. _We can't change this, it's too hard_.
I think the solution is a small device with an embedded secret key. All it has to do is sign data [secondary: show text, wireless, usb, etc].
For example, to complete a transaction, a store asks you to sign this:
[
VISA Credit Transfer
"here's a one-line ad because we just can't help it!"
amount: 12.34$us
buyer: John Doe
seller: Matt's Grocery Store
date: August 7, 2008
buyer public key: 09 f9 11 02 9d 74 e3 5b d8 41 56 c5 63 56 88 c0
seller public key: 4B 3D BA 71 3B D8 56 43 2B A7 E8 F4 69 CA C5 5A
seller transaction id: 594864purplebunnies
protocol version: 1
]
Then the store also signs it, and sends it and the signatures to VISA, or whoever.
The beauty here is that the security is now entirely encapsulated in a) the signing device, and b) the plaintext format for requesting credit.
In the example I have given the buyer only has to check that the amount is correct because all other modifications give them free groceries. The store only needs to ensure they match the format specified by VISA, and that the buyer's signature is valid. VISA takes most of the work, checking that the format is correct, the signatures are valid, the transaction id is unique for the seller, the buyer has enough credit, etc.
I'm sure there are holes, but it's a hell of a lot better than what we have now.
BSD is about giving to the world.
GPL is about changing the world.
A secure voting system, in the sense that every person can keep their vote a secret AND check that their vote has not been tampered with, requires every person to have a public/private key pair. It will be years, maybe decades, maybe never, before we start getting those kind of IDs.
While you're at it, complain about university students not making their books less expensive. Your beef is with publishers. They aren't the entire scientific community.
Unless it says the number is prime (you have to trust there are no factors) or gives factors that aren't primes.
That's actually not true. 'Optimal' play means you are at least as good as every player, it doesn't mean you are the best against every player. You can sometimes do better against poor players by making moves that good players would counter.
A computer with a perfect model of their (non-optimal) opponent will do better in the long run than a computer just playing the probabilities.
The same reason insulation must display a resistivity value: so the customer knows what the hell they're buying. It might not be illegal to sell mature games to young kids, but it's illegal to put a teen rating on those games (or it should be!).
It has to not look suspicious. Spaghetti code _always_ looks suspicious.
I suggest a BIGGER PLANE to surround the smaller one with.
Of course then we'll need bigger bigger plane in case they get a bigger plane.
Spontaneous generation is a type of abiogenesis, but abiogenesis is not spontaneous generation. Abiogenesis refers to any theory about how life arose from non-life. Lookup abiogenesis on wikipedia for more.
It's becoming more commonplace for "creationist" to by default mean "young earth creationist", and that is what the GP was most likely referring to (definitely seems that way based on the context).
Strictly speaking, you're correct because "God created the universe" and "the theory of evolution is true" can both be true.
Math does not go together with multiple choice tests. Many math problems are hard to solve but easy to check, meaning it can be much faster (and easier) to just run each answer backwards than solving the original problem. As a simple example: suppose you're asked to integrate sin(x)*cos(x). Since differentiation is easy, you just differentiate all the answers and see that sin(x)^2/2 differentiates to sin(x)*cos(x). You could do an entire test on indefinite integration and *never even perform integration*. (Another example: prove X vs which proof proves X)
Of course some math problems don't have that 'easy to check' property, and they are more appropriate for multiple choice tests. Definite integration is an example of such a problem. But even then multiple choice tests are easier, because you can catch your own errors by comparing against the possible answers (not to mention the non-negligible chance of just guessing the correct answer).
I expect that even if it did work, you would end up increasing the rolling friction of the tire. That would cut into gas mileage.
Well, assuming you can't extract the private key from the chip, and assuming you can't get one of the keys used to make authorized private keys from the manufacturer: you can still crack the games.
At some point either the game or the hardware needs to check the TPM. Guess what: we can modify the game to not perform the check, and we can emulate the hardware except without the check. Emulating the TPM isn't even necessary.