According to Schneier, even quantum computing won't be able to brute force 256 bits in any meaningful time. I don't know enough about it to have what could be called an informed opinion, but I suspect that the kind of algorithmic breakthrough that would allow QC to do much better would have to be accompanied by a breakthrough in thermodynamics.
yeah, but ten years ago we were trying to hold down processing time for legitimate work. At this point the key lengths that can be handled pretty easily are unbreakable not because the cpus aren't fast enough but because thermodynamics says they can't become fast enough.- https://www.schneier.com/blog/...
Govt: "Unlock this, we have a writ" Victim: I can't Govt: Precedent says you have to Victim: I really can't Govt: welcome to contempt of court Victim: But it's just flat out not possible. It's not because I don't want to, I _can't_. Govt: *snore*
If you don't include time and processing power, then the only answer is "doesn't work". That said, your point that there's a difference between "holds up for a week" vs "holds up for my lifetime" vs "holds up for the planet's lifetime" vs "holds up for Keith Richards' lifetime" is valid, but given how common and workable "holds up for the planet's lifetime" is, it doesn't seem unreasonable to use that as a baseline and fall back to the boolean "it works (for effectively ever)" or "it doesn't work (for effectively ever)" for most cases (there's stuff without enough cpu to handle that baseline, certainly, but these days it's not usually something that's used for emails and web browsing, it's something being used to open your garage).
from what I've read the FBI prefers the latter but would accept the former. However, Cook has said that law enforcement around the country has already said they have hundreds of iPhones they want appel to unlock if the FBI wins; if that's so, I don't think destroying the tool is going to be a viable option.
which does them no good whatsoever without the particular key storage chip from that phone. Otherwise they'd just clone it a hundred times and throw 10 attempts at each clone.
It probably isn't... legally. The print is probably still legally the property of 20th Century Fox (or whatever that company's name is now), just tagged as "lost" in the records, and I recall hearing that "restoration" counts as making a new work, which would be derivative, and therefore infringing...
he has stated that his sample (admittedly limited) shows one result. You have asserted that a larger sample shows a different result. Both can be true; it simply points up that his sample is limited and apparently not representative of the whole population.
As I recall, the key factor of that decision was that IR scanning was not widely available to the public, and the idea was that the cops shouldn't have too much more capability than normal people without going through due process. http://abcnews.go.com/US/story...
https://www.law.cornell.edu/an... indicates that the protection is against testimony. I assumed (but did not clearly indicate) that they already know you have a diary in a lockbox, and in fact where the lockbox is, at which point what they need is a warrant, not your testimony. If they don't know about it, then you telling them about it would be testimony.
That said, the phrasing I originally used was based on http://lawcomic.net/guide/?p=2... and the discussion on previous pages, and could probably be expressed more precisely.
Doctrine about the 5th amendment is that the government is not allowed to force you to create evidence (by speaking or writing in answer to a question). It is not considered to protect already existing evidence (your diary in a lockbox). There's been a kerfuffle lately about whether requiring you to decrypt some already existing piece of data constitutes creating evidence or merely accessing existing evidence. However it is not inconceivable that an actually accurate mind reader could be considered "accessing existing evidence", just like reading your fingerprints, dna, or blood content.
I don't see what you're accepting. If the AI is at fault, Google is paying. If the AI is not at fault, and you're not at fault, the other person's insurance is paying. It doesn't look like you're on the hook unless you personally are at fault, which is as it should be.
lack of seed money, most likely. After all, 3 of those 4 got rich doing something other than managing their portfolio, and Warren Buffett is at the very high end of adviser skill (based on performance) and still took a long time to get to where he is.
yep. But unless you can point to a name for the relationship based on a quadratic equation (I haven't found one), I figure following the trend of "the relationship is named after the equation" is the way to go, because "exponential relationship" has a specific definition, which is that the variable is in the exponent, not the base.
Slashdot Mirror
Clothing. Particularly warm or loose clothing. And macaroni and cheese. That's even known as "comfort food"!
also to make it harder for them to maneuver by weighting their shields.
According to Schneier, even quantum computing won't be able to brute force 256 bits in any meaningful time. I don't know enough about it to have what could be called an informed opinion, but I suspect that the kind of algorithmic breakthrough that would allow QC to do much better would have to be accompanied by a breakthrough in thermodynamics.
yeah, but ten years ago we were trying to hold down processing time for legitimate work. At this point the key lengths that can be handled pretty easily are unbreakable not because the cpus aren't fast enough but because thermodynamics says they can't become fast enough.- https://www.schneier.com/blog/...
Govt: "Unlock this, we have a writ"
Victim: I can't
Govt: Precedent says you have to
Victim: I really can't
Govt: welcome to contempt of court
Victim: But it's just flat out not possible. It's not because I don't want to, I _can't_.
Govt: *snore*
If you don't include time and processing power, then the only answer is "doesn't work". That said, your point that there's a difference between "holds up for a week" vs "holds up for my lifetime" vs "holds up for the planet's lifetime" vs "holds up for Keith Richards' lifetime" is valid, but given how common and workable "holds up for the planet's lifetime" is, it doesn't seem unreasonable to use that as a baseline and fall back to the boolean "it works (for effectively ever)" or "it doesn't work (for effectively ever)" for most cases (there's stuff without enough cpu to handle that baseline, certainly, but these days it's not usually something that's used for emails and web browsing, it's something being used to open your garage).
... in the sixth circuit.
from what I've read the FBI prefers the latter but would accept the former. However, Cook has said that law enforcement around the country has already said they have hundreds of iPhones they want appel to unlock if the FBI wins; if that's so, I don't think destroying the tool is going to be a viable option.
which does them no good whatsoever without the particular key storage chip from that phone. Otherwise they'd just clone it a hundred times and throw 10 attempts at each clone.
Part of the process of going from PIN to decryption key is in hardware, and they only have one of that chip.
... in the voice of Patrick Stewart.
It probably isn't... legally. The print is probably still legally the property of 20th Century Fox (or whatever that company's name is now), just tagged as "lost" in the records, and I recall hearing that "restoration" counts as making a new work, which would be derivative, and therefore infringing...
That's the only argument Cox has standing to make - they have to show that _they_ are being harmed; otherwise their opinion is legally irrelevant.
he has stated that his sample (admittedly limited) shows one result. You have asserted that a larger sample shows a different result. Both can be true; it simply points up that his sample is limited and apparently not representative of the whole population.
Your statement really only challenges the "[that] I know" qualifier.
since it makes no difference whatsoever to my behavior, I shan't bother.
yep. Wish we had one.
As I recall, the key factor of that decision was that IR scanning was not widely available to the public, and the idea was that the cops shouldn't have too much more capability than normal people without going through due process. http://abcnews.go.com/US/story...
https://www.law.cornell.edu/an... indicates that the protection is against testimony. I assumed (but did not clearly indicate) that they already know you have a diary in a lockbox, and in fact where the lockbox is, at which point what they need is a warrant, not your testimony. If they don't know about it, then you telling them about it would be testimony.
That said, the phrasing I originally used was based on http://lawcomic.net/guide/?p=2... and the discussion on previous pages, and could probably be expressed more precisely.
Doctrine about the 5th amendment is that the government is not allowed to force you to create evidence (by speaking or writing in answer to a question). It is not considered to protect already existing evidence (your diary in a lockbox). There's been a kerfuffle lately about whether requiring you to decrypt some already existing piece of data constitutes creating evidence or merely accessing existing evidence. However it is not inconceivable that an actually accurate mind reader could be considered "accessing existing evidence", just like reading your fingerprints, dna, or blood content.
I don't see what you're accepting. If the AI is at fault, Google is paying. If the AI is not at fault, and you're not at fault, the other person's insurance is paying. It doesn't look like you're on the hook unless you personally are at fault, which is as it should be.
lack of seed money, most likely. After all, 3 of those 4 got rich doing something other than managing their portfolio, and Warren Buffett is at the very high end of adviser skill (based on performance) and still took a long time to get to where he is.
yep. But unless you can point to a name for the relationship based on a quadratic equation (I haven't found one), I figure following the trend of "the relationship is named after the equation" is the way to go, because "exponential relationship" has a specific definition, which is that the variable is in the exponent, not the base.
I think technically it ought to be called "quadratic", but not many would understand that.
and particularly considering that this was just selecting delegates to go to the county-level caucus.