You're the one trying to convince me that his argument makes sense:-)
No, I wasn't. I merely defended my description of it as a "Pascal's wager".
So I see, after reviewing the thread.
I do take issue with your characterization of Pascal's wager as a form of innumeracy. I would agree that the assumption that life extension has infinite value is a fallacy of some sort, which might be characterized as innumeracy, if you squint.
> At least, they do if they're actively being attacked. If you assume that someone is trying to brute force your password, even a fairly long, complex password is only likely to stand up for a few months.
A fairly long, complex password is likely to stand up for millennia against brute force.
Wishful thinking.
At least by most people's definition of "fairly long, complex" -- but still reasonable to type and to remember -- password cracking is eminently accessible, though not (yet) cheap.
A ten-character password, containing a completely random selection of alphabetic, numeric and symbolic characters has about 61 bits of entropy. That's already beyond what most people are prepared to deal with, so consider this calculation an upper bound and reduce it by two or three orders of magnitude (minimum!) for the average real-world password.
According to this article an Amazon EC2 instance with GPU-based cracking can test 3.488 billion passwords per second. At that rate, it would take just short of 300,000 hours to search the entire password space, about 34 years. That's not trivial, but it's hardly "millenia". And, of course, password cracking scales perfectly, so you can use 34 times the resources to do it in one year, or 408 times the resources to do it in one month, or 300,000 times the resources to do it in one hour.
At the rate mentioned in the article, $2.10 per hour, it would cost ~$313,000, on average, to crack a password. That's substantial, but assuming it declines per Moore's Law (which wasn't about $/cycles, but close enough), in 10 years it'll cost just over $3K, in 15 years it'll cost about $300, and in 20 years it'll cost about $30.
Of course, good systems can make the attack more expensive by iterating the hashing operation to increase the cost of each password tested. But, still, the point is that the most complex passwords that people can readily handle are within the reach of a serious attacker, and this situation is just going to get worse.
Google has taken a step forward with their two factor options (via say, a cell text) but that's not really a practical option for many small web sites.
Small web sites can just use Google's OpenID for authentication, and tell their users to turn on two-factor on their Google accounts.
Alternatively, if your userbase has smartphones, you can use Google Authenticator to allow your users to use one-time passwords, without any involvement from Google. And it's not all that difficult to implement SMS-based second-factor auth yourself either. There are libraries for all of the major languages.
This is an excellent case for a PKI. Users should generate a public-private key pair, and provide the public key to the web site upon sign up. Extra authentication steps could be done at setup (web of trust a la PGP, known entities, a la X.509, callback texts, whatever). Users would sign a login blob with their private key to authenticate.
That would work, but I think OpenID is a simpler and more flexible solution. It just lets sites all over the web take advantage of the authentication solutions already built by Google, Yahoo, Facebook, Blizzard, etc. And people who don't trust those can always build their own authentication servers, or use one they do trust.
Changing a secure password offers no additional security. Its not like they wear out.
Yes, they do.
At least, they do if they're actively being attacked. If you assume that someone is trying to brute force your password, even a fairly long, complex password is only likely to stand up for a few months. This is one of the rationales for changing passwords periodically.
However, if you're really worried about that, you absolutely should not use the same password for multiple web sites. Because every site you use it with sees the plaintext password every time you log in, even if they store it properly salted and hashed. So it only takes one unscrupulous admin and your "strong" password becomes known. The OP says he adds some site-specific bits to his common password, but unless that's done very well, it adds nothing. And even when done well it doesn't add very much security, if the unscrupulous admin is clever enough to guess that's what's being done.
As an example, look at mailman, the mailing list manager. Not only did it store the plaintext password, it mails it to you monthly.
That's not because the developers of mailman were idiots. It's because they assumed that the users were not idiots, an assumption that was probably reasonable in the early days of mailman. Mailman has always told users that they should not use an important password because it would be e-mailed to them monthly. The idea was that your mailman account is very low-value, and so it made sense to use a weak password, and it made sense to e-mail password reminders because so many users forgot their low-value passwords.
The reason mailman changed was because the developers discovered that users were, in fact, idiots, and commonly used the same password that they used for, say, their on-line banking account, rather than making up a throwaway password which they didn't even have to bother to remember because it would be e-mailed to them monthly. Thus, mailman has to take pains to secure the user's password, not because the mailman account needs protection, but because all of the other accounts that use the same password need protection.
The argument is based on the assumption that there is a possibility of an outcome with infinite value.
Which is a perfectly valid assumption in the case of Pascal's wager, since that's exactly the expected payout under the Christian theology with which Pascal was familiar.
If there is a defect in Pascal's wager, it's in the assumption that an unknowable deity offers a known reward for known actions. Essentially, Pascal's wager takes as a given that if God exists, God is of a certain form. But if you can't know if God exists, how can you know what God is like?
The actual labeling of this outcome as say, "Heaven" rather than say, "live longer" is irrelevant.
No, it's not. Living longer has finite value, because it has a definite and unavoidable end. Heaven is purported to be endless, and therefore can offer infinite value. There's a big distinction there.
Keep in mind that the original AC poster said "What is 5% weighed against infinity?" He really was assigning an infinite value to living longer.
Erroneously assigning an infinite value to living longer.
Yes, as in this specific case one does a Diffie-Hellman key exchange before an attack. But when proofing anything against bruteforce style attacks on assumes that it's going to be an offline attack.
Depends on the context. Structuring the protocol to eliminate off-line attacks and then implementing countermeasures to defeat on-line attacks is a common and perfectly valid strategy. Particularly when you want to have human-usable keys, as in this case.
We assume worse case in cryptography research, not best case and hope someone doesn't work out how to make it offline.
No, we don't assume worst case. The right way to build cryptographic security systems is to define the threat model, identify the avenues of attack and implement necessary threat mitigation countermeasures -- with an appropriate level of conservatism, of course. Always assuming the worst case tends to produce systems which are unusable in practice, and often still insecure if the due diligence wasn't performed on the threat modeling side.
The fundamental structure of WPS is fine, and within that structure there's nothing wrong with an 8-digit PIN. The problem here is that protocol defects effectively reduce that 8-digit PIN to a 4-digit PIN. Even that might be okay, given appropriate countermeasures to slow a brute force search, but those countermeasures weren't implemented.
The average person might enjoy it, but a geek will struggle in the truly claustrophobic walled garden that they have erected. I actually was crying 30 minutes after opening the package, because it was essentially worthless to me. (Thank god for people having their hands on it a month ahead of me, and doing all the work in rooting it.)
I just helped my sister root her Nook Tablet. Later that night, while she slept, it auto-installed an update that un-rooted it, and (I believe) renders it for the moment un-rootable.
Indeed, and it sounds more like a programming flaw than a platform flaw. If you need a hash function with cryptographic properties, don't use MD5. It may not always be obvious, but if you work with unverified user input, chances are that you need some level of cryptographic strength.
This isn't an MD5 issue. I doubt that most of these hash table implementations use any cryptographic hash function (why bother? Cryptographic collision resistance is expensive and not generally relevant to hash table uses), but even if they used iterated SHA-512 interspersed with RIPEMD, TIGER and WHIRLPOOL in a pseudo-random sequence, the implementation would still be vulnerable.
How? Well, all the attacker has to do is to predict with reasonable accuracy the size of the hash table. Since the attacker almost certainly has access to the exact code that is running on the web server, this is trivial. Suppose that the initial hash table size is always 32. This means that the attacker just has to find a whole bunch of names whose hash values are all the same mod 32. Finding full collisions is effectively impossible. Finding values with the same five-bit suffix is trivial. At some point the hash table may grow, and the attacker will have to adjust... but the attacker knows exactly when this is, and exactly how to adjust to it.
A simple countermeasure that comes to my mind is to prepend a fixed, but site-specific string to each name before hashing it. This would randomize the hash values, destroying the property the attacker worked so hard to create. Or even choose a different random prefix when the hash table is constructed, store it with the table, and apply it to each string inserted or looked up before hashing it.
The key is to make the hash output unpredictable to the attacker, so he can't choose inputs that hash to specific values.
There are probably other countermeasures, perhaps simpler ones.
They don't do anything if nobody reports you. They don't actively seek out non-compliant accounts unless you are under 13 and are dumb enough to put in your real age. They don't care.
Is there anything here to suggest that Google does actively seek out non-compliant accounts or address unreported policy violations?
I'll bet in this case that someone flagged the pic as inappropriate and that's the only reason it got looked at.
It is Pascal's wager, which is a form of innumeracy. To argue that even a slight increase in the odds of survival has infinite payout, has some serious implications for our lives.
I don't see how that's really related to Pascal's wager, which is the idea that, assuming it is unknown whether or not God exists (but assuming the Christian God, or similar), it's a good idea to bet that He does... because if you're right you gain eternal rewards and if you're wrong you don't lose much. In Pascal's wager, the successful bet does in fact provide infinite payout. In the case of decisions made to extend life, the payout is guaranteed to be finite.
So your only defense now is that I can't capture a packet and do an offline attack?
Ah, sorry, I missed this part.
Yes, that's exactly the problem. You can't capture a packet and do an offline attack. There is no packet to capture which will enable you to test many values offline. Each attempt to guess the PIN must be an on-line attack, transmitted to the router.
Actually. it is achievable, it's not about speed it's about being extremely parallel, if I try a different key across each 200+ processor cores.
You only have one router to test against. Every key you want to try must be transmitted to the router, as part of a multi-step protocol. I don't care how many cores you have, the router is the bottleneck.
This is different from a situation where you have, say, a hash of a password and can parallelize hashing operations, trying to find a password that hashes to a known value. Ditto for brute forcing a cryptographic key space searching for one that decrypts a known ciphertext to a known plaintext (or a plausible plaintext). That sort of attack is eminently parallelizable. This case isn't.
I can try 8 digit pin (0-9 only?) in mere seconds on modern hardware just a bruteforce.
Only if the router is stupid. A proper implementation should at a minimum impose a second or two delay after failed attempts, and a good one should implement exponentially increasing delays.
Actually, "mere seconds" is likely impossible even without any delays. 10^8 values tested in, say, 10 seconds, means you have to be able to test 10^7 keys per second -- that's ten keys per microsecond. Given wireless protocol overheads, inter-frame delays, etc., plus the fact that the router hardware isn't tremendously fast, I seriously doubt that's achievable. Hours, definitely, minutes, maybe, seconds, no way. With exponential backoff, and without additional weakness in the protocol, an attacker is basically out of luck.
The 8-digit PIN is fine, assuming the rest is done right.
But if someone is unhappy for 6 months they are less likely to take the early upgrade at the 18 month mark.
Or more likely to take the upgrade as soon as they can get it. A few months of frustration might be just what's needed to get them to watch that upgrade date and jump on it as soon as they can.
Not really. That would imply that changing the default configuration to something else would fix the problem, but it doesn't. The only thing that fixes it is disabling WPS. Well, I suppose setting a really long PIN -- but the default is 8 digits which most people would expect is reasonable anyway. If the protocol didn't leak information about the PIN, or the device didn't allow brute force searches, this wouldn't be a problem.
This isn't a default configuration problem, this is a security protocol defect coupled with an implementation error.
Well.. it's a textbook example of worker alienation - one of the results inherent in the conflict between worker and capitalist.
Or, in this case, worker and government agency. Most likely. Most trade colleges are state-funded, and the private ones tend to be better-funded than described here.
Because if you don't have that right, you are not a free man, you are a subject. And as such, your rights and your life can be taken at any time the people who are your masters decide to. This is not theoretical. See Apartheid. See a hundred other things like that and worse.
See Ghandi.
Indeed, do.
"Among the many misdeeds of the British rule in India, history will look upon the Act depriving a whole nation of arms as the blackest."
- Mahatma Gandhi, "Gandhi, An Autobiography", M. K. Gandhi, page 372
Ghandi was not opposed to self-defense, either personal, social or national. He chose non-violence partly because he believed it was more moral, but mostly because it was the only viable strategy available to him. The British had already stripped the Indian people of arms in the Indian Arms Act of 1878 (which is what he was referring to in that statement, not, as some claim, to their refusal to allow Indians to be soldiers in WWI), so violent action wasn't likely to work. Had his people been armed, he may well have chosen a more direct route to ousting the British. I don't think that's likely; he was committed to non-violence and the situation in India was amenable to it. But he was also a pragmatist and recognized that there are situations in which those unwilling to defend themselves will simply be exterminated to no beneficial effect.
Again, IANAL, but "bear" arms presumably means, you know, to actually carry them. (That is, in fact, the definition of the transitive.) Although the SCOTUS has yet to decide on this issue, it's pretty clear cut to me that we ought to be able to carry guns basically anywhere per the constitution.
Yes, but you definitely don't have the right to go onto private property without invitation. If the owner of that property says "You may only enter my property if you are unarmed" then you have to leave you weapon behind in order to enter legally. It's your choice whether you do that, or to stay off their property and armed (or to break the law by trespassing, in which case you'd better be ready to deal with the consequences).
You are correct, but remember that there is a substantial amount of private property which is open to the public, and there are extensive limitations on what the owners of such property can restrict. In legal terms, any place of business that opens its doors to the general public is called a "public accommodation", and it may not exclude any person on the basis of their membership in a protected class, per federal law. In addition to that, many states further limit the authority of public accommodation owners by law. For example, the state of Utah (where I lived until recently, so I know the laws well) specifies that anyone accused of trespassing has a defense if the location was open to the public at the time and the person was not interfering with the owner's business. Court rulings have applied additional limitations, especially around the area of free speech.
With respect to firearms, the situation varies widely: In some states, business owners can post signs which have the force of law, meaning that it's a crime to carry on property that is posted "no guns". In many other states, such signs have no force, not even to make "violators" trespassers, at least until they've been personally asked to leave, and have refused. (In Utah, many suspect that a person who refuses to leave when asked to remove an openly-carried firearm is not trespassing, but it has not been tested in court.)
My point here is just that with respect to public accommodations, the extent of the owner's rights are much less clear. We know owners give up some rights by opening to the public. Whether or not the right to exclude armed customers is among the rights given up hasn't been determined at a national level.
I just watched the videos and it's not clear to me one way or another.
It is clear that the Ady Gil was under power, though moving slowly, while the whaler was moving at a good clip. It's also clear that the whaler wasn't at all upset about having collided with the Ady Gil. I don't see any evidence that either boat made any attempt to change course or speed.
I think it was a game of chicken. Either side could have blinked, but neither did. The captain of the Ady Gil thought the whaler would blink, in order to avoid an incident. The captain of the whaler thought the Ady Gil would blink, because it was obvious that a collision between the 13-ton composite-hulled Ady Gil and the 628-ton steel-hulled Shonan Maru 2 would destroy the trimaran and do nothing to the whaler. I also suspect that both hoped for a collision: the Ady Gil in order to create an incident, and the Shonan Maru 2 in order to crush a pest.
I can't be the only one who finds the idea of patents for " Mining of User... Data to Identify User[s]" a bit unnerving....
Of course, I'm one of the apparent minority who tend to adhere to the concept that privacy is still a right.
From a privacy perspective, patents like this are a good thing. Why? Because if individual companies systematically lock up the rights to invade your privacy in various ways (assuming this is an invasion), then they'll all be restricted to violating your privacy in only the ways they have patented. The only legal way for them to violate other aspects of your privacy is to cross-license from or collaborate with the competition. This would slow innovation in privacy violation as much as it does in other areas -- perhaps more, if this is the standard that must be met to obtain a patent.
From a patent perspective, of course, this is a case of using obvious information to draw obvious conclusions, which would seem to be pretty much the definition of "obvious". Actually, that same obviousness makes me think that calling this a violation of privacy is... excessive. People have always drawn obvious inferences like this. Every time you interact with anyone you give them information about yourself.
Slashdot Mirror
You're the one trying to convince me that his argument makes sense :-)
No, I wasn't. I merely defended my description of it as a "Pascal's wager".
So I see, after reviewing the thread.
I do take issue with your characterization of Pascal's wager as a form of innumeracy. I would agree that the assumption that life extension has infinite value is a fallacy of some sort, which might be characterized as innumeracy, if you squint.
> At least, they do if they're actively being attacked. If you assume that someone is trying to brute force your password, even a fairly long, complex password is only likely to stand up for a few months.
A fairly long, complex password is likely to stand up for millennia against brute force.
Wishful thinking.
At least by most people's definition of "fairly long, complex" -- but still reasonable to type and to remember -- password cracking is eminently accessible, though not (yet) cheap.
A ten-character password, containing a completely random selection of alphabetic, numeric and symbolic characters has about 61 bits of entropy. That's already beyond what most people are prepared to deal with, so consider this calculation an upper bound and reduce it by two or three orders of magnitude (minimum!) for the average real-world password.
According to this article an Amazon EC2 instance with GPU-based cracking can test 3.488 billion passwords per second. At that rate, it would take just short of 300,000 hours to search the entire password space, about 34 years. That's not trivial, but it's hardly "millenia". And, of course, password cracking scales perfectly, so you can use 34 times the resources to do it in one year, or 408 times the resources to do it in one month, or 300,000 times the resources to do it in one hour.
At the rate mentioned in the article, $2.10 per hour, it would cost ~$313,000, on average, to crack a password. That's substantial, but assuming it declines per Moore's Law (which wasn't about $/cycles, but close enough), in 10 years it'll cost just over $3K, in 15 years it'll cost about $300, and in 20 years it'll cost about $30.
Of course, good systems can make the attack more expensive by iterating the hashing operation to increase the cost of each password tested. But, still, the point is that the most complex passwords that people can readily handle are within the reach of a serious attacker, and this situation is just going to get worse.
Living longer has finite value
I'm not the person you have to convince.
Keep in mind that the original AC poster said "What is 5% weighed against infinity?" He really was assigning an infinite value to living longer.
Erroneously assigning an infinite value to living longer.
Tell him, not me.
You're the one trying to convince me that his argument makes sense :-)
That's not because the developers of mailman were idiots. It's because they assumed that the users were not idiots,
Uh, doesn't that make the developers of mailman idiots? How stupid would you have to be to make such an assumption about users?
Because it was a very reasonable assumption up until the eternal September.
Google has taken a step forward with their two factor options (via say, a cell text) but that's not really a practical option for many small web sites.
Small web sites can just use Google's OpenID for authentication, and tell their users to turn on two-factor on their Google accounts.
Alternatively, if your userbase has smartphones, you can use Google Authenticator to allow your users to use one-time passwords, without any involvement from Google. And it's not all that difficult to implement SMS-based second-factor auth yourself either. There are libraries for all of the major languages.
This is an excellent case for a PKI. Users should generate a public-private key pair, and provide the public key to the web site upon sign up. Extra authentication steps could be done at setup (web of trust a la PGP, known entities, a la X.509, callback texts, whatever). Users would sign a login blob with their private key to authenticate.
That would work, but I think OpenID is a simpler and more flexible solution. It just lets sites all over the web take advantage of the authentication solutions already built by Google, Yahoo, Facebook, Blizzard, etc. And people who don't trust those can always build their own authentication servers, or use one they do trust.
Changing a secure password offers no additional security. Its not like they wear out.
Yes, they do.
At least, they do if they're actively being attacked. If you assume that someone is trying to brute force your password, even a fairly long, complex password is only likely to stand up for a few months. This is one of the rationales for changing passwords periodically.
However, if you're really worried about that, you absolutely should not use the same password for multiple web sites. Because every site you use it with sees the plaintext password every time you log in, even if they store it properly salted and hashed. So it only takes one unscrupulous admin and your "strong" password becomes known. The OP says he adds some site-specific bits to his common password, but unless that's done very well, it adds nothing. And even when done well it doesn't add very much security, if the unscrupulous admin is clever enough to guess that's what's being done.
As an example, look at mailman, the mailing list manager. Not only did it store the plaintext password, it mails it to you monthly.
That's not because the developers of mailman were idiots. It's because they assumed that the users were not idiots, an assumption that was probably reasonable in the early days of mailman. Mailman has always told users that they should not use an important password because it would be e-mailed to them monthly. The idea was that your mailman account is very low-value, and so it made sense to use a weak password, and it made sense to e-mail password reminders because so many users forgot their low-value passwords.
The reason mailman changed was because the developers discovered that users were, in fact, idiots, and commonly used the same password that they used for, say, their on-line banking account, rather than making up a throwaway password which they didn't even have to bother to remember because it would be e-mailed to them monthly. Thus, mailman has to take pains to secure the user's password, not because the mailman account needs protection, but because all of the other accounts that use the same password need protection.
The argument is based on the assumption that there is a possibility of an outcome with infinite value.
Which is a perfectly valid assumption in the case of Pascal's wager, since that's exactly the expected payout under the Christian theology with which Pascal was familiar.
If there is a defect in Pascal's wager, it's in the assumption that an unknowable deity offers a known reward for known actions. Essentially, Pascal's wager takes as a given that if God exists, God is of a certain form. But if you can't know if God exists, how can you know what God is like?
The actual labeling of this outcome as say, "Heaven" rather than say, "live longer" is irrelevant.
No, it's not. Living longer has finite value, because it has a definite and unavoidable end. Heaven is purported to be endless, and therefore can offer infinite value. There's a big distinction there.
Keep in mind that the original AC poster said "What is 5% weighed against infinity?" He really was assigning an infinite value to living longer.
Erroneously assigning an infinite value to living longer.
Yes, as in this specific case one does a Diffie-Hellman key exchange before an attack. But when proofing anything against bruteforce style attacks on assumes that it's going to be an offline attack.
Depends on the context. Structuring the protocol to eliminate off-line attacks and then implementing countermeasures to defeat on-line attacks is a common and perfectly valid strategy. Particularly when you want to have human-usable keys, as in this case.
We assume worse case in cryptography research, not best case and hope someone doesn't work out how to make it offline.
No, we don't assume worst case. The right way to build cryptographic security systems is to define the threat model, identify the avenues of attack and implement necessary threat mitigation countermeasures -- with an appropriate level of conservatism, of course. Always assuming the worst case tends to produce systems which are unusable in practice, and often still insecure if the due diligence wasn't performed on the threat modeling side.
The fundamental structure of WPS is fine, and within that structure there's nothing wrong with an 8-digit PIN. The problem here is that protocol defects effectively reduce that 8-digit PIN to a 4-digit PIN. Even that might be okay, given appropriate countermeasures to slow a brute force search, but those countermeasures weren't implemented.
Because the point of taking my wife to a movie is to get us both out of the house.
The average person might enjoy it, but a geek will struggle in the truly claustrophobic walled garden that they have erected. I actually was crying 30 minutes after opening the package, because it was essentially worthless to me. (Thank god for people having their hands on it a month ahead of me, and doing all the work in rooting it.)
I just helped my sister root her Nook Tablet. Later that night, while she slept, it auto-installed an update that un-rooted it, and (I believe) renders it for the moment un-rootable.
I think she's going to see if she can return it.
Indeed, and it sounds more like a programming flaw than a platform flaw. If you need a hash function with cryptographic properties, don't use MD5. It may not always be obvious, but if you work with unverified user input, chances are that you need some level of cryptographic strength.
This isn't an MD5 issue. I doubt that most of these hash table implementations use any cryptographic hash function (why bother? Cryptographic collision resistance is expensive and not generally relevant to hash table uses), but even if they used iterated SHA-512 interspersed with RIPEMD, TIGER and WHIRLPOOL in a pseudo-random sequence, the implementation would still be vulnerable.
How? Well, all the attacker has to do is to predict with reasonable accuracy the size of the hash table. Since the attacker almost certainly has access to the exact code that is running on the web server, this is trivial. Suppose that the initial hash table size is always 32. This means that the attacker just has to find a whole bunch of names whose hash values are all the same mod 32. Finding full collisions is effectively impossible. Finding values with the same five-bit suffix is trivial. At some point the hash table may grow, and the attacker will have to adjust... but the attacker knows exactly when this is, and exactly how to adjust to it.
A simple countermeasure that comes to my mind is to prepend a fixed, but site-specific string to each name before hashing it. This would randomize the hash values, destroying the property the attacker worked so hard to create. Or even choose a different random prefix when the hash table is constructed, store it with the table, and apply it to each string inserted or looked up before hashing it.
The key is to make the hash output unpredictable to the attacker, so he can't choose inputs that hash to specific values.
There are probably other countermeasures, perhaps simpler ones.
They don't do anything if nobody reports you. They don't actively seek out non-compliant accounts unless you are under 13 and are dumb enough to put in your real age. They don't care.
Is there anything here to suggest that Google does actively seek out non-compliant accounts or address unreported policy violations?
I'll bet in this case that someone flagged the pic as inappropriate and that's the only reason it got looked at.
It is Pascal's wager, which is a form of innumeracy. To argue that even a slight increase in the odds of survival has infinite payout, has some serious implications for our lives.
I don't see how that's really related to Pascal's wager, which is the idea that, assuming it is unknown whether or not God exists (but assuming the Christian God, or similar), it's a good idea to bet that He does... because if you're right you gain eternal rewards and if you're wrong you don't lose much. In Pascal's wager, the successful bet does in fact provide infinite payout. In the case of decisions made to extend life, the payout is guaranteed to be finite.
So your only defense now is that I can't capture a packet and do an offline attack?
Ah, sorry, I missed this part.
Yes, that's exactly the problem. You can't capture a packet and do an offline attack. There is no packet to capture which will enable you to test many values offline. Each attempt to guess the PIN must be an on-line attack, transmitted to the router.
Actually. it is achievable, it's not about speed it's about being extremely parallel, if I try a different key across each 200+ processor cores.
You only have one router to test against. Every key you want to try must be transmitted to the router, as part of a multi-step protocol. I don't care how many cores you have, the router is the bottleneck.
This is different from a situation where you have, say, a hash of a password and can parallelize hashing operations, trying to find a password that hashes to a known value. Ditto for brute forcing a cryptographic key space searching for one that decrypts a known ciphertext to a known plaintext (or a plausible plaintext). That sort of attack is eminently parallelizable. This case isn't.
I can try 8 digit pin (0-9 only?) in mere seconds on modern hardware just a bruteforce.
Only if the router is stupid. A proper implementation should at a minimum impose a second or two delay after failed attempts, and a good one should implement exponentially increasing delays.
Actually, "mere seconds" is likely impossible even without any delays. 10^8 values tested in, say, 10 seconds, means you have to be able to test 10^7 keys per second -- that's ten keys per microsecond. Given wireless protocol overheads, inter-frame delays, etc., plus the fact that the router hardware isn't tremendously fast, I seriously doubt that's achievable. Hours, definitely, minutes, maybe, seconds, no way. With exponential backoff, and without additional weakness in the protocol, an attacker is basically out of luck.
The 8-digit PIN is fine, assuming the rest is done right.
But if someone is unhappy for 6 months they are less likely to take the early upgrade at the 18 month mark.
Or more likely to take the upgrade as soon as they can get it. A few months of frustration might be just what's needed to get them to watch that upgrade date and jump on it as soon as they can.
Same old thing, default configuration is bad.
Not really. That would imply that changing the default configuration to something else would fix the problem, but it doesn't. The only thing that fixes it is disabling WPS. Well, I suppose setting a really long PIN -- but the default is 8 digits which most people would expect is reasonable anyway. If the protocol didn't leak information about the PIN, or the device didn't allow brute force searches, this wouldn't be a problem.
This isn't a default configuration problem, this is a security protocol defect coupled with an implementation error.
Well.. it's a textbook example of worker alienation - one of the results inherent in the conflict between worker and capitalist.
Or, in this case, worker and government agency. Most likely. Most trade colleges are state-funded, and the private ones tend to be better-funded than described here.
Because if you don't have that right, you are not a free man, you are a subject. And as such, your rights and your life can be taken at any time the people who are your masters decide to. This is not theoretical. See Apartheid. See a hundred other things like that and worse.
See Ghandi.
Indeed, do.
"Among the many misdeeds of the British rule in India, history will look upon the Act depriving a whole nation of arms as the blackest." - Mahatma Gandhi, "Gandhi, An Autobiography", M. K. Gandhi, page 372
Ghandi was not opposed to self-defense, either personal, social or national. He chose non-violence partly because he believed it was more moral, but mostly because it was the only viable strategy available to him. The British had already stripped the Indian people of arms in the Indian Arms Act of 1878 (which is what he was referring to in that statement, not, as some claim, to their refusal to allow Indians to be soldiers in WWI), so violent action wasn't likely to work. Had his people been armed, he may well have chosen a more direct route to ousting the British. I don't think that's likely; he was committed to non-violence and the situation in India was amenable to it. But he was also a pragmatist and recognized that there are situations in which those unwilling to defend themselves will simply be exterminated to no beneficial effect.
Again, IANAL, but "bear" arms presumably means, you know, to actually carry them. (That is, in fact, the definition of the transitive.) Although the SCOTUS has yet to decide on this issue, it's pretty clear cut to me that we ought to be able to carry guns basically anywhere per the constitution.
Yes, but you definitely don't have the right to go onto private property without invitation. If the owner of that property says "You may only enter my property if you are unarmed" then you have to leave you weapon behind in order to enter legally. It's your choice whether you do that, or to stay off their property and armed (or to break the law by trespassing, in which case you'd better be ready to deal with the consequences).
You are correct, but remember that there is a substantial amount of private property which is open to the public, and there are extensive limitations on what the owners of such property can restrict. In legal terms, any place of business that opens its doors to the general public is called a "public accommodation", and it may not exclude any person on the basis of their membership in a protected class, per federal law. In addition to that, many states further limit the authority of public accommodation owners by law. For example, the state of Utah (where I lived until recently, so I know the laws well) specifies that anyone accused of trespassing has a defense if the location was open to the public at the time and the person was not interfering with the owner's business. Court rulings have applied additional limitations, especially around the area of free speech.
With respect to firearms, the situation varies widely: In some states, business owners can post signs which have the force of law, meaning that it's a crime to carry on property that is posted "no guns". In many other states, such signs have no force, not even to make "violators" trespassers, at least until they've been personally asked to leave, and have refused. (In Utah, many suspect that a person who refuses to leave when asked to remove an openly-carried firearm is not trespassing, but it has not been tested in court.)
My point here is just that with respect to public accommodations, the extent of the owner's rights are much less clear. We know owners give up some rights by opening to the public. Whether or not the right to exclude armed customers is among the rights given up hasn't been determined at a national level.
So, your assertion is that people buy whale products not because they like them but to protest the actions of Sea Shepherd?
I'll freely admit that I know nothing about this, but based on my experience of human nature that seems... very unlikely.
While sea sheppards were definitely harassing the whaling vessel with Ady Gil, it is hard to watch the footage and not see it as that the Ady Gil was rammed by the whalers. http://www.youtube.com/watch?v=Brw6JN0lQXY&feature=related
I just watched the videos and it's not clear to me one way or another.
It is clear that the Ady Gil was under power, though moving slowly, while the whaler was moving at a good clip. It's also clear that the whaler wasn't at all upset about having collided with the Ady Gil. I don't see any evidence that either boat made any attempt to change course or speed.
I think it was a game of chicken. Either side could have blinked, but neither did. The captain of the Ady Gil thought the whaler would blink, in order to avoid an incident. The captain of the whaler thought the Ady Gil would blink, because it was obvious that a collision between the 13-ton composite-hulled Ady Gil and the 628-ton steel-hulled Shonan Maru 2 would destroy the trimaran and do nothing to the whaler. I also suspect that both hoped for a collision: the Ady Gil in order to create an incident, and the Shonan Maru 2 in order to crush a pest.
I can't be the only one who finds the idea of patents for " Mining of User... Data to Identify User[s]" a bit unnerving.... Of course, I'm one of the apparent minority who tend to adhere to the concept that privacy is still a right.
From a privacy perspective, patents like this are a good thing. Why? Because if individual companies systematically lock up the rights to invade your privacy in various ways (assuming this is an invasion), then they'll all be restricted to violating your privacy in only the ways they have patented. The only legal way for them to violate other aspects of your privacy is to cross-license from or collaborate with the competition. This would slow innovation in privacy violation as much as it does in other areas -- perhaps more, if this is the standard that must be met to obtain a patent.
From a patent perspective, of course, this is a case of using obvious information to draw obvious conclusions, which would seem to be pretty much the definition of "obvious". Actually, that same obviousness makes me think that calling this a violation of privacy is... excessive. People have always drawn obvious inferences like this. Every time you interact with anyone you give them information about yourself.