4 ms·

> I mean, why not though? Why is it OK for the government to have any secrets?...Maybe we shouldn't be sending out special forces teams or covert operatives. T

by jarrett 13y ago

> I mean, why not though? Why is it OK for the government to have any secrets?...Maybe we shouldn't be sending out special forces teams or covert operatives. This gets at the heart of the issue. As I mentioned in my parenthetical above, the arguments in favor of operational security are only relevant if you assume operations are happening in the first place. If you philosophically reject all warfare, espionage, and law enforcement, then there is no legitimate reason for a government secret. However, it's no light matter to reject those things altogether. It implies a society whose structure is fundamentally different from our own. I'm not saying it's impossible (nor am I saying it is possible), just that we're now tremendously magnifying the scope of our claims. We're no longer talking about a particular government policy (nuclear launch decisions), but a complete rewrite of society from the ground up. > Or perhaps systems should be built with redundancy and decentralization. It shouldn't be up to any one person to make the decision to launch a nuke, or any other attack. We have methods such as Shamir Secret Sharing that would require multiple people to agree to something before being able to provide something to authorize a use of force. That doesn't address the problem the White House faced in the Cold War. Namely, how do you ensure the nuclear deterrent is credible even when lines of communication have broken down? If you rely on Shamir Secret Sharing, then retaliation would be impossible if the secret holders are unable to communicate with each other. > Unless that "hypothetical, worst-case scenario" is the government going awry and surveying it's entire populace and killing its citizens without going through any process set down by the constitution of said country. In that "hypothetical, worst-case scenario" is it really OK to be giving the government more power and leeway while decreasing the transparency? You're shifting the meaning of "worst-case scenario." I used the term to refer to a physical threat, external to the government, which the government is responsible for defending against. You're using the term to refer to a pathology of the government itself. So, the scenario you envision is not the one to which I was referring.

jimktrains2 13y ago
>If you philosophically reject all warfare, espionage, and law enforcement, then there is no legitimate reason for a government secret. Law enforcement doesn't need to be based on secrecy. I never said you have to broadcast that you're investigating someone, just that the information about who, how, and why you're investigating someone be reasonably public (going through real courts to get warrants and such). As for war, if we're in a defensive war we can chat about not letting the army put their immediate plans up for all to see. Ditto with the case about the Olympics in a sibling thread. I see the usefulness in _short-term_, someone's-life-is-in-danger (not because the state put them there), secrecy. > That doesn't address the problem the White House faced in the Cold War. Namely, how do you ensure the nuclear deterrent is credible even when lines of communication have broken down? If you rely on Shamir Secret Sharing, then retaliation would be impossible if the secret holders are unable to communicate with each other. The basic idea would be to have redundancy in channel, location, mode, since we're talking about a case in which the chain of command's top has been ripped off leaving many higher-ranking officers at the top of the new chains. An attacker wouldn't be able to take out all (all routes for all modes all over the US) of the communication network simultaneously, especially if we're not talking about superbly massive bombings in which case there would/should be detection. Also, to be credible, you're assuming the US is telling other's about their system (so that the USSR knows/will think that it can't simply take out the president and there won't be retaliation). Otherwise it's all just based on speculation and in terms of deterrent doesn't matter what the actual system is. So instead of saying "in a communications breakdown everyone does what they think is best, so someone is going to retaliate" it's "there can't be a communications breakdown because of redundancy in mode, route, and location; so in the event of a collapse of the chain of command, it can only collapse so far and hence retaliatory action is still possible." This just seems better in practice as well. There is the Dr. Strangelove example of someoene just taking power, and one can also imagine that someone innocently loses communication and assumes the worst. > You're shifting the meaning of "worst-case scenario." I used the term to refer to a physical threat, external to the government, which the government is responsible for defending against. You're using the term to refer to a pathology of the government itself. So, the scenario you envision is not the one to which I was referring. Why must it be external, though. Both you and I are concerned about how the government should operate in a manner that protects the populace; I'm just saying that from the populace's POV, the government is/can be an external threat. Look at Kiev right now. Yes, there will be (exceptional) times when secrets are needed; I get that. Those should be exceptions, and not the rule, as it feels like it is now. I just think that erroring on the side of transparency, and not doing actions that would otherwise be limited by that transparency (e.g. covert operations) is the better option in terms of dealing with external _and_ internal threats than is secrecy. Also, an ideal would be for emergency measures to simply flow from the normal operational measures. In the nuke example, forcing some sub-set of commanders to agree that they've received orders to launch nukes (and by implication of participation agree with them and have verified them) before nukes could be launched. The emergency situation is now simply that instead of confirming an order from above, they confirm that they all agree that the situation is such that they should launch.
- jarrett 13y ago
  > Law enforcement doesn't need to be based on secrecy. I never said you have to broadcast that you're investigating someone, just that the information about who, how, and why you're investigating someone be reasonably public (going through real courts to get warrants and such). There's a difference between judicial oversight and public oversight. The court may issue a wiretap authorization, yet that will probably remain secret for as long as it's in effect. Otherwise, the wiretap will be largely useless. You can't go to your local courthouse and demand a list of currently active wiretaps. It's not public record. > The basic idea would be to have redundancy in channel, location, mode, since we're talking about a case in which the chain of command's top has been ripped off leaving many higher-ranking officers at the top of the new chains. An attacker wouldn't be able to take out all (all routes for all modes all over the US) of the communication network simultaneously, especially if we're not talking about superbly massive bombings in which case there would/should be detection. What you're proposing is maybe practicable today. I doubt it would have been at the time in question. Technology was very primitive then, in comparison to what we have now. > Also, to be credible, you're assuming the US is telling other's about their system (so that the USSR knows/will think that it can't simply take out the president and there won't be retaliation). Otherwise it's all just based on speculation and in terms of deterrent doesn't matter what the actual system is. The assumption is that the enemy has some insight into your system. In which case it really does matter what your system is. > Why must it be external, though. Both you and I are concerned about how the government should operate in a manner that protects the populace; I'm just saying that from the populace's POV, the government is/can be an external threat. Look at Kiev right now. Again, you're arguing about a different issue than the one I am. I agree that a government can become an enemy of its public. But while that's a plausible scenario, it's not the one I was analyzing. Thus my comment that you're shifting the meaning of the vocabulary to refer to a different concept.

by jarrett 13y ago

jimktrains2 13y ago

>If you philosophically reject all warfare, espionage, and law enforcement, then there is no legitimate reason for a government secret. Law enforcement doesn't need to be based on secrecy. I never said you have to broadcast that you're investigating someone, just that the information about who, how, and why you're investigating someone be reasonably public (going through real courts to get warrants and such). As for war, if we're in a defensive war we can chat about not letting the army put their immediate plans up for all to see. Ditto with the case about the Olympics in a sibling thread. I see the usefulness in _short-term_, someone's-life-is-in-danger (not because the state put them there), secrecy. > That doesn't address the problem the White House faced in the Cold War. Namely, how do you ensure the nuclear deterrent is credible even when lines of communication have broken down? If you rely on Shamir Secret Sharing, then retaliation would be impossible if the secret holders are unable to communicate with each other. The basic idea would be to have redundancy in channel, location, mode, since we're talking about a case in which the chain of command's top has been ripped off leaving many higher-ranking officers at the top of the new chains. An attacker wouldn't be able to take out all (all routes for all modes all over the US) of the communication network simultaneously, especially if we're not talking about superbly massive bombings in which case there would/should be detection. Also, to be credible, you're assuming the US is telling other's about their system (so that the USSR knows/will think that it can't simply take out the president and there won't be retaliation). Otherwise it's all just based on speculation and in terms of deterrent doesn't matter what the actual system is. So instead of saying "in a communications breakdown everyone does what they think is best, so someone is going to retaliate" it's "there can't be a communications breakdown because of redundancy in mode, route, and location; so in the event of a collapse of the chain of command, it can only collapse so far and hence retaliatory action is still possible." This just seems better in practice as well. There is the Dr. Strangelove example of someoene just taking power, and one can also imagine that someone innocently loses communication and assumes the worst. > You're shifting the meaning of "worst-case scenario." I used the term to refer to a physical threat, external to the government, which the government is responsible for defending against. You're using the term to refer to a pathology of the government itself. So, the scenario you envision is not the one to which I was referring. Why must it be external, though. Both you and I are concerned about how the government should operate in a manner that protects the populace; I'm just saying that from the populace's POV, the government is/can be an external threat. Look at Kiev right now. Yes, there will be (exceptional) times when secrets are needed; I get that. Those should be exceptions, and not the rule, as it feels like it is now. I just think that erroring on the side of transparency, and not doing actions that would otherwise be limited by that transparency (e.g. covert operations) is the better option in terms of dealing with external _and_ internal threats than is secrecy. Also, an ideal would be for emergency measures to simply flow from the normal operational measures. In the nuke example, forcing some sub-set of commanders to agree that they've received orders to launch nukes (and by implication of participation agree with them and have verified them) before nukes could be launched. The emergency situation is now simply that instead of confirming an order from above, they confirm that they all agree that the situation is such that they should launch.

jarrett 13y ago
> Law enforcement doesn't need to be based on secrecy. I never said you have to broadcast that you're investigating someone, just that the information about who, how, and why you're investigating someone be reasonably public (going through real courts to get warrants and such). There's a difference between judicial oversight and public oversight. The court may issue a wiretap authorization, yet that will probably remain secret for as long as it's in effect. Otherwise, the wiretap will be largely useless. You can't go to your local courthouse and demand a list of currently active wiretaps. It's not public record. > The basic idea would be to have redundancy in channel, location, mode, since we're talking about a case in which the chain of command's top has been ripped off leaving many higher-ranking officers at the top of the new chains. An attacker wouldn't be able to take out all (all routes for all modes all over the US) of the communication network simultaneously, especially if we're not talking about superbly massive bombings in which case there would/should be detection. What you're proposing is maybe practicable today. I doubt it would have been at the time in question. Technology was very primitive then, in comparison to what we have now. > Also, to be credible, you're assuming the US is telling other's about their system (so that the USSR knows/will think that it can't simply take out the president and there won't be retaliation). Otherwise it's all just based on speculation and in terms of deterrent doesn't matter what the actual system is. The assumption is that the enemy has some insight into your system. In which case it really does matter what your system is. > Why must it be external, though. Both you and I are concerned about how the government should operate in a manner that protects the populace; I'm just saying that from the populace's POV, the government is/can be an external threat. Look at Kiev right now. Again, you're arguing about a different issue than the one I am. I agree that a government can become an enemy of its public. But while that's a plausible scenario, it's not the one I was analyzing. Thus my comment that you're shifting the meaning of the vocabulary to refer to a different concept.