2 ms·

Thanks for the context and the extra detail. It was so interesting I thought I'd poke around a bit more on my own. As for (3), please note that this involves i

Thanks for the context and the extra detail. It was so interesting I thought I'd poke around a bit more on my own. As for (3), please note that this involves intelligence, counterintelligence, or international terrorism. So now we're back to FISA, the PATRIOT act, NSA doing SIGINT on domestic/foreign internet traffic, etc. [insert long discussion here about the necessity of having intelligence and counterintelligence and the conflict with an also-necessary open society] More to the point, there's no way to determine that the process would be followed correctly. (This is the same problem the FBI is having with warrantless wiretaps: the law's wording and the actual practice is much different, and there are no independent outsiders monitoring what's going on) This is a good link to chase down. It directly relates to both FOIA and national security. In this case, you can't even ask about the general interpretation of a law by the government. http://www.zdnet.com/blog/btl/newspaper-sues-government-to-reveal-secret-patriot-act-interpretation/60407 http://www.zdnet.com/blog/btl/newspaper-sues-government-to-r... "...The New York Times is suing the U.S. government for refusing to divulge how its law enforcement interprets the Patriot Act...." Note that this isn't the detail of any particular record. This is simply explaining what the law is. If you can't go there, good luck trying to chase down the actual application of the law in any specific circumstance. I'm not getting into the overuse of "national security" to cover all sorts of things, or how you can target the FBI with the statute and end up pulling in all of DHS with the application. These are old discussions and probably best left for another day. Just wanted to point out that you could drive a Mack truck through the holes in that statute. It sounds very limited in theory. In practice, based on past experience, it doesn't look that way at all to me.

tptacek 15y ago
To simplify this conversation: I don't think we should have secret laws under any circumstances. I do think it's legitimate for the federal government to set up countermeasures to attempts to fish for the identities of confidential informants. I think there is some information relevant to counterintelligence and counterterrorism that should be classified and some that shouldn't, and I think more stuff is classified that shouldn't be than the other way around. I think that it's wholly reasonable for the federal government not to have to confirm the existence of heavily classified documents. I think there are many classified documents whose contents are not appropriate for disclosure, but whose existence is.
- DanielBMarkham 15y ago
  We don't disagree at all. The only thing I would add is that for every system of control there should be a counterbalancing system of inspection and audit. Congress surely isn't up to the job, there is no IG that I'm aware of that works at the National Security cross-agency level, and even if there was the results couldn't be made public anyway. For all of those needs that we agree on, you can't leave the government an open-ended blank check. It just doesn't work. FOIA is supposed to be part of the control process, but if we can't even learn what the policies are for documents that we then are asking about, FOIA is not working. Constitutionally, there are supposed to be hard limits to the amount of control the government has. If, for instance, J. Edgar Hoover was keeping files on Elvis Presley, sooner or later we'd get access to them and institute laws constraining the application of domestic counterintelligence operations. Laws and rule changes that prevent this "sooner or later" process from happening are bad. Really bad. At the very least there should be a firm time limit on whether or not to admit documents exist -- perhaps 10 or 20 years. That's not optimal, but it would at least allow for the system to self-inspect and adapt and evolve ever so slowly. (I still remain _very_ concerned that the theory we are talking about and the actual application are two completely different things, but I guess the point of this rule change is that we'll never know.)

tptacek 15y ago

To simplify this conversation: I don't think we should have secret laws under any circumstances. I do think it's legitimate for the federal government to set up countermeasures to attempts to fish for the identities of confidential informants. I think there is some information relevant to counterintelligence and counterterrorism that should be classified and some that shouldn't, and I think more stuff is classified that shouldn't be than the other way around. I think that it's wholly reasonable for the federal government not to have to confirm the existence of heavily classified documents. I think there are many classified documents whose contents are not appropriate for disclosure, but whose existence is.

DanielBMarkham 15y ago
We don't disagree at all. The only thing I would add is that for every system of control there should be a counterbalancing system of inspection and audit. Congress surely isn't up to the job, there is no IG that I'm aware of that works at the National Security cross-agency level, and even if there was the results couldn't be made public anyway. For all of those needs that we agree on, you can't leave the government an open-ended blank check. It just doesn't work. FOIA is supposed to be part of the control process, but if we can't even learn what the policies are for documents that we then are asking about, FOIA is not working. Constitutionally, there are supposed to be hard limits to the amount of control the government has. If, for instance, J. Edgar Hoover was keeping files on Elvis Presley, sooner or later we'd get access to them and institute laws constraining the application of domestic counterintelligence operations. Laws and rule changes that prevent this "sooner or later" process from happening are bad. Really bad. At the very least there should be a firm time limit on whether or not to admit documents exist -- perhaps 10 or 20 years. That's not optimal, but it would at least allow for the system to self-inspect and adapt and evolve ever so slowly. (I still remain _very_ concerned that the theory we are talking about and the actual application are two completely different things, but I guess the point of this rule change is that we'll never know.)