[Goose]

“Unclear” is my favorite description of international affairs. For all the leaks, we still know very little. Even people in the know have trouble predicting anything with reasonable accuracy. It’s not like a controlled science experiment where you can attribute change to a particular cause with a high degree of certainty, let alone predict how the next experiment will turn out.

Very true! Unfortunately many decisions made even at high levels are greatly influenced by who presented data last or how the decision maker was feeling that day. Many almost "random" effects. That is why war, among other international activities, is never certain. Unfortunately humanity seems to always forget that and then is totally surprised when things don't go as expected :-).

One other fact that we in the US always forget. Each president will get the intelligence community he "wants", no matter how hard he tries to maintain an "objective" organization. It will take about two years, but it will happen. It is gravity. The organization "learns" to pander to the prejudices of it's master. It is a subtle but very real effect, and influences outcomes. In almost all cases, there are opinions all over the map about any complex intelligence question. The "chosen" result is greatly affected by the above mentioned learning. So far, nobody has figured out how to prevent this.

Another byproduct of this effect is that when some prediction of the intelligence community is proven by events to be "wrong", there is a search for the cause of the error. Inevitably somebody somewhere in the intelligence community did "get it right", and in retrospect it is "obvious" why that conclusion should have been reached. There will be a "reorganization" to "fix the problem". Of course, it won't do so. This is not to say all improvements are futile, but in fact most of them are.

IMO the attempt to make data more accessible to more analysts has accomplished little good. The subject matter experts (in their stovepipe organizations) always had access to the necessary data, and when they did not, they took action to get the data. There are lots of "facts" that can be connected after an event that nobody would have connected before the event happened. Even if somebody did, nobody would believe them and it would go nowhere. For example, there was a Colonel that exactly predicted the German offensive that became the battle of the bulge. Nobody believed him. There were initiatives after 9/11 to spread data more widely and break down the "stovepipes". IMO the result has been the destruction of information security without any real benefit. Thus the current questions about "need to know". If we intend that virtually any random person can make a connection between facts that (up until the person made this great leap of insight) are totally unrelated, how can there be anything such as "need to know"? IMO we should back down considerably from that dubious "goal".

[Goose]

“Unclear” is my favorite description of international affairs. For all the leaks, we still know very little. Even people in the know have trouble predicting anything with reasonable accuracy. It’s not like a controlled science experiment where you can attribute change to a particular cause with a high degree of certainty, let alone predict how the next experiment will turn out.

Very true! Unfortunately many decisions made even at high levels are greatly influenced by who presented data last or how the decision maker was feeling that day. Many almost "random" effects. That is why war, among other international activities, is never certain. Unfortunately humanity seems to always forget that and then is totally surprised when things don't go as expected :-).

One other fact that we in the US always forget. Each president will get the intelligence community he "wants", no matter how hard he tries to maintain an "objective" organization. It will take about two years, but it will happen. It is gravity. The organization "learns" to pander to the prejudices of it's master. It is a subtle but very real effect, and influences outcomes. In almost all cases, there are opinions all over the map about any complex intelligence question. The "chosen" result is greatly affected by the above mentioned learning. So far, nobody has figured out how to prevent this.

Another byproduct of this effect is that when some prediction of the intelligence community is proven by events to be "wrong", there is a search for the cause of the error. Inevitably somebody somewhere in the intelligence community did "get it right", and in retrospect it is "obvious" why that conclusion should have been reached. There will be a "reorganization" to "fix the problem". Of course, it won't do so. This is not to say all improvements are futile, but in fact most of them are.

IMO the attempt to make data more accessible to more analysts has accomplished little good. The subject matter experts (in their stovepipe organizations) always had access to the necessary data, and when they did not, they took action to get the data. There are lots of "facts" that can be connected after an event that nobody would have connected before the event happened. Even if somebody did, nobody would believe them and it would go nowhere. For example, there was a Colonel that exactly predicted the German offensive that became the battle of the bulge. Nobody believed him. There were initiatives after 9/11 to spread data more widely and break down the "stovepipes". IMO the result has been the destruction of information security without any real benefit. Thus the current questions about "need to know". If we intend that virtually any random person can make a connection between facts that (up until the person made this great leap of insight) are totally unrelated, how can there be anything such as "need to know"? IMO we should back down considerably from that dubious "goal".