Your comment and parent are starting the dialog that need to happen. Classified networks like SIPRnet are a neat idea, but it's ultimately people who are responsible for securing the information. And at some point, you have to say no harm, no foul.
There's a hallucination that classification is still used in the way it was intended. System administrators, like Snowden, have to demonstrate the airgaps in their SIPR systems, meanwhile, plenty of traffic is intentionally not classified, so as to avoid dealing with SIPR. The current intelligence systems classify the front page of the New York Times, yet the humans can type whatever they're gonna type into whichever keyboard.
And this happens all the time outside computers too. Some congress-critter goes on a submarine ride and accidentally spills the depth of the dive, which up to that point had been a state secret for 30 years. Who's to blame? Is there blame?
Classification is a government sanctioned land-grab and you've got everyone from the President to 18 year old kids grabbing the land. If we obeyed this law in its strictest sense, and applied it as broadly as possible, government would be in vapor lock.
We have mechanisms for machines to trust each other, and perhaps for people to trust machines, and machines to trust people, but you can't code a human to trust another human. At some point it's a judgement call.
Thus far, we have a small number of people at the top of the intelligence community using computing trust to prevent their people from using resources the intel higher-ups don't trust. But that
1) prevents people from using resources, so they're weakening their own people, and
2) effectively encourages the people who are trying to do their best, to not tell the boss what they're doing. They just do it and hope no one bothers to do anything about it.
I hope I'm not the only one who sees how these two effects ultimately weaken the whole enterprise.
There's a hallucination that classification is still used in the way it was intended. System administrators, like Snowden, have to demonstrate the airgaps in their SIPR systems, meanwhile, plenty of traffic is intentionally not classified, so as to avoid dealing with SIPR. The current intelligence systems classify the front page of the New York Times, yet the humans can type whatever they're gonna type into whichever keyboard.
And this happens all the time outside computers too. Some congress-critter goes on a submarine ride and accidentally spills the depth of the dive, which up to that point had been a state secret for 30 years. Who's to blame? Is there blame?
Classification is a government sanctioned land-grab and you've got everyone from the President to 18 year old kids grabbing the land. If we obeyed this law in its strictest sense, and applied it as broadly as possible, government would be in vapor lock.
We have mechanisms for machines to trust each other, and perhaps for people to trust machines, and machines to trust people, but you can't code a human to trust another human. At some point it's a judgement call.
Thus far, we have a small number of people at the top of the intelligence community using computing trust to prevent their people from using resources the intel higher-ups don't trust. But that
1) prevents people from using resources, so they're weakening their own people, and
2) effectively encourages the people who are trying to do their best, to not tell the boss what they're doing. They just do it and hope no one bothers to do anything about it.
I hope I'm not the only one who sees how these two effects ultimately weaken the whole enterprise.