[This is an attempt to express something, not from a technical level, but more from a sense of awe after working with the dynamic picture of the network for a long while. I hope it is meaningful at least to those who have been around a while. In a way it’s an attempt to express the ineffable.]
“Give me a place to stand and with a lever I will move the whole world.” – Archimedes
I’ve watched and participated in quite a few discussions about attacks, snooping, ways to compromise security, etc., on the SAFE Network.
One thing that I see consistently–which isn’t necessarily bad in those instances, but should be considered–is that such are most often posed without a well-defined context, due to being in a sort of “different universe” than what has existed before.
I grasped early on that probably the most important security feature inherent in the basic organization of the network is that there is “no place to stand” in order to leverage unpermissioned knowledge from the network as it functions.
For instance, it’s easy to say “If an enemy with unlimited resources could control or have collaboration with a bunch of ISPs they could . . . .” or "If I can analyze traffic over a section of the network, then . . . ", etc.
This is fine for sake of thought experiments, to a degree, because “who knows what some quantum computer might be capable of,” etc. But actually showing how that sort of thing could be accomplished is almost always extremely suppositional, not reflecting anything like what it would take to execute the exploit.
I think that what doesn’t get appreciated till one has looked at it from many angles (and realizes that it is a different paradigm) is the absolutely diabolical arrangement of blind interdependencies, emergent order from deterministic chaos, etc., that is inherent in the network design.
There is literally “no place to stand” to determine anything of value, even from within the network. You would have to maintain access to a huge number of points simultaneously to know anything at all, and that would only be for that transient instant, because the scene is constantly in motion. The landscape will be different in the next moment.
And that’s assuming that one could coordinate all that data from some external perspective–and then, would whatever learned have any real value? I do believe the only profitable vector for compromise will be the individual hardware one is using. Keyloggers, literally looking over someone’s shoulder, etc. Other things which might be determinable, would be of almost no value, especially considering the cost and the inability to target specific users.
A lot of you will know exactly what I mean and perhaps can thilnk of a better way of expressing it.
It is not a wonder that such a feat has proven difficult to pull of in the real world, but it’s really all there in the theory. And since it’s been worked out logically, it WILL be built–it IS being built.
Since I’ve been watching I HAVE seen a few microscopic pinholes exposed, which might give a lucky observer or exploiter some minuscule piece of knowledge about what’s going on–probably nothing that could be actionable in the real world, but still . . . Then I’ve seen those pinholes welded shut with solutions that make the design even more elegant, rather than just a patch.
Hope this resonates with some.
What are you doing…? I said I would hold my tears for the release date. 
What did you doooooo!!! Not cool…