Wired networks are a lot more difficult to attack than wireless. It's way more cost-effective to just use existing infrastructure. Layer 1 doesn't give anyone any information, and Layer 2 is hardly usable across networks, and you can encapsulate layers 2-7 in a tunnel, so really there is no need for new infrastructure. We just need a critical mass of network nodes.
You're misunderstanding the crux of the issue. Think of the NSA as a man in the middle. Every signal you send is allegedly stored for later processing. At that point, it doesn't matter how many layers you have encapsulating your data -- it can still be analyzed and decrypted given enough time and resources.
The crux of the issue is that somebody thinks the NSA is limited to using existing infrastructure. They aren't. You will be more successful in your aim by simply using the existing infrastructure in a way that is impractical to attack.
The most secure method of transmitting a message is a one-time pad. There is no way to analyze or decrypt it without its mated random pad. However, for internet traffic, this is horribly impractical. So you use alternate methods with some tradeoffs to achieve security that is practically impossible to attack.
The only benefit to alternate infrastructure is to work around government-imposed sanctions on traditional internet services, such as Iran, Egypt and other countries shutting off internet access. Project Byzantium is one way to work around this.
I agree with the crux of your message, but references to "one time pads" are an insidious red herring. They aren't actually secure against anything but the simplest passive adversary, and do little to further an understanding of the scope of modern cryptography (public key distribution, message integrity).
It's not a red herring. I was comparing the commenter's idea that "with enough time you can crack anything" to a system which cannot be cracked and how you wouldn't want to use it for this purpose anyway. I was not attempting to further the understanding of the capabilities of modern cryptography as I don't have all day to sit here and rehash whole textbooks on crypto.
Also, what active adversary are you implying would defeat the one-time pad? Lead pipe cryptography? That pretty much works for everything.
A one-time pad XORed with a message doesn't provide integrity, and therefore can't reliably secure an arbitrary protocol. An active attacker can easily flip any bit in the message. An easy example would be flipping a debug bit (presumably it started off 0) so that the server responds in plain text. We could blame this weakness on the protocol, but the point is that without message integrity we do not have a solid abstraction to build on top of. Referencing OTPs as unbreakable obscures this and several other problems that modern public key cryptography actually solves.
One-time pads (and all symmetric crypto) can never be user friendly - every pair of parties looking to communicate has to meet in person and exchange keys. Pop references to OTPs create false intuitions that real encryption takes significant work to exchange large eys, only secures preexisting links, and is only ever done by those with super serious secrets (spies, children in treehouses, spies, etc).
I also didn't mention the hardware random number generator required to use it properly. Getting a hardware RNG for each machine on the internet is probably more difficult than implementing authentication with OTP. But I did say it was a horrible idea for internet traffic, didn't I? =)
Give me a cryptosystem, and i'll give you a reason nobody should ever use it.
