Jul 4 19:36 [raw]
Quantum key distribution (QKD) is a ruse. It is a ruse to convince government agencies and corporations to abandon classical encryption for a pie in the sky key exchange mechanism that can't be trusted. If the NSA lets this ruse go unabated the consequences could be disastrous to western nations' intelligence and data security. The initial seed of QKD must be exchanged using PKI. This is the weak link, and the link that the ruse promoters want all communications dependent on. They use high-sounding jargon from quantum mechnics to sell this snake oil. Current popular PKI schemes all have the weakness of relying on the intractibility of factoring. The other intractible problem, discrete logarithms, is a much harder problem and is used much less frequently. Logarithms do not follow any fractal pattern, whereas recent research has shown that prime distances (factors) do exhibit a definite adherence to fractals, which means they can be graphed by polynomials. The quantum hype is pushing to get everyone to trust in this weak link, while promoting it as Quantum OTP. It is not OTP if a factor-bound PKI is involved to exchange the key generator. It is still mathematical crypto, which has never been proven secure. Don't buy into the ruse. Use real OTP by physically exchanging one-time-pads, and discard pads after use. If you use the QKD's PKI step to exchange OTP seed material, you might as well just use PKI to exchange symmetric keys.
Jul 5 03:56 [raw]
I can't agree more! I'm a physicist and I've taken courses on quantum information theory. It's still just as possible to do a MitM attack by capturing both classical and quantum channels. *sigh* I don't even think that general purpose quantum computing (as needed for cracking RSA / DSA / ECC) is a realistic possibility, to be honest. And even if it is, we already have classical crypto algos that are believed to be quantum-resistant!
Jul 5 09:40 [raw]
Be careful. The pro-quantum trolls will crucify you for this blasphemy. I also doubt the idea of quantum computers being physically possible. I do believe we could simulate quantum states, but not actually have real quantum states under human or machine control. I appreciate your feedback. I do hope you will be lurking (and chiming in) on the chans for a long time to come.
Jul 5 15:43 [raw]
> Current popular PKI schemes all have the weakness of relying on the intractibility of factoring. The other intractible problem, discrete logarithms, is a much harder problem and is used much less frequently. Some ECC schemes use discrete log problem over curves and binary fields, but none of the curves and fields can be trusted and the keys are too short to be taken seriously. Even high-end consumer hardware can't brute force these short keys, but $100 million black-ops math mammoths are likely breaking them in realtime. NSA can easily afford a data center running thousands of networked, multi-core processors and GPUs. That would rip through bitcoin keys and other ECC short-key systems. Elgamal is a good scheme if it relies strictly on the discrete log and uses really big keys. The weak link is the prime factorization problem. If your public key is the product of two primes you must be naive to think the government hasn't thrown billions at that problem. The fact that the number has factors is the security problem. The public key should be a prime number, and the private key should be a large field of moduli related to the number, or vice-versa. If someone discovers a way to factor huge numbers efficiently (don't assume they haven't, or you are assuming security which is wrong thinking) then all communications using RSA-type schemes are compromised for all time. The discrete log problem is a matter of iterations rather than numerical structure so it is a MUCH harder problem when dealing with big keys. Recently university researchers published a report in which they discovered and proved a series of repeating, overlapping fractals underlie the entire sequence of numbers to infinity. This means that there is real potential to discover a mapping of polynomials to large numbers and quickly discover their prime factors.