Hacker News Reader

I don't get how this is better than keypair cryptography.

RSA cryptography can be broken quickly by quantum computers. There are other public/private key schemes that might avoid that problem, though none are widely used. Some of these rely on NP-complete problems instead of merely BQP problems like factoring.

Quantum cryptography, if correctly implemented, should be theoretically unbreakable, not merely "hard" to break -- much like a one-time pad.

I know of no one time pad algorithms for message authentication, which is a critical component of basically all real-world cryptographic systems. IMHO talking about bare XOR as "theoretically unbreakable" misleads newbies as to the important concepts in modern cryptography.

Basically, your marketing materials get to say that you implement the cool new thing.

Current QC accomplishes shared key generation only (so it's similar to a public key primitive like DH). Bulk ciphering and MAC are still done by classical symmetric algorithms. Forgoing the fibre and only using the pre-shared key would yield similar security properties.

Reply on news.ycombinator.com