4 ms·
> Saying "they are public so it's ok" is technical oversimplification. It's not. The whole idea of the scheme is that you can publish them everywhere with no r
by hobarrera 11y ago
> Saying "they are public so it's ok" is technical oversimplification.
It's not. The whole idea of the scheme is that you can publish them everywhere with no risk.
- dchest 11y agoOMG, people! "Public" in "public key" doesn't mean that you should share it with the whole world, it's just because it's an antonym of "private key", meaning that the other party you want to communicate with doesn't need to have a pre-shared secret key with you. Other party may be the whole world, but may be not. How public key is used is up to a protocol and threat model: it may as well be secret, and yes, in some protocols and threat models security can depend on public key being secret, and it's not wrong, and not "security by obscurity".
- oleganza 11y agoIn case of Bitcoin pubkeys you don't want to publish them anywhere for privacy reasons (that turn into real physical security reasons once you have enough BTC). Also, notice how pubkeys are normally hidden under hashes and revealed only at spending time. This drastically limits attacks on ECC if/when some weakness in curve math is discovered. And if quantum computer is invented tomorrow, people can safely transition funds to a new signature scheme by introducing 2-phase commitments to safely reveal (now crackable) ECDSA pubkey after another transaction was made with commitment to concrete signature (preventing double-spend attempts using a cracked key).