[Cryptography] Ynt: A new, more efficient consensus protocol

[Vincent Strom]

If CBDC has a central design then it could expose the central node to hacking, or it could be operated unfairly etc. (If you are saying that such things can be avoided even with a centralized design, please elaborate.).

The advantage of having a decentralized design is that the CB's operation can be completely secure and transparent if performed through a smart contract. As a user, one would enjoy all the same features of a cryptocurrency because no trust is involved. Perhaps one could be averse to the ideology but the benefits of traditional monetary policy (to achieve price stability) and of legal backing are hard to ignore.

I am not necessarily advocating the use of central bank currencies over crypto (or vice versa) but I am simply saying that here is a protocol that is capable of turning CBDC into crypto (modulo ideological issues). Moreover, I am saying that this protocol (now thought of as crypto) seems to perform well on many metrics e.g. cost (compare to PoW), security (compare to PoS), fairness and possibly scale.

Let me now respond to more concrete issues that you raised.

1. Yes! the reward should scale with N. Thanks for pointing this out. I have made the correction. But note that even with a billion people, the breakeven reward is 20 cents!

2. This point is addressed in footnote 4. Whether a transaction qualifies as a lottery or not can be tested by any node. This is not user specific. The criteria is that the transaction should have some minimum fee and that its hash should be less then some threshold. One can imagine some altruistic nodes doing this job and maintaining a separate mempool for lottery tickets (a single such node suffices). This mempool will grow only as 10 lotteries/s. If no one on the network is an altruist then every user has to do this job for themselves and would require the hashrate comparable to transaction generation rate ~ 2K or more. This is still significantly less than what is required for mining. Moreover, the requirement is capped at this rate. The required hashrate does not increase as more hashpower joins the network.

Regards,
Vincent
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://www.metzdowd.com/pipermail/cryptography/attachments/20210115/b70bcf1d/attachment.htm>