3 ms·
> For one, I'd be ok with Tor traffic being de-prioritized in an attack since that is commonly where an attack is coming from anyway. The hardcoded list of "ba
by Bae8anoh 11y ago
> For one, I'd be ok with Tor traffic being de-prioritized in an attack since that is commonly where an attack is coming from anyway.
The hardcoded list of "bad" peers is effectively worthless (it's already hopelessly out of date), and the additional service which downloads a new blacklist from a centralized website is completely insane. The reality of the situation is that if anybody with criminal intent wants to attack the Bitcoin network you need exactly two IP addresses (one v4, one v6) to completely disable all new incoming connections on all nodes in the network. This patch doesn't change that fact, nor that criminals often have botnets with unlimited access to new IP addresses every second of the day.
It doesn't achieve the stated design goal, and introduces new vulnerabilities which aren't stated in the commit.
> All of his other proposals are just that, proposals and "lets talk about this and the problem". I hate to think if I proposed/discussed a bad solution to a problem and suddenly I'm part of some conspiracy theory.
Bitcoin isn't like any other software on earth, it can't exist in fragmentation or in consensus incompatible forks. Operating on a fork of the software with soft consensus changes or simply operational changes that do not affect consensus are completely fine, that happens today under the assumption that you are somewhat at risk if you run software that is even slightly different to anybody elses. Some nodes in the network for example support different P2P commands, and that's fine because the P2P network is not part of the consensus at all. The linked discussions are mostly harmless, it is extremely positive that all scenarios and eventualities are debated out in the open.
Prompting companies and individuals to attempt a hard fork of the network without consensus is another matter entirely. In the light of that, the previous discussions lose their innocence somewhat.