I feel there is an important point between the lines here: a central authority is well positioned to stop bad actors by working outside the definition of “expec

I feel there is an important point between the lines here: a central authority is well positioned to stop bad actors by working outside the definition of “expected behavior” of the system. P2P systems do not get this by design. For load bearing infrastructure like DNS, escape hatches to stop bad actors putting humanity’s hive mind at risk is a handy tool that has been used effectively. There are the philosophical questions of “how do we define a bad actor” and “who gets to decide what’s acting in bad faith?” But many botnets don’t suffer from that ambiguity, a well deployed/utilized botnet can take down load bearing internet infrastructure. Being able to stop future threats without requiring migrating a P2P network _seems_ like a feature, and reminds me a bit of common law. In law, we know there will be cases we can’t anticipate upfront and leave the courts room to interpret. P2P seems to still be in a state where we have to identify and protect against every form of “bad actor” upfront in our game theory in order for our network to be stable moving forward.