Farcaster Devs

Reading the Snapchain doc: https://warpcast.notion.site/Snapchain-Public-0e6b7e51faf74be1846803cb74493886

Going to cast all my stupid questions in this thread as they come up 🧵👇

>The right number is probably 10 - 20 globally distributed write hubs

What was the reasoning behind this statement?

Needs to be sufficiently large enough that 1/3 of the network can safely fail at any time (so rolling updates don't threaten the network's uptime), but not so large that the consensus algorithm ends up being over encumbered and cannot handle volume (a real issue in any pBFT approach and why you see the pBFT-based networks touting 100k tps only have a handful of validators — validator count is inversely proportional to throughput under pBFT approaches)

BDFL of Quilibrium, Eng @ Farcaster, ex-Coinbase, always opinionated, never hydrated

/quilibrium

A nation state actor would focus on the weakest link (backends for the clients) first. A PoS based system immediately gives a direct correlation for cost of attack, in that a nation state considering resource expenditure may find it cheaper to just eclipse the network by buying validator positions.

Utilizing another network's cryptoeconomic security can make sense, but it has to be aligned, e.g. intersubjectivity

Have you looked into PoS mechs that emulate PoW via slots where block producers are elected using a rng? Eg. Ouroboros 

I think 20 feels too low because a nation-state level actor could take down 10 nodes if they had strong desire