We need better FUD. SVM is ironically a modular kernel design. Everything is a separate program, and all programs are isolated from each other.
Cryptographic ops are so performance sensitive that they are implemented as native programs. Native programs canโt be over the air upgraded so they have to go through the full node upgrade path.
I'm not saying it wasn't a multisig. I just saying it's not a rollup. When I was in college I had built a multisig with alt-da. That doesn't make it a rollup.
.@_Dave__White_ asked me why I keep going. Itโs basically because of:
1) itโs challenging. The problems are really hard so my brain is fully engaged solving them. 2) itโs competitive. HL ethereums etcโฆ are all driven by smart people solving the same hard problems. The environment is dynamic, no two days are ever exactly the same. 3) I can see the impact of my design decisions in the competition.
Imho, this is basically how to retain great people through the ups and downs.
Crypto founders should be doing a pre-mortem exercise once a quarter. If the general markets tank 95%, what do you do? 1) what product would you be grinding on for those 18 months until things rebound? 2) whatโs the best team to focus on it? 3) do you have enough cash in short term t-bills to do it?
Decentralization test is objective, not a bunch of subjective hand wavy bs:
1) can a minority unilaterally block the majority from making changes or functioning then itโs not decentralized
2) can the majority block the minority from exiting or forking its own chain (for example: withhold data, emergency upgrading the bridge contract) then itโs not decentralized