P2P + ZK - blockchain offers significantly superior decentralization, efficiency, privacy, verifiability, censorship-resistance and permanence

The *only* thing blockchains are good for is strict global consensus

Alas, many apps are being compromised into blockchains because of financial incentives over rational ones

Real World Ethereum 🌎 
building /daimo ✳️ 

Blobs are already pruned so not retrievable on Ethereum, history will be pruned with EIP-4444, and eventually state with state expiry/hibernation

Indeed, we're relying on P2P systems to not forget data Ethereum has & will

Some obscure files are forgotten is an incentivization problem, not one inherent in P2P systems

Agree with those except "permanence".

In practice every single byte that has ever finalized on Ethereum is easily retrievable today, and that will remain true even with danksharding.

I've tried many P2P protocols (dat, zeronet back in the day, torrents ofc) and links broke / showed "0 seeds" all the time

Technowatermelon. Elder Millenial. Building Farcaster. 

nf.td/varun

You'll probably like the new blog post that @v is writing.

Ties into Justin Thaler’s observation that we focused as an industry on verifier-optimized tech because of it, while prover-optimized tech can be super meaningful

Known for the everywhere.computer (/ec), /unglish (unglish.co.uk), the /secure-design WG, WalletUncon, Causal Islands + drumming in all-genre power trios.

"Less than forever but more than 15 minutes" is Bitcoin

I keep saying strict, global *timely* consensus, because I suspect there are other ways to have strict consensus globally if you are willing to wait less than forever but more than 15 minutes.

We need to go back to our hackers roots and do key signing parties 🎉 

Then we need to build private by default social graphs with selective disclosure.

Does the fact that "strict global consensus" are all built on top of financial incentive, affects the apps onchain?