This is an idea that's been floating in my head for months, and I figured it was reasonable enough to get some people to look at it and poke holes in the logic.
https://gist.github.com/christophebiocca/5341b2b55fb16f09c7ec
TL;DR: With the right merkle commitment structure (and full-nodes that index all transactions), we can achieve full-node security without full, independent validation of the blockchain. Unlike compact fraud proofs (which require miner participation and/or hard forks of the protocol), this approach can work with today's blockchain structure, and be used by a subset of the network (which is good for minority clients).
But I don't know if I'm overlooking something obvious here, so I'm requesting peer review.
https://gist.github.com/christophebiocca/5341b2b55fb16f09c7ec
TL;DR: With the right merkle commitment structure (and full-nodes that index all transactions), we can achieve full-node security without full, independent validation of the blockchain. Unlike compact fraud proofs (which require miner participation and/or hard forks of the protocol), this approach can work with today's blockchain structure, and be used by a subset of the network (which is good for minority clients).
But I don't know if I'm overlooking something obvious here, so I'm requesting peer review.
Last edited:
