Ethereum protocol fellowship week 5 & 6

Unfortunatelly having a hard time writing a weekly updates so for this time I'll wrap up 2 weeks to one. Hopefully next week I'll do it by the end of the week and not after 2 weeks.

Here are updates from my side:

  • Decided to stick with Verkle Tries in Java lib and still work on project proposal related to verkle tries in-circuit
  • Trying to understand Pedersen commitement and pedersen hash which is actually a same thing - it's simple a multi scalar multiplication of eliptic curve points. So it's a1*A + a2*B + a3*C… where a1,a2,a3 are scalars and A,B,C are eliptic curve points.
  • Pedersen hash is used for computing a verkle trie-key (5 EC points MSM) and pedersen commitment is used for computing a commitemnt to 256 children in a trie (256 E.
  • I have proposed for a call with OnlyDust contributors and Karim from Besu team where we brainstormed and talked about what is needed for Verkle Tries library. Other fellows participated, Naman and Agnish. Also few other people from OnlyDust community: Matteo, Thomas and I think Dinesh was also on the call too, and maybe i forgot somebody.
  • Made a progress on project proposal. It's still not ready but here's a draft: https://hackmd.io/l2XvogKoQOCH748T1rdWZw?view
  • Been chatting with Kev and Ignacio, they are helping me regarding what's the best way to execute this project.
  • also been chatting with people outside of EF on what's the best way to do this

So plan is to be involved and help on verkle tries in java, but also contribute to verkle tries in-circuit. I consider both things part of core protocol because Ethreum future is zkSnarks. Or verifiable computation - i think that's a better description.

Few more things to add on:

  • I was thinking about whether the crypto primitives used in Verkle tries can be improved. So if would think of PCS (polynomial commitment scheme) which is additively homomorphic, post quantum secure and zk-snark friendly, that could probably be a better thing to have it in the EIP of verkle tries. Stumbled upon this paper: https://eprint.iacr.org/2022/1368.pdf but it seems like this is not performant in-circuit, that's the conclustion after chatting with Lev: https://github.com/Lev-Stambler/CP22-function-commitment
  • I'm currencty in Palo Alto, California, US, been meeting people from the EF and from blockchain community. It's been fun and i'm talking to people regarding EF cohort - i'm trying to promote it and make people interested in understanding ethereum core protocol and contributing to it. It's great that is "permisionless".
  • it's been really fun and i'm so happy that i'm doing this fellowship
Select a repo