Proposal for a Zcash Governance Bloc (“zBloc”)

Governance
29

Today the LCWG discussed some topics around FROST and NDFM use case was brought up.

While some argued that it was too soon to propose tools, and moreover that it wouldn’t be suitable because of some requirements: signature attribution and robustness, I think otherwise

The following represent my views and not LCWG’s.

I think that any NDFM that has a k-of-n kind of disbursement mechanism already outlines a threshold and FROST should be used before proposing any other signature mechanisms.

Given that a NDFM requires a k of n approval to disburse funds there are some important aspects to consider.

Requirements and open questions:

  • Attribution: do we want to know who voted for what? Is that important or is it just tabloid magazine material?
  • robustness: is FROST optimistic protocol not robust enough for a k-of-n with small k and n’s?
  • synchronicity: what’s the timeframe to vote?
  • privacy: what privacy properties do we want?
  • chain activity, yes or no: do we want the NDFM v1 be on chain? or is it off chain ok? what has to be on chain? all of it? or just the final k-of-n signature?

A case for FROST:

  • attribution: according to @conradoplg Verifying Shares can be made public to actually have signature attribution
  • Robustness: FROST provides a single round protocol by participants sending commitments to the coordinator beforehand.
  • synchronicity: what’s the timeframe to vote? I believe that deliberation of the different parties should be async but disbursement approval shall be done asynchronous within a human compatible timeframe.The world is round and crypto is decentralized. This is hard because of timezones.
  • privacy: do we want recipients to be private and shielded? or shall these transactions be shielded but viewable by a viewing key? can a grant recipient in an oppressive regime be put in danger by this?
  • chain activity, yes or no: NDFM v1 should be as simple as it can be. It will already be better than direct funding. There’s an improvement from scratch. NDFM v1.1, 1.2, etc should iteratively be improving on its robustness and incorporate feedback from experience iteratively and regularly until there’s a very stable version of NDFM.
  • Reputation: It will hurt Zcash a lot if we don’t use our own tools. We developed FROST for the world to use but it’s not “good enough for us”? that does not look well.
5 Likes