Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Activity
    Sina Mahmoodi
    @s1na
    Hey @jamesray1, re discussion in Meta-tx/Harbour-MVP#14:
    I agree that round-robin is not safe, and in my comment I didn't mean to use VDF with round-robin.
    Imagine we have an array of relayers, use the output of VDF to select the index of the relayer who should propose a block now. The VDF itself should be computed over the blockhash of N blocks ago.
    It's not only the implementation complexity of RANDAO. It also has some issues from what I gathered, regarding the last revealer. I noticed that for Eth2.0 they've pivoted away from RANDAO
    James Ray
    @jamesray1
    There are 35 results for RANDAO in the current spec: https://github.com/ethereum/eth2.0-specs/blob/master/specs/beacon-chain.md.

    There are probably issues around using a public piece of data for an input to the VDF.

    Honest proposers are expected to keep their local entropy private until their assigned slot.
    https://ethresear.ch/t/minimal-vdf-randomness-beacon/3566

    James Ray
    @jamesray1
    I guess it may be to do with that other proposers may jump in to someone else's slot and use the public piece of data to compute the VDF, potentially more quickly the assigned proposer.
    James Ray
    @jamesray1
    James Ray
    @jamesray1
    These are listed as possible rework/additions:
    • Harden RANDAO against orphaned reveals
    • Introduce a RANDAO slashing condition for early leakage
    Sina Mahmoodi
    @s1na
    sry, I meant away from pure RANDAO
    hm, let's say an opponent has specialized hardware and can calculate the output of VDF faster
    it still won't change the output, right? if they compute fast enough, they can again withhold block for example
    i'll follow the question there
    James Ray
    @jamesray1
    Yes, AIUI they shouldn't be able to change the output, just decide whether or not to withhold the block based on the VDF output and what is more advantageous to them.
    James Ray
    @jamesray1
    Justin replied.
    Sina Mahmoodi
    @s1na
    Yeah I read that. Security requirements of the beacon chain are really high and warrant adding complexity to gain security. I'm not sure however, how much weaker block hashes are, and whether they suffice for a L2 solution such as this one or not. Maybe we can simply outline multiple solutions, each with their own trade-offs.
    All things considered, how do you think the PoS-like mechanisms compare to other ones mentioned in Meta-tx/Harbour-MVP#3?
    James Ray
    @jamesray1
    Generally I would be leery about simpler PoS mechanisms since even for meta-txs where txs may be usually be low-value, failed txs will affect UX, reputation, user adoption, growth, etc. Talking about PoW mechanisms seems kinda futile since Ethereum is moving towards PoS, and you are building layer 2 infrastructure on Ethereum.