Proof of Stake Consensus participants should realize some of the upside from relays done by Validator Nodes

Mining works because people are greedy. To protect a PoS chain, consensus participants need to make enough to make it “worth” participating, but nothing more.

If consensus participants were only exposed to transaction fees on the network, I do not think the financial incentive of being a consensus PoS node would be high enough. There is research that if bitcoin were to run just on fees, they would be between $18,700 and $108,700 per transaction.

This is part of the dilemma of bitcoin - If it is a store of value, there will be fewer transactions and fees will be higher. For it to be some form of money, fees would need to be substantially lower but with much higher throughput. Bitcoin cannot scale to that level, and layer two solutions like Lightning avoid paying miner fees.

Pocket Network can be run without a blockchain. Validator nodes can check each other’s transactions in a p2p network and receive fees directly from developers. As a result of the economic model, the blockchain part is required to keep the integrity of fees earned, since they are minted and not directly transferred by the developer using a node’s services. I bring this up to highlight the importance of those staking POKT to keep consensus of the blockchain. It is a critical layer that keeps the system running, and as a result consensus participants should earn some percentage of the relays completed.

With that said, the following is my proposal for consensus nodes receiving some amount of reward from relays:

A couple assumptions first:

  • 100% of fees for transactions are burned
  • Blocks with no relays have some mechanism to allow for the block to propagate with no reward

With these assumptions we can state that the only way for PoS to work is if consensus participants are exposed to some of the upside from relays completed by Federated Nodes. It is an open question as to how much consensus participants should be allowed to earn; but let’s assume that 1% of the total block reward goes to consensus participants.

This allows the incentive of the consensus participants to scale as the network scales in relay volume. By exposing them to some of the upside from relays, it aligns all participants of the network in the same direction.

Please clarify the definition of a “consensus participant”.

  1. Maintains a full copy of the POKT Blockchain? (Passive full node/wallet)
  2. Relays POKT transactions and serves blocks? (Full node)
  3. Creates POKT blocks?
  4. Has non burnable stake? (Dash type staked node)
  5. Has stake at risk of forfeiture?
  6. Provides other services beyond a standard full node?
  7. Participates in the Pocket relay service as a service, dispatch or validation node?
1 Like

Yes, the difficulty with establishing a reward is that it is difficult to determine the true utility of a given action at a given point in time with 100% confidence interval. Reminds me of Heisenberg’s uncertainty principle.

But demoninating the reward with a stablecoin rather than a volatile coin would help to reduce risk/uncertainty and more closely approximate the reward to the true utility over time). If the stable coin is demoninated against an increasingly diverse bucket of assets (with the ideal bucket representing the entire global market, thus having an optimal Sharpe ratio), the more stable it is, and less risk for more return.

That’s also assuming that you’ll use Federated Nodes, which are more centralized due to the high stake requirement and which I am generally leery of. You can introduce pooling, which could potentially decentralized with PoS e.g. like Rocket Pool, but obviously a lot of effort is required for R&D and ensuring security. I strongly suggest to consider other designs without requiring high stake, centralized nodes, and I will think about this.

He may have meant it to be a catch all term for any participant that helps to establish consensus in the network. Similarly, I have used the term securitors in the context of Ethereum to refer to a variety of participants including validators, notaries, block proposers, however securitors can be a misleading term since any participant that could act badly.

A consensus participant is a full node participating in PoS consensus, and not relay validation. I am proposing a model where Developers, Service Nodes and Validator Nodes participate in consensus. There should be a minimum stake for PoS. There may be a possibility for Developers to have a significantly smaller minimum stake that counts towards their throttle. Once anyone reaches the PoS threshold they can opt to participate in consensus as well.

  1. Yes
  2. A developer could be a PoS consensus participant and not relay transactions
  3. Yes
  4. All stake should be at risk of burn
  5. Yes
  6. Or is using the network as intended (Developer)
  7. See 6

With this said, I do feel there is an merit to the idea of allowing anyone to participate in PoS consensus, not just those contributing to relays.

I agree. I think that anyone with skin in the game should be able to participate in consensus decisions in a degree (hopefully) equal to the amount of risk that they are taking, and / or value of service they are performing. That includes:
A. Active Developer stake: clearly taking a risk.
B. Service/Validator and dispatch stake: Clearly providing a service.
C. Hodled stake: I can’t think of any way to differentiate hodled stake from inactive developer stake. (Can you?) Therefore it should be treated equally even if we’d prefer not to reward the two activities at the same rate.

2 Likes