I haven’t had time yet but I plan to read through the report and respond specifically to where I disagree. But I wanted to go ahead and respond to this since the question was also coming up in the discord…
concerns re centralization and governance capture
and quoting the NodeSet report:
Over the last year, NodeSet has spoken with numerous commercial node operators, including some of the largest in the industry. Many such organizations are interested in taking advantage of RP’s leveraged model to generate incremental returns, and the only thing holding them back has been RPL, which they cannot carry on their balance sheets due to volatility or legal concerns. Independent operators do not have this issue, so the current tokenomics helps keep RP decentralized and distributed. Today, large institutions may engage in bespoke RPL lending agreements. Still, there is limited RPL to be lent, limiting their potential individual impact on the protocol and preserving one of RP’s most significant differentiators: a decentralized operator base.
I disagree that our proposal introduces centralization/governance concerns.
It sounds like the assumption is: “By keeping the RPL requirement, we can prevent centralization since centralized actors won’t buy RPL”
Even if we take that assumption as true for a minute, my response would be:
Under our proposal:
- A large wave of (centralized) Eth Only NOs join
- They have zero governance power (they don’t own any RPL)
- If we hit our self limit of 22% and still have outsized NO demand we can lower the “node_operator_share” close to 0% and instead make most protocol revenue go to “voter_share” which requires RPL
- Now we have solved the problem in the same way that the original assumption suggested (require RPL to earn rewards, and centralized actors won’t buy RPL, so they are incentivized to leave and go elsewhere)
If you have qualms with “node_operator_share” being too high (too attractive to EthOnly Node Operators), we can start by giving “node_operator_share” a lower cut of commissions and instead give that to “voter_share” which requires RPL at stake, which resolves the issue under the original assumption.