Given all the recent drama surrounding a certain “DAO” and the controversial and borderline self sabotaging decisions being made via governance by holders, it’s a good time to look at the governance models in DeFi, and if they’re actually beneficial at all.
https://twitter.com/litocoen/status/1488488938310414336?s=20&t=onW2SvO5UwRDE8GC_BeihA
For a new age technology, DAO governance models are pretty similar to traditional governance, albeit digital and on-chain. Either its a simple 1:1 model, a many to 1 delegation, or elect a representative who in turn votes on proposals like Synthetix’s Spartan Council etc. The ENS and DYDX airdrops were memed for the responsibility, and while well-received, not everybody cares the same amount about governance. And that’s alright. Economic outcomes need not be aligned with participation in governance. A whale with 1% of the supply who sees massive financial upside may have a fraction of governance activity than somebody with 0.001% of the supply, but with innovative ideas for project growth.
With evolving structures, there is a need for evolved models of governance that better reflect the true nature of decentralization. Simple weighted voting gives too much power to whales who can push through proposals beneficial to them. Sometimes, the whales don’t even care about the governance, they’re simply in it for the gains. The veToken narrative recently is a good experimental model, but still can cause large disparity between holders of different size. Gearbox Protocol’s newly introduced governance model uses a Reverse Voting Escrow Model. A unique way at trying to reduce weightage of team members and initial contributors in favour of the community.
https://twitter.com/bout3fiddy/status/1487866886117376003?s=20&t=s8GNQpG-eY1MT_mqHjkJyw
A crucial element being missed here is governance activity. Regular participation in governance activity, voting, forum discussions, maybe even specific discord channel engagement should carry a weightage in voting, a la Proof of Governance.
A model of fair governance by ElasticDAO experimented with incentivizing/penalizing holders to take part in governance and giving rewards for partaking. But this can be expensive to the treasury. Governance activity should be independent of economic incentives. Instead it should let the holder accumulate a form of social credit, which could possibly be carried over to governance of other projects. Adding this up with Vitalik’s recent post on Soulbound governance, you get a form of score based off governance activity.
https://twitter.com/Andotlas/status/1486783306565709824?s=20&t=VeRo_06MX7D3pU9wuSTLcg
Let’s assume on a periodic basis, holders of XYZ token get airdropped a non-transferrable NFT, with holders above an arbitrary duration eligible for it. The process could be made more efficient if holders have to “Opt-In” to governance. Each NFT would have the same basis score (ex: 400) starting off, which puts all holders on the same level. The score can then be updated depending on weekly/monthly governance participation of holders, which could then correspond to their weightage of votes (which rises to a fixed extent, assume 1.5x-2x). Differenty types of action could be attributed a different factor to the score, executed unanimous proposals being the highest, followed by any proposal execution, positive feedback to forum engagement, forum participation, so on and so forth.
Using an NFT over a simple non-transferrable ERC-20 allows for the addition of more detailed properties and information. If technologically hard to execute, the NFT can be replaced with X amount of non-transferrable ERC-20, that can be rebased positively or negatively depending on various actions.
Participants can be penalised for missing a specific action, and could bring down their score. Projects could event set a minimum lower limit of score required for voting on proposals, and the only way to get back up to that threshold could be via forum engagement/discussions. Essentially establishing participation in discussion as a need for voting if you’ve previously failed (multiple times) to partake in voting.
Custom properties via NFTs could add information about whether the holder is a high value discussion participant or do they create high value proposals which are executed, etc. With DAOs being divided up into various sub sections based on specialization, you could even see a rise of prominent specialized participants with a quantifiable score for their activity across projects.
To Sybil this system, one would require engagement across multiple wallets, and not just any engagement, it would have to be productive. There is also no economic incentive aligned with Sybilling this system, except to push through beneficial proposals. But multiple wallets with no/low history of engagement would be of very little help and weightage.
The method used by TempleDAO to break up the community into Enclaves, along with the usage of its Observance bot can be a model that can be looked to track participation. Observance also linked wallets to Discord IDs, making Sybil a bit more harder.
I still stand unsure of how much of this is technologically executable, but quantifying governance participation in a project can be beneficial for all parties. Projects want an innovative community, and innovative community members should be able to have a say without depending on their financial ability. At some level, I imagine these governance NFTs can become akin to honorary NFTs of collections made for CT influencers. Members with high governance scores would be wanted by projects, and this makes them easily recognizable too.
None of this is meant as a way to create a barrier to governance, but we’ve seen that not everybody even cares about participating, and clearly not everybody knows what the right decision for the project is. It can be argued that sometimes the binary of right and wrong may be blurred, but a trusted/high-value member’s view of the take should ideally hold more weightage than somebody who hasn’t participated at all, and should not matter on how many tokens they hold.