Uniswap Delegate Reward Initiative - Cycle 3 Discussion

Governance-Meta
21

I agree that increasing the slots to 50 would be a valuable addition @Argonaut . However, @stablelab’s reluctance to increase the number of delegates, combined with the current delegates not being incentivized to vote for this expansion, suggests that it may be too late for this program. It has already been cartelized. Why would the delegates vote to decrease their pay and increase competition? Only increasing their compensation would make sense for the delegation.

A few points to consider:

  1. The current criteria structure incentivizes two main components: creating proposals and voting. Creating proposals distinguishes many delegates from others, so I would argue that this is the most important criterion for remaining in the cohort as a tiebreaker. UNI DAO is currently incentivizing proposal creation and the passing/voting of proposals.

  2. How many people are setting the point criteria for this program? Is it just @Doo_StableLab? Is it a group of 2 or 3 other groups deciding? Is it the all the delegates deciding?

  3. Aside from the Layer 2 deployments/incentives and the base fee tier lowering proposal, most proposals seem self-serving, aimed at creating paid committees with little oversight or promoting/charging for a delegate’s software business (e.g., charging for analytic analysis using proprietary software on previously passed proposals, such as @stablelab’s Forse Analytics).

  4. If the current delegate structure remains unchanged over the next 12 months and a fee switch is enabled that ties into delegation, there will be very few options for new delegators to choose from. This would 100% entrench the current small number of VC-backed delegates into a governance takeover, as they would not only have the VC delegates but also a significant wave of uninformed delegators. Once this occurs, the delegation could vote to allocate themselves as much as they want from the UNI treasury for UNI incentives. UNI incentives for this program will grow as self-importance increases.

  5. Keep in mind we are also paying a committee to research Uniswap Treasury management. @Doo_StableLab and @AbdullahUmar proposed this committee, and it has @karpatkey @pennblockchain. Who are also paid as part of the delegate reward initiative!

Cartelization can occur purely through the incentive of wanting to be paid more! Monopolies and cartels emerge naturally through economics, not by delegates communicating behind the scenes.

The biggest risk to the UNI treasury is the cartelization of delegates! I have personally witnessed this through EOS. A cartel “strategy” undermines a project’s long-term viability, as value leaks to third parties without accounting for value alignment or empowering core users. Legal uncertainty, and legal avoidance has molded the DAO onto a path of failure.

UNI DAO would benefit from value-aligned proposals from delegates who share its values. Currently, we are incentivizing proposal-making and voting from a majority* of non-aligned delegates (e.g., NOT: liquidity providers, Uniswap application developers, or frequent Uniswap Protocol traders/users).

Objective:

Avoid cartelization at all costs; increase the number of delegates to choose from, with a high priority on value-aligned delegates—especially if a delegate fee switch is eventually enabled!

Just as a16z played a role in not stopping the Compound governance attack. I believe they may also be unintentionally complicit in a Uniswap delegate cartelization attack over the next year, purely due to economic alignment. This program in its current form is a canary in the coal mine.

2 Likes
22

I don’t think you saw we are against increasing the compensation. We believe there are other focuses to improve on.

Either way, the point about aligment and quality was to reference your point so a bit confused why you are criticial of your own idea. The idea is to have delegates that are aligned. Also, in practice, in case for example, we divide it by 50 delegates, unless there’s strong restraint, the delegates will likely ask UniswapDAO for higher budgets.

23

I am considering the long-term implications of the structure of this program. This program does not address double dipping in committee compensation, conflict of interests, lacks critical analysis requirements of the proposals being voted on, and shows little effort toward a merit-based solution. I am okay with the costs of this program increasing if its goal is to combat cartels, enhance competition, and attract value-aligned delegates. A governance mechanism needs to be designed, similar to how incentives are provided to LPs. However, the premise of how this program is run is the antithesis of decentralization.

Who is the “we” making the decisions? Is there a vote? You propose a 30% reduction in compensation; are you determining how much is paid in the program as the proposal leader, and compensated delegate?

Please feel free to DM me, I feel I have stated all I can on the risks of this program.

1 Like
24

It’s great to see many discussions happening around this topic. We would like to propose an idea of expanding the number of delegates eligible to receive rewards in Cycle 3 and introducing a tiered system to accompany this change, which is somewhat similar to ideas from other delegates.

Problem Statement

Under the current criteria, it’s anticipated that many delegates will end up with the same score in the next cycle. This is because the number of delegates considered reasonably active, based on existing standards, is expected to increase. There’s a possibility that many delegates will cluster around 6 or 7 points due to points earned from activities like voting, which even new delegates can achieve.

Objective

As a fundamental premise, we believe that the goal of this program is to enhance the overall health of the Uniswap DAO. From this perspective, let’s examine the proposed measures.

Firstly, we think that increasing the slots in the Delegate Reward Initiative can simply boost the DAO’s activity level. This is evident from the fact that new delegates, including us, had started participating since Cycle 1.

However, merely increasing the number of slots might only inflate the budget, and providing the same amount of rewards to delegates who may be less active than those already selected might not be cost-effective.

To address this concern, we believe that introducing a tiered system would be beneficial. This system would adjust the reward amounts by dividing delegates into different ranks. By implementing this system, we can maximize the program’s effectiveness by providing sufficient rewards to delegates who have been and will continue to be highly active, while also expanding the Reward Delegate slots to include new delegates beginning their activities.

Agreement with GFX’s Proposal

We agree with GFX’s idea to change the reward amount per delegate based on performance. Introducing a subjective component into the compensation structure can incentivize delegates to provide more value to the DAO.

If expanding the budget is acceptable, perhaps we can increase the rewards for the upper-tier delegates while keeping the lower-tier delegates’ rewards at about the same level as now. This approach allows us to reward highly active delegates more generously without significantly increasing the overall budget.

Specific Measures

We think it would be best to discuss specific numbers after gauging reactions to the overall direction of this proposal.

As a general idea, we are considering increasing the budget rather than operating with the same amount as before, aiming to limit the increase to only what is necessary. Regarding the tiered system, unlike models in other DAOs like Sky (Maker), which have significant disparities in rewards, we envision a structure where both tiers receive rewards close to the current amount ($6,000).

For example, we could expand the number of delegates in this initiative to 20. Delegates who score higher based on the current criteria (e.g., 8 points or more) could receive higher rewards (around $10,000), while delegates with fewer points could receive rewards similar to the current amount (around $6,000). This approach could maximize Uniswap DAO’s activities and quality of them while keeping the budget in check.

Alternatively, we could consider making the top tier consist of the top 10 delegates.

We welcome any opinions or questions regarding the above idea.

1 Like
25

As mentioned in the last cycle’s discussions, I think you should have the UNI earned vest over time, like @eek637 also brought up.

For the subjective measurement, the delegates should not be ranking each other. You could have pure UNI token holders (no delegated UNI) vote on which delegates they feel have done the best work. Or even have people vote with their LP shares. This puts more power to direct economic contributors.

-Erick

26

That’s a list of fair concerns to address and view as well.

I do think it’s better to work within certain budget limit to optimize within because once again, DAO treasury is a public good. The slippery slope I am worried about is it becoming bureaucratic where instead of checks and balances, it becomes bureaucratic. So if the budget is let’s say 1 million per year just for simple demonstration. It would be better to agree to allocate 300k to tools and mechanisms needed to enforce such and 700k to delegates as rewards rather than 1 mil to delegates and then additional budget for tools and mechanisms.

It’s of course the delegates that make the decision whether to accept such version when it comes to vote. No perfect system but will incorporate various feedback to have a model for the next cycle when it gets closer. The focus of Cycle 2 was increasing access for new delegates as you mentioned (maintaining high participation rate for past 6 months is very difficult while for newer delegates, it only counted for when they joined with understanding there will be another cycle which meant they were likely to vote more actively) , which resulted in the newly rewarded delegates being all new delegates less than 4 months joining.

So if this is the “cartel” as you mention, why are we trying to dilute or restrict ourselves? We understand that skeptics will always view with skepticism, but we do believe DAO’s long term interest is also interest of all involved parties.

27

Thank you for your insights and we also understand that criticisms are needed and beneficial for governance

1 Like
28

Having reviewed the insightful feedback from fellow delegates, I’d like to share my perspective on how we can balance effective governance with broader participation and alignment within uniswap community.

I understand the concerns about potential collusion, conflicts of interest, and the concentration of delegate representation. If unchecked, these issues could lead to outcomes that harm the DAO’s integrity. However, imposing excessive restrictions might inadvertently hinder beneficial participation. We need to find a balance that promotes diversity and encourages new delegates from various backgrounds to contribute effectively.

While oversight is important, we also need to promote and support more participation. This brings me back to the Value-Aligned Approach—which I strongly support. But before determining what those values are, I believe we should first reflect on the role of a delegate. In my view, key responsibilities include:

  • Active Participation: Regularly voting on proposals and engaging in governance activities.

  • Staying Informed: Keeping abreast of the DAO’s developments and understanding the implications of proposals.

  • Supporting Initiatives: Being available to create or back meaningful initiatives that advance the DAO’s mission.

  • Providing Insightful Rationales: Offering thoughtful explanations for voting decisions, which help shape community sentiment and guide voters—especially during critical moments.

By focusing on these core responsibilities, we can align delegates’ actions with the long-term health and success of the DAO.

I agree with @eek637’s suggestion of implementing vesting for UNI tokens. Vesting aligns delegates’ financial stakes with the DAO’s long-term health, incentivizing sustained commitment and discouraging short-term opportunism.

Ensuring Separation of Interests:

In some DAOs, rules prevent service providers from applying for committee positions to ensure a clear separation of interests. Adopting similar guidelines could help mitigate conflicts of interest within our governance structures, fostering greater transparency and trust.

The central challenge is how to effectively measure the value of delegate contributions. I partially agree with @GFXlabs’ Delegate Rank model, which resembles systems like Coordinape. However, unlike Coordinape, GFXlabs’ model proposes anonymity, which could increase the risk of collusion behind the scene.

I believe transparency in ranking and reasoning is essential. Just as we publicly justify our votes, we should also be open about why we rank certain delegates highly. This openness can discourage collusion and promote accountability. If we opt to this delegate ranking model, I suggest it should be for distributing additional bonuses pool, while maintaining a lower fixed compensation based on straightforward, objective metrics.

Tracking On-Chain and Off-Chain Activities:

To gain a complete picture of delegate contributions, it’s crucial to track both on-chain and off-chain activities:

  • On-Chain Metrics: Voting records, proposal submissions, and other blockchain-based actions. This is an area where us (Curia Lab) is actively working; tracking and analyzing on-chain governance data to provide insights and transparency.
  • Off-Chain Metrics: Forum engagement, proposal feedback, and overall governance involvement. We are also exploring methods to effectively capture and analyze off-chain activities to help the DAOs we currently work with. By integrating both on-chain and off-chain data, we aim to offer a holistic view of delegate contributions.

By capturing these activities, we can develop a comprehensive understanding of each delegate’s impact. Importantly, all metrics should be fully transparent and traceable, allowing delegates to verify how their scores are calculated and address any discrepancies.

We can draw inspiration from existing frameworks, such as:

  • Arbitrum’s upcoming Delegate incentive program v1.5: This model offers a detailed approach to evaluating delegate quality through proposal feedback and other contributions. It emphasizes comprehensive evaluation criteria that capture both quantitative and qualitative aspects of delegate performance.

Again, I believe we should establish clear, accessible criteria for evaluation:

  • Objective Metrics: Maintain fixed compensation based on measurable activities like voting participation or impactful feedback frequency.
  • Qualitative Assessments: bonuses based on the quality of contributions, evaluated through transparent processes.

This approach would ensures fairness while incentivizing delegates to go above and beyond in their roles.

2 Likes
29

We would like to share with the community a suggestion that we have to make on this subject, which we believe should be evaluated in order to correct an inconsistency that we have observed in the last two cycles:

Increase the Weight of Voting Participation Metrics:

In the previous cycle, the points assigned for low voting participation were very generous. For instance, someone could have a 70% voting participation in off-chain polls and an 80% participation in on-chain polls, yet the penalty would be only a reduction of half a point, while attending a single community call or attending a conference or in-person Uniswap event would grant you a whole point. The main responsibility of a delegate is to vote, as doing so reduces the risk of a governance attack. Therefore, we shouldn’t consider a delegate with poor voting practices a good candidate for the program.

A better approach would be to have tiers that range from 100% voting activity down to 70%, decreasing the number of points proportionally until reaching zero at 70%. A formula like this can be used:

Participation Points = ((Voting% - 70%) / 30%) * Points for the Category

For example, in the case of 80% voting and 2 points available for the category, the total participation points would be:

((80% - 70%) / 30%) * 2 = (1/3) * 2 = 0.67

In the previous cycle, this level of participation would have received the full amount of points.

Consequently, as we said above, the main responsibility of a delegate is to vote, so we understand that the weight of participation in both onchain and offchain voting should be increased in comparison to other activities that, although relevant, should not have an equivalent weight to direct participation in governance.

6 Likes