Upgrading the Community Staking DAO Allocation Delegation Strategy

Hello everyone!

I’ve just published a three-month report on the results thus far of Regen Foundation’s Community Staking DAO Allocation delegation strategy.

In short, with the soon-to-be-expanded validator set, we need to upgrade our policy, and we need your input!

Option 1—keep it as is, delegating to the 6th through 50th top validators, compensating for ranking and commission.
Option 2—expand current delegation algorithm to the new validators, staking to the 6th through 75th top validators, keeping
Option 3—any other idea you can think of!

I’ve raised this issue a few times before, and I don’t feel we’ve yet come across community censuses here.

We should be having an AMA on this as well, tentatively scheduled for the 19th (more on that once we firm up a date).

Please add your comments and proposals and discussion below!

8 Likes

Bliss Dynamics definitely supports option 2, an expansion of the current algorithm to the 6th-75th top validators.

We strongly believe that incentivising more validators to participate in Regen Network is good for the ecosystem because it attracts stakeholders of various skillsets to the project. Having sais this, motivating them to contribute further is also important which is why we have been supportive of an accreditation process where validators should justify their contribution to the project in order to receive any foundation delegation. It’s sad to see people gaming the system in other projects to profit and provide no further benefit.

We don’t believe that any extra compensation for new validators should be necessary due to missing out on the first three rounds (as described in the extreme case in the report) as this is the reward for those who have contributed well so far, although we would like to see a rolling delegation update in future (we’ve talked about reassessing this each year) where existing delegations are recalculated so that we don’t see early validators resting on their initial work and stop providing benefit to the ecosystem, but also allowing newcomers at a later time to benefit from their contributions.

3 Likes

Bliss,

I don’t know what the current algo is, but support your “accreditation process”.

Further, Notionally speaking, notional considers its contribution to be this:

and

and

Thanks and look forward to signing some regenerative blocks!

PS: I just read the three month update.

Notional thinks that it can provide better for Cosmos communities with a flat 9.69% rate across all networks.

1 Like

I personally support option 2, however I strongly believe we need to develop a stronger framework for tracking community contribution, reputation, validator ops auditing etc, then work to update the delegation strategy to consider criteria and have a variable delegation rate according to merit. However I dont believe we should wait on expanding the delegation algo to include new validators.

This statement is not a statement of policy from RND inc, or my position as a foundation board member, but as a community member.

Generally speaking I am excited to four things: more validators, better decentralization of stake across validators, validator engagement with our domain specific value prop providing services to the community beyond block validation, and lastly: validator engagement with governance and upgrades of the protocol in service to regen network’s mission to create market mechanisms to drive natural climate solutions.

Towards those ends I believe three things must be true to achieve more decentralized stake distribution: (this is a comment that is more broad than the question at hand in relationship to Foundation delegation strategy of the CSDAO allocation)

  • Regen Ledger must have an on-chain validator reputation system
  • Regen Ledger must leverage groups module to create governance system for regen registry and eco-credit asset curation, and governance of community spend pool and other key functions of the ledger efficiently with representative democratic governance instead of unicameral token holder voting on ever topic.
  • Regen Ledger must embrace a quadratic stake model to incentivize distribution and decentralization of delegations.
5 Likes

Hi all,

I began to be part of the validator set quite recently, but sharing my opinion on this in case it’s worth it.

I think option 2 is a nice starting point, it helps with decentralization and incentivizes validators to be affordable and willing to join, and in return those can give back to the community in terms of: support, building tools, governance, promotion, etc.

I agree with @Gregory_Regen comment.

3 Likes

Hello everyone! I support option 2 - this will motivate all new validators. But I would like to add that there was some kind of accumulative points system, perhaps for activity in the form of project promotion.

We just hosted an AMA on Discord. I was joined in a panel discussion by my fellow Board Members at the Foundation—Kei Kreutler (Strategy at Gnosis) and Gregory Landua (CEO at Regen Network Development) as well as Chris Remus (Validator with Chainflow).

I took some notes, which I’ll share here:

Kei, Gregory, and Chris are all in support of option 2. Kei would like to see the development of a reputation system. Chris is interested in tracking validator performance; he pointed out that often times people equate the validators with the most stake as the most performant, and this simply isn’t the case (many validators with small stake are still very well-run and big contributors to the community).

I pointed out that the question at hand is how we upgrade our current delegation strategy, but mentioned that this is an iterative process, and we can certainly continue to upgrade as this turns into a redelegation strategy. For example, I see it as unlikely that we’ll be able to pull together a comprehensive reputation system over then next few weeks, but this could be something that is integrated by the time we get to relegation.

Gregory expressed interest in a quadratic cap model—a way of addressing validator centralization at the consensus mechanism layer. Of course, this is a big undertaking, and is more of a mid-term goal. Gregory’s interested in focusing on other ways of achieving a health Nakomoto coefficient in the near-term.

I raised the fact that, the Foundation can’t do anything directly to remove stake from top validators, but we can make a difference in the distribution of staked tokens.

The conversation then turned to barriers to entry for the bottom of the set. Currently, the 74th and 75th validators only have 10 REGEN delegated. This means the barrier to entry is extremely low. Chris pointed out that this shows significant dedication on behalf of these validators—they’re losing money doing this, so they must be in it for other reasons (such as caring about our mission at Regen Network around planetary regeneration). It would be nice for the Foundation to consider delegation all the way down through the 75th validator, but how much should we delegate to number 75, as that will effectively set the barrier to entry for other validators?

We also discussed the race to the bottom that other validator sets have seen. Osmosis had to implement a minimum commission of 5% because otherwise big validators would go down to 1% or 0%. A recent network governance vote to lower that to 1% failed because of the way it biases towards centralization. Chris mentioned that it is possible this race to the bottom hasn’t happened yet on Regen because we’re still a bit under the radar. He mentioned that on Solana, it is very clear who is operating validators to attract stake, versus those focused on own profits (they’re 100% commission, no name, just an address). Apparently Akash will only delegate to people at 3%; Chris feels that this is low.

Another related point we discussion is how are commissions used? For example, Akik Takat has committed to reinvesting profits into the mission of the ecosystem, so a higher commission is justified, especially if these commitments can be audited.

Robert brought up that Cardano hasdynamic block rewards, to have more decentralized validator set, and that Sunny had been highlighting this in Cosmos, but that discuss has fizzled out. Robert spoke to the need for education: avoid anonymous validators!

There’s a need for digital identity. One example Dan mentioned is Aneka’s integration with KeyBase.

Dan gave an example of an exploit here, how with Hive blockchain, while still on Steem, Justin Sun came in, created a bunch of validators, delegated to himself, essentially took over the blockchain. (More on that here.)

Discussion then turned to the question of, should Community Staking DAOs run their own validators? Kei responded that this could happen eventually, and that we’re going to start with a buddy system (DAOs paired with existing validators).

Paul shared that he wants to run a validator because it helps justify spending attention on the network, and Chris echoed this sentiment. Chris said that we should do everything we can to make operating a validator accessible. Chris got into PoS because PoW was becoming inaccessible—now there’s the same trend in PoS.

That’s the extent of the notes that I was able to take during the call. If anyone else has something to add, please jump in here!

4 Likes

I just wanted to point out that I quite agree with the minimum commission. Delegators usually make their validator pick decision based on total stake amount and commission.

Small validators then tend to compete with the lowest fees possible (even at 0%) in order to be able to attract delegations and be able to climb on the validator set, we’ve suffered from this and currently running a validator at 0% fee in other network.

I think setting a minimum threshold is a good measure but has to be backed by a DAO delegation strategy that supports small validators to grow, if not they won’t be able to compete because delegators would make their decision mostly based on the delegation amount only.

I also think it’s a nice idea to have measures for the delegators to make their validator pick decision not only on delegation amount and comission…reputation, performance, community support, building tools/documentation/…, making regenerative agricultural tasks, governance, project promotion, etc.

Akash validator set ordered descending by commission, shows the great competition in terms of validator commission

2 Likes

Right now our algorithm rebalances based on commission down to 3%. In other words, someone with a 10% commission will get half the delegation of someone with 5%—so that revenue stays flat. Would you recommend that the Foundation discontinue factoring in commission (one concern has been that this could incentivize high commissions; the current arrangement is neutral in regard to validator economics). Another possibility would be to pass a governance proposal for the network setting a min (such as 3%). Would love to hear your thoughts here.

Also comment that there are other networks that has a top up mechanism to allow distribute the users delegations more evenly, but I guess this is not implemented con Tendermint/Cosmos.

The top up mechanism consists that the delegations that are above a certain delegation threshold are rewarded with a lower percentage. For example:

  • delegation top up threshold: 100k
  • tokens delegated til 100k rewards: 25%
  • tokens delegate over 100k rewards: 15%

that way, when the threshold of a validator is passed, the more staked tokens in validator the worse for the delegators because the overall rewards decrease tending from 25% to 15%

1 Like

Yes—we were discussing these sorts of mechanisms on the call today (and the identity requirements to effectively implement). There are a number of members of the Regen community that are inclined towards this direction in the long run. What networks have you seen this work well in?

I think it’s good to rebalance the way it is now, but maybe not in a linear fashion. I would implement some kind of curve (quadratic, …) that would benefit a bit the ones with the lower commission to avoid encouraging high commissions.

Anyway I think it would be better to set a low steep curve on the algorithm, and try to adjust it with time.

An important point to have in mind in my opinion is that validators with lower commission would get more DAO rewards than validators with higher ones; having a lower commission also would attract more delegators, so it could turn into kind of a rabbit hole for all validators setting the commission to the minimum (3%).

Seems like a non easy thing to tune.

I think the best solution would be to take into account at least: the commission and the delegation amount. The lower the commission the better, the higher the amount the worse.

Elrond Network is an example of this top up mechanism.

All these points makes great sense and are from real experiences. I am also in favour of option-2. I believe option-1 is not a best choice as there was priority for all the early supporters in initial delegation plan. Also the initial supporters got token incentives too. So now it’s better to reconsider that and consider all the validators (6-75) for delegation. We can always consider long-time supporters as a factor but it should not be used for capping IMO.

As @Gregory_Regen pointed out, decentralization, governance activity and contribution to the network & community should be top priorities when designing the delegation strategy. But as the systems are not in place (or takes more time) to measure/quantify this value now, we can go with old model (but consider full validator set). We can may be enforce min-commission, max-commission requirements for becoming eligible for community staking DAO delegation.

2 Likes

Thank you everyone for you contributions! There’s a clear consensus in the thread here to go with Option 2—expanding delegations to include validators 6 through 75. Read more here.