Detalhes da Proposta
Framework for Central Entities in Decentralized Governance
atlashub
Resumo
Central entities shall not exert undue influence under the label of a decentralized system.
That is, entities who own, manage or control considerable funds in the form of assets that are decisive for on-chain governance shall not dominate decision-making.
E.g.,
🔷 neither IOG nor Emurgo nor the Cardano Foundation shall act as (dominating) DReps or voters in the Catalyst Program
🔷 votes should not be decided by single entities with large assets.
This proposal covers the issue of the role central entities should take in a general sense. It may eventually evolve into several CIPS or governance actions to be implemented on-chain as algorithmic measures.
Motivação
Resolving the contradiction between central elements in a system and the idea of decentralization.
Counteracting detrimental forms of behavior and intrinsic forms of of system failure.
Specifically the founding entities shall adhere to a code of good conduct and practice given their prominent role.
As an example: It should not be the case that CF votes with a voting power of 180M ₳ (obtained by "fiat lux" or "deus ex machina" with the genesis distrubution - an amount not earned through contributions or investments) in a Catalyst Fund where the average voting power of all other participants together is about 50M ₳. This situation effectively means that whatever CF votes for eventually wins and renders the whole voting pointless.
In particular, the following specific problem areas are addressed:
✅ Keep central actors from undermining decentralization
✅ Ensure the diversity, quality and quantity of on-chain activities and initiatives
✅ Prevent self-interest and power games from dominating or derailing development
✅ Maintain direction, stability and growth for the ecosystem
✅ Foster transparent decision-making and allocation of funds
Justificativa
Possible measures for implementation:
(Currently, this is an ongoing collection of ideas. Some may be better, others less suitable, some may have potential, others may prove detrimental upon closer examination.)
🛡 Voting power limit for DReps: Implementing an upper limit for voting power for DReps as described below in a general sense. There are DReps that centralize a substantial amount of voting power acquired as influencers or acting as agents of a founding entity.
🛡 Enabling cummulative voting as opposed to simple weighted voting in appropriate cases (such as Project Catalyst): Instead of voting for any number of proposals with full voting power, participants have to split their voting power among the proposals they want to vote for. For example, a central unit with a voting power of 180M can only use it once on a single proposal or has to distribute it across several proposals. This reduces the influence of big players and is also multi-wallet resistant (i.e. voters splitting their wallets for voting cannot gain any advantage from this). Additionally, cummulative voting counteracts the effects of purchased votes as voting power can only be sold and used once.
🛡 A (relative) upper limit for voting power: If this limit is exceeded, voting power is restricted or participation in voting is excluded. This may be set generally or for specific votes. The limit should not be set as an absolute value (without empirical evidence) but rather as percentage of the total voting power used in a vote to determine the outcome. Example: If the voting power of an actor exceeds 30% of the total voting power used, it is reduced to this value as a limit.
Although this approach may reduce excessive voting power, it may not eliminate undue influence (unless participation is not completely excluded which may also be inappropriate).
Example: If a single voter uses a voting power of 180M ₳ against the rest of the community with a voting power of 50M ₳, the dominant voter with a reduction to 30% still has a voting power of 0.3 * (180M + 50M) = 69M, which is enough to decide the outcome.
What remains is that this method can encourage participation as it becomes more likely for the rest of the community to stay competitive against voters with a substantial amount of voting power.
In the illustration with a dominant voter that has his voting power reduced to 30% of the total voting power used voting remains fair, if the rest of the community has at least 43% of the voting power of the dominant voter [0.3 * (x + y) = y which gives y as (0.3 / 0.7) * x]. When it comes to the example of a (single) dominant voter throwing a voting power of 180M ₳ into the ring and there is a reduction to 30% of the overall voting power, the rest of the community has to bring a voting power of more than 77M ₳ to obtain a fair result.
🛡 Relative upper limit with balanced voting power: If the voting power of a participant exceeds a certain proportion of the total voting power used, it is reduced to that limit and the surplus is distributed proportionally to the other voters. This may eliminate undue influence. [Details and description TBD.]
🛡 Inclusion of historical values: Taking into account active stake from past epochs (snapshots) as an averaging or weighting mechanism may remove wrong incentives for splitting wallets as an immediate bypass of limitations in voting power.
🛡 A separate class of addresses for central entities (such as the founding entities) or institutions that hold a guarantor position. This class only allows the rights in decentralized governance that are appropriate for these actors. The rights associated with these addresses may be adjusted over time or for specific votes.
🛡 Measures to prevent or make it more difficult to vote multiple times with different wallets or identities (needs to be defined).
🛡 Self-commitment by the founding entities to abstain from certain votes (such as voting in Project Catalyst) or roles (being a DRep) - a mechanism for this has to be defined.
🛡 The deposit of a collateral that will be retained in the event of violations (needs further definition or refinemet).
🛡 (Caution: here live mighty dragons) The possibility of suspending, annulling or rectifying votes in the event of irregularities (needs further definition or refinemet).
Registre seu Voto
Comentários (34)
I like and appreciate the ideas presented in the proposal. I think a solid solution will probably involve all parts suggested for mitigating over-powered voting, and probably more. We might also add ranked voting as well, which allows for people's desires to be expressed without the risk of losing out on any single vote.
A potential problem with cummulative voting is that this limits the little person far more than the rich. A person with 10k ada having to distribute their vote reduces their power, whereas someone with 10m ada can vote for a lot of things they want to pass. It does limit the 10m ada voter, but far less than the 10k ada voter.
In a scenario of someone
cummulative voting is a really good idea, but the others I think it is a centralization idea to help descentralization, in other words, a contradiction
One way to do this is to have a power limit to a DRep. Once a DRep reaches the power limit then no more ADA can be delegated to them.
Sorry to be a "Debbie Downer" on this. Most of what you are proposing requires digital identity enforcement in order to achieve some notion of 1 person == 1 vote. In my opinion, a better way forward is to continue along the pathway to multi-resource consensus and then create hybrid voting systems where individual voting power can be limited based upon a privacy preserving digital identity system. See Minotaur: https://dl.acm.org/doi/pdf/10.1145/3548606.3559356
This proposal seems to involve a considerable amount of handwaving, introducing draconian measures under the guise of good intentions. Instead, I urge us to consider a more constructive approach. The founding entities are indeed a significant concern, but rather than imposing strict regulations, why not require these entities to shard their holdings into a capped amount per address (e.g., 1M or 10M ADA per address)? By doing so, they could delegate each address to different decentralized representatives (dReps) within the ecosystem. This method would not only decentralize governance power but also respect the principle that ADA grants governance rights to all holders indiscriminately. To enhance the effectiveness of this approach, we should establish a public platform that tracks all founding wallets and their distribution of dReps. This transparency is crucial in building trust within the community and ensuring accountability from those in power. Moreover, while this solution promotes decentralization, it also mitigates the risk of concentrated influence by founding entities. It allows for a more equitable distribution of governance power without alienating any stakeholders. In terms of implementation, it would be essential to develop clear guidelines on how sharding would work and ensure that it is technically feasible within the existing Cardano framework. Challenges may arise regarding the management of these addresses and their delegated dReps; however, with proper oversight and community engagement, these hurdles can be effectively addressed. Ultimately, my proposal aims to foster a more inclusive governance model that aligns with Cardano's foundational values while addressing the concerns surrounding the influence of founding entities.
I agree with the general sentiment, but we are heading deep into "should or should not do" territory instead of "Can't be evil, won't be evil" because none of this is enforceable on the protocol level and Delegates at the Intersect Convention have just voted on "All ADA holders are the same" - As for founding entities this train of limiting power has left the station the moment Genesis ADA was moved into circulation, so now we are talking blacklisting wallets for certain on-chain-actions, which opens this whole ecosystem up to a range of shenanigans I am not comfortable with, especially not with the current climate between entities. As stated in the proposal, setting limits does not solve anything at all, because you can always split wallets, we have been through this with Multipools already.
There is another argument to make here: In case we ever see real attempts of whales (lets say Binance for example) messing things up, I do trust in the founding entities to act in the best interest of the ecosystem and step up with all their voting power in governance. CF has explicitely stated they will do so in the first Constitution vote if the need arises. A security measure we should not abandon lightly, just because some of us disagree with CF voting in Catalyst as they did. In the end they were just tired of convincing huge projects to build on Cardano (as is their mandate) just to see them loose the Catalyst vote to some random Youtuber with a huge following. Quite understandable really.
As for Catalyst the solutions are very different from governance: their voting isn't even happening on Mainnet, so they can set whatever different voting mechanism they like. And by "they" I mean us, because going forward Catalyst will have to convince dReps and/or Intersect (or whoever puts Catalyst in their budget proposal) that their system benefits the right players in this ecosystem.
1 ADA = 1 ADA 4ever
i agree with the framework. Decentralization is everything when it comes to crypto.
I completely agree with this framework. Decentralization is at the core of blockchain governance, and allowing central entities to dominate decision-making contradicts this principle. Implementing checks and balances to prevent undue influence is crucial for maintaining the integrity and fairness of the system. This proposal is a significant step in the right direction.
Interesting idea.
technically yes
Only joined looks very scientific
How do we vote as a deligate? I want everything as decentralized as possible. AI is coming we need the people in charge
Not sure if this belongs here, as a general governance mechanism for the Cardano Blockchain. This shall be proposed and voted within Catalyst as a particular project. If not we will be debating here any particular project, which is almost impossible nor desirable in my POV.
agree with the sentiment, would like to see more discussions on the specific numbers
agree with the sentiment, would like to see more discussions on the specific numbers
The most democratic mechanism is a relative upper limit with balanced voting power because: Balances the influence of big and small players. It motivates the wider community to participate. It maintains decentralization without completely excluding large players. The second strong candidate is cumulative voting, which promotes more strategic decisions and reduces the risk of manipulation.
I’m glad to see this conversation getting started. It’s difficult to answer these questions and even more challenging to determine a fair voting mechanism. Additionally, what qualifies as a central entity? Is it limited to founding entities, or does it include large CEXs, DApps, etc.? As far as voting mechanisms are concerned: The limits on voting power in proposals can easily be bypassed through wallet splitting. I tend to slightly favor cumulative voting, but I believe it would only work for funding initiatives like Catalyst. I don’t see how it could be applied to governance actions, such as parameter changes. Lastly, the argument that large entities with ‘skin in the game’ (due to significant investments) should have the right to exert influence because of their vested interests is a compelling one. I'd love to have this conversation moved forward with input from more people and with deeper conversations that cannot be conducted through the comments section here.
On a more technical aspect, your proposal should define very clearly what you mean by central entity... You seem to define it by examples limited to funding entities, but any company may be a central entity e.g. SPOs (whatever it is a single or MPO), any DEFI protocol team org, ... does it relate to the legal definition of the specific location the organization is registered ? (In such case, Binance and Binance US may be different entities), does it relate to the extent of voting power ? (if I would have 1B, I would feel natural that my voice is heard more than someone have 10000 ADA, skin in the game...), does it relate to the way ADA were acquired (airdrop, giveaway, bought), ... clearly defining the framework will clarify extent and limit of mitigations and will give a better view of what could be reasonably done...
Even if I understand concerns at the origin of the initiative, I am questioning why resources should be spent on proposed mitigations... I mean that governance voting power distribution is not more at risk of centralization than staking power distribution to secure the chain (which is even more critical), and none of actions above would prevent bad actors to use multiple DReps to bypass some of the proposed mitigation (the same way nothing prevent MPOs to drive multiple single SPOs). Cardano decentralization took time under the oversight of funding entities, and I strongly think that it will take time to better distribute voting power, however, I equally strongly think that community effort should be oriented towards ppl on-boarding rather than towards the development of complex rules and railguards which may not garantee what is the desired outcome... Finally, let put things in perspective, Avengers MPOs stand for 1.39B stake, Binance for 786M, and I am feel that CF, IOG and Emurgo have more aligned interests with community than CEXes to ensure cardano future, therefore, I amequally fine with hearing their voice as part of our governance if it can help to counter balance CEXes voice
I agree that we need such measurements, and rather sooner than later. I think your proposals are all good starting points but they need further discussion. However I do not yet see how a relative upper limit (or even a concave voting power function) could resolve the problem of just splitting wallets (since as far as I know any diminishing/concave function is subadditive). However, large holders like the CF are probably encouraged to not engage in activities like splitting wallets for more voting power, and therefore the relative upper limit could indeed be a viable solution. But if there are "bad actors", can the splitting problem only be solved with some sort of oversight, identity integration or collateral deposit/escrow (which is all not optimal)? As for the cumulative voting option, could it introduce the problem that 1 ADA != 1 ADA depending on whether it was already used for voting? I was thinking about a voting-cooldown, that new and previously empty wallets could not vote right-away. However this seems more like introducing an unnecessary difficulty for new users that is easy to overcome by just waiting once. Some sort of community encouraged self-commitment by the founding entities could certainly be helpful, maybe even with some sort of "escrow"/deposit that the community could vote to send back to the treasury in case of violation?
Be good to run a few inconsequential topics for vote to engade partisipation and observe results
Glad this converstaion is happening. Here are few questions we should be asking ourselves as we consider this. 1. - If upper limits were implemented how would that impact the decisions of other large organizations and investors entering or considering developing on our platform? At the end of the day this is still a field of dreams and are anticipating a crowd to show up soon, so lets make it a warm and inviting place for big and small alike. 2. - Can it be enforced? Any solution that can be easily bypassed should not be considered. Limits on wallet size or identifying wallet owners both fall into this catagory.
I wholeheartedly agree with the Motivation provided in the proposal. I will be very interested to see how it evolves and shapes things to come, particularly give recent turns of events.
This is provisional work in progress that needs to be refined in order to become a concise guideline. Deliverables include
- What are "central" entities, i.e. how are they precisely defined (first ideas for this do exist),
- What is expectet from these entities, what is permissible for them?
- How is that enacted? That clearly needs input from the community.
Você está pronto para participar?
Construindo juntos para impulsionar a Cardano.