Discussion Content: Creds & Governance at POKT Network

As we start to prepare our proposal for the governance upgrade we wanted to provide this overview and summary to allow people to engage in discussion around specific objections or concerns with this design.

We welcome you to attend the community call next week (Thursday 15th) where we will walk through the system, invite feedback, and open the floor to anyone to ask questions or get the information they feel they need to be better informed about this important DAO decision.

We hope to see you there!

TL;DR

POKT Network’s proposed governance upgrade is built on three principles:

  • Governance power cannot be bought, it must be earned.
  • You earn governance power by driving impact, and it should grow with your impact.
  • This requires a lot of thought “under the hood” but should be automated as much as possible to be frictionless for participants - you work to earn your governance, but claiming what you’ve earned should be easy.

Why do we need this upgrade?

There are four main challenges with the current system (more info in our original post here):

  • Gaps in representation: current paths are not appropriate for many stakeholders (investors, grantees etc) and outright exclude others (i.e. anons who cannot verify via Discord selfies).
  • Inertia in representation: earning a vote takes a long time and once it’s earned it persists even after you stop participating. In the last 12 months we had only ~5 new DAO voters.
  • Imbalances in representation: it’s much easier to earn a vote for some types of stakeholders compared with others, so we get imbalances.
  • Manual processes: the manual nature of the current system produces frustration and can deter vote seekers.

How do we define impact?

There are many different ways you can positively impact the project.

We look at impact in 3 areas, or “houses”:

  • Builders: drive impact by building the ecosystem and protocol. These people are shipping code, building products / tools, winning POPs (RFPs), running experiments, and much more. They earn governance for these types of impact and hold significant weight (40% of total).
  • Stakers: drive impact by productively using the POKT token to support or secure the network. These people burn tokens as Gateways, stake POKT as Servicers or Validators, provide liquidity to our wPOKT pool, and much more. The network cannot operate without stake, so they too have significant weight (40% of total).
  • Citizens: drive impact by understanding and promoting our mission, values, and vision. Every voter must show this base level of understanding. So every voter (Builder or Staker) must also be a Citizen. But since barriers are low, being a citizen carries less governance power (20% of total).

Because any member of the community can earn power across any / all of these houses, we called the system ‘3D Governance’.

How is that simple?

There are a couple of things you need to do at the start but, after that it’s (mostly) automated.

No more manual actions, bottlenecks, or follow-up.

Your governance power accrues automatically as you ship code, stake in LP pools, win an POP or do any of the many many things that positively impact the project (more examples later).

It will also fade if you become inactive.

Here is the process:

  1. Create a Gitcoin Passport. Retain your anonymity while we make sure that nobody can game the system by using multiple identities.
  2. Create a digital ID on mygateway.xyz . Connect your wallet(s) and any other sites you wish to the Gateway protocol; this provides the infra for Cred issuance and automations.
  3. Earn your Citizenship Cred by completing 2 automated ‘quests’.
  4. Automatically receive Builder Creds. When you launch a successful proposal, win a POP, complete a Socket, have a PR merged, etc. - including within the 12 months before launch of Creds - all of this will automatically be captured and issued to your ID or wallet.
  5. Automatically receive Staker Creds. When you run your own nodes or stake with a third party your credentials are also automatically issued to you.
  6. Vote in Snapshot. Exactly the same as today - to vote you simply use your Ethereum wallet and vote. Your voting power is automatically calculated based on your Creds.

Into the detail

As anyone who has ever built anything knows, simple is HARD and it has to be supported by a LOT of deep thought, detailed design, and careful execution.

This section is for those who want to dig into that.

Perhaps the biggest challenge we sought to tackle was to ensure that the system is capture resistant and that power is balanced fairly across individuals and groups.

This balance is achieved in part through:

  • Individual Voting Weight
  • Collective Voting Weight

Individual Voting Weight recognises that all impact is not equal and creates the granularity needed for power to grow with your impact.

For example:

  • Weighting of 5 points for shipping a PR to one of the main POKT Network repos (because of the importance of this to the performance of the protocol).
  • Weight of 1 point for being approved to open a Quick Grant (Socket) and an additional point if it’s open for more than 2 self-reporting periods.

While there are many more examples of this individual weighting, and many more ways to earn voting weight as a builder, each is just a representation of this same “impact = what?” question.

Collective Voting Weight sets constraints on how much voting power a group can have in the DAO. For example:

  • We plan to give 20% of voting power to Citizens, 40% to Builders and 40% to Stakers at launch. As such, no single group can achieve a simple majority on its own.
  • Stakers represent all of demand, supply and liquidity and, within the 40% allocated to Stakers, we plan to sub segment further:
    • Servicers & Validators collectively holding 16 percentage points
    • Gateways collectively holding 12 percentage points
    • Liquidity providers (LPs) collectively holding 12 percentage points

This is the full and complete set of collective voting weight constraints.

We also shared more high level information on the concept of citizenship, personhood, and this idea of productive capital in our blog posts.

You can also find the full and complete detail in our forum posts about the concepts of 3D Governance (PGOV1), Modular Implementation (PGOV2), and Personhood (PGOV3), as well as the Citizenship Specification (PGOV4), Builder Specification (PGOV5), Staker Specification (PGOV6), and Voting Specification (PGOV7).

Final Thought on the Future

You may have seen us talking about Modularity in Governance.

We’ve used that approach to design this system in a way that will allow us to iterate as we learn, in a modular fashion, without breaking the system. We don’t know as much today as we will tomorrow, and we need to retain flexibility while not undermining our legitimacy.

This means that, if you disagree with one of the parameters referenced above (e.g. 40% to Stakers), you don’t need to reject the whole system.

All of the parameters are designed to be calibrated over time.

More on the modular governance approach here.

Once again, we strongly encourage you to attend the community call next week where we will share a walkthrough of the system, welcome further feedback, and open the floor to anyone to ask questions or get the information they feel they need to be better informed about this important DAO decision.

We hope to see you there!

Final Notes:

It is worth noting that the governance power in the Staker house does not include all token holders; it is only linked to productive capital (staked capital). Holding a token is not enough, it needs to be deployed / staked to support the network.

It is also worth noting that, within the Staker house, unlike raw token voting (where each token held correlates to more power) we apply a method that is similar to quadratic voting in order to flatten the curve and more evenly distribute the balance of power.

10 Likes

Thanks for this clear summary of the 3D voting system!

I currently do not have an strong yes or no opinion, I lend towards the yes, because the system seems logical and representative. I do not have any objection on its concept, however, I’m yet to go through the important details like implementation and deployment. I will later comment on this once I;m ready.

In the meantime I would like to read the opinions against this system, in the last community call there was some talk around this and @Cryptocorn made a pool in the unofficial telegram group around vote weighting (not related to 3D gov I think). It would be great to read those opinions in a more “formal” medium, such as this forum thread, otherwise it is only noise IMHO.

5 Likes

Thanks for this, @RawthiL!

Would love to hear your thoughts when you have a chance to dig in to the details.

And it would be awesome if @Cryptocorn, or any of the community, could help bring some of that discussion in here / to the call on Thursday.

I firmly believe that together we’ll get to the best outcome.

1 Like

I got a concern about the “Stakers”. The 3rd party node running host providers / staking pools are staking a huge amount of POKTs for their customers. What if a 3rd party staking service provider leverage its customers’ POKTs to forge the fake voting power? Will we have any countermeasures?

4 Likes

Thank you for compiling this information.

I do plan on reviewing the material once more. However, currently, I believe we should redefine the Stakers category to be even more specific by distinguishing between node runners and node stakers. One actively contributes to the network, while the other solely receives dividends. Similarly, I propose adjusting the voting power to reflect this discrepancy with LPs having a different weight compared to Gateways.

My suggestion is as follows:

  • Node Runners: 16%
  • Node Stakers: 6%
  • Gateways: 12%
  • LPs: 6%

I consider this breakdown to be equitable while still encouraging network decentralization.

3 Likes

Thanks Bruce. It is worth clarifying the mechanism for staking so everyone is clear how it works and where power will be distributed.

Non-custodial: With non-custodial staking, the owner retains control of the relevant wallet and can connect it to their Gateway ID (completed by signing a transaction in Node Wallet). They will individually control how that stake votes.

Custodial: With custodial staking, the stake is aggregated via the Service Domain. This means that the owner of that service domain (the node provider who is owner of the relevant DNS) will be able to attach a wallet to their own Gateway ID and with that they can vote with all of the custodied stake within the service domain they control.

In answer to your question above, it is not possible to “forge” stake for the purposes of governance as stake is activated through wallets, but it is quite important that all stakers (and DAO voters) understand this difference between custodial and non-custodial staking.

4 Likes

Question on Co-owned Nodes

For node runners, if more than one person owns the service domain, e.g., a jointly owned corporation, will all owners be eligible for Staker Creds (as only seems reasonable)? And if so, would the custodial stake attached to their nodes be divided equally for the purpose of voting weight calculation?

1 Like

As mentioned, I put up a poll to get a ‘feel’ of what the community thought about the appropriate vote weight delta should be between experienced builders/operators in our ecosystem, and newbies that have passed the citizenship test on Day 1.

I’m doing more research on this topic, looking at other DAOs and what published research papers there are, but we are boldly going where no DAO has gone before, and so we are mixing in best educated guess with what associate data we have.

  1. my gut feeling is that there has to be a higher delta between major contributors to the system and someone on day 1 who just finished the Citizenship CRED. In a more mature ecosystem we might achieve that if there are thousands of people sharing the Citizenship 20% but only dozens sharing the Builder’s 40%. That might be an optimistic scenario, what happens if there are 100 citizens (roughly our current engaged population) and 30-50% of those are builders (current subset of engaged population who would qualify imo). That would give a ~4x delta in favour of builders. Is that optimal? What should we aim for? How do we work around distortions in voting power at various points of maturity?

  2. what can PNF share about how they came to the weighting conclusions? As mentioned above, I appreciate that resources are scarce, but understanding how these numbers were arrived at would help question them/defend them to find an optimal balance.

  3. IMO- I would personally put more weight to Builders, less weight to both Stakers and Citizenship.

  • Citizenship should rightly be not exhaustive to obtain, but voting power should correspond to this limited amount of engagement needed to be included in the voting pool.
  • It seems that people are already happy to stake, LP, run nodes etc due to dividend rewards/commercial business incentives. So without the ‘need’ to offer additional governance incentives, why offer the same amount of voting power as those who actively contribute to the system in the builders category? I would lower the total voting weight to the stakers category, especially as it seems undemocratic that 3rd party custodial Node-runners are to accrue such am immediate voting weight (although I appreciate this is partly flattened by using a system a kin to quadratic voting- something I support).
  • Building gives the most impact, and so should have the most weight in my opinion, as an additional incentive to bring builders into the ecosystem and to benefit from those most involved in the ecosystem.
4 Likes

I’ve been looking at the mechanisms more closely, I think I can answer some stuff here

While I agree with this, the only difference that we can see on chain is custodial vs non-custodial. There is no way to distinguish a node that is being managed and run by the owner of the stake (a real node-runner) from a node that’s being run for a client. We could argue around common delegated addresses, output addresses or service domains, but this is no hard information, a node can have delegated addresses or a different output address just because the node runner likes to organize his nodes like that and this node runner wont be different from any other node that is run for a 3rd party.
Since we cannot make a clear on-chain difference between the quality of the owner of a given node, then it is difficult to assign different weights to them.
It is worth noting that, to some extent, the sum of the square root of the stakes makes nodes with unique domains to have enhanced voting power. So, if the real node runners use an unique service domain for their nodes, then they can claim this stake with some enhanced voting power.

No, the (custodial) stake of a given service domain can only be claimed by a single citizen.
There is no mechanism to divide the stake in a trust-less manner that is no gamable (that I can think of), and hence it cannot be part of the automatic system that is being developed.

The need for giving voting power to people holding stake is a long standing issue, I would not say that we are only happy with that. In fact I believe that part of the 3D governance model was done specifically to cover this need. We can discuss their weight, but they require some degree of representation.

I would also like to know more about this, but I think that they are rather arbitrary (this should not be interpreted as an attack).
It would be more intestesting to know if we can comeup with some examples that we want the model to fullfill and then adjust these numbers, for example:

  • how many citizens should be needed to equal one onlshanky-grade builder?
  • how much staked tokens should be needed to equal the power of Grove?

Things like that will give as boundaries that we can use to tune the voting power share among houses and intra-house classes.


Some general note:
Please do not use “quadratic voting” or “quadratic sum” for the voting power combination mechanism, as this is easily mistaken with the opposite:

  1. Quadratic sum: a^2 + b^2 + c^2 + … + n^2
  2. Sum of the square root of elements (no mathematical term for this): a^(1/2) + b^(1/2) + c^(1/2) + … + n^(1/2)

We use the second expression in the voting model, which dilutes the weight of larger numbers in the summation. The first achieves the exactly opposite effect.

8 Likes

Disclaimer
As an external observer and co-founder of How To DAO, a DAO Governance collective, and Bankless Consulting, I provide feedback based on my experiences and the deep knowledge I’ve garnered from various governance designs. While I may not be deeply familiar with the internal processes and standards of Pocket Network, my comparative insights could offer valuable perspectives on your proposed governance structure.

Feedback

I appreciate the thoughtfulness of the proposal and the effort to empower various stakeholders within the Pocket Network ecosystem, including builders, stakers, and citizens. The approach aligns well with a vision of inclusive and effective governance, similar to successful models for product-based DAOs.

Regarding the transition to the new governance structure, I am curious whether this will reset voting powers or if there will be a dilution over time from existing voting rights. Clarity on this matter would help community members understand their roles and contributions moving forward within the Pocket Network ecosystem.

The emphasis on incentivizing builders is particularly commendable. Pocket Network ensures that product development is closely tied to the overall strategy by rewarding those who directly contribute to the ecosystem’s growth; if this is not the case, it could cause significant issues in the product development roadmap.

While there has been extensive discussion on the role of stakers, core team members adequately address those concerns, and I have no further comments.

However, I would like to highlight the role of citizens within Pocket Network. Allocating 20% of the voting power to citizens seems fair, but there might be an opportunity to reconsider this allocation. It’s essential to ensure that non-technical contributors, who can significantly impact the network through other means, are sufficiently empowered. Their involvement in design, promotion, and strategic guidance is invaluable, and recognizing their contributions could further enrich Pocket Network’s ecosystem and increase the revenue significantly.

Furthermore, I like that influence cannot be bought but must be earned. It maintains integrity within the governance model, even though it is not a standard in DAOs.

But you also raised the concern that some of the voters disengage at some point. Therefore, I would consider the option for delegation, which could enhance participation without compromising these values. Allowing community members to delegate their voting power to representatives aligned with their views could foster more coordinated and impactful decisions. You can also limit the amount of delegation compared to their own holding so you don’t encourage professional politicians but only empower already existing stakeholders.

In conclusion, while supporting the proposed governance structure and its principles, further clarification on the transition process, increased empowerment of non-technical contributors, and the potential for delegation could enhance Pocket Network’s governance model. These adjustments would ensure a more inclusive, dynamic, and effective governance system that truly represents the diverse interests and talents within the community.

Thank you for considering my feedback. I look forward to continuing this constructive dialogue and contributing to the refinement of Pocket Network’s governance model.

5 Likes

Thank you @puncar . Really appreciate you sharing your perspective!

Just one small clarification. Builders in our definition is not solely technical. It includes anyone who helps build the project in any capacity - from thought leaders, governors, docs contributors, bounty completions, quick grant recipients, and many more! :heart:

5 Likes

Amazing - yes thats definitely important and its great to learn that you have been thinking about it and included also non-technical contributors.

1 Like

Hey @puncar (and other members of the community) - using AI as a tool is really exciting, and I get that it’s a useful way to get things on paper and help organize thoughts. I have a personal bias against it which I’m aware colors my thoughts, but I wanted to put in writing that using AI to write a response (a lengthy one, at that) wastes everyone’s time, and I am strongly against it.

Please consider writing in your own voice - it lets us know who you are, how you think, and what’s important to you. Robots cannot do that for you, so you’ll have to actually take the time and write it.

In general I will be calling out instances of this as we don’t want our forums overrun by robots, and if need be, moderating cases of it.

1 Like

Hey all (esp @puncar ) - I wanted to apologize for my comment above. I made many assumptions and responded in a way that is, at heart, against the POKT community DNA. It’s the opposite of having an open and welcoming community for newcomers, which as Head of Community I should be a beaming example of and not setting the tone like this. So I am sorry.

I’d like to avoid derailing the actual convo at hand about governance, but wanted to publicly say that this isn’t the way to a healthy community and I’ll do better. If anyone has concerns my DMs are open. <3

5 Likes

Two issues I’d like to discuss in the forums/on tomorrow’s call:

  1. In the original CREDs forum post from October ‘23, Citizenship was given 10% voter weighting, and Builders/Stakers 45% each. In the latest iteration of 3D Governance, this has been changed to 20% for Citizens and 40%/40% for Builders/Stakers.
    I’d be interested to know why this change was made - what was the thinking behind this? What do other members of the community think is the optimal weight for Citizenship?

  2. I’d be interested to hear views on taking out Governance as part of the Builders House and creating its own house. It seems that by keeping Governance within Builders, it reduces the weight that builders hold as the House could be largely full of Governance experts over technical/builders.

Further, I feel that Builders are the single most important component of our ecosystem, and should hold the most weighting in our design. Diluting their vote to overlap with Governance I think is wrong. While I appreciate the argument that building Governance IS building, I think these are different enough concepts that they should be housed separately.

Suggest solution for the above:

I start with a caveat that I appreciate that 3D Governance has been thought about for 6 months+ and suggested weighting changes are a potentially significant departure from the MVP. However I think a better system would be:

Builders: 40%

Stakers: 30%

Governance: 20%

Citizens: 10%

This goes back to the original design that gave Citizens a more limited amount of voting power until they start to contribute to the ecosystem. Builders are the most important part of Pocket and are rewarded likewise. Governance is separated from Builders and given it’s own limited vote weighting. Stakers are still given a substantial amount of voting power without creating an overly plutocratic system where large POKT holders have undue influence over governance.

1 Like

Proposal Builder CREDs Query

And from: PGOV5 Builder Specification:

Builder CREDs are received “automatically” according to this (Feb 9) post. But the earlier post on Builder Specification says a manual request by the Person is required. What’s changed?

1 Like

re: the manual request, the system requires you to collect a data asset. We’re working to see if that’s possible (and feasible) for me to issue automatically as the person reviewing + approving, or if the person who has earned it will need to basically click a ‘claim’ button.

It’s looking like the latter - you’ll need to claim it but it’ll be awarded for having completed it. I think Ben will touch on this more in the community call in about 90 minutes.

2 Likes

A little late commenting here. Applaud the efforts on Governance innovation to design a mechanism for Governance to be earned, not bought. Will spend some more time digging into the various links in this post but one immediate question that comes to mind is:

Where would an organisation fit into that is contributing on multiple fronts: sockets, gateways, staking, PNF partnerships, Gateways etc. would this all be linked to a single delegate identity thats on behalf of all contributions/contributors or would each individual contribution be treated and therefore engage in Governance independently?

In Developer DAO’s case, we’ve collectively contributed across various fronts (PNF Partnership, Members submitting Sockets, events [1,2]) and we plan to propose further ways to align with POKT and support the Networks success. Curious what is the best way to move forward with this so we can have a strong DAO-to-DAO relationship and become more active contributors to POKT’s Governance (and potentially vice versa if this is of interest to POKT DAO).

4 Likes

Thank you for sharing your thoughts @kempsterrrr - apologies for the slow response.

We had a really good discussion around Creds and Ben is currnetly working on a v2 that we’ll be sharing back with the community. Hopefully it will answer some of your questions. :slight_smile:

2 Likes

UPDATE - Please read

Thank you to everyone who attended the recent community call to discuss the governance upgrade. If you missed it, you can find the recording here.

We want to provide the following update based on the call and subsequent discussions with voters and community members.

  • We recognise that the original design created a cognitive burden. This led to genuine concerns around capture resistance that we want to avoid. Below, we propose some changes that we hope to resolve this.
  • Otherwise it felt as though we were generally aligned on the implementation and technical architecture.

We are therefore reducing the scope back to our common ground so we can move forward on implementation. We want to reiterate that this upgrade is being provided in an MVP form. We hope that perfection does not become the enemy of progress as we move to a vote.

Summarising the final proposal we plan to bring:

We will outline a simplified final proposal to the DAO next week. The new proposal is for 2 governance “houses”, the primary one for Builders (of all sorts) and the second one for Stakers.

The Builder’s house would hold 80% of the vote weight and be governed on a simple 1-Person-1-Vote basis (1P1V). The time after which you lose your vote, if you become inactive, would be determined by the size of your impact.

The Staker’s house would hold 20% of the total vote weight and this would then be divided equally between supply (nodes and LPs) and demand (Gateways). Vote here would be token-weighted but with a square root function for more equal representation.

Citizenship would be a pre-requisite to vote in either house, but without any voting power in and of itself.

As in the previous iteration, parameters are flexible and designed to be optimised over time, but with the total number of parameters significantly reduced from the original proposal. We will retain the automations and platform improvements from our original proposal.

Confirming the final changes:

After assessing the concerns raised, we want to reaffirm our commitment to POKT’s best-in-class 1P1V Proof of Participation model. We plan to do this through the following changes:

  • Citizenship is required to become a DAO voter, but we will remove the “Citizen’s House” from our governance design to reduce complexity.
  • Proof of Participation is really about Proof of Impact/Contribution. This will be the primary source of power in the DAO and all voters within the Builder/Impact house will have 1 Vote. We will use the decay functions so the highest impact contributions have the maximum power decay period (12 months)
  • We will retain the proposed Staker house but simplify it further with an equal split between supply stake (nodes, LPs) and demand stake (gateways). It will retain the square root function to provide more equal representation of stake.
  • We will now have only two houses. The balance between the two houses is to be controlled by a single IMPACT:STAKE Ratio parameter which we intend to set at 80% Impact and 20% Stake at launch.

Reasoning for these changes:

  • Citizenship is an important part of onboarding and governance but the citizen house added complexity that can be abstracted away.
  • We heard clear feedback around the simplicity of a 1P1V system. We will retain this but use decay to more effectively recognises high impact contributions over time. This enfranchises new contributors without the complexity of weighting.
  • We heard the concerns about the risks of introducing staked tokens to governance. However, stake plays an important role in any system “owned and governed by its users”. We don’t want to kick this question further down the road and plan to bootstrap small and controlled experiments with the creation of the Staker house. We will further assess the role of stake in governance only when we have collected more evidence and data.
  • By having one parameter - the Impact:Stake Ratio - we have the ability to further tune the system without creating cognitive overload in understanding the power distribution. Everyone can (mostly) just focus on driving impact.
  • With these changes, we think we have created a strong MVP and truly believe this latest round of community engagement and feedback has led to a better and safer version of the system to emerge.
  • We will provide an update on governance for discussion at the end of each three-month cycle so that the community can be an active design partner in the continued evolution of the system.

If you have any further questions, please share them here in the forum or reach out to me directly.

10 Likes