Counterproposal to xGov

Abstract

In order to further the decentralization of the Algorand, there is a clear need for the measures to be voted upon in the Governance to be put forward by entities besides the Foundation. The current proposal of Governance period #2 suggests these entities to be part of another layer – xGov. Despite the implementation of xGov not yet defined, this post argues that adding another layer to the Governance is a step away from decentralization. It can lead to a creation of a closed xGov lobby, creates a barrier of entry for new entities, and goes against the principle of equal power of each and every ALGO. A counterproposal is made for a system that satisfies the same requirements with less assumptions, while also preserving the inherent equality of ALGO and the current openness of the Algorand, which furthers wide engagement and innovation.

Introduction

In order to further the decentralization of the Algorand, there is a clear need for the measures to be voted upon in the Governance to be put forward by entities besides the Foundation. The current proposal of Governance period #2 [1] put forward by the Foundation suggests these entities to be part of another layer, referred to as Expert Governors (xGov). The current proposal does not yet define the details of the actual xGov system and leaves its exact implementation to be defined in later governance periods.

Despite the implementation of xGov not yet defined, this post argues that the general idea of adding another layer to the Governance through xGov is a step away from decentralization because it can lead to a creation of a closed xGov lobby, creates a barrier of entry for new entities, and goes against the principle of equal power of each and every ALGO.

Furthermore, as an alternative, the post suggests a system for proposals to be put forward in the future satisfying the same requirements of: longer commitment, higher stake, aggregation of experience and anti‑spamming. However, the suggested system preserves the inherent equality of ALGO and the current openness of the Algorand, furthering wide engagement and innovation.

xGov – A step away from decentralization

The xGov will not only have the power to propose measures. They will also act as intermediaries when approving a proposal to be passed to a general vote. Their power will be represented by experience‑weighted stake. For this group to act for the good of Algorand, it has to be assumed that the majority of this subgroup’s stake will be held by good actors. If that is not the case, a good proposal will be censored by the bad actors or even a bad proposal put up for a general vote. To discourage this, the xGov are supposed to show their good intents with a longer commitment of their stake.

However, a bad actor could infiltrate the xGov and act “good” for a while to gather larger stake through experience. If a good actor exits the xGov for whatever reason, their experience would reset. Therefore, malicious actors could gain the upper hand over time. Moreover, if a new good entity enters the Algorand and wants to make a proposal, it would not be able to catch up with the longer committed entities of xGov, which could censor the new entity. This essentially means a possible creation of a closed xGov lobby. If that were to happen, there is no way back to a different Governance system since the lobby is the one which would need to propose a measure for change.

Furthermore, in the current proposal [1] it is stated that “xGov program will be governed by a set of ground rules, protecting the rights of all governors and the community at large”. While the suggestion of how this will be implemented is not yet defined, it means there will be an authority judging and enforcing these ground rules. This inherently goes against decentralization. Hence, each and every proposal should be allowed, and the system itself set up in a way to discourage malicious proposal. However, if a malicious proposal is brought up regardless of this measures, the existing assumption that the majority of Governors act for the good of Algorand must suffice to prevent its approval.

Moreover, the xGov system contradicts one of the core principles of Algorand – the equal power of each and every single ALGO. The power (thus value) of ALGO held by xGov entities would be different from the rest due to the experience‑weighting, e.g. the ALGO held by a long‑term xGov would be worth more than the rest since it holds more power in the Algorand by having higher influence on putting new measures forth.

Counterproposal

The counterproposal expands on the system used in [2]. Anyone at any time can propose a new measure. However, before that measure is put up to a vote, a predefined commitment limit (CL) must be reached. The commitment does not necessarily have to come solely from the entity that made the proposal. Any other entity can show support by adding to the commitment, thus contribute towards reaching the required CL for the proposal to be put up to a vote. It is up to the proposer to find enough of this initial support.

The vote itself has besides the options of “for” and “against” the proposal also an option “against with veto”. The last option is used to express strong disagreement with the proposal in case its quality is unacceptable or the content malicious. If this option wins, the commitment is not returned to the supporters of the proposal (but could e.g. go towards future governance rewards). This mechanism prevents spamming of low‑quality or malicious proposals, and is based only on the already existing assumption that the majority of Governors act for the good of Algorand, i.e. without malicious intents to steal the commitments backing good proposals.

If this mechanism was used in Algorand, the proposal would be put up for a vote the following governance period (GP) after the CL is reach. However, instead of having the commitment based solely on stake, it can be a time‑weighted commitment. Such a mechanism would allow smaller entities to still reach the CL, just a longer commitment would be required.

To explain this on an example, assume that it was agreed upon that a proposal is put up to a vote if 1M ALGO is committed for 180 days. This would present a CL of 180M AD (ALGO*days). If an entity proposes a measure and gathers the support of 1M ALGO, it would be put up to the vote after 2 GPs (where each GP is 90 days). However, if the proposer has only 0.1M ALGO backing, it would take 20 GPs. But in the meantime it could gather additional support, reaching the limit quicker, e.g. gathering support of an additional 0.4M ALGO after 5 GPs would fulfill the limit after overall 8 GPs.

The ALGO committed to a proposal could be used for participation in governance while waiting to reach the CL (the same way as suggested for xGov). Moreover, an exception could be made such that ALGOs committed to governance could also be used for the initial support of any proposal without causing the wallet to fall out of governance (to keep this short(er), the implementation is omitted).

Furthermore, to ensure commitment of the proposers towards long‑term benefit of the Algorand, not all committed stake would be returned right after the proposal passes but have a cooldown period. Nevertheless, the stake could still be used for further participation in governance or for initial support of other proposals.

Conclusion

The counterproposal provides a solution to the same problem and fulfills the same requirements as the envisioned xGov system but without adding another layer, which requires an additional assumption that a subgroup of entities will act for the good of Algorand. According to the Occam’s razor principle, it follows that the counterproposal should be the preferred solution.

Furthermore, the proposed system provides a more direct participation, has a lower barrier of entry for new entities, furthering wide engagement and innovation. Moreover, it keeps the Algorand’s core principle of equal power of each and every ALGO.

References:

[1] Algorand Foundation. “xGov: an Expert Governors System for Algorand”. 2022. Accessed on 13.02.2022. Available at: https://algorand.foundation/news/xgov-expert-governors-system

[2] Tendermint Inc. “Concepts”. Accessed on 13.02.2022. Available at: Concepts | Cosmos SDK

7 Likes

Thank you for this well thought-of opinion, this is exactly the kind of discussion that we are hoping to have with the wider Algorand community, if option A is approved by the vote.

We believe that many tenets in this counterproposal are compatible with the xGov approach. For example, giving everyone the ability to come up with a proposal (and leaving it to the proposal to drum up support for it) is consistent with an xGov layer for upvoting these proposals to put them on the ballot. We look forward to continuing the discussion and incorporating community suggestions like this one in the Governance framework.

4 Likes

Could the Foundation please elaborate on why it deems necessary for the proposers to be part of another layer of governance (since this was not addressed in the current proposal [1])?

As pointed out in the post, it seems unnecessary (and even potentially damaging for Algorand) to require another assumption for the governance system to operate on; especially because the same requirements for making and approving proposals could be met without the addition of another layer.

1 Like

Hi Anon, what do you think about Introduction - VoteCoin ?

Fully implemented governance standard on algorand for DAOs…

I have just recently published a refinement of the counterproposal to xGov in another thread.

1 Like

I think you have good ideas, but somehow i have feeling algorand needs to solve other issues first…

You write too much, but I have not found a mention about how should be governance implemented. In my opinion the crucial thing is the delegation of voting power.

If you want any DAO to work efficiently, you must have possibility to ask any time any question. And DAO must not go bankcrupt with making the decisions onchain, which means the cost efficiency must be good for dao as well its users.

If one person can delegate his voting power to someone else, DAO makes better decisions because the sum of voting power is high. If DAO requires every member to vote in each vote, members will get discouraged because it might take time to study the proposals. Thats why i believe the delegation should be implemented, so that if i trust you, i delegate my voting power in finance to you and i can vote only for the decisions that i really want to vote on…

After the DAO agrees to this type of knowledge based representative pure democratic voting solution, than we can discuss how to motivate people to participate in the DAO decision making…

I agree that the governor who study the proposals and always votes should get better rewards than the person who just delegates his voting power, but i agree that algorand foundation should not compete the DeFi space with the governance rewards and locking mechanism to lock algos…

Am i the only one here who thinks that DAO should make more decision than 1 per quarter?

1 Like

i have feeling algorand needs to solve other issues first…

I strongly disagree. Current priority should be to setup a system for everyone to make proposal to be put to vote. Let me explain why:

  1. Results of G2 showed a wide support for this. But it should be done the right way (which I think xGov is not per se).

  2. After we are all able to make proposals, it will be easier to address all the other issues because multiple people could actually work on those and propose concrete solutions.

  3. Based on the quality, provided reasoning, and (lack of) engagement in public discussions for the previous Governance proposals from the Foundation, it seems they are preoccupied with other things to lead the Governance.

  4. The need for wider community to be able to make proposals is also clearly seen in the thread by the amount of community responses vs. the ones from the Foundation. Until a system for considering proposals by the public is in place, we will be seeing what is happening in the thread. A lot of good ideas, but unstructured and without concrete proposals.

You write too much

The topics are complex. If we do not precisely express our ideas, we will get such incomplete proposals as are the ones made by the Foundation.
If you see concrete suggestions on how I could explain my ideas in a more succinct manner, please feel free to reach out and we can work together on those.

I have not found a mention about how should be governance implemented.

I did not address this simply because it has already been solved. We have a voting system in place. I could agree that it might not be perfect and better solutions can be made. But that is not a priority, since even if you come up with a better system than we have now, you cannot get others to consider it since you cannot put up a proposal to vote.
Fruther, I believe that the solution I suggest on how to decide which proposals can be put to a vote, is in its essence independent of the actual voting procedure.

In my opinion the crucial thing is the delegation of voting power.

I would prefer not to endorse delegation of voting power since it inherently makes the system more centralized. The main reason I believe in Algorand is its elegant solution to the consensus - PPoS, which enables direct participation.
Delegation of voting power will/is already spinning up naturally because some do not want the hassle associated with taking decisions. That is fine, but it should not be endorsed at the core of the Governance.

If you want any DAO to work efficiently, you must have possibility to ask any time any question.

I agree that current Governance mechanism is very rigid in this regard. But this is again a separate topic.

And DAO must not go bankcrupt with making the decisions onchain, which means the cost efficiency must be good for dao as well its users.

I believe long-term Governance will not be rewarded. It will simply be a voluntary civic responsibility. The costs of voting are anyhow minimal - just a transaction.
Currently, it is a good thing for the participation in Governance to be rewarded because it engages more people. This will not be necessary once the community grows. It is also a part of tokenomics, but again a completely separate topic.

Am i the only one here who thinks that DAO should make more decision than 1 per quarter?

I completely agree that 1 decision per quarter is inefficient. Again, the issues is firstly in a lack of proposals because not enough people have the option to work on those. How to solve this, should be our top priority.

2 Likes

I support @anon proposal. Governance is the most important utility and biggest power of the network. We can’t underestimate it.
I was sharing similar ideas in the recent Evolving Algorand Governance - #160 by robert thread.

However, I think that:

  • governance should be more fluid - restricting proposals only to gov periods is a too big limitations
  • locking tokens for more engagement makes sense - however I would use liquid staking solution here.

@anon idea looks to be an evolution to Cosmos x/gov – which is battle tested.

One more thing which is good to consider is expenditure proposal process - where in a case of a crisis we are able to establish a fast gov process to submit changes.

Please see my reply to your similar post in another thread.

This is an excellent point. I think the above implementation could accommodate this as well simply by having e.g. another commitment threshold level for urgencies (CLU). If that level were reached, the proposal could be put up to a vote right away. But in this case, I think the CLU should be about 10x larger than normal CL not to end up being overused.

1 Like