PPoS xGov Proposal

I am dissatisfied with current xGov implementation progress. And also with the proposals I have seen so far on this forum.

Here is an xGov proposal that embraces Pure Proof of Stake as a guiding philosophy and decentralises the process, whilst ensuring it is streamlined and effective.

  1. Anyone with a wallet with over X tokens can submit a proposal for next Governance vote. (The amount X is to avoid spam. The amount X to be decided via Governance from a range of options). This is called an xGov Proposal.

  2. Anyone with a wallet can view xGov proposals. Proposals are automatically and exclusively sorted by largest balance per submission wallet. So proposals from wallets with larger balances get seen the most.

  3. There is a period of time for feedback from community on proposals before xGov voting begins, so the proposal can be amended. During this period proposing wallets cannot fall below the balance of the wallet when the proposal was made. This is to avoid proposal spam from the same token pool.

  4. At the end of the proposal/feedback period proposers are rewarded from a xGov proposal reward fund, proportionally to ProposalStake / SumOfAllProposalStakes.

  5. At the end of the feedback period no new xGov proposals can be made.

  6. EVERYONE can then vote on each proposal, as Yes or No, if they wish. A Yes vote adds the number of Algos in the voting account to the Tally for that proposal. A No vote does not influence the Tally.

  7. Sorting is updated according to Tally.

  8. The proposal with the largest Tally at the end of xGov voting period gets submitted for voting in Algorand Governance.

  9. Once the winning proposal is selected, for every proposal a wallet voted on, and as long as the balance of the voting wallet was maintained, they get a Reward proportional to NumAlgosInWallet / (TotalXGovVotingAlgos * TotalNumOfProposals).

  10. A fresh proposal/feedback period begins, during which new proposals can be made and amended.

PS: Open to ideas for improving this proposal. Will update over time.

PPS. Also receiving a lot of feedback from AlgorandOfficial Reddit, which I am applying as above.


This is a good idea and needs to highlighted. xGov is essential to algo’s success and the ability to reject the foundation’s gov proposals. Governers need a real choice and the ability to say no.

1 Like

Choosing a single wallet for governance would hurt competition and discourage the development of new wallets. This would lead to centralization based with only 1 viable wallet.

There also should be some criteria for creating a proposal as it could lead to too many proposals for people to navigate and focus on the most important issues.

Proposals need to be able to be tweaked as feedback is given. A good idea shouldn’t be tossed because the proposal contained one detail that the community disagrees with.

The second two points could be handled by sub committees of elected xgovs.


Forgot to post this as a reply. Still new to forum.

Point taken on the first. Perhaps I don’t need to specify Pera.

The filter is that the proposals from the largest wallets get seen first. Perhaps there could be a lower limit on Algo ammount in a wallet, to be able to submit a proposal to avoid spam. Amount to be voted on by Governance.

There could be a period for feedback on each proposal before xGov voting begins.


This is nice on the surface I think and it’s good of you to take the time to write it up.

I think two issues with “1 coin 1 vote” are:

Firstly I already think exchanges like Binanace and Coinbase skew governance. Imo they don’t have the right to vote their coins but have to in order to get the rewards to distribute to their users.

I’d personally like to see an option C: “no preference” on votes for this reason that anyone holding coins for others is expected to take.

And yeah this proposal would make that issue bigger.

Secondly it doesn’t recognise usage as well as holdings. So someone who is doing 10k transactions per day but holds 1 Algo has 0.1% the voting power of someone who does no transactions but holds 1000 Algo. That doesn’t seem right to me, imo incentivising usage is more important than just holding and we want the most prolific users of the network to be in control of it.

I also personally am a small fish so dislike “1 coin 1 vote” for that reason as I’d personally prefer a system where if you prove you have knowledge and commitment you can get more voting power. But maybe that’s selfish.

1 Like

Opps, again forgot to post as reply…

Disagree entirely. Algorand is all about Pure Proof of Stake.

In general I like the idea you are going with but I do see one issue with the suggestion that only the proposal that receives the most votes is put to the general vote (or in a more general case, N ideas with the most votes). This way, a whale could censor other ideas by spamming their proposals just to get their own always to the general vote. Hence, I’d prefer to have a constant limit that a proposal must achieve. This could result in a less predictable number of proposals that are passed to the general vote but would be a more transparent process. This then leads essentially to referendums, which has already been suggested e.g. initially during G2 or in a updated form during G3 but unfortunately never gained traction. That suggestion goes even closer to the principles of PPoS by considering time-weighted stake, i.e. if one is committed to supporting a proposal for a longer period, it would eventually get to the general vote even if the supporting stake is small. This way, even ideas brought up by small holders could be heard if their conviction is strong. Another difference to your suggestion is that multiple wallets can contribute towards reaching the limit X, which fosters collaboration.

1 Like

As an improvement to fixed sorting I would suggest using a random order that is weighted by the number of algos in the wallet.

This means that even mega-whales with 50% of algos would only have their proposal ranked as #1 half of the time, while someone with 1% of algos would have 1% probability of getting their proposal ranked as #1 (which is enough to get the proposal seen, if the ranking was randomized each time a user views the proposals).

And if many smaller fish make identical proposals (as a mechanism to pool their algos behind one proposal), for example ten people each with 1% of the algos, then that proposal would have 10% probability of being ranked #1. (Of course 1% of all algos is not small fish by any means. I just used it as a round number for illustration purposes.)

1 Like