Feedback request on proposed Q4 2022 governance measures

This was my instant reaction and I’m a but surprised more people didn’t have this reaction. Suppose I want to allocate 90K ALGO to buy NFT art. I just make Option A allocate 80k and Option B allocate 100K, and I get what I want and the voters have no say. That is exactly how this entire proposal reads.

Doesn’t Algorand have a “Red Team” that reviews drafts before publishing here? I would imagine someone on an internal team would have had such criticism and it would have been addressed prior to being pre-published here, but maybe this is a 1 or 2 person part-time job at AF writing these things??

2 Likes

Yeah, the more I have thought about it, the worse it became, not only as a general way of doing votes, but also the issue 3 seems kinda fishy.

If I wanted to launder some money, I would go the same road.

Just create some NFT and buy from yourself, or the people, you want to bribe, classic in art already and now the foundation wants a way, to easily be able to do it.

I know, I know - this is just my skeptical brain, but honestly, if you they truly just want to help the little guy and buy some NFTs from small artist (what I’m even not so much against (well I see some merits to it)), they should at least address the money laundering problem by explaining, how there will be checks and balances and so on. But no - not at all.

Honestly, after this current voting system, I’m considering not DCAing anymore into Algorand, this is just a to big of a red flag for me. Well - at least, I will stop it for now and see, what will happen from it.

1 Like

In a general sense, I think the overall options are leading - as in, there isn’t a “do nothing” option; instead, it’s leading to something that the foundation wants, just at different scales. I think the voting options or measures should always have a ‘do nothing or deny’ choice.

Option #3, supporting the NFT creators. I don’t see why it has to be an allocated amount from the foundation. Instead, if you like the art and the utility, purchasing the NFT as an individual (not the foundation) would help just the same. Tweeting or sharing what you like or enjoy about a specific NFT can help ‘support’ the creators as well.

I think option #3 has to be carefully considered.
What has been deemed a good NFT vs not?
What parameters is the foundation considering when concluding yes, this NFT deserves support?
Would the foundation disclose what NFT projects it supports and why?

You see, this has the opportunity to manipulate the market - not saying it will, but the option is there.

These are just some of my initial thoughts on the governance period.

2 Likes

The biggest issue of these proposals is that there is no option to refuse any of them. If you do not agree with the concept you are only left with two options that do not allow a negative vote.

This is just stacking the deck.

Reform Governance!!!

1 Like

Introduction
Analyzing the Algorand community
Many people are showing the .algo mark, but
There were few factors affected by influencers
On Twitter, there was only price talk and unnecessary talk like “we are the best”.
If you’re truly community-building, we need a few people to keep community
(People who are not working for Algorand or who are not Foundation people)
Appropriate rewards must be given to induce this.
The way I think is similar to the direction of securing Defi liquidity with compensation obtained through participation in existing governance.
It is to give appropriate compensation to those who build communities through compensation.
Existing ambassador systems are less obligatory and it is difficult to intuitively see what activities they are doing. In fact, I don’t even know who they are.

Main subject
It uses the same method of applying for voting rights by applying according to the current governance period, and delegates your voting rights to each representative.

  1. Representative
    Anyone can become a representative and form a delegation, and they form their own proxy voting group.
    The formed proxy voting is entrusted with the Algo voting rights of delegators and fulfills the voting role.
    1-1) There is a minimum delegation $Algo limit to get rewards (prevents multi-cherrypickers)
    1-2) There is a maximum delegation $Algo limit that can be rewarded (adjusted via saturation)
    1-3) If you fulfill the role of representative, you can receive part of the reward through commission.
    (Representative role and promotion of community formation)
    1-3) The decentralization of voting rights can be promoted through parameter adjustment.
    (Adjust the maximum and minimum delegated Algo quantity through parameter adjustment)
    1-4) Fees can be changed, and malicious attacks can be prevented by depositing representative collateral
    (Adjustable from 0 to 100%, fees and collateral of the delegation transparently disclosed on the blockchain)

  2. Delegator
    2-1) Delegate your right to vote to the delegation so that you can cast your vote instead
    2-2) Possible to vote for opinions different from those of the delegation delegated by himself (re-vote system under the delegate himself)
    2-3) Can be delegated to the desired delegation
    2-4) If the fee is too much, you can run your own delegation or conduct your own governance vote and use the existing system.

This method was conceived from the community method built through the free competition system through Cardano’s SPO (Stake Pool Operator) community.

In the case of Algorand, the PPOS method is used, but since rewards are not currently given and rewards are distributed through governance, a free competition system is established through voting rights to compete (good faith) among representatives and promote a sustainable growth system for the Algorand community. is to create

Content-Based Bibliography
https://docs.cardano.org/learn/pledging-rewards
https://docs.cardano.org/learn/stake-pools

1 Like

We are grateful for all your feedback, especially those of you who took the time to write well-considered counterproposals and observations about the measures for this governance period. We agree on the importance of having a healthy dialogue with the community on governance measures, and we are committed to improving in this area going forward.

You gave us 3000 submissions - including comments, upvotes, views, and likes - to our posts on the Algorand Forum and the Algorand Official subreddit with suggestions and positive and negative sentiments about all measures. Thank you!

After two weeks of lively discussions and careful deliberation, we have consolidated and taken on board your many disparate views on how to frame voting questions. Based on this feedback, we are amending two proposals: Measure 2 to allow the community to vote down a fund allocation for xGov’s distribution and Measure 4 to allow the community to vote down a fund allocation for the NFT Collection. Elaboration of the revised Governance Period 5 Measures is below:

— First Measure <<< This measure has been updated >>> See latest post here

Governance Rewards

During 2022, it became clear that a high yield in General Governance was inhibiting the growth of our DeFi ecosystem. Well-considered community comments called for the complete removal of General Governance rewards so that such funds might be available for more productive use. On the flip side, others argued that reducing non-participation governance rewards with immediate effect - from 70.5MM Algo per quarter to zero - could run the risk of destabilizing the ecosystem, with potentially unintended consequences.

We agree we need to continuously examine incentive mechanisms and be diligent stewards of foundation resources. Therefore, the proposed measure represents a compromise between the 2022 rewards regime and those of the opinion that paying general governance rewards is wasteful.

At the last governance period, the community approved the redirection of 7MM Algos DeFi rewards for Q4 2022. We have observed the positive impact that this change has affected, especially in present market conditions, and therefore see value in both restating support for DeFi as well as providing a route for the community to (again) redirect a portion of general governance rewards to DeFi, should it wish to do so.

Therefore, taking the above points together, Measure 1:

  • Materially reduces the overall level of rewards emissions, from 70.5MM to 45MM Algos per quarter
  • Only allocates rewards for the next two periods (versus the entire year of 2022). That will give time to assess the impacts of reducing General Governance rewards and redirecting resources, without locking in the incentive mechanism for a longer term, should it prove to be suboptimal.
  • Provides an instrument for the community to allocate 10MM Algos of incentives between (1) DeFi participation and (2) general soft locking and voting.

In subsequent governance periods, we look forward to broadening the scope of ecosystem participants who can receive incentive rewards, including node operators, oracles and other infrastructure providers.

Tl;dr;

The reviewed Measure 1 asks governors to vote on allocating the extra 10MM Algos to either General Governance or DeFi rewards. This measure does not include a “No” vote.

— Second and Third Measures

Previously Measure 2

The initial Measure 2 offered a choice between 1MM or 2MM Algos to be allocated for a Community Grant pilot program to be moderated by the xGov.

Community feedback on this measure converged around the lack of an option to decline this allocation altogether, as well as two main themes:

  • Lack of clarity around xGov implementation; and
  • Allocation size.

These are fair observations, and we would like to provide more context.

xGov implementation
Work on an xGov ARC is underway and we expect to release a draft for comment in the weeks following Decipher. The public discussion phase will aim to gather your feedback to fine-tune the design and implementation timeline for the program.

Allocation Size
Our intent with this measure is to provide an allocation big enough to be meaningful and generate community engagement, without unduly risking resources on what will initially be an early-stage MVP implementation.

Should the implementation prove successful, we anticipate bringing further measures to expand the scope of this program in the future, allowing the xGov program the ability to upvote projects that foster the growth of the ecosystem.

Tl;dr;

Measure 2 asks governors whether or not we should allocate funds to the xGov community grants pilot program. Then, in Measure 3, governors vote on whether to allocate 1MM or 2MM Algos for the MVP. Measure 3 is only relevant if Measure 2 is passed.

—— Fourth and Fifth Measures

Previously Measure 3.

After reviewing feedback around this measure - extensive, despite the relatively lower amount of resources involved - we would like to clarify the proposal and restate the rationale for the Algorand NFT collection.

First, we should state that the term “NFT” is broad, and is conceived of as covering the full range of Algorand NFTs - from art & collectibles to music, gaming etc. The idea is to showcase a diverse range of assets, communities, capabilities, and the utility that Algorand NFTs offer. We agree that it’s more clear to refer to an “Algorand NFT Collection” than “Algorand Foundation Art Collection”, so we’ll go with the former term from now on.

Second, we envisage that the development of selection criteria for specific works will be driven primarily by the community, initially at least via a committee consisting of Foundation and community stakeholders. Future oversight, decision-making, and management will ideally - in the Foundation’s view - be folded into the xGov process, a DAO, or some other community-led structure. To be robust and effective, such mechanisms would need to be well thought through and developed via extensive community involvement and consultation.

The benefits we would hope to see from this initiative, if passed, include:

  • Offering a forum to display Algorand NFTs.The collection would be showcased in both virtual and on-site events, giving select projects/creators the stage and marketing visibility, and showcasing Algorand’s robust range of NFTs to people within and outside the ecosystem.
  • Providing support to existing creators on Algorand. Purchasing select pieces and/or emphasizing primary sales and shuffles is just one example of this kind of support. Other examples may include providing grants for strategic/high-profile initiatives, commissioning work, providing creators yield via fractionalization and/or collateralization or other support mechanisms.
  • Onboarding new users into the Algorand ecosystem. For example, running campaigns, shuffles and giveaways to attract collectors from other chains and incentivizing them to create new wallets.
  • Engaging with Algorand’s existing NFT collectors. Active governance is one example, but the collection committee could collaborate with ecosystem players (i.e. DeFi platforms, NFT marketplaces) around awareness, “learn to earn (NFTs)” programs etc.

Finally, we heard many voices calling for a choice on whether to pursue this project at all (the original drafting only contained a choice of whether to fund 300K or 600K Algos). We have responded by providing such an option.

Tl;dr;

Measure 4 asks governors whether or not we should allocate funds to seed the establishment of a Community curated NFT Collection. Then, in Measure 5, governors vote on whether to allocate 300K or 600K Algos for the seed collection. Measure 4 is only relevant if Measure 3 is passed.

6 Likes

Thank you for thinking carefully about this and taking on board all the community feedback, that’s really healthy and really appreciated. :slight_smile:

One thing re an official art gallery is that there could be an incentive system like “if you send an NFT to the art gallery for free, and it meets our standards, then we’ll display it on the front page for a while.”

That helps creators by giving them exposure, helps showcase what is happening on Algorand and allows the foundation to build up a big collection without spending a lot on it.

2 Likes

Thanks for your post. Just a few notes.
Introduction:
I don’t know who you’re following on twitter in the .algo community, but almost no one I follow talks about price action, and only when something really volatile is happening. So if you don’t like that, maybe ask around and see who else you’d like to have show up on your wall. I’m @averagezen and I follow a lot of great #algofam if you want to check out who I follow.

As for compensating non-Foundation employees for posting on social media…we have a wonderful project in $VYBE that does exactly that for Twitter, Youtube, Reddit, TikTok, and possibly others I can’t recall.

This governance we included Defi liquidity and LP tokens in governance.
The Foundation (and soon xGov) compensates the ecosystem through grants to builders in the community already.

Existing Official ambassadors…we won’t talk about that right now.

Main Subject:

  1. Representative
    Sounds like xGov…that’s what we’re working on and is discussed in this post. There will be more details very soon.

  2. Delegator
    This option is also currently available (if I understand you correctly) in rudimentary form through Folks Finance liquid governance on Algorand and could be adopted or adapted in any number of ways through the xGov system or by compartmentalization for protocols involved in governance.

Thanks for the information on Cardano.

Posted my review to twitter:

Thanks for this update! I think it demonstrates that the Foundation is considering and acting upon community feedback and adds to the legitimacy of this forum, in particular.

I appreciate the pivot on these measures, which are directionally aligned with my own vision for the evolution of Algorand governance. Although I would advocate for faster shift to investment in active builders instead of passive holders, making that transition in a measured way feels stabilizing and wise.

Please prioritize the definition, creation, and mobilization of the xGov program, which, if I extrapolate from the vanilla governance program and feedback thereabouts, seems to be the obvious place we need to go as a community.

1 Like

Thank you so much for your detailed reply!
From your writing, I must have had a narrow view.
See you on Twitter!
my handle is @algorand_whale

1 Like

Where are the excess 24.5MM ALGO from Measure 1 going to be redirected? Will the foundation be keeping them in their treasury? And how does this effect the original distribution plan to eliminate the role of the foundation to distribute the tokens by 2030?

2 Likes

Dear Foundation:

As long standing Algorand community supporter and someone who holds a lot of Algo, Measure 1 is particularly concerning for several reasons:

  1. How could the Foundation arbitrarily decide to reduce Governance Rewards by 25M+/quarter, without explicitly explaining to the community where the funds are redirected to be used for? The increased incentive for other activities (DeFi, NFT etc) are tiny compared to the reduction of the standard governance rewards, we view this as a “skillful” way for the Foundation to cut overall community incentive without good reason.

  2. Even if the Foundation wants to incentivise other activities like DeFi or running nodes, why would the funds have to come from the funds/ALGO already committed/allocated to Governance Rewards? As far as we can tell, Foundation still have 1B+ Algo in its treasury, and why wouldn’t you allocate additional resource from your treasury for other initiatives? How are you planning to use the 1B+ Algo for then? Please remember: Foundation’s goal is to fully decentralize and disappear after 2030, but not to run a for-profit organization like venture funds. The lack of budget planning and communication to the community is extremely concerning.

  3. All this being said, if you really wanted to push this Measure out, you should at least include an option for the community to vote “status quo”. In our opinion, a “status quo” vote option should always be available, otherwise this gives the Foundation the super power to change any existing measurement as they see fit.

We are passionate about the Algorand technology and ecosystem, but we are extremely concerned about how some of these governance voting measurement are being proposed. The example of proposed Measure 1, if passes, will slash Algorand’s “staking yield” by almost half - making it the lowest yielding L1 token out there. Given the lack of liquidity and overall crypto market sentiment, we are extremely concerned this might trigger a sell-off on Algo.

We urge you to reconsider the proposal before it’s too late.

6 Likes

I’ve followed you back! I appreciate the feedback and I’m glad I could help offer a different perspective!

To address your points I’ll have you refer back to some previous posts in this thread that touch on these subjects, but I’ll give you the tl;dr here if you wish.

  1. In @Adri follow up post here you can see the reasoning behind this, but I’ll recap: High rewards for governance are diluting $Algo without providing any benefit to the ecosystem. The reduction in rewards should make it possible to earmark funds to pursue the Foundations directive of expanding Algorand growth in new ways that can be expressed via the xGov system (which would still be within the Governance budget ideally, which would negate any conflict with using Governance funds for a Governance program). This second part may or may not be directly related but even if it isn’t, the dilution from high rewards for no risk was not good for Algorand for a lot of reasons that require their own thread.

  2. The funds from governance being used to incentivize DeFi are being used to incentivize liquid governance enabled DeFi programs and those utilizing liquidity providing protocols with $Algo in them. To be frank, Algo in DeFi is of more benefit to the ecosystem than Algo in vanilla governance. The reward should be larger imo. As far as node running (which is not currently on a proposal) one of the main criticisms of Algorand is lack of decentralization. True or not, node runners incur real world cost beyond the purchase of $algo and secure the network, which is of value to me and I’m willing to give up a few easy $algo if it means a stronger Algorand.

  3. There were “No” options added to the other votes. This was not added to Measure 1 and that should speak to you on how vital this Measure is in the current market conditions.

In all honesty, anyone protesting the cut to governance rewards does not have a very good knowledge of the Algorand ecosystem and the projects on it. The opportunity to make a MUCH LARGER yield is there, not in governance. If that’s your sole reason for objection, then perhaps there are better options that are probably less upsetting, that you can still use to support the best L1 around.

1 Like

For measures 2 and 4 can you add a line saying this may be proposed again in the future?

I would hate for the ideas to be shot down and never be able to be voted on again.

1 Like

Hi @averagezen, thanks for your reply but it doesn’t address any of our key questions to the Foundation. We found your argument not convincing.

Foundation’s path to completely decentralization and disappearance after 2030 is not clear and its budget of how to use the 1B+ ALGO still in its treasury not transparent. Nobody is denying the importance of DeFi, which exactly makes the Foundation’s small additional allocation to that incentive looks pathetic - why don’t you allocate all the 25M+ ALGO you scaled back from Governance Rewards to DeFi incentive then and some? Why does the Foundation decide to decrease the overall incentive rewards?

Beyond all these, do not give the community an option to vote “status quo” is against the spirit of decentralization, democracy and the purpose of a Foundation. The Measure might or might not be vital IN FOUNDATION’S MIND, but after all it’s for the community to decide by voting. If you truly believe this is the right proposal, why don’t you spend more time convincing the ecosystem supporters than engineering taking away their ability to vote no? Are you not confident the community feel the same way? How does the community have any confidence the Foundation wouldn’t arbitrarily change anything else in the future “just because it’s vital”?

In all honesty, anyone proposing removing the ability for the community to vote “no” does not have a very good knowledge how decentralization, community and democracy work. You might be working for the Foundation, but if that’s your sole reason for the argument, then perhaps there are better roles that suits better.

4 Likes

I’m definitely not working for the Foundation, let’s clear that up right away please. I’m sorry you found my reply unsatisfactory, although I do point to reasoning given (not by me, just paraphrased by me except where I note my opinion) for the decreases. As for the transparency on exact amounts and why such a particular methodology was used or not used, those are excellent points. I believe some form of answer could be produced based on a given set of metrics or logic, but what would that change? The lack of decentralization that you speak of is being addressed one piece at a time through various strategies, including the xGov implementation which, if done correctly, solve many of the issues you address here involving Foundation influence on voting options. I’m not at all sure why you assume I’m rooting for anything other than the best outcome for the Algorand blockchain as I see it. If you had read my thread on twitter that I link in this thread, you would know that I disagree with a few of these proposals myself.

You are talking about 1 measure amongst 5 now as if the outcome of this one particular proposal will be the judge of Algorand as a whole and the Foundation as well. It’s not just the Foundation that believes this is vital. @AlgoNautilus posted a thread proposing to INCREASE DeFi rewards even more. Who are these ecosystem supporters that you think aren’t being spent enough time or resources on? Who do you think is getting these resources that shouldn’t be? Why? I find your argument unconvincing. Mine can be backed up by numbers…here: Algorand TVL - DefiLlama

1 Like

No reason to argue with him, somehow, he cannot understand the simple importance of having a NO-vote in a voting proposal, which does not change, by saying “But The Proposal Is Important, Hence, We Removed The No-Option”.

As if this would be assuring at all.

The same discussion with him is done multiple times already in this thread :smiley:

I do like your takes overall :+1:

Just wanted to add - the only option I could live with a NO-Option would be the x-Gov one, where we already had a vote on in the last voting-session. So Im kind of irritated, why they added one for this but not one for the new first one.

But in the end, they at least listened partly to my (and others) concern about the missing NO-option and added some. I would still have kind of preferred to not having to explain the need for one in any voting session, but at least there was a change - kudos Foundation for that :+1:

1 Like

I very much understand the importance of a “No” vote. This is also a forum for governance debate, so I see nothing wrong with a different opinion. Just because we don’t agree doesn’t mean anyone is wrong. I will be casting a few “No” votes myself this quarter. Thanks for your replies and input!

2 Likes