Proposal Discussion: Bounties + Reputation

realitycrafter

Note: This is a more formal introduction of the concept of a merit/reputation system that has been bouncing around the governance channel in various forms over the past week or so. Shout out to @0xMega#0028, @Joseph_Eager#0579, @CrystalGravy#5077, @tnorm#7124 and @RobertoT#5565 for contributing to this conversation and helping to shape the concept into more concrete terms.

In the interest of enabling our members to contribute autonomously to the development of materials and projects, as well as to recognize their contributions we propose a Bounty and Reputation system for ATX DAO. This system will extend the existing Committee Budgets to include in these seasonal budgets an allocation of an ATX Reputation token (REP). REP will be awarded by Committee heads to DAO and community members that make contributions of their time and talents to our organization's ongoing projects and initiatives. This REP can later be traded in for membership minting access or outright membership. It can also be leveraged in the future by the DAO for other forms of recognition for stellar members.

Clarity enables Committee heads to create tasks and assign token based bounties to them. Completion of these bounties can then be verified by authorized users and rewards itemized + distributed in regular intervals by our treasury. Committee heads are recommended to set both UST and REP based rewards for tasks and to allow bounty hunters to choose which reward they would like to receive when submitting the completed assignment for review.

This system solves for two issues (as identified by 0xMega):

Existing members have equal skin in the game in terms of identifying contributors and rewarding contributions to the DAO
Non-members can contribute value immediately with a quantifiable outcome that counts toward full membership.

We suggest that REP be launched as an ERC-20 on the Polygon network, as this will enable distribution at scale and a seamless integration with Clarity as they launch their support for the Polygon L2.

We suggest the following transition

Onboard DAO members onto an internal Clarity bounty board. Committee Heads will create tasks with UST rewards (distributed at set intervals according to the Committee Budgets proposal) and/or delayed REP rewards (tallied and tracked until token launch + airdrop).
Launch REP tokens and open an external bounty board with USD/REP rewards
Begin a round of membership minting where REP can be exchanged for access+membership

Would love to hear thoughts on specific numbers for distribution here. I'm imagining something like...

200 REP = Membership mint access (just holding, no burn/swap)
1000 REP = Swapped/burned for membership so that we have a clear conversion from membership pricing.

512mace

Why would DAO members want REP tokens? and if they don't want REP tokens, why are they being offered UST? Does offering UST seem fair to people who are contributing without UST incentives?

What are the standards for valuing tasks in REP/UST? Who is going to be assigning this value?

For committees where tasks are not being completed by others, are committee heads going to be paying themselves?

SamPadilla

I echo some of the questions Mace brought up above. I def see the need for a bounty system but want to understand the details. I have a few questions below and my thoughts for each.

- What is the criteria for assigning value (both in REP and UST) to the tasks?
Maybe each task valuation proposed by the committee head should be a proposal the community approves or not. Or perhaps the members attribute a minimum and a maximum amount to a given task and the committee head gets to choose the value within that range.

- Will members receive only UST for the bounties they complete or also REP? If both, what is the value of REP for members? Any expectation of implementing REP as the new measure for voting power?
I see the value in having members be only paid in UST for their bounties and non-members be paid only in REP if there is no plans on attaching extra voting power based on REP. If there is a future in which (extra) voting power is attributed via REP (which itself deserves a lot of deliberation), then the nature of the task should define the payment.

- What happens when bounties are not completed?
No ideas here. Open to thoughts.

I want to stress that I am very skeptical and cautious of governance systems in which the governance tokens are sensitive to monetary contributions. I have no issues with REP being considered only for memberships, but if there is a future in which it starts affecting voting power, we should deliberate extensively on the matter.

realitycrafter

SamPadilla

Requiring a proposal for each task is incredibly inefficient. A natural market will form just based on which tasks are being performed and at what prices. The goal here is to decentralize the work being done and to open up access to contribution opportunities. If we were to do a proposal for every task, no clear market would ever form. Admittedly, it is a rough and slow market, given that the Committee heads are a human approximation of the market makers, but perhaps we can find ways of scaling this further. Certainly once enough REP has been distributed, folks with their own REP might be able to set up tasks for DAO members on a more custom bounty board.

In the proposal as written, members would choose between the UST and the REP. But this is a suggestion for Committee Heads, not a rule. REP will be a tool alongside the Committee UST budget for heads to incentivize productive activity.

Keep in mind that REP is also valuable for community members interested in becoming fully fledged DAO members. It is a way to access governance rights, albeit indirectly. It is a tool for measuring contribution that has been given free of charge.

When bounties are not completed, the incentives are not aligned and the rewards should be increased.

realitycrafter

Initially DAO members might want REP tokens as a way to guarantee that they can secure mint spots to friends. We haven't defined all of the use cases for REP, but I can see it becoming a foundational piece of how we operate and determine who has provided value to the DAO. So some of the accumulation may be speculative.

In short, REP is a way for us to track who has given to the DAO without receiving back from it. It is a way for us to track who should be most rewarded in the future when we find ways to give back to our community.

Initially I would expect the value of REP to be set by the amount of REP that we determine someone could swap for membership. The past pricing and expected future pricing of our membership will allow us to approximate value for our first couple tasks. The value of REP over time will be determined by the Committee heads over time as they assign tasks. If tasks are only being completed for UST and aren't being completed for REP, perhaps the ratio should be adjusted. If a committee head is running out of REP and not UST, perhaps it should be adjusted in the opposite direction. Putting all tasks on the bounty board allows people to choose and for a natural market to form around the value of tasks.

One way we could approach the payment of committee heads is to pay them a fixed amount of REP at the beginning of the season, and perhaps a variable amount of REP distributed across the Committee heads based on voting from the rest of the DAO at the end of the season.

realitycrafter

One more clarification: The UST for these bounties is coming directly from the Committee budget. It is a way of leveraging our funds into more value for the DAO albeit on a smaller scale than most of our full sized projects + events.

Things like: getting a design system put together, writing internal guides and reviews, writing articles for external publication, operating a particular necessary role. These are all important things that it would be so helpful to have way of measuring and tracking in terms of contribution. Or just things that we should be outright paying for!

josephf

I think giving points as reputation score is a good idea. These reputation points grant status (like stackoverflow) but don't give a guaruntee of extra voting or monetary value. I've talked to other community building experts and they say that paying purely in money can destroy a community.

There maybe be some tasks that are paid in both UST and REP points, but most tasks are paid in pure REP. Does Clarity allow dual token reward tasks? I don't know that we even need an on chain ERC20 token yet if there is a system that can reliably track the workflow of: 1)task creation, 2)assignment, 3)verified completion, 3)payment.
I think a discord bot that allows points to be given would be an option to consider rather than a token since we use Discord so much, it is easy and we don't want these points to be transferable. For example the points discord bot: https://top.gg/bot/484395597508509697

Perhaps a member has to earn a certain number of REP points before they can start earning UST. Only committee heads can earn UST at first.

realitycrafter

josephf

I don't necessarily see an issue with allowing people to transfer these points, but I dont think we should create a liquidity pool or incentivize one. The use case of getting friends in is reason enough to allow for transfers.

I do like the idea of requiring a certain amount of REP before accessing UST tasks. Perhaps we can have reputation levels and tasks that require a certain REP level to access. This way more important tasks are done by folks with a vested interest in the DAO.

josephf

An Alternative suggested name instead of REP could be "THX" for "Thank you points" which is kind of a fun symbol. (If this is not an ERC token the three letter symbol doesn't matter)

Instead of a discord bot, perhaps a google spreadsheet would be sufficient. After task are marked in clarity as "ready to pay" a member of the operations/treasury committee adds the points to the spreadsheet or database. Everyone else has view-only access to the database or spreadsheet.

realitycrafter

josephf I do like THX as well, although I feel that REP might be more suitable if we expect members to be exchanging it among themselves.

It may actually be less overhead to launch a token since Clarity automatically integrates with wallets as user IDs. As a token we also have a greater range of options for wide spread usage and plug + play use cases.

512mace

This seems like a good idea to me and I appreciate the time and effort that was put into thinking this through because it is definitely something that the organization needs. It takes a lot to come up with a process like this and I think it will be a useful addition - thanks!

If were to post a bounty for brand guidelines at X REP and someone were take the task and complete it but their work was unsatisfactory, would they get REP points anyway?

realitycrafter

512mace

This is a great question. One way we could approach this is to have deputized reviewers in each comittee that review the work done before accepting it. A set of clear acceptance criteria would be essential to ensure that we're accomplishing the goals and needs of the DAO.

We could also include a graded scale in which reviewers assign a grade to the work completed and the bounty is paid out to a percentage tied to the grade. A 1-5 scale with 20% intervals perhaps?

realitycrafter

Some of these solutions I'm suggesting aren't automatically built into Clarity, but would need to be imposed by process. So Clarity doesn't have a dual token bounty or a graded scale system built in, but we could leverage other fields or comments to build on top of the existing system.

josephf

What do you think about the idea of trying to tie REP points to hours of work? Where we pay something like 25 REP points per hour. Time is the most democratic unit, we all have the same amount each day. (and some tasks could pay 25 UST+25 REP per hour)

I think this is how effective in-game economics work as well. There is a certain amount of time you have to put in to level up, and everyone who plays the game knows that time (and expertise) were required to achieve higher levels. (or you purchased an item from someone who put in the work).

This also creates a base value for REP tokens.

We could use a tool like toggl.com to record time worked in a verifiable way. This might be too impractical to implement initially in all cases, but we should think about ways to make the system trustless, robust and scalable which might eventually mean some type of integrated tooling for verifying time put in.

We will also need committee heads to review quality of work.

brennanmulligan

If we plan on allowing this token to be transferred, I think that we should be very careful about how this fits in with the larger mechanism of the DAO, especially any current plans for other ERC-20 tokens to be released.

I think that a REP token would be much better suited to being non-transferrable, much like a person’s actual IRL reputation. This takes a lot of the stakes out of the tokenomics, and would make it much easier to offer both UST and REP rewards on tasks. In addition, there would be less potential overlap for the city coin alternative Roberto briefly mentioned last night.

We could still use it as a gating mechanism for NFT minting (perhaps this is the mechanism by which we award early / free mints), but if it is not transferable we can be more sure that the person getting this special access to the DAO is the same person who actually completed the work.

Also, I am not sure whether we should allow REP to be spent by a member to grant a non-member access. Should someone be able to jump to the front of the line because their hardworking friend wants them in the club, even if they have not contributed themselves? Would someone being “nominated” via REP in this way still need to apply or would they be able to mint automatically?

All in all, I do think this is a good idea and an important one for us to execute on. Huge props to Jesse for putting this together.

brennanmulligan

One more thing. How would tasks be added to each committee’s backlog?

Would any DAO member be able to propose a task? I think this would be a useful means of decentralizing ideation / requirement generation. From there, would only the committee head be in charge of assigning REP bounty to each task? If so, would there be an appeal process by which someone could dispute that bounties?

There may be conflicts if a committee head and committee members have different opinions on how work should be prioritized, so it would be wise for us to consider a dispute resolution process - though that may just be accounted for in the fact that committee heads are elected, and their seat could be contested in the next election cycle.

josephf

Does clarity have a view to show how many points have been earned per person?
Maybe that is all we need for now. that would make the REP points are non-transferable if they were only tracked in Clarity. This would also mean no gas required.

I think we are ready to start creating tasks and asking for volunteers, at least on trial basis.

There is going to be a bit of work for the committee heads creating, assigning, and verifying tasks. We need a dedicated project owner to run our REP points bounty program in clarity. This role could be our first bounty task.

Ncasares

Love all of the dialogue here. A few thoughts I'll chip in:

REP vs UST
I'm not a huge fan of allowing tasks to be completed in exchange for UST. I understand that it gives the committee more purchasing power but, IMO, REP is an option for aspiring members to contribute value and receive recognition prior to joining the DAO. The payment in REP is akin to sweat equity that can be redeemed at a later date for a membership token. My concern with UST-based payment is that it could drive a mercenary culture where people pick off tasks for UST even if they have no intention of joining the DAO. That doesn't mean there isn't a place for that kind of contribution but those types of relationships feel better aligned with a vendor/contractor mentality.

REP monetary value
I agree with the other comments about staying away from assigning monetary value to REP. Assigning monetary value (directly or indirectly) will almost certainly drive short-term thinking and speculation. I'm also not convinced that the REP system needs to be on-chain. If REP is available in open markets, it defeats the purpose of rewarding the actual contributor (something that should not be transferrable).

Task sizing
This one is tough because of the variable nature of work. One thought here is that we could use a sprint task approach and stipulate that all bounty tasks need to be assigned a point value (t-shirt size, Fibonacci, whatever). Tasks that are to big (or too small) to fit in point buckets will need to be decomposed or grouped. This creates a little more work for the committee but it has several benefits: 1) the only agreement needed to instantiate work is the alignment on bucket sizes and typical tasks. This is something committee heads should be able to coordinate and calibrate over time 2) a potential contributor can better evaluate the amount of effort required to complete a task 3) The additional upfront work of sizing tasks will force committees to clearly articulate the work to be done and establishes a vernacular for effort across the DAO. 4) A system that standardizes REP against common task sizes allows each committee to better estimate and plan their work against their REP allocation.

Task completion
I think we should have a consistent process for reviewing and accepting work. The rubric should be decided upon by the grading committee but a consistent process is key for setting good expectations with the community. In cases where work is rejected, we should consider a baseline reward for situations where a contributor tried to complete the tasks but the work wasn't up to par. This will hopefully incentivize a growth mindset across the DAO that encourages "leveling up" instead of "failing".

realitycrafter

I've distilled more of these concerns and questions into a larger revised proposal here:
https://app.clarity.so/atxdao/pages/1b0db5f0-0b40-44c5-8a5f-47e248bc4e2c

REP Proposal: Reputation and Bounty System

In the interest of enabling our members to contribute autonomously to the development of materials and projects, as well as to recognize their contributions, we propose a Bounty and Reputation system for ATX DAO.
This system will extend the existing seasonal Committee Budgets to include an allocation of Reputation tokens (REP) for each Committee. REP will be awarded by Committee Heads to DAO members and community members who make contributions of their time and talents to our organization's ongoing projects and initiatives.
Note: Healthy token ecosystems have both faucets (ways in which tokens are distributed) and sinks (ways in which tokens can be spent or used). We expect more faucets and sinks may be developed in later proposals, but will be providing a baseline set of expectations in this proposal.

What are the uses for REP tokens? (Sinks)

At its core, REP is a way for us to track who has given to the DAO without receiving back from it. It is a way for us to track who should be most rewarded in the future when we have opportunities to give back to our community.

Membership: Access to membership minting or traded in for outright membership.

Community members holding 200 REP will have guaranteed early mint access in our next membership round
Community members will be able to swap 1000 REP for a membership NFT (no other cost)
The past pricing and expected future pricing of our membership will allow us to approximate initial value for REP. At the rate of [1000 REP = Membership], with our most recent round of Membership valued at 0.78704 ETH, and the current price of ETH (on 3/26/2022) ≈ $3100...

1REP ≈ $2.44 (for the purposes of obtaining membership)

Contributor Levels: A way to display and track contributor status.
During our next round of membership minting, an additional minting portal will be opened for current DAO members to convert their REP into Contributor Levels. These levels will be used to track who our most valuable contributors are and gate opportunities/access to these contributors. Contributor level will be displayed on discord and checked when assigning tasks in Clarity.

Contributor LVL-1 NFT
- Trade in value: 500 REP
Contributor LVL-2 NFT
- Trade in value: 1000 REP & LVL-1 NFT
- Members holding a LVL-2 NFT can take on tasks with UST bounties.
Contributor LVL-3 NFT
- Trade in value: 3000 REP and a LVL-2 NFT can be traded in for a LVL-3 NFT
- Members holding a LVL-3 NFT can take on tasks with UST bounties.
- Members holding a LVL-3 NFT are official representatives of ATX DAO for the purposes of interviews and public speaking engagements.

Note: While these are the base uses for the REP token, this system can be leveraged by the DAO in the future to provide a variety of other forms of recognition for stellar contributors.

How will REP tokens be distributed? (Faucets)

Bounty Board

Committee heads will use Clarity to create tasks and assign REP bounties to them (see "Task Sizing" below)
- Committee heads can also assign UST bounties to tasks, however it is recommended that only these tasks only be available to members with Contributor Level 2 or above
  (see "Contributor Levels" in the sinks section below).
Members can acquire and submit tasks at biweekly Committee meetings
Committee heads will rate the success of completed tasks, rationale and associated payout percentage (see "Task Success Ratings" below)
Completion of these bounties can then be verified by users with the 'reviewer' role.
Rewards will be itemized + distributed in regular intervals by our treasury
- CSV backups of approved rewards will be kept at least until the end of the following season.

Committee Budgets

Each Committee will receive a 5000 REP budget per season.
This budget can be increased for a particular Committee in the current season via a 'Budget Increase' proposal. Budget increases carry forward to subsequent seasons.

Task Sizing

Small: Quick, simple tasks taking no longer than an hour or two
- Rewards: 10-20 REP
Medium: Tasks that require several days or specific skills to accomplish
- Rewards: 20-100 REP
Large: Tasks that require weeks, coordination of a team or very valuable skills to accomplish
- Rewards: 100-500 REP
Ongoing: Regularly occurring sets of tasks that are assigned as a role to a committee member, lasting over a month.
- Rewards: 500-2000 REP

Task Success Ratings

Excellent: 100% of Reward
Adequate: 80 - 100% of Reward
Minimum: 50 - 80% of Reward
Lacking: 0 - 50% of Reward
In cases where work is rejected, we should consider a baseline reward for situations where a contributor tried to complete the tasks but the work wasn't up to par. This will hopefully incentivize a growth mindset across the DAO that encourages "leveling up" instead of "failing".

Committee Head Payments
Committee Heads will be paid the first half of their REP salary when they are elected or continue in their role at the beginning of a season. The other half of their REP will be distributed on a sliding scale based on community voting at the end of their term.

How will this system be rolled out?

We suggest the following transition

Committee Heads will create tasks with associated REP rewards (tallied and tracked until token launch)
Onboard DAO members onto our internal Clarity bounty board.
Launch REP tokens and open our bounty board up to external community members
Announce our next round of membership minting where REP can be exchanged for access+membership

Token Technical Details

REP will be launched as an ERC-20 on the Polygon network. (This will enable distribution at scale and a seamless integration with Clarity as they launch their support for the Polygon L2)
Initial Token Supply : 1,000,000 REP Tokens

Errata

Q: Is there any expectation of implementing REP as the new measure for voting power?

A: No. In fact, to prevent this from occurring through the conversion of REP into membership NFTs (which currently provide one vote each to the holder), we suggest the following change to our snapshot system:
- Every member has one vote, regardless of the number of membership NFTs they hold.
- Genesis members who were recognized for their contributions with multiple membership tokens can give them away to add new members or convert them back into REP points by donating these tokens back to the DAO.

Q: What about all the work that folks did before the Reputation system? Will there be a retroactive airdrop?

A: REP will be used as a measure of all work going forward from the Austin On-Chain event. Efforts have been made by the core team to track and quantify the work done before this event.

Q: Why is Reputation a transferable token? Wouldn't reputation be better suited to being non-transferrable?

A: Technically all of our membership NFTs are transferable, so preventing the transfer of Reputation would only slow transfer capabilities. The transfer of Reputation is a way to vouch for another member of the community. It also leaves the door open for a system to form at some point in the future where community members can fund their own bounty tasks with Reputation rewards.
- Note: We will not be contributing to or endorsing the creation of a liquidity pool. Given the low circulation of these points and regular conversion of them back into NFTs, we expect that if a liquidity pool should form, slippage would make it unusable.

Q: What happens when bounties are not completed satisfactorily?

A: See the "Task Success Ratings" section.

Q: What is the value of REP for people that are already DAO members?

A: See the "Contributor Levels" section.

Q: How would tasks be added to each committee’s backlog?

A: Committee heads are responsible for creating bounty tasks and maintaining their backlog according to the approach described in the "Bounty Board" section

Q: Will there be a dedicated project owner to run our REP points bounty program?

A: Each Committee head will be responsible for creating, reviewing and approving tasks for their Committee. Our treasury committee will be responsible for payouts on a monthly basis.

Proposal Contributors

Significant contributions were made by Nick Casares (0xMega) and Joseph F (JosephF) for the "Task Success Ratings" and "Task Sizing" concepts. Discussion and input from CrystalGravy, Brennan Muligan, Mason Lynaugh, Roberto Tamalas, Sam Padilla and Weaver informed some of the thinking and language throughout the rest of this document.

josephf

Very interesting thoughts here on this v2 proposal Jesse, nice work. I think we need to figure out how to simplify it so that it will be managable. One idea to allow for fewer+larger tasks is to allow teams of people to work on a REP task, and split the REP tokens among themselves as they see fit. That would make it easier for the Treasury team and for Committee Heads to manage and verify task completion. I think if we create tasks that are small 10-20 REP, that might be hard to manage.

How long is a season? 1 month? 1 quarter? 1 week? I would vote that 1 season = 1 month.

Regarding this statement: "This budget can be increased for a particular Committee in the current season via a 'Budget Increase' proposal."
I don't think budget increases should be decided by public proposals. Frankly, all of our public proposals always pass and so it is likely that every committee's budget will just keep growing and growing. This inflation would degrade the value of REP. There needs to be a budget allocation meeting where committee heads debate over allocation of a fixed supply of REP each month. One committee doesn't get more without another committee getting less. There can be some inflation per month consistent with ATX DAO growth, but this needs to be optimized and not allow hyper-inflation.

Clarity is going to be difficult in its current state because it doesn't have a way to Total up tokens rewarded per committee. How do we verify that each committee is not giving out more than their 5000/season budget? Do we also track all tasks in a google spreadsheet, a database or a more established tool like expensify/quickbooks?

Who is going to mint the REP token? Clifton? Jesse? Engineering? Should we mint the REP token on Eth mainet and bridge over to Polygon so that REP has more credibility/security and so it can be bridged to other networks more easily?