Ratification of Profit Definition for Round 4

[system]

Ratification of Profit Definition for Round 4

Impact = profit

Optimism’s Retroactive Public Goods Funding (Retro Funding) has been built around the principle that positive impact to the collective should be rewarded as: Impact = profit. This principle serves as a North Star to motivate the creation of a more productive and sustainable ecosystem.

In previous Retro Funding rounds, each Citizen conducted their own assessment of impact and decided how to calculate profit by considering the past rewards received by projects and deducting that amount from impact to arrive at a final OP allocation. This non-standardized approach sparked various debates within the community about whether external funding (e.g., VC funding, Optimism grants, or external grants) should be deducted from a project’s overall public goods funding reward.

Round 4 Definition

In Retro Funding 4: Onchain Builders, a definition of profit will be universally applied to the retro round. Badgeholders will no longer need to calculate profit, but instead, a collective definition will be applied and subtracted from each voter’s assessment of impact as such:

Impact - Past transfers = Award in OP

The possible categories considered as Past Transfers are:

• Grants from Optimism
• VC funding
• Grants from other ecosystems
• Revenue.

All data collected on this categories in the Round 4 application process was for the timeframe of Jan 2023 - May 23rd, 2024.

---

Deliberative Process

Over the past month, a randomly selected subset of badgeholders experimented with a deliberative process to determine this profit definition for Round 4. You can access the public documentation of this process below. We encourage you to review this documentation so you have the full context to vote on this proposal.

• Session 1:
  • Pre-reading Info Kit
  • The plenary session recording for Session 1 is unfortunately unusable
  • Forum recap
• Session 2:
  • Pre-reading Info Kit
  • Plenary session recording (doesn’t include breakout rooms) - Kenncode: 6R2W*ORW
  • Forum recap
• Session 3:
  • Pre-reading Info Kit
  • Plenary session recording (doesn’t include breakout rooms) - Kenncode: U1uN9!6L
  • Forum recap

This Miro board was used to collect input in all three sessions.

---

Proposed Round 4 Profit Definition

Vote 1

Participating badgeholders were asked to determine the profit definition taking into account the specific category definitions and limitations on data collection as it applies practically in Round 4. It is important to understand the nuance and complexity of the practical implementation of defining profit which are distinct from philosophical debates about inclusion or exclusion of each category. Two votes occurred during the deliberative process experiment. The first was on which categories to include in the profit definition. The results are below:

Based on the Collective’s standard 51% approval threshold, these results determined that a deduction weight would only be applied only to Optimism Grants.

Vote 2

The second vote was on the deduction weight that should be applied to Optimism Grants, as broken into three sub-categories:

• Token House growth grants, which are passed on to end users
• Token House builders grants, which are locked for one year
• Foundation grants and Missions
• *Past Retro Funding is not included as the evaluation period begins after the end of Round 3

The responses resulted in a proposed profit definition of Round 4 as outlined below:

A breakdown of votes is shown below:

Ratification

If you wish to ratify the below definition, which will be globally and automatically applied to the voting result calculation in Round 4, please vote “Yes” on this proposal.

Impact - ((10% * Token House Growth Grants) + (45% * Token House Builders Grants) + (25% * Foundation Grants)) = Award in OP

The OP amount deducted from the impact rewards will be returned to the Retroactive Public Goods Funding treasury.

If you do not wish to ratify this definition, a fallback definition proposed by the Foundation will be globally and automatically applied to the voting result calculation in Round 4 instead. This definition is:

Impact = Award in OP

The deliberative process experiment surfaced the amount of complexity, nuance, and effort required to accurately define profit and the challenges in verifying self-reported profit data which is oftentimes not publicly disclosed. In light of these learnings, the Foundation believes it is premature to accurately calculate profit and instead proposes solely focusing on impact in Round 4. The fallback profit definition below, will be implemented if this proposal is not ratified.

Voting will take place on Snapshot and will run from June 20-26 at 19:00 GMT. Quorum will require 30% of badgeholders to cast a vote with an approval threshold will be 51% of votes cast.

[katie]

Ultimately this is disincentiving projects that are building in the OP ecosystem so I don’t agree with this initiative. If we are going to say impact = profit it should not come with any caveats, including prior grants. I will be voting against this proposal.

That being said, I appreciate the deliberative process and thoughtfulness put into this by everyone involved!

[Gonna.eth]

This is my own opinion and does not represent the Grants Council or Govnerds.

@system

45% * Token House Builders Grants

While I like the deliberative process and understand the need for an “Impact” definition, during my time as a reviewer in the Grants Council, I always considered the grant to be 50% of the OP needed to deliver their milestones and encouraged projects to seek the other 50% on RetroPGF once the impact was made. Also, builders granted projects get exposed to market conditions for 1 year before unlocking their grant at their own risk. During season 6 grants will be limited by tiers only, lifting the 50k OP cap. I would not like to see projects asking for more during the grant process because their impact won’t be properly rewarded.

@system

10% * Token House Growth Grants

Growth grants are under a no-sale rule, every last OP token given under this rule must go to end users. So basically projects work for “free” during the distribution process. RetroPGF is about rewarding impact, and I liked the approach of working on impact metrics Retro4 is taking but I believe this acts against what Retro is set for.

@system

25% * Foundation Grants

I don’t know how this type of grant is delivered and lack the necessary context to give feedback.

I strongly believe voting on metrics is experimentation enough for Retro4 and we should wait for results before applying these formulas. I’ll vote against it.

[mastermojo]

I am voting no. I believe we should not be reducing retro funding for projects that are building in the Optimism Ecosystem. On top of that optic wise it does not look good if we reduce projects that received Optimism Grants, but not grants from other ecosystems (Solana, Arbitrum, etc) and VC funded projects.

[jackanorak]

This is admittedly unfair, but before I give myself the time to go on at length as to why I find this to be the case, I’ll simply go early on record as calling this an enormously ill-informed proposal, one that will lead to perverse outcomes. I urge people to vote NO on this.

Builder grants are to get a thing built. Critical milestones are there to show that the thing has been built and that there have been attempts to make the thing useful. If the critical milestones aren’t met, no payment.

Benchmark milestones reflect actual impact, but you can whiff all the benchmarks and still get paid. Why? Because builder grants aren’t compensation for impact; they’re an ecosystem investment in the hope that stuff has an impact later on. For analogy: would a startup book investments as income?

Growth grants don’t even go to the teams, a distinction that people need to fully internalize. I hope the implications of this fact are a little more obvious to the reader.

Foundation grants are more or less like builder grants in that they’re speculative, but I can’t speak to them. Good question to ask the Foundation here is: why 25%? Seems arbitrary.

Attempting to apply weights like this without complete understanding of what these transfers are for — without consideration of the other types of ‘revenue’ these projects get, no less — just goes to penalize projects that have been successful.

Can we please just focus on measuring and rewarding impact without clumsy purity tests like this and the open source thing? I promise there’s more R/R there.

[MattGov.eth]

agreed 100% here @jackanorak - captures my thoughts very well

we’re only now testing around the usage of proper metrics, like sequencer fees, but this looks like a path to actively punishing those who’ve been invested in optimism from the beginning

[spengrah]

I will be voting Yes. This is actually a really tough call (and because of all the factors involved, difficult to analyze quantitatively), but I think the benefit of the discounts defined by this proposal is marginally greater than the drawbacks. That said, I would not be upset if the outcome of this specific vote was No.

I want to take this opportunity to articulate my current perspective on Impact = Profit, which is driving my evaluation of the present proposal as well as how I hope retro funding evolves over time.

Primary principles

These are my primary principles:

1. A project’s Impact is the value it created for the Optimism Collective. It’s important that this be measure over a concrete time period t. Perhaps a better label would be Impact_t.
2. A project’s Profit is the reward it receives for delivering that Impact. This should really be time-bound as well, ie Profit_t.

In other words, Impact_t = Profit_t.

Reward, not profit

I actually don’t think that “profit” is the right label. Profit is a net amount, e.g. reward minus cost. The amount of reward a project receives for delivering impact should not be a function of the cost incurred to delivery it. We want to promote efficient allocation of resources, and so we should not be explicitly covering the cost of delivering impact. If projects can deliver impact for less cost than the reward they (expect to) receive, then they will continue delivering it.

Impact_t = Reward_t

Revenue replacement

Instead, we’re looking for something that behaves more like revenue. Previously, I’ve argued that the OP awarded to a given project is best understood as a form of replacement for the revenue that project would have generated if it had not been a public good. I still believe this is roughly correct.

The potential discounting of other funding sources that has been the topic of this deliberative process is a helpful frame for understanding what I mean.

We can express Reward_t — again, this is what is currently labeled as Profit_t — as the sum of the various forms of rewards available to a project. I think the categories used in this deliberative process are pretty good for that purpose, with one exception:

• Grants from Optimism, which breaks down into
  • Token House Growth grants (THGG)
  • Token House Builders grants (THBG)
  • Foundation grants and missions (FG)
• Grants from other ecosystems (OG)
• Revenue (REV)
• Retro Funding awards (RFA)

That one exception is VC funding. I would broaden that category to any kind of investment with the expectation of economic return, not just investment which comes from venture capital funds. At its most basic, investment — including venture — is a the purchase of a claim of a project’s future profit in exchange for an injection of funds that are expected to help the project achieve said profit in the future. Investment itself is not a reward. When a project receives an investment, the stakeholders in that project expand to include the investors. Any profit the project receives — including OP from retro funding — naturally is (eventually) divided up among those stakeholders.

Without significant up front capital, many of the most important and impactful projects would not exist. It would be a huge shame if only organizations earning sufficient profit from other activities were able to contribute impactful ecosystem goods to the OP Collective.

Retro Funding Awards

Another way to say this is that to maximize the amount of positive impact for the OP Collective, that impact needs to be incentivized — ie, rewarded — sufficiently and in proportion to the amount of impact. The total amount rewarded includes all forms of rewards, including grants from both Optimism and other ecosystems as well as revenue. Often — largely because of the market failure related to public goods — that impact goes under-rewarded. It is retro funding’s job to ensure projects are appropriately rewarded for the impact they deliver by topping up that amount with OP awards.

Mathematically, that looks like this — where A, B, C, D, and E are within the range {0 to 1}:

Impact_t = Reward_t = (A * THGG_t) + (B * THBG_t) + (C * FG_t) + (D * OG_t) + (E * REV_t) + RFA_t

Which is equivalent to this:

RFA_t = Impact_t - (A * THGG_t) - (B * THBG_t) - (C * FG_t) - (D * OG_t) - (E * REV_t) + RFA_t

So, we need to figure out the correct values of A, B, C, D, and E. This deliberative process resulted in the following values:

• A = 10%
• B = 45%
• C = 25%
• D = 0%
• E = 0%

I believe A, B, C, and D (and likely E, to a degree) should be non-zero. So this proposal — while likely imperfect — is better than the alternative.

[spengrah]

@jackanorak

Builder grants are to get a thing built. Critical milestones are there to show that the thing has been built and that there have been attempts to make the thing useful. If the critical milestones aren’t met, no payment.

Benchmark milestones reflect actual impact, but you can whiff all the benchmarks and still get paid. Why? Because builder grants aren’t compensation for impact; they’re an ecosystem investment in the hope that stuff has an impact later on. For analogy: would a startup book investments as income?

The main difference between a builder grant and an investment is that the grant expects a non-financial return while an investment expects a financial return. We can see how this makes a practical difference by looking at a hypothetical comparison between a) a builder grant and b) an investment, both of the same amount X, given by the Token House to the same project that delivers impact of amount Y.

Let’s say that they delivered more impact than they received, ie Y > X.

In A, the project receives X as a grant and delivers Y impact.

• If retro funding discounted builder grants (for simplicity at 100%), the project would receive (Y - X) award from retro funding. In total, it would receive Y reward for Y impact, satisfying the criterion of Impact = Profit.
• If retro funding did not discount builder grants, the project would receive Y award from retro funding. In total, it would receive X + Y reward for Y impact. That’s too much! Because retro funding is not unlimited, some other project would necessarily go under-rewarded for its impact.

In B, the project receives X as an investment and delivers Y impact.

• If retro funding discounted investments (for simplicity at 100%), the project would receive (Y - X) award. That’s too little! The project team may well decide that building for the OP ecosystem is not worth the trouble, and pivot to something else in order to generate a return for its stakeholders’ investment.
• If retro funding did not discount investments, the project would receive Y award. In total, its reward for delivering Y impact is also Y, satisfying the criterion of Impact = Profit. Note that Z goes back to the Token House.

@jackanorak

Growth grants don’t even go to the teams, a distinction that people need to fully internalize. I hope the implications of this fact are a little more obvious to the reader.

Its true that the growth grants don’t go directly to the teams, but they are effectively subsidizing their marketing budgets. Certainly paying your users OP to execute transactions against your contract will result in greater sequencer revenue generated by your contract!

[jackanorak]

@spengrah

The main difference between a builder grant and an investment is that the grant expects a non-financial return while an investment expects a financial return. We can see how this makes a practical difference by looking at a hypothetical comparison between a) a builder grant and b) an investment, both of the same amount X, given by the Token House to the same project that delivers impact of amount Y.

The framework you have is misapplied and unnecessary because builder grants aren’t given to generate impact. No returns, financial or otherwise, are expected or compensated. This is explicitly the case. No double-counting results from retro funding.

@spengrah

Its true that the growth grants don’t go directly to the teams, but they are effectively subsidizing their marketing budgets. Certainly paying your users OP to execute transactions against your contract will result in greater sequencer revenue generated by your contract!

The goal of a growth grant is to socialize a marketing budget in the interest of the Collective, not subsidize a private one (as in many cases, builders aren’t seeking profits). Consider: dev gets a growth grant to encourage use of a product beneficial to the ecosystem, and they generate no income from this product. Are you suggesting that retro funding should deduct this stimulus? What exactly are we doing here if so?

I’m afraid you’d going to dig in here so will preemptively wish you a good day and remind you that there’s still time to vote NO.

[Gonna.eth]

@spengrah

In A, the project receives X as a grant and delivers Y impact.

• If retro funding discounted builder grants (for simplicity at 100%), the project would receive (Y - X) award from retro funding. In total, it would receive Y reward for Y impact, satisfying the criterion of Impact = Profit.
• If retro funding did not discount builder grants, the project would receive Y award from retro funding. In total, it would receive X + Y reward for Y impact. That’s too much! Because retro funding is not unlimited, some other project would necessarily go under-rewarded for its impact.

The issue with this logic is that is missing time. X is approved for the project and locked for one year until you complete the milestones. After being exposed to 1 year of market conditions and if you completed all your milestones you get X which is mainly the 1-year cost assumed by the applicant (their own fund invested in Optimism) and at the same time that year gives the applicant a historical impact to apply for Y.

[spengrah]

@jackanorak

The framework you have is misapplied and unnecessary because builder grants aren’t given to generate impact. No returns, financial or otherwise, are expected or compensated. This is explicitly the case. No double-counting results from retro funding.

From the perspective of the OP Collective, which presumably wants to allocate its own resources as efficiently as possible to achieve, rewarding a project twice for the same impact is not desirable.

I do agree, thought, that the expectations should be clear. If they are not clear currently — and in such a way that likely meaningfully influenced project behavior, then I’m ok with keeping that out of this proposal (which I guess means No on this proposal).

@jackanorak

The goal of a growth grant is to socialize a marketing budget in the interest of the Collective, not subsidize a private one (as in many cases, builders aren’t seeking profits). Consider: dev gets a growth grant to encourage use of a product beneficial to the ecosystem, off which they generate no income. Are you suggesting that retro funding should deduct this stimulus? What exactly are we doing here if so?

I am suggesting that rewarding impact generated by stimulus provided by the Collective would be an inefficient use of the Collective’s scarce resources. Certainly the entire amount of the stimulus should not be deducted — the project must have done some valuable work to convert the stimulus into impact — but IMO it doesn’t make sense to reward projects for impact that the Collective paid for.

@jackanorak

I’m afraid you’d going to dig in here so will preemptively wish you a good day and remind you that there’s still time to vote NO.

Hopefully its clear to all that we are both debating in good faith with the best outcomes in mind for the Collective.

And as I mentioned in my original post, I don’t think a NO outcome here is bad. It’s probably a 52%/48% proposition in my view. I may change my vote to NO if I hear arguments that flip those percentages.

[Gonna.eth]

@spengrah

From the perspective of the OP Collective, which presumably wants to allocate its own resources as efficiently as possible to achieve, rewarding a project twice for the same impact is not desirable.

It is not double rewards, it is the return of their own time and funds invested (grant) and retroactive rewards of their impact (RPGF)

[spengrah]

@Gonna.eth

The issue with this logic is that is missing time. X is approved for the project and locked for one year until you complete the milestones. After being exposed to 1 year of market conditions and if you completed all your milestones you get X which is mainly the 1-year cost assumed by the applicant (their own fund invested in Optimism) and at the same time that year gives the applicant a historical impact to apply for Y.

This is a very good point. The locking really does mess with the intended mechanism of retro funding, doesn’t it

In fact, I would strongly recommend to the OP Collective that they change how grant funds are delivered in order to harmonize with retro funding.

[kaereste]

@spengrah

am suggesting that rewarding impact generated by stimulus provided by the Collective would be an inefficient use of the Collective’s scarce resources. Certainly the entire amount of the stimulus should not be deducted — the project must have done some valuable work to convert the stimulus into impact — but IMO it doesn’t make sense to reward projects for impact that the Collective paid for.

actually, I think it makes perfect sense. With grants we’re incentivising builders to push for specific targeted initiatives that we see as beneficial for the collective. It only makes sense to reward them if those initiatives do provide impact. In my opinion grants and retro funding should amplify, not cancel each other.

[spengrah]

@Gonna.eth

It is not double rewards, it is the return of their own time and funds invested (grant) and retroactive rewards of their impact (RPGF)

Grant locking considerations aside (which I do agree complicates things greatly), in my opinion the Collective should not be in the business of reimbursing for time spent. Rather, it should be in the business of paying for / incentivizing / rewarding impact. It is through that lens that I view grants as a reward for value delivered.

@spengrah

In fact, I would strongly recommend to the OP Collective that they change how grant funds are delivered in order to harmonize with retro funding.

Perhaps one change to consider is to lean into grants explicitly as up-front funding. Size grants sufficiently to de-risk the contribution by the recipient and rely on retro funding to reward recipients for the fullness of impact created. In theory this should mean grants can be smaller and therefore less risky to the Collective to send with shorter (or no?) locking requirements.

[spengrah]

@kaereste

actually, I think it makes perfect sense. With grants we’re incentivising builders to push for specific targeted initiatives that we see as beneficial for the collective. It only makes sense to reward them if those initiatives do provide impact. In my opinion grants and retro funding should amplify, not cancel each other.

If I get an OP grant of X to build something that is expected to create X impact, why should I be entitled to an additional X in retro funding?

To be clear, if what I built ends up creating 2X impact, then I should definitely get an additional X in retro funding.

[kaereste]

@spengrah

If I get an OP grant of X to build something that is expected to create X impact, why should I be entitled to an additional X in retro funding?

you don’t get a grant to create X impact, you get a grant to deliver A, B, C outcomes as described in critical milestones. That’s why it makes sense to provide additional reward if A, B, C created X impact. You can think about it as a performance bonus.

[spengrah]

@kaereste

you don’t get a grant to create X impact, you get a grant to deliver A, B, C outcomes as described in critical milestones. That’s why it makes sense to provide additional reward if A, B, C created X impact. You can think about it as a performance bonus.

I see what you’re saying, though I think you probably mean “outputs” instead of “outcomes” (the latter is basically equivalent to “impact” ). But the purpose of those outputs in the first place is to create impact, no? In other words, the Collective is only interested in having somebody produce those outputs because it anticipate that they will create positive impact.

At this point, I will admit that this is sufficiently nuanced and potentially confusing to maybe warrant punting on the whole question for this round. I really do think the Collective should strongly consider changing how grants work to be a better and clearer fit with retro funding.