Introduction

<aside> 👉 Some motivating examples from a protocol designer’s perspective:

• Governance contributions. In a token-voting-governed smart contract, the designer wants to incentivize governance professionals to produce the most valuable proposals, participate in constructive debate over other proposals, justify their votes and vote in a coherent way. Rather than compensate whoever has most delegation, the network wants to compensate more those who demonstrated more valuable contributions, eg. with growth proposals which had more impact, more in-depth research and arguments over critical updates, more success in helping reach agreement on more divisive changes, etc.
• Public goods funding. Part of a L2 network’s governance is a commitment to distribute part of its sequencer fees to public goods built upon its stack. The designer wants that distribution of funds is maximized towards builders of public goods which displayed most welfare increase (and/or user perceived value) for its ecosystem. Rather than, e.g. gathering ratings from a Citizen House which weakly optimizes for expertise and has no guaranteed alignment, the protocol designer wants to leverage the most expert community members and enforce alignment with a shared vision.
• Airdrop farming. A protocol leverages incentives to bootstrap participation. The designer wants to incentivize participants in the most neutral way, namely those who do not display aggressively opportunistic farming behavior that could distort the protocol (like entering in liquidity position then abruptly removing liquidity once rewards are claimable). Rather than simply applying fixed reward rules, the designer wants to incentivize behavior in a more fine-grained and constantly-improving way, so as to leverage more and more data and capture a wider and wider set of misaligned farming behaviors. </aside>

Rationale

Overall, these cases have in common that a designer (which includes the governing body) should rather optimize the allocation of the resources under control of the network in order to maximize a given objective (eg. growth in governance contributions, sustained growth in DeFi farming, welfare produced by PGF).

Usually, blockchain or on-chain protocols are designed so as to optimize resource spending for security, which is controlled by the protocol itself (e.g. difficulty adjustments in Bitcoin). But furthering goals agreed on within a network is usually left to governance decision systems which sit outside of the protocol. See https://vitalik.eth.limo/general/2021/08/16/voting3.html.

We aim to include contributions and incentives within the purview of the designer and provide economic guarantees on the alignment of those contributions with an objective.

<aside> 👉 In our examples, contributions are defined in the following way:

• Governance contributions: proposals documents, complex proposals which require spending and project management (e.g. Uniswap Foundation’s protocol fee activation proposal), feedback on existing proposals, votes.
• Public goods funding: public good projects that comply with predefined criteria.
• Airdrop farming: protocol participation, typically, in the form of liquidity provision or trades. </aside>

We are left with two questions:

• How to define the objective within a network?
• How to further the objective with cryptoeconomic guarantees?

In the following, we will assume that the protocol designer’s main purview is treasury allocation (including monetary inflation) and aims at increasing the number and quality of contributions to the protocol that are aligned with the network’s objective.

Objective definition and evaluation

The most usual definition for an objective at the network level is network value, often represented by the market capitalization of a token (utility token or governance token with financial rights). Other measurable objectives can be defined as well (eg. TVL, funding, carbon footprint…). Qualitative, non-measurable, objectives are often embodied in a mission statement or a constitution.

Depending on the nature of the protocol and the maturity of the network, the protocol designer might choose a different way to define the objective.

In the case of measurable objectives, contributions’ alignment can be evaluated via prediction markets that measure the probability of increase in the objective quantity. The most direct application of prediction markets would be a futarchy where such contributions are enacted whenever the market odds are in their favor.

When the objective is of a qualitative nature, prediction markets can’t be directly applied. We will explore below how to leverage a jury to evaluate alignment of contributions under such conditions.

<aside> ☝ Some real-world example objectives include:

• A constitution document, like the MakerDAO constitution.
• A set of objectives for contributions, as the ones defined in Optimism’s Retro PGF rounds.
  • These can notably include metrics, hence producing measurable objectives. But usually, these measurable objectives only partially represent the decision space. </aside>

A radically different approach consists of letting network participants themselves define the objective, thus replacing the protocol designer’s role. This option is not explored in this document.

Network-defined objective and external evaluation

The protocol designer defined the objective either: