On Governators

The following is a brain-dump ahead of a Metagov Govbase call (7am 29 Sep 2023 AEST) in which we will discuss the concept of governators.

A validator is a machine contributor with a human minder. The contributions a validator makes include proposing and attesting blocks. When a validator fails to perform — either through downtime or malicious behaviour — rewards are paused and penalties may be imposed. These boundaries are prescribed by the protocol rules.

A governator is a human contributor with a machine minder. The contribution the governator makes is ensuring that risks are mitigated and roadmaps are met. A governator operates within the governance surface of the protocol, which may include changes to the protocol rules from time-to-time (which becomes a dual governance system as validators take part by enacting software upgrades). Rewards and penalties are issued in accordance with the outcomes of the governator’s contributions.

The above definitions are my own, derived from my ethnographic research into validators and governance processes in Proof of Stake (PoS) blockchains.

My definition of a validator is fairly standard, except perhaps the contributor component. As an ethnographer, I look at experiences and actions rather than mechanisms, which has led me to see validator rewards less as incentives to encourage deposits and more as payment for performing tasks, including producing blocks and securing the network. The tasks are repetitive and automated but the attention of the human minder is an essential component.

My definition of a governator requires a bit more explanation. PoS blockchains have so far neglected to reward outcomes within the governance surface — at least at the protocol level [1]. Smart contracts exist that can make some administrative tasks easier, and there are some tools that assist DAO members to record their contributions, including activities that are intended to produce outcomes. The term governator specifically applies to the human-machine configuration that contributes to governance and receives rewards in PoS blockchains that have been designed with onchain governance.

I am still tracing the genealogy of the term ‘governator’. Josh Lee (Keplr and Osmosis) used it in the context of validator governance at the 2023 Blockchain Infrastructure Forum, suggesting that validators need to be able to delegate votes to governance experts, who should receive a portion of validator rewards in return. Josh sent the group a link to a Governator Manifesto post and thread initiated by Jacob Gadikian (Notional DAO), which credits Jim Yang and other Osmosis figures for developing the idea. Zaki Manian had floated similar ideas to me after a Validator Commons forum in 2022. It also comes up on various forum posts, although not always using the ‘governator’ terminology.

In these conceptions, a governator, like a validator, is paid in protocol rewards. This is in contrast to dominant governance models in which Foundation teams pay protocol DAO delegates (such as in Maker) or DAOs that make protocol decisions (such as C1 on Juno). When the idea was discussed in Vienna it was meant as a way to for validators to defer governance responsibilities to individuals or entities. It is therefore an evolution of on-chain governance in that it seeks to release those validators who do not vote (typically due to liability concerns or lack of capacity) to delegate their voting power.

Potential problems

A governator that is a delegate of validator votes could maintain plutocratic governance dynamics. If validators who have a large portion of stake are able to select a representative then it means that those who are receiving those delegations can be captured by the capital interests of the validator and/or their investors.

Such a model also encourages the worst of the politics-media nexus, favouring PR strategies and possibly election cycles. The result is that diversity of ideas/worldviews, experience and careful behind-the-scenes work goes unseen while the loud, confrontational or popular attain power.

One way to deal with this is for governance rewards to be sent to a pool, with rewards then allocated based on contributions (a version of this occurs with Polkadot’s OpenGov). Validators who wish to participate in governance can do so alongside others who are not running validators. It’s likely that many validators would continue to participate in governance as validators have an interest in the longevity of chains whereas the token holders who delegate to them are more likely to be in it for short term gains.

The opportunity

The opportunity of blockchain institutions is to reward contributions that are recognised and valued, as opposed to rewarding for publicity or showing up (other systems these older formats already exist, such as labour contracts and advertising revenue). Various groups have built and experimented with contribution tools, including SourceCred, Govrn, and Coordinape (see my ethnography of SourceCred here, which explores the dynamics of contribution-based governance). Existing DAO tools do not provide an out-of-the-box solution for governators, but they do demonstrate ways of doing that lean towards recognising contributions as an organisation function of permissionless systems.

Another frontier is experiments with LLMs in organisational contexts (such as BlockScience’s), which may evolve into the machine helpers that are able to see if a governator’s processing of ideas, anticipating risks and offering expertise result in stability, fairness, resilience or partnerships (or other indicators typically used by boards and committees to monitor and evaluate). This is an area that Metagov is beginning to explore.

For governators to work they need to achieve corporate governance objectives such as dealing with agency problems (minimising the power of insiders). A significant problem with validator governance power is that validators have been known to vote against decisions on one chain that go against their financial interests in another chain — a problem that will get worse if/as app chain ecosystems grow. The smart contracts that assist governators could automatically excise them from decisions where there is a conflict of interest and issue penalties for failing to declare or deal with such conflicts. The extent to which a governator’s actions are public is an important question to consider. Reputational tokens could allow governators to prove their experience or attributes and possibly do so in a privacy-preserving way if desired (including credentials such as legal expertise, geographic and gender diversity). Puja Ohlhavar’s work is a good starting point for considering privacy, reputation and collusion-resistance.

Conclusion

When designing governators we need to consider:

• A tool (the machine helper) that is able to see someone’s contributions and tie these to achieved outcomes for the protocol.
• An automated means to prevent conflict of interest.
• Demonstrate credentials and experience.
• Minimise the influence of political public relations and media.
• Possibly enable teams to take on governator duties whilst ensuring these do not become a backdoor for collusion.

From a theory perspective, if governators do eventuate, they provide a chance to observe a system that departs from both representative democracy and corporate governance. The former achieves consent for one group to rule over others, while the latter can help deal with agency problems but can also create its own fiefdoms within protocols. The purpose of governators should be to enable coordination in permissionless systems via the governance surface. If governators are conceived and designed as contributors rather than token holders or citizens then the emphasis shifts from choice (politics) to action (contributions). Governance then becomes a matter of value creation rather than preferences.

[1] Optimism has a version of this already in that they provide retroactive funding for governance activities, using network revenue (from sequencing) for rewards. Kleomedes is a DAO validator that is building tools for governance rewards using validator income.

I am a Professor at RMIT University in Melbourne, working across the RMIT Blockchain Innovation Hub and the Digital Ethnography Research Centre. I am also an Associate Investigator (AI 🤖) of the ARC Centre of Excellence for Automated Decision-Making and Society and a Research Director in Metagov. I acknowledge the support of the Australian Research Council, FT19010372. I use this Medium blog for work-in-progress ideas and reflections.