As a new badgeholder, but someone who has been very active in reviewing projects and observing the process, I’d like to offer a few spicy takes 
This is a repeated game
We need to zoom out and remember this is a repeated game. If five cycles from now we are still rewarding the same projects with the same proportions of funding, then what’s the point of all this governance overhead?
These rounds should be reflexive to there being different forms of impact that get rewarded in different proportions over the life of this game.
- Do we want to reward protocols that attract lots of new users to the superchain?
- Do we want to reward efforts that deter scammers and make it easier for people to bring more assets onchain?
- Do we want to reward projects that promote decentralization and network resilience?
These proportions might change over time because the relative impact of these efforts may change over time for a fixed value set. These proportions might also change as our value systems become more well-defined and thought out.
The impact categories and the majority of lists give very little signal to what forms of impact are most valuable. That has to change if we want to win a repeated game.
No one ever got fired for putting geth at the top of their list
There’s a risk that we are too conformist and lose the benefits of having 150 people with different perspectives and values sets.
It’s good to have a culture of going outside our comfort zone and advocating for projects that have had an outsized impact in less obvious areas. Props to the badgeholders who have done that work. Even more props to those who have dug up dirt on projects that aren’t a good fit for RetroPGF.
I’m NOT saying that geth shouldn’t be on people’s ballots or that it should be low in the final allocation. I’m only saying that there’s a huge benefit to people having different values sets and they should vote assertively.
If the impact you think is most valuable to Optimism follows a power law distribution, express it that way.
Impact is not pornography
Finally, there’s a certain mindset that is prevalent in the public goods space that I want to push back against. It’s the belief that measuring impact is like defining pornography.
If you don’t get the reference, there’s a famous court case where the judge said (about pornography):
I shall not today attempt further to define the kinds of material I understand to be embraced… but I know it when I see it.
If we believe it’s impossible to get better at quantifying impact, then we should go back to a much simpler form of governance, where you have a council of experts who decide what’s a public good and what isn’t, what’s impactful and what isn’t, what’s good funding and what’s bad funding, etc. In that world, you have a lot of ring-kissing and lobbying efforts directed at the inner circle, asking for their blessing, etc.
Wait! That’s how public spending works in traditional governments. We don’t want to go there…
Not going there means that we have to accept that our priors about what is impact and what is profit will always need updating after every round.
Not going there means accepting that there will be some questionable projects that slip through the review process BUT it’s worth it because hidden gems will come in this way too.
Not going there requires the experienced people with battle scars and strong views on impact to do the meta-level thing and help turn that into a framework that can be applied more consistently. Stop criticizing the process on social media and draft an improvement proposal.
Not going there means giving constructive feedback to the various efforts such as the RetroPGF 3: Impact Evaluation Framework, Pairwise, and some of my technical lists that are trying to bring rigor to the process.
I genuinely believe this is the most interesting economic experiment in the world and would hate to see the outcome of it being that we reinvent effective altruism.
The correct things are easy to write about but very hard to actually do. The collective needs to empower people who are going to do that hard work and get behind them.
/end rant