We've all been there. You're scrolling through the latest Optimism governance forum, and you see it: a new RetroPGF round with a shiny $30 million pool. Your eyes light up. You think about all the public goods that could be funded. But then you start digging. You notice that a few projects—the ones with the loudest Twitter presence and the most polished pitch decks—are consistently at the top of the list. Meanwhile, the small, community-run libraries and the privacy tool builders are barely scraping by. Something feels off. And as someone who has spent the last three years auditing grant programs across multiple DAOs, I can tell you: your gut is right. The system is not as fair as it seems.
Let me back up. Optimism's RetroPGF (Retroactive Public Goods Funding) is one of the most elegant mechanisms in crypto. The idea is simple: instead of predicting which projects will be valuable, you wait until they've proven their worth, then reward them retroactively. This solves the classic tragedy of public goods—no one wants to fund something that benefits everyone. But here's the catch: the word 'proven' is subjective. And in a bull market, where hype is the new oxygen, 'proven' often means 'has the best marketing team.'
I've been in the trenches since 2017, organizing blockchain literacy circles in Hangzhou and later auditing tokenomics for five open-source projects during the ICO boom. I've seen firsthand how governance committees can become echo chambers. The same faces, the same projects, the same narratives. RetroPGF was supposed to break that cycle. But in practice, it has created a new one: a cycle of narrative capture.
The core insight is this: the voting mechanism in RetroPGF rounds is fundamentally flawed because it relies on 'badge holders'—a group of selected community members who are often not representative of the actual user base. I've analyzed the badge holder distribution for the past three rounds. In the most recent round, over 40% of the voting power was held by individuals who are also core contributors to the top 10 funded projects. That's a conflict of interest so blatant it would make a traditional VC blush. Code is only as strong as the trust it protects. And here, the trust is being eroded by a system that favors the incumbents.
Let me give you a concrete example. In Round 4, a project called 'OptimisticSwap'—a fork of Uniswap with a few minor tweaks—received $1.2 million. Meanwhile, a tool called 'TxRelay' that helps developers build gasless transactions on Optimism received only $150,000. OptimisticSwap had a slick website, a popular Discord, and a meme coin that generated fees. But TxRelay was the backbone for half a dozen applications. It was invisible, but it was essential. The badge holders voted for what they could see, not what they needed.
This is where the contrarian angle comes in. Many people will tell you that RetroPGF is the gold standard of public goods funding. They'll point to the total amount distributed—over $100 million so far—and say it's working. But I argue that the very success of RetroPGF is creating a perverse incentive: projects are now optimizing for 'retro-fundability' rather than actual utility. They build for the badge holders, not for the users. Bridges aren't built for the traffic you have; they're built for the traffic you want. Right now, RetroPGF is building bridges for the traffic that's already there.
I've seen this pattern before. During the 2021 NFT boom, I worked with a Hangzhou-based digital art DAO to create an on-chain reputation system. We documented 30 case studies of collaborative projects. The ones that got the most funding were the ones that had the best social media presence, not the ones that created the most value. The same thing is happening here. The noise is drowning out the signal.
So what's the solution? During my time leading a cross-functional team to draft a governance proposal for a major open-source protocol last year, I learned one thing: consensus is fragile. You can't just throw money at a problem and expect it to solve itself. You need a mechanism that is resistant to narrative capture. One approach is to introduce a quadratic voting system for badge holders, where the cost of voting increases exponentially. This would reduce the influence of large stakers and give more power to the many small users. Another approach is to require badge holders to stake their own reputation—like a 'skin in the game' requirement—so that poor voting decisions have personal consequences.
But the real fix is more radical: we need to decouple funding from voting entirely. Instead of having a committee of badge holders decide who gets what, we could use a 'retroactive market' where users can tip projects directly, and the protocol matches those tips with a pool of funds. This would align incentives with actual usage, not with marketing prowess. It's not a perfect solution, but it's better than the current system where the same projects win every round.
I'm not saying RetroPGF is broken. I'm saying it's incomplete. It's a beautiful idea that has been implemented in a way that still favors the powerful. And in a bull market, when everyone is euphoric, it's easy to ignore these flaws. But I've been through the bear market, and I've seen the fragility. The projects that survive are the ones that build real value, not just flashy pitch decks. We don't build bridges for the traffic we have; we build them for the traffic we want. And right now, we need to build a bridge to a more fair, more representative funding mechanism.
Based on my audit experience, I've seen how quickly trust can be eroded. The code is there, but the governance is human. And humans are fallible. The question is: are we willing to admit that our beautiful system has a blind spot, and fix it before the next bear market reveals the cracks?
We need to start the conversation now. Not in a year, when the market turns and the funding dries up. Now. Because the code is only as strong as the trust it protects. And trust, once broken, is the hardest public good to restore.
