Here is the article for those interested. Vitalik shows how some systems like EOS and Bitshare lead to bribes.
Note also that the above is not a criticism of all on-chain voting; it does not rule out systems like futarchy. However, futarchy is untested, but coin voting is tested, and so far it seems to lead to a high risk of economic or political failure of some kind - far too high a risk for a platform that seeks to be an economic base layer for development of decentralized applications and institutions.
In Vitalikâs article, the âbadâ voting systems elect a set of the X highest ranked candidates to receive funding. You do not need to be the best to receive funding, so as long as you can bribe enough people to be in the set, bribing can be profitable even if the majority is honest.
This is not the case in my proposal, if a recipient contract doesnât get the majority, it doesnât become the recipient. This makes bribing way harder (as bribers would not earn anything until they can bribe the majority).
In his article, Vitalik also talks about asset futarchy (see this article on how it works). I am a huge prediction market enthusiast and an alternative proposal could be:
No validator voting, just use asset futarchy to determine the best splitter contract. I.e. let people trade on ETH price conditional on moving or not moving to the new splitter contract and only move if it would lead to a higher conditional ETH price.
Because validators have incentives to set the rate in order to optimize
E[Extra Value to ETH] - Loss of staking revenue
They would not be incentivized to move the funding rate past its optimum.
Actually the proposal leads to underfunding as the extra value to ETH is shared among all ETH holders, while the loss of staking revenue is only for stakers.
Here, I am starting to like more the idea of taking part of the burn instead of staker revenue as this would not lead to underfunding and it may be more socially acceptable (contrary to my initial assumptions) from what I read on X.
It is hard to know exactly what will be needed in the future (quantum resistance? more scaling due to AI demand?) but I am pretty sure more work will be.
Indeed, the EF currently operates without much âcompetitionâ. In this proposal, weâd see teams with faster delivery get more funding and urgency would increase.
If there are better development methodology, we should fund those.
Burned fees may actually not be that bad as it doesnât underfund (cf my previous post). I see more people suggesting that, so maybe we should have an updated proposal.
If we whitelist recipients, we lose credible neutrality. However there may be something doable at the social layer:
The idea would be that the splitter contract only funds mechanisms (and not final recipients directly) such as quadratic funding, protocol guild, asset futarchy, tx fee optimization futarchy, etc.
In this case, it would be more clear if moving to a specific splitter contract is âan attackâ or not. Iâd be in favor for this to be part of the proposal even if it is only at social layer and not at the technical one.
Actually it is made to be the simplest possible at protocol level. From the point of view of the protocol it is just âDo we change the recipient address from X to Y?â. The voting complexity of finding the best splitter contract according to a set of preference is done at the client level (and even then, it can still be quite simple).
I think itâs a good idea, my initial thought was that people would be opposed to it as it would reduce the supply reduction (i.e. lead to more ETH supply), but feedback doesnât align with my initial fears.
Indeed, itâs very unlikely that this proposal would implemented as is, but we had to propose something to open the Overton window.
Yeah, competition between recipients should lead to that.
Yeah, âdeflate lessâ is basically equivalent to âinflate moreâ. If we still want low/negative issuance, this could be done in an hardfork which also decreases issuance (JerĂ´me de Tychey is pushing on that).
I think we agree on the diagnosis more than it seems. Your point about it being unfair to ask people to contribute for free to enrich ETH holders deserves to be followed to its conclusion.
What youâre describing is a structure where developers bear the cost of public goods while extractors capture the value those public goods create. The profits are privatized, the costs are socialized. This is how the current global financial system operates, and Ethereum has inherited its logic wholesale â every successful application built on it so far has been an extraction apparatus. DeFi, MEV infrastructure, token issuance, and now treasury companies directing protocol research toward their balance sheets. The more Ethereum succeeds under this model, the richer the extractors get, and the poorer the people maintaining the infrastructure become.
This is why funding mechanisms cannot fix the problem, however well-designed. Redistributing a portion of extraction revenue back to contributors is the same structure as government taxation â and it faces the same capture dynamics MicahZoltu identified: whoever controls the allocation controls the money. The thread above illustrates this precisely. Every proposed solution adds a layer of mechanism design, and every layer creates a new surface for capture. This problem is not solvable at the allocation layer.
It is solvable one layer below: at the application layer. The root cause is that every major application on Ethereum is extractive. The solution is one application that is explicitly non-extractive â a system where value cannot be silently detached from the relationships that created it, where extraction is structurally unprofitable at the monetary layer.
Refusing to censor how others use the infrastructure is necessary and right. But when the people who maintain the public goods also build their entire livelihood on moral neutrality, the structure becomes unsustainable.
That means the ecosystem needs an application with a clear value stance that generates enough economic activity to sustain the entire contributor community. And that economic activity cannot serve the same extractive purpose as the existing financial system does and as every application built on Ethereum so far has done â it has to be a self-sustaining exchange system built on fundamentally different values and social relationships, one where value stays anchored in the real interactions that created it.
Since âprivatize profits, socialize lossesâ is arguably the most pressing structural problem in the real-world economy, an application designed to oppose it would carry enormous real demand. The transaction volume, gas consumption, and capital inflow from such a use case would make infrastructure maintenance naturally profitable â the way Linux kernel contribution is naturally profitable because the world recognizes Linux as critical infrastructure. The funding problem disappears when the ecosystem has a purpose the world actually needs.
The idea that existing financial institutions will migrate their operations onto Ethereum is a fantasy â they have no reason to move to infrastructure they donât control. The only way Ethereum captures real economic activity is by running a system that resolves the contradictions the current financial order cannot escape. Let the new world run on Ethereum, and Ethereum thrives. Try to serve the old world, and Ethereum becomes a redundant database with a gas fee.
I am an emissary form the distant land of Ethereum Classic. Some of us noticed this thread, and that we had a similar debate in 2021. In the end, we decided not to implement a protocol layer treasury.
I wrote extensively on the topic at the time, and while not everything applies these days to ETH, some of the arguments against the treasury are outlined at ethereumclassicclassic[dot]org/#blockchain-seppuku
The main thing is that protocol neutrality is of supreme importance, and if you bake in a little bit of subjectivity like a DAO or voting system, it will probably tend to collapse in on it self and snowball into centralization. You also burden holders with additional responsibilities they didnât sign up for.
The fact that a system âcouldâ potentially be captured is not a reason not to do it.
If we were excluding those systems we would not have modern societies, governments, companies, non-profits, cooperatives, all can be captures under particular circumstances. We canât just reject a system because it âcould be capturedâ, but should work to make systems the most resilient against such risks.
Note that not doing anything also has capture risks, as if the protocol canât pay its workers, some other actors may and would be the ones having power of it (capturing it).
Maybe this is why we need futarchy. I think going full futarchy may be premature but once the system becomes proven, this could be an option.
I think weâre talking past each other slightly. My argument has nothing to do with whether this specific mechanism could be captured.
The problem originates at the application layer. Nearly every major application on Ethereum is extractive by design. The value created by public goods flows into applications that privatize the profits. Even if your proposal worked perfectly, the funded developers would build public goods, and those public goods would continue to be consumed by extractors to capture value that flows right back out.
This creates a power dynamic that funding mechanisms cannot counteract. Extractors accumulate more resources. Resources translate into influence over what gets built next. The development agenda drifts toward what serves the largest holders. We are already watching this happen: the Foundationâs budget is being cut by 40% while treasury companies with holdings worth tens of billions are stepping in to direct protocol research toward institutional adoption, because that is what serves their balance sheets.
A stake-weighted voting system would formalize this dynamic. A small number of coordinated large holders, unified by a single financial interest, would cast a bloc vote pointing in one direction. The remaining validators, dispersed individuals with scattered preferences, would dilute each other. The allocation would converge on what the largest coordinated bloc wants, cycle after cycle, each round reinforcing their position as returns flow back into more stake and more voting power. The end state is an ecosystem whose research agenda is disproportionately aligned with the financial interests of its largest extractors.
The only way to escape this trajectory is to make Ethereum the infrastructure for something fundamentally different from what the current extractors need. An application that could redefine the beneficiaries of the current blockchain ecosystem.
If that application embeds distribution into the monetary layer itself, for example a universal transaction tax that funds a universal basic income for every verified human, with escrow and dispute resolution built into the transaction layer, it gives people a reason to anchor their economic lives to the system.
When people can earn, spend, invest, and survive inside a system that provides fundamentally better condition than the outside world, they stay. When they stay, they generate real economic activity: transaction volume, gas consumption, capital inflow through the reserve channel.
That economic activity is what transforms Ethereumâs position entirely. Right now, many among the Ethereum community are trying to convince financial institutions to migrate their operations onto it, competing on speed, cost, and compliance against infrastructure those institutions already control. That is ridiculous and self-contradictory. If a genuinely compelling application runs on Ethereum, one that solves THE problem the existing financial system structurally cannot solve, Ethereum stops competing for adoption and starts generating it organically. It becomes foundational infrastructure the way Linux became foundational infrastructure: indispensable because something the world depends on runs on it.
In that world, maintaining Ethereumâs protocol is no longer a public goods problem that requires funding mechanisms and allocation committees. It becomes one of the most valuable engineering activities on the planet, the way maintaining the Linux kernel is. Companies, service providers, and investors whose livelihoods depend on the system running will compete to fund and contribute to its maintenance, because the cost of letting it degrade exceeds any contribution they could make. The funding problem disappears once the ecosystem has a purpose that makes contribution self-evidently worthwhile.
The question is whether such an application can exist. I linked a whitepaper in my previous reply that attempts to sketch one out.
I think the question of whether the Ethereum community can become an open-access order may be a false goal. Every governance design, every contribution system, every allocation mechanism will relocate the access boundary somewhere. The community is a group of people with unequal resources and competing interests, and that will remain true regardless of the rules.
Where open-access can actually be achieved is at the monetary layer, through an application running on Ethereum.
Consider what a universal transaction tax embedded in money itself would do. A tax high enough to make lending structurally unprofitable eliminates the need for credit scoring, compliance infrastructure, and bankruptcy courts, because there is no debt. Without debt, the entire apparatus of capital intermediation that determines who gets to participate in organized economic activity loses its reason to exist.
No incorporation required to receive investment, because collaboration structures are permissionless smart contracts with on-chain operating data visible to everyone. No bank account required, because the monetary system is the bank. No auditors required, because every transaction is publicly verifiable. The barriers fall away because the mechanisms that created them, lending and the intermediary infrastructure it demands, are gone.
And the tax revenue itself flows into a universal basic income for every verified participant, so that anyone can unconditionally accumulate the capital needed to participate, without requiring prior wealth, credit history, or institutional permission. Open-access is achieved through the structure of money itself.
This can only work on a platform where the contracts enforcing the tax are truly immutable, because anyone whose intermediary business model is threatened would lobby to change the rules. That rules out any chain where a small group of validators can be coerced into altering state. The reserve asset must have no issuer who can freeze it, which rules out anything backed by stablecoins. The execution environment must be credibly neutral across jurisdictions, because a government that considers its financial intermediaries threatened will attempt to shut the application down.
These are structural requirements that only Ethereumâs infrastructure meets. If such an application succeeds, Ethereum becomes foundational infrastructure the way Linux is: indispensable because something the world depends on runs on it. The governance of the Ethereum community becomes as irrelevant as the governance of the Linux community is to the billions of people whose systems run on Linux.
I linked a whitepaper a few posts up that sketches out what such a system might look like in detail. I think you would find sections 6 and 7 on organizational primitives particularly relevant to your research on contribution systems.
One more thought building on your point about contribution systems.
I think the reason goes deeper than governance design. Under the current monetary structure, value can be transferred at zero cost, fully detached from the relationship that produced it. This is what makes capture possible in the first place. An insider who extracts value from a contribution system can move that value away from its source instantly, invisibly, and without friction. The tools for betrayal are built into money itself.
This creates a dynamic where even well-intentioned participants are pulled toward extraction. In a system where anyone can capture value and walk away at zero cost, cooperation becomes the dominated strategy. If you contribute honestly while others extract, you subsidize their extraction with your work. The rational response is to extract before others extract from you.
The contribution system degrades, and the usual diagnosis is âbad governanceâ or âinsufficient capture resistance.â The actual cause is that the monetary substrate makes betrayal free and cooperation costly. Under these rules, designing a capture-resistant contribution system is like designing a boat that stays dry in a flood. The water is coming from below.
This is not a blockchain-specific pathology. It is a condition that has been present in every human society since the invention of money. Blockchain did not create it. What blockchain created is the first tool capable of changing it, because immutable smart contracts can enforce monetary rules that no human institution could be trusted to maintain. The reason blockchain has not yet fulfilled this potential is that every project so far has treated the monetary structure as a given and built on top of it, rather than recognizing that the monetary structure itself is what needs to change.
The only way to change this dynamic is to change the payoff structure at the monetary layer. If every transfer of value carries a real cost, betrayal stops being free. If all value flows are transparent and on-chain, betrayal stops being invisible. If the proceeds of every transaction automatically flow to all participants through a universal basic income, the private gains from betrayal are socialized.
When betrayal is expensive, visible, and automatically redistributed, cooperation becomes the dominant strategy. The prisonerâs dilemma dissolves, because the three conditions that sustain it, zero-cost defection, private capture of gains, and unobservability, are eliminated simultaneously by the structure of money.
This is why I believe the answer to your research question, how to build contribution systems that donât degrade into limited-access orders, lies at the monetary layer rather than the governance layer. And it is why Ethereum is the only platform where this can be implemented: the rules must be enforced by immutable contracts on a censorship-resistant network, because any entity whose intermediary profits are threatened by this change will attempt to shut it down or alter the rules.
This would be the case with a naive voting system like first past the post, not with the proposed design.
The proposed designed probabilistically converges toward the Condorcet winner if there is one (Smith set in case of Condorcet paradox, but this is also fine). So there is very little advantage to strategic voting, therefore coordination of large actors would not give them an edge (again, I emphasize that this is because I proposed a âgoodâ voting system, using first-past-the-post would lead to what you describe).
Stakers would not need to coordinate, but simply to specify their preferences to their staking clients or validators.
On the contrary, the status quo leads to a prioritization of the interest of the large actors.
Assume:
a. 10 small actors which would each get 1 if a project is funded. Funding the project costs 6.
b. 2 large actors which would each get 5 if a project is funded. Funding the project costs 6.
For a. it is very unlikely for the project to get funded, as it would require the coordination of at least 7/10 actors to be net positive (here they would benefit 1-6/7). And most actors would want to be on the non contributing set (which gets straight up 1).
But for b., coordination is easy, both actors know that if they coordinate, they get 5 - 6/2 = 2. While if they donât, they get a guaranteed 0.
Opt-in contributions favor large actors, while the propose systems doesnât.
Letâs not forget the EF 2024 report [1] about treasuries across the ecosystem, where e.g. Optimism had 3.5B, Uniswap 3.1B, Mantle 2.5B, Gnosis 1.8B, EF 970M, and MakerDAO had 301M, while UniSwap funded the Ethereum ecosystem with 10.7M (LOL), EF with 240M, and MakerDao 76M. So the distribution is totally skewed. We have to work on finding ways to force extractive systems to contribute back. Funnily enough itâs also in their best interest. Itâs the so-called tragedy of the commons, biting us. I find this proposal a step in the right direction.
Just my 2 cents.
[1] Itâs the report-2024.pdf file under Ethereum Foundationâs website. Itâs worth a look. Very stark, and very sad.