Kevin,
There’s a lot to love about your vision and obviously I am a huge fan of it. I hope GitCoin will become one of the most important models in the world for coordination and I hope to be there right along side you helping you figure it out.
In that spirit of supporting you all, I think there are some key things about the best (most socially beneficial, most scalable and most profitable) business model for GitCoin that this post doesn’t do a great job of capturing. It runs to the core of the post, because it relates to the way that it misses the essence of plurality and what it means to relate as a part of a plural infrastructure. I think this arises from falling into certain conceptual traps that are very common in the blockchain world and that I hope you can move beyond to achieve your ambitions.
Let me start by trying to illustrate why I don’t think the vision literally as you sketched is realistic or desirable. Then I’ll turn to a trajectory that makes more sense and try to sketch how the technology and business strategy needs to evolve to follow that trajectory.
Let’s imagine that we tried to run QF/GitCoin on a truly global scale based on an open PoH identity system. Let’s divide how this could play out into a few cases:
- You manage to get a large chunk (let’s say 50%) of the world population actively participating. In this case, it seems very likely that there will be at least public good (say related to climate) that will get at least a $1 contribution from at least (say) a billion people, given that even without matching some disasters have elicited responses of this magnitude. QF “wants” to match this will a quintillion (10^18) dollars, so even if you had a constrained multiplier of 1/10,000, you’d still need to raise more money than all of global GDP to do this, and in the process you’d end up with a multiplier that made matching basically useless to any normal cause. In short, a global QF process is simply infeasible.
Furthermore, even if you managed to somehow make the math compute and get control of most of the world’s resources (neither of which seems desirable or feasible), the allocation of these resources would almost certainly be decided by which existing organizations in the world have the greatest combination of coercive power and internalized solidarity…China, for example, which has much higher social cohesion that liberal democracies and more surveillance powers would almost certainly manage to an order of magnitude more contributors and an order of magnitude higher contributions to any cause it favored, capturing essentially all matching funds. Now, in principle one could imagine policing such “collusion”, and I’ll return to this below, but given the sheer array of potentially “colluding” groups in the world, doing this by hand rather than by basic changes to the architecture of QF would force GitCoin to become precisely the kind of centralized surveillance state it abhors. And hoping for a world without organizations of solidarity other than through QF seems both entirely vain and deeply contrary to a spirit of pluralism that seeks to build on the accomplishment of existing means of coordination, not replace all of them. - You don’t managed to get a large chunk participating (say less than 1%). In this case, Proof of Humanity is of almost no use in deterring Sybil attacks. While it ensures each person gets only one vote, 99% of the planet is available to be recruited by an actual participant to create an effective fake account for them. One would then be stuck with either a ridiculously small matching pool that wouldn’t attract such attacks or being sucked dry by Sybil attacks.
- You’re in the 1-50% range. In this intermediate range you’d end up with a mix of the first two problems.
In short, the basic set up of QF/GitCoin cannot and should not scale in an open, PoH based way globally. This doesn’t mean there isn’t something awesome going on here that should scale, and I’ll return to it soon. But it does mean something quite a bit stronger than the first reaction a lot of the audience here is likely to have: - These issues are not a result of a flaw in QF as a solution to the decentralized public goods framed as coordinating individual self-interest towards common goods. QF is the optimal solution here and anything else based on the same open, PoH, individualistic thinking is going to run into the same mix of problems: either it will be too weak to do much, be open to Sybil attacks or eat up the whole world/be controlled by existing coordination mechanisms.
- In fact, these issues apply, albeit in somewhat less extreme forms, to other applications of PoH. Let me give two examples. If you try to do validation by PoH, either you are going to need like 50% of the world running validation nodes, which seems wildly unrealistic, or you’re going to be open to Sybil attacks by people recruiting large numbers of those who aren’t currently running nodes. If you try to do UBI by PoH, either you’re going to need a large chunk of global GDP and/or the payments are going to be so small they will only make a difference in the lives of the poorest people on earth, who anyway are extremely likely to be subject to some form of coercion or unable to access their PoH-based account in the first place. The open PoH approach makes it impossible to, as other UBI studies have done, target some local community with large UBIs without requiring control over a large fraction of global resources.
What’s going wrong here and what does it tell us about how to do things better? Fundamentally, the framing of the problem is missing a key element. It assumes that it finds individuals in an atomized state, caring only about their individual self-interest, without any institutions that can help solve free rider/public goods problems. It then provides incentives to overcome these. Yet, if we really lived in a world like this, civilization would pretty much not exist and we’d be in a Hobbesian war of each against all. Thankfully we don’t; there are a range of institutions (yes, I used that taboo word) at a range of scales, both formal (like governments, unions, corporations, etc.) and informal (like feelings of patriotism, altruism, kin ties, racial solidarity, etc.) that avoid this outcome, and there always have been from the dawn of humanity. Because these institutions exist, the problem of coordinating a billion people is not really a million times harder than the problem of coordinating a thousand people, as the atomized QF approach assumes. It is probably closer to being on logarithmic scale (say 6 times harder).
Now you can think what you will of these institutions and how well they perform their assigned roles and I know that most in this community will tend to dismiss them out of hand. But even if you don’t love them, you cannot ignore them for three reasons: - Practical: Even if you don’t like these institutions, there’s little to no chance they are going to completely disappear soon enough that GitCoin and other things in the Web 3 space can safely ignore the impacts they have on behavior. If you do, you are simply solving the wrong problem, potentially with catastrophic effects as noted above.
- Ethical: It would be the height of arrogance to imagine that we are confident QF or any other Web 3 mechanism will always and everywhere lead to better coordination than these existing institutions without even experimenting with them significantly. Thus, we should not want to wipe out these other institutions, if ever, at least until we’ve had a chance to play with these mechanisms in a range of setting of increasing ambition. But as we do, we had also better hope these existing institutions don’t collapse, or we’ll end up in Mad Max along the way. Thus, we must plan to coexist and even cooperate with them, not for the world after they disappear. And if you really believe in pluralism like you suggest, that pluralism should be far broader than just “other stuff floating around the blockchain ecosystem”…there are more things in Heven and Earth, Kevin, than are dreamt of in your philosophy.
- Stability: Even if, somehow, you did end up in a world where all other institutions disappeared and QF ran everything, you wouldn’t stay there long. The whole point of QF is to form cooperative communities. And one of the most reliable features of human social psychology is the tendency to form bons of affection and solidarity with those you engage in a common project with. So the minute you start running QF, you will see “institutions” form, as a result of the mechanism itself, and these will then undermine the operation of QF as noted above.
Ok, so, Mr. Gloomy Face, riddle me this: how has QF/GitCoin not turned into a disaster so far? And is there any path to scaling it successfully?
I would submit that the success of GitCoin so far has been largely due to the mismatch between the way this post imagines GitCoin and the way it operates in practice. It is not really a global open system. Instead, it has operates in specific communities, of similar scales, that roughly approximate the members thereof having nearly symmetrical social distance and the administrator having sufficient knowledge of the community structure to use various bespoke mechanisms to police “collusion” (viz. solidarity outside the mechanism), as well as potential Sybil attacks not just in terms of non-humans, but also people who don’t belong in the community and are being roped in just to help someone win.
But while this closes one avenue for growth for GitCoin, I think it opens another that is even more exciting. All around the world there are organizations and institutions like those mentioned above that a) want to see a better, more responsive and dynamic provision of public goods within them, b) lack information about which goods have the greatest value and c) know quite a bit about social structure internally which would allow them to police the way GitCoin has in the domains it knows. A natural business model is to create a 3-sided market between donors, sponsors and projects within a range of these institutions to strengthen them rather than replace them. For example, most companies have major problems with different divisions with diverse objectives and KPIs failing to provide common infrastructure that supports the company wholistically (my office at Microsoft largely exists for this reason). If QF could be thoughtfully applied, say, to representatives of these divisions, it could surface the needs for shared infrastructure.
Does any of this require or even comport very well with a blockchain? Not particularly. Given that the intention is not to have something that is global or open even in the PoH sense, where identity is deeply bound to community and most naturally to accounts that the company has already, the overhead associated with blockchains doesn’t seem to be a very good fit. But, at some level, who cares: you have a great business model that can do all kinds of good. Don’t get overly committed to a particular tech stack.
Now, eventually, you should want to go beyond these specific applications to specific communities if you can get them to work; you should seek, say, local or state governments that want to subsidize the creation of public goods across local communities, companies, etc. within the locality or state. These somewhat larger entities will have some read on community structure, but probably a not to easily directly police issues around collusion, participation, fake accounts etc. So, you will want algorithmic structure to help do all this.
But the basics of QF won’t even come close to getting you there. First, PoH won’t do much to police the boundaries of participants, as the community will be a tiny fraction of the world and you’ll need some “proof of community/membership/authority”. Second and even more importantly, membership of that community won’t be the primary thing you want to track anyway…you’ll want to be able to reduce matching for common projects between those who already are coordinating: those in the same companies, same ethnic group, same religion, etc. But even with PoH or something of that sort, you’re not going to get rich community information like that and algorithms to run on it. That’s a big step from where we are. I have an essay (“Why I am a Pluralist”) coming out soon about how it might all work in open standards, and my piece on the RxC blog on Intersectional Social Data hints at it. But it would certainly be as big a break from the current “Web 3 ecosystem” as that is from Web 1, Web 2 or perhaps even pre-internet world.
Perhaps this all sounds a bit daunting…perhaps it is. But it is also exciting and suggests that GitCoin and the space it is cracking the door of is even more important and ambitious than you suspect. I truly believe this type of thinking can transform and empower every person and organization on the planet, that it is core to our future and that we’ve just started to imagine how it might all work. Think, for example, of how we can do QF over time with those alive now, those that have died and those that will live in the future to address the long-term future of humanity?
You’ve got the Apple IIc. It portends a revolution. But not far down the road are the GUI, the internet, mobile computing, etc. Keep and open mind. Not all the answers are even imaginable now. But if you do, GitCoin has a real chance to change the world.
Best,
Glen