Online Governance Surfaces and Attention Economies | Ronen Tamari | Michael Zargham | Rena O'Brien |

And we are back with another exciting panel for you all. So, as many of us love, know, and love crypto for decentralization, decentralization is great, but it offers a lot of drawbacks, especially when it comes to coordination. And so, on this next panel we're going to be talking about governance and how we make decisions in blockchains. So, for our next panel we've got online governance surfaces and attention economies hosted by Nathan Schneider from the University of Colorado, Boulder. Please give a round of applause.

Hello. Good to see you all and as as my teammates here file in, I just want to say first of all that this is a talk that comes out of work that we've been doing for the past couple of years at Metagov. Metagov.org, you can learn about the work that we're doing around imagining futures and building futures for online governance. So, this project is the result of a a collaboration between me, Kelsey Nab, and Ronin Tamari, and Michael Zargham.

Kelsey can't be here today. She's on another continent, but we're also joined by Ren O'Brien from Gitcoin to tell us whether we're on to anything here. But, the focus here is something that a bunch of us have observed over the years working around governance in Web3 and beyond. Many different kinds of context where we see attention as a critical focal point for the challenge of governing our communities and organizations. And this builds on a challenge that goes way back.

For, you know, many generations people have tried to imagine what it would be like to have a world that is governed by its participants from the ground up, a kind of deeper democracy, a kind of deeper participant involvement, whether you call it the the cooperative commonwealth or the you know, the DAO revolution. But if we had that kind of world, if we won, what would we do? How would we manage to participate in all those DAOs, all those cooperative companies, or whatever it is that we're engaged in? We imagine very quickly that we're going to run out of attention. And especially so when we're working in the context of digital networks where information flows so freely and so wildly.

So this talk is exploring some some research that we've been doing into thinking how about how people are are already working around attention in the design of governance and how we could do that more intentionally and more clearly. So we looked at the theory of commons governance, like the literature, the kind of OG Elinor Ostrom, Yochai Benkler, and other researchers that were theorizing about the commons early on. And what we noticed, and this is kind of one of the sparks that triggered us to work on this project, was that they actually didn't reckon with attention as a limiting factor, even though they were very much about participatory governance and thinking even some of them in the digital age when we had a lot of online communities, they weren't really noticing that attention was going to become a limiting factor. In practice, we kind of know how this turned out. They were kind of web one theorists.

And then web two was what happened when information overload essentially broke human attention. And what happened was then that platforms stepped into this void and they made huge investments in the infrastructure, the tech that basically outsources our our attention governance uh to these platforms. And we all know that that didn't turn out so well. Web three is trying to step in and say "We're designing the online governance systems for ourselves, but what we saw was that if we want to do that, we really need to reckon with attention again." And this will require a lot of investment uh both in the tech the infrastructure, and thinking more about the social aspects that the web two platforms didn't do.

All right. So, here in the paper we made an effort to sort of demonstrate this organization as more of a living system where the participants who are applying their attention towards making various kinds of decisions in turn uh affect what takes place and the outcomes that are wrought. And an important aspect of this this model that we we've laid out in the drawing is that the the participants exist within an action arena or in a circumstance where they have some choices. First of all, are they participating in governance at all? They could be participating in other communities, doing other things.

They may simply have better things to do. But assuming that they've decided that they want to be part of the participatory base that governs the the community or organization they're part of, they also have multiple ways to engage or modes through which they interact with governance. This could be voting, it could be discussions on forums, it could be um going in and starting a new project in some of the more bureaucratic organizations. But the governance surface is this set of ways that you can channel that attention into like effective action within the org or action at all, maybe not necessarily effective. Whether it's effective depends on what happens.

And so there are outcomes and those outcomes have impacts or effects on themselves feeding back around as well as effects on other stakeholders, maybe those that are not participating in the action arena. So using that model we went and had a look at different organizations and the ways in which they intervened in their attention economies. So they they sort of um we we recognized that in DAOs this is a very common pattern of using token voting. That's a mechanism by which decisions are taken, maybe funds are dispersed from a treasury or certain activities are given sanction to proceed. And in this case there's a bottleneck around votes.

How do you get people to participate in voting? How do you get them to become informed? And so each of our three case studies, Gitcoin, DAOstack, and Governator proposed interventions at the site between the token holders and the token voting in order to reduce the attention costs of participation in that process, ideally without muting or eliminating their effects. So they're adding less attention but still being heard, um that their opinions and preferences are still being manifest. And so in Gitcoin there was a stewards-based system, in DAOstack there's a boosting mechanism, in Governator they were proposing individual agents that could vote or recommend votes on the holders' behalf.

And what we found interesting about this is that using the previous structure you can see that the interventions, while different, are all in the same place, which tells you something about what the pain point is perceived and experienced by the members. But we also noticed that these interventions don't necessarily improve the situation. You often introduce a new mechanism to help improve the direct arrow from token holders to token voting, and if that attention cost is saved in the immediate sense, you may still not have saved the attention cost when you account for operating or maintaining your intervention. So, there's other efforts that go into educating and curating stewards. There are other efforts that go into sort of calibrating or financing a boosting mechanism.

There are other efforts that go into maintaining the ML ops or security of an AI agent. And so, it's not clear that the attention costs are a net savings, but that's something that differs across the individual projects. And then, finally, having made those observations, we came back and we developed some heuristics, things that participants in DAOs, as well as um folks trying to design them, can kind of think about when they're architecting their DAOs. So, what are the modes of um interaction? So, the ways in which you can go about participating in governance and how much attention or what kind of attention do they require?

What are the organizational processes or technical processes that manage or distribute attention? What information is provided to participants? So, how do they become informed and how does it expect their expectations? Are you dropping a ton of documents and expecting everyone to read them? Are you providing nice summaries and back-linking to reports?

Information design can play a big part in the attention cost of informed participation. Incentives, so you're spending your own time and effort, like that labor um can actually be something that is incentivized. In some cases, that can create positive effects or negative effects, but what are the incentives and how are people responding to them? And ultimately, most and most importantly, feedback. How does the the decisions that are taken um get recognized as affecting the outcomes that occur, which feeds back on people's future decisions to allocate attention and to participate, and even feeds back on our organization itself in terms of revising the way that we structure these systems.

Yeah, maybe just to add to that. I think one of the sort of this isn't you don't see it here, but we we were playing with this term of like return on attention where we see participants investing attention in a space and expecting to get something in return and to sustain an organization, you kind of have to have that return on attention be positive. So, people have to feel like they're getting more than they're putting in and and this is your this heuristic helps us kind of see where the attention is being invested, how much attention are we asking people to invest and then what they're getting out of it. So, it's helping us become more intentional about designing our organizations and how people are investing attention. We think is really crucial for designing the the communities of the future.

Thanks. One really critical takeaway that we um uh that is kind of a fundamental uh observation here is that attention is already become a problem in Web3. And this is something that that was observed from some of the earliest developments in building tooling for uh communities in this context. People were already designing, as in our case studies here, to try to figure out how to how to tackle the challenge of uh attention economies. How do you uh manage uh and direct the attention of potentially thousands of token holders?

And this is not a new problem. You know, I've been doing this work in conversation with work I'm doing with large cooperatives, for instance. Organizations that are not on chain, but have participatory voting and and things like that. And they too have to think about a lot of the same challenges here. And they've come up with solutions.

For instance, here in Colorado, we have 42 rural electric cooperatives around the state and they uh bring a person in a robot suit to their annual meetings and have really good food. You know, that is one tool for attention economies that attracts people in, it draws them in for participation in governance beyond just the the act of voting. A few things that we hope to see this provocation encourage. First of all, is more intentional design around uh stakeholder attention in governance, both technically and socially. Some of the projects that we saw being less effective in addressing these concerns were the ones that simply tried to put a technical fix on the problem.

And the ones that seem to be making progress were the ones that recognize this as a social challenge as well as a technical challenge. This involves recognizing that when you encounter unexpected behavior in your governance system, like for instance, very low token voting percentages or kind of mob-like behavior, the kinds of failure modes that we see all over the place in Web3, that rather than what I often see, which is kind of blame the the the victims of the attention economy, blame the people who aren't voting or something like that, instead takes that in as data. Takes that in as an opportunity to reconsider the design of the attention economy in your governance system. Take it as feedback. And expect attention economies to evolve.

You know, the attention economy of a small, early-stage community is going to be very different from a large, mature, you know, high high-activity community. It's going to have to be designed very differently differently. You're going to have to make sure to channel attention intentionally in very different ways as an organization evolves. And so make sure that you're set up to intentionally shift those attention economies as as the organization evolves and as the kind of realities of those economies change. These are some thoughts we'll leave you with for now, but next we're going to turn the conversation over to Reyna O'Brien, who is executive director of Gitcoin, is in the trenches of the attention economies of governance and web 3.

She's going to share some reflections on her experience leading an organization in this context, and then we'll have a conversation. Take it away. Whatever's comfortable. Thank you. I talk and and talk and Thank you.

Um if you haven't had a chance to read the paper, I will say it's a fantastic read. Um there's a ton of great information in there if you're looking at your own organization as well as some of the history of the attention economies. Um I'll give you a little bit of a brief overview and first again just want to say thank you for having me on the panel. You guys are always great. I wish we could have just recorded our conversations in the green room and played that um cuz we got a little spicy back there, but I joined Gitcoin in 2023 as chief of staff.

And at that point we had already started to see a dwindling in terms of our attention economy within the community. When the DAO was formed, we had some fantastic contributors who were super excited about governance and token holders and liquid democracy and we're going to change the world through public goods funding. We still kind of are. Um 67 million gone through to the program over the last x amount of years. Um but as these gentlemen have pointed out in these slides and in the paper itself, there's a whole bunch of different uh mechanisms that we want to look at when we're thinking about our community, how we're doing governance.

And what I found when I came in is that the governance people that love doing governance weren't looking at what our community was doing. And figuring out frameworks that they thought would work great internally and then rolling them out to a community that didn't have the time, the effort, or the incentives to care about what we were doing. So, started off with liquid democracy. We were able to give delegations to people to be able to vote on the proposals and the conversations that were happening in the forum. And what happened is that most of that was just given to internal contributors.

Great, that's fantastic. The people that I trust to make the decisions are literally the ones making decisions internally. I have no more context on what I can do about a budget or what the crew is doing internally, so I'm okay with that. We attempted to take in a couple of different versions of this to make it more accountable. And this was something that's mentioned in the paper and that myself and Z have talked about.

We have the health cards. Well, that's fantastic. I can see a score of how much somebody participates in governance, but do I know if they're good at making decisions? I can tell you over the last 3 years, given that we've gone from nearly 70 contributors down to 22, probably not all of them were making fantastic decisions. And with that being said, we went from these large, over-the-top figuring out structures internally and centralized and try to roll them out to a community that had no idea what our mission statement was and then wondered why governance was breaking as professional delegation companies are coming into the space.

They're all fantastic. They're great, too. But, you know, for 8K to 10K a month, they're going off to the Arbitrums and the Uniswaps to push through compelling proposals on what to do on a protocol. Do we think that they care on what we do with a budget request? Not really.

So, incentivizing the contributors with information that's actually interesting and compelling to them, especially in an age where both personally and professionally we are drowning in information and starving for wisdom, doesn't really make a ton of sense. We've gone back to both out of like a forcing function in terms of having the capacity to run governance, a keeping the lights on method. And what I thought was really compelling as somebody who does give a about governance is that it didn't matter that we were only keeping our lights on. Um it didn't seem like anything was breaking because the proposals that were going up were created by internal contributors. Um it was getting voted on by internal contributors and then getting passed.

And every so often somebody would dip into the forum and yell at us for GTC utility. Fine. Like, sorry. I'll do what I can with that governance token. And we are trying to do that.

So, it does make it more compelling for the people that are participating in governance. But where we've now come to in 2025 is a bare-bones version of, "Hey, we would like to have some uh intelligent, informed governance delegates that can tell us where we're going wrong, and we can start to build it back up again from first principles." Which is a very long way to go around, especially given the tenure of Gitcoin, but basically building that framework because we've tried all of the other mechanisms that involved 17 people, three goats, 4 days on the forum, and you had to have a certain badge holder, and you had to know what our blood types were to basically saying, "We don't have the capacity or the knowledge to understand what we need to fix and what we can govern and build within the community to give you funding to do that, to figure out how it won't take you forever to get that funding. You guys know better than we do. Please tell us what we should be doing.

Um and so that's where we've come to today. We still have the steward health cards. I still get asked who's a delegate from people that created governance. I still have to check the forum to like see what's going on with people that I'm not sure are even like a delegate. Um and so going back to these first principles is where Gitcoin is today.

Um and I'm just keeping an eye on the timer there. How am I doing, Nathan? You're great. Take all the time. Okay, fantastic.

Fantastic. Um I'm trying to think if there's anything else that I could cover up here in terms of Gitcoin and what we're doing. Michael? I was going to say you could just pester us if you want. If you don't want to just be required to talk 10 minutes.

I mean, I could riff up here. I could tell you all the things wrong with all the organizations. Let's go. I'm good with that. I'll pass it back over.

Thank you. All right. We'll turn it over to a discussion about some of the things we've been talking about and you can also join us for further discussion at 3:00 in the community hub on the third floor. We'll we'll hang out there and we'll hand out pirate copies of the paper that Rina was talking about as well. So but to get us started, I wonder if if if anybody wants to bring up like any particular cases where you see instructive things going on that we might want to break down around attention economies, whether they're cases in the paper or or beyond.

I mean, one thing that comes to my mind from from this is you know, outside of the the crypto world is my own experience in like community governed organizations in my life. Right? For instance, my credit union is like the largest mortgage lender in the region up in Boulder. It's called Elevations Credit Union and like I I've been to one annual meeting. I I've even met with a board member, but like I the last thing I would want to do is like participate in decisions about like who should get a mortgage, right?

And yet there is still an accountability structure. You know, there's still a sense that that organization is ultimately accountable to to members like me and there are mechanisms for doing that. On the other hand, you know, uh, there are there are projects like for instance I I I co-founded like a a little cooperative social media server uh, some years ago called social.coop. Like I love going to those meetings because we're doing something new and it's fun and and I feel like I have something real to contribute.

So I I think about that combination as being a recognition there there's always going to be a topology and we shouldn't expect a kind of one size fits all for our attention economies. So I I would I would build on that that I think one thing that's often missing in the DAO space is a like adequate separation between what is a governance concern and what's an operational concern. And this is important in the attention economy context because frankly like you're asking for the labor, the attention of your community and if you're asking them to do things that are not actually governance, they're ops. So the the difference between like making high-level decisions about what you're trying to accomplish or appropriating funds, budgeting and again if you're doing high-level participatory budgeting, but if it's day-to-day operational things, then like as long as you've done the work to secure the buy-in, get the goals, uh, relegating something to an operational layer is actually a way of separating um, you know, governance attention from ops attention, giving people the opportunity to govern without asking them to operate. And I I want to throw this back at Rena cuz we've talked about quite a bit the ways in which DAOs missing this split make both worse.

What did you say we managed to make ops and governance. We managed to make both of them worse at one point. Um, like how how you do that is virtually impossible. Um that that is a great point. Um and we did talk about this and as well as what you're doing at Block Science, which I found very compelling, which is that tension between the operation side of things and the decisions that should be made within the community.

Um and I I don't think there's an answer for it, to be fair. I do think that there's a balance. Um one thing that you mentioned that you're doing well at Block Science, which I think applies to this, is that there's trust, accountability, will, and skill. And figuring out where those decisions should lie for some of these, um I think before just setting up a random framework and a snapshot and a tally vote is really really useful. Um and for me, we talked about this as well, like a little bit from the type of contributors or the people that you're governing needs to be considered as well.

A protocol builder is not going to want to be governed in the same way that somebody's contributing research is. It's very very different in the the incentives, the contributions that they're doing, and then the trust and accountability in terms of what they're shipping. And I guess to provide a little bit of clarity. So, Block Science is a roughly 30% R&D firm. We've been working in the Web3 space for over 7 years.

And for out of somewhat randomly, we were actually a service provider of Gitcoin when it was a company during the time it became a DAO, which was a ride. Um but uh no, concretely, like we ended up with even with a lot of people who are very like Web3 motivated um and and to some of what Nathan was saying, like we have a largely sociocratic structure. We have circles or we actually we call them councils. They they are accountable to each other for core business function. There are people who are part of multiple of them.

The councils themselves are more governance-like, and then within them we have more operational activity, various activities that produce those business functions. And people within those councils are um accountable to each other and up to their group. And at the scale of 30-odd people, it's wonderful, but I also don't recommend that as a scalable method beyond that. We we've kept it at the size we have precisely because we like that culture and that style of organization. And so I I'd be careful about trying to take that model and scale it beyond that, you know, few dozen people.

Yeah, I I wanted to add something about the the kind of diverse kinds of people that participate in these communities and recognizing that. And for me, one of the questions coming into the the study was we kind of felt what extractive attention economies look like when we're thinking about things like platforms and doom scrolling and it's kind of like addictive and we just feel like we're being sucked uh our attention's being sucked dry. And then you know, we're in year of regenerator regenerators, right? This this ETH Denver. And I think I was trying to figure out what does a regenerative attention economy look like?

And it's a question that I'm it's still kind of I'm thinking about. Uh and I I think it kind of feels more like permaculture as opposed like monoculture where we're being like mined for attention or yeah, you know, strip like the industrial agriculture. It's more like a permaculture where there's many different kinds of contributions, many different kinds of ways of recognizing people's contributions. People feel heard, they feel seen. Uh they feel like they're not wasting their time.

I think there's also something about how AI helps with all of this and doesn't destroy all of this cuz I think there's also ways that it really does a lot of harm and we haven't really talked about it at all. But yeah, maybe permaculture attention is kind of what I'm thinking about. One of the considerations that I have and I'd be curious from this group as well is um without going into like the what am I version of it, but how do you have a regenerative attention culture when you've got different types of people uh contributing both internally and externally in an organization as well as across the ecosystem. And I I that is one of the bigger questions and I you see the fragmentation with a lot of the organizations and the x y z like you've got 13 DAO frameworks and now you got 14. Like great, uh fantastic.

Um and and again that that drowning in information version of it because like I said we we this happens personally and professionally um and from like an emotional emotionally intelligent leadership space like I take care of myself, I take care of my people, I take care of my organization and I haven't even gotten a chance to see what's happening outside of that. So it's both that attention economy from how leadership has to approach it as well as the entire organization and I do think that AI can be a tool but I I've said this a couple times this week and I'm sure a lot of people are sick of me hearing me bang on about it but like that tool has to be the right tool for the use case and I don't think based on your heuristics that we're looking at it in the right way right now. One word that we bounced around a lot in this process too and wrestled with is justice. You know, what is attention justice? And and one uh stream of thought that informed our work was was uh feminists uh thought around the way in which particularly emotional labor and labor of attention is unequally distributed in in many of our communities and that too is a really uh you know, critical frame for thinking about what's regenerative.

How do you make sure that people are able to be justly rewarded for the attention they're providing and that their attention is not being exploited? Yeah, I I want to kind of click on that because especially in DAOs we have a tendency to do this thing where if we can't metricize it, we can't value it and there's a certain kind of emotional labor, care work, support that even in an environment that's heavily metricized can't actually be metricized and or shouldn't be or when it is it often causes more harm than good. And so like making sure that we hold space in our ways of valuing within DAOs for kinds of activities that are in some ways resistant to that instrumentation. And if we insist on certain kinds of instrumentation, then we also insist on devaluing that kind of work. Well, we're about at the end.

Um, thank you so much for uh uh to my collaborators for working on this project. Rana, thank you so much for being part of this conversation. Again, you can join us in the community hub on the third floor to continue the conversation at 3:00. And um, please if this is the kind of the kind of discussion that you geek out on and you really like to participate in more, please uh check us out at uh uh metagov.org and and see the many, many projects around uh building a future of governance that uh that we and the many, many more people in this community are are engaged in.

Thank you all so much.