How Decentralized Can Governance Be?

Transcript

Okay. >> Can we get everybody who's outside inside? Yeah. You who's standing up walking out with the backpack. >> Stay. This is going to be good. >> It's a good one. >> We promise. We promise. Thank you. I see you in the hat going for the door handle. It's right. Come back in here. >> Okay, I think we can start things and people will be joining I guess. Uh so well uh for the people that's here, I guess this is an interesting topic. Uh we've been talking all day long about the tradeoffs uh with decentralization, centralization,

uh how to navigate this ex spectrum, right? So we're talking about uh stacks, operations, the technical layer, but also human coordination uh at the end. Um so I just want to give the opportunity to the panelists to for a quick intro and what it's also important is to know uh what are their lenses uh where are they looking uh at when they look at the centralization and what they think the centralization is. So I pass it to Matt here.

>> U thank you uh my name is Matosh. I lead the governance team over at the arbitrum foundation. I've been in this space for um over 11 years now. Um and governance or what has become governance is something that that has gotten me into the space in the first place. Um so I'm really happy to see that you know we're getting some some uh love in the space for for for governance related events. Uh at last I would say I'm retro the uh ecosystem coordinator at Sky. Um, the lens I usually apply to decentralization is modularity. Like how easy is it to rip and replace certain parts of the system.

>> Dennis, um, co-founder and CEO at Tally. I think decentralization for us means a lot of different things and I think maybe it it's probably very different than what the market thinks it might be or what the government thinks it might be. Um, something that I've thought a lot about lately is can we have decentralized organizations without state charters, which is a pretty kind of like a hot topic idea, but maybe worth digging into. Hey everyone, um, or the few of you here. I'm uh, Ian, I'm uh, um, the tokconomics lead at Aragon. Um, and I think the lens we look through at decentralization is like can onchain

assets be a viable asset class and like like is decentralization functionally useful to tokens being a viable asset class. >> Okay, thanks for the for the quick intros. Um so continuing with the reasoning that that Denison just put together uh the centralization can be seen can be understood can be understand from different lenses. So it's different to see the centralization from a delegate perspective from a stakeholder perspective from a founder perspective and I know that this is something that came out when we were discussing about this panel. So I would say uh if any of you have some takes for

example uh on the view of these different stakeholders that give shape to a governance I I mean I think at this point everybody knows that like decentralization and governance for its own sake is is is a mute point. It's a it's a waste of time and focus and energy for an organizational perspective. And like I I'm I'm I'm pitching for a new slogan for us which is like like build something worth owning because because the focus of governance and decentralization for for the past I would say four or five years in the space has been like how you how you split the pie and like it took so much

attention and focus from like like how how do you actually cook it, right? like like build something that people would care enough to participate in whatever like loops you you have them jump through to call it governance, but like like first build something that is valuable enough so people would care and and and participate and then you know on top of that you obviously want to layer like like real tangible financial incentives because money is why people do things in the real world. So it's it's like build a viable business, layer financial incentives and then everything else comes secondary.

Building on what Ivan said, I think you know governance is only really possible when you have a protocol or a treasury to govern. Uh and that is also where the the whole debate of decentralization comes into into play. I believe it's very you know very much tied to to how much on or where on the spectrum of subjective subjectively correct outcomes you are meaning if we are talking about something that I would personally still consider governance like ordering transactions within a block uh the consensus there is very straightforward is you know it has subjectively correct outcomes and there is no no wondering about that if we're

now talking about how to bootstrap an eosystem system using a treasury that is you know completely non-technical governance. The the the the you know the the the subjectivity of that decision making will dictate how much decentralization should be present within that uh within that space. But when it when we're talking about these different whats right of of the decision-m system, I think um you know it's also important to understand that that some things like for example immutable permissionless protocols will require the same level of of of of solid and resilient governance to even function. Uh whereas if we're talking about subjective treasuries, well then it's more about um you know finding the

so-called wisdom of the crowd and seeing the expertise in in every participant's minds uh come together and and come, you know, create good positive outcomes. >> Has the wisdom of the crowd ever created good positive outcomes? Because I I would like push back really hard on that one. >> Please do. Please do. That's what we're here for. uh I would say the whole history of humanity as it concerns uh the the you know democratic decision-making process has been an exercise in that and you could argue that you know at times it wasn't perfect it was far from perfect and it has led to tyrannies uh but there are instances in our history that that say um it is

valuable right and there is um you know actual value in what what Aristotle in in politics describes described as a feast being better funded out of many pockets rather than one. Like you can have a a a a a space of of decision makers where uh the expertise of them comes together in in in different shapes and sizes than if you were just to rely on even one, you know, insanely good expert.

I think it's actually interesting to look back at the history of the corporation to sort of like divine a little bit of like wisdom here and it's really interesting when you look at before the creation of the sort of like joint stock corporation um you had partnerships and you know it's kind of interesting that one of the things we've dealt with over the past four years is being you know targeted as a general partnership right and there was a period of time when general partnerships were the preferred form of organizing human capital and labor because the theory was that if you had these joint stock corporations or these organizations where there are many owners and many participants um you would get a problem of lack of participation and poor

participation right in the days before we had you know state charters for for uh the corporation general partnerships basically if you you know messed up like the the king would take your house and maybe your life and all your stuff often all of you guys would go down um with whatever you know mess up one of you made. So the partnership was preferred because you knew that you were bound by the outcome of your decisions. Thus you had aligned incentives and that's why also partnerships were rather small, right? You're only going to trust a certain number of people and you really want them to have this like skin in the game which would be you know unlimited liability. uh and the common belief at the time was that uh corporations join

sock corporations and I would extend that and say Dows would lead to a world where you had low participation and the the participants didn't have enough ownership in it to really care which is really I think very instructive for what actually happened right with a lot of organizations I think one of the largest complaints in the past four years is that you have this low participation you have this lack of like incentive alignment you have this lack of material link to what particip ipation looks like. And I think when we we think about Dows today, um, you know, this is something I was I was talking to someone just a few moments ago. We're kind of in this weird interpace moment where Dows are struggling to be a corporation of

some kind. And the originators of many of the Dows are struggling with the Dows and their own corporate entity, right? Because, you know, many of Dows were created as a way of kind of creating a new space for this token to exist and potentially capture some sort of value without um having this like liability issue. >> It's called regulatory arbitrage. Don't beat around the bush. >> I'm beating around the bush here. Uh regulatory arbitrage. We never did regulatory arbitrage. Never. Um and and so I think that's like a pretty important thing. Like yes, it is important that we govern something like the onchain that's a protocol. Um, yes, it's important that we we think through how these structures work, but I would actually pose a kind of question to the

panel uh around like the long-term longevity of Dows as we see it, right? Like why aren't public corporations Dows? In fact, if you really look at how most Dows operate today, they look exactly like how Tesla's operated. They had a big vote on whether or not to give Elon Musk a trillion dollar pay package. That looks exactly like what a Dow would look like, right? Have we been calling them Dows, but really they've just been corporations the whole time? >> Yes. >> All right, that was easy. We're done here. >> I I I think everybody here raises really good points, but going back to Ivan's uh that's probably where I would uh ground myself. The how how is governance valued, right? Is

governance a product? Like that's a question that we've been asking ourselves a lot. And if it isn't a part of the the value creation mechanism then why is it there? So um we I had this conversation with a few people in optimism in like 2022 and in the their most recent state of the evolution of optimism they published um it basically crystallized it um like the value of governance aside from cash flow and and like the stuff that we avoid because it it it lands in like a security zone, right? like the pure value of governance over infrastructure

is capture resistance and and and and it's actually a function of how like I I want to have enough of a say into how this infrastructure invol evolves so no other party can change the infrastructure in a way that benefits them over me. So it's in essence insurance and then actually like but but that's not even it like like it's it's actually even like like in terms of pure monetary perspect from a pure monetary perspective like like you have a cap on that which is like what's what's the switching cost for me.

So let's say I'm an app that is built on a certain rollup, right? Like my competitor comes in and starts to like bid to change a rollup parameter that's going to benefit them like like I'm I don't know like um I'm a wallet and like somebody's like oh like we're going to prioritize flow to like my competitor then I want to block this right. So, so I need enough voting power to be able to block this decision, but that matters only if for me the switching cost is high enough. So, I would actually care, right? If the switching cost for me is low, I'll just swap out for another infrastructure provider and be like, good luck, right? So that's the value

from like a like like the the the the most logical path from governance rights to like financially quantifiable value that doesn't require a lot of like hypothetical nation state dynamics of like what could a shoulda would have like like if you have to draw a straight line and and we know that capital allocators have a pretty short attention span. you want to drive a straight line of like why is this valuable? That's the the like the most straightforward answer.

>> Can can I push back on that just a little bit? Um I think like you know I optimism I think is a good example there. I think a lot of people talk about like capture resistance and that's the point of decentralization. Um but I would like to flip that around a little bit because a lot of the argument is just like we need capture resistance so someone come doesn't come and do something that I don't want them to do. like is that not already kind of like capture I want to design a system so that nobody can like take away my control and I think that's relevant in this case because most of these decentralized entities have a disproportionate amount of the decision-making power allocated in the creators in the founders and the founding team and the labs entities right and what they're looking for and

you know this is maybe a spicy take what they're looking for is a system where there's decisionm that's not threatened And if they were truly corporations, then that would be a lot of clearer line, right? They would be like, I'm the owner. I'm Zuckerberg. I have 52% like you know, this is how it is. You know, you're on the boat or you're not. Um, but you know, with the the you know, cuz you can capture Bitcoin, right? You just have to have enough money and like have enough miners and you capture Bitcoin, right? And in in these decentralized systems if there like capture like I've come to kind of believe that capture resistance is like it's a little bit nonsensical right because if if if no one can gain control

you're kind of saying well we we all have to like agree on everything but what we've seen is there's that kind of like just deadlock in many times right so I don't know I want to like push back on like the capture resistance part where maybe we actually need a little bit more capture and the switching cost be the the decentralization component of it, right? There should be a free market of competiting competing tools in like L2 land. There are other L2s, right? So maybe that's the more fair decentralization layer like that that you should be able to more freely move, right? Maybe that's the the piece. Maybe I'm agreeing with you here.

>> Yeah, you are kind. >> Yeah. Yeah, actually now. Okay, now I'm back. I agree. So it seems that the centralizing stuff it's buying us problems. Uh but just to shift a little uh the line that we were discussing. Um so we are talking basically about trade-offs right uh in centralization versus decentralization. What are those tradeoffs? Those trade-offs uh and who are the winners and who are the losers in that game? >> Anyone that wants to take that?

Definitely the delegates are the losers. They've worked for years. They've worked for years and not gotten properly compensated for their work. >> I I I think I would look at decentralization specifically uh versus centralization for running an enterprise organization or a protocol as as the the cost of context almost. So, it came up in like a workshop earlier today of like, oh, well, you know, things have to be super clear or the entry paths have to be uh permissionless. And I I wouldn't necessarily agree with that. I I I think there's a clear distinction between permissionless and decentralized. Uh the value that

decentralized represents for us is the opportunity to take advantage of the pathways that are defined to start participating in the system. not even necessarily with voting but just participating within the governance layer of you know operating and taking advantage of the infrastructure that's there when it's more centralized the the opportunities are potentially just as fuzzy or just as unclear as the the centralized systems but the the value of that is not necessarily having the opportunity to make an impact to change the rules to somebody's

preference or not. But it's to to just know that the the the infrastructure there is again like modular where one part of the system isn't going to fail and take down all the work that you're investing in. It's it's intended to be replaceable or it's intended to have some competition around it. So if something better does come along, depending on if it's a weekly voting cycle, annual voting cycle, it's it's able to take advantage of the growth of the market or the growth of the participants that are interested in it and and establish somewhat of a baseline from the incumbent to say, "Hey, we are better than this. Let's put in the effort to to take advantage of the opportunity." Then

>> I think an interesting framing dilemma is is always present when discussing decentralization as it's you know often times positioned against efficiency especially as we um think about treasury um governance where the dilemma may not even exist. In fact, uh it is a a a matter of onchain legitimacy at the end of the day and I would point towards representative political systems where um you know the uh the the the elections uh define who will be owning the actual legislative or or or decision-m process for the next four years or 5 years. Um

and during that time there is very little recourse for uh for citizens to to take the uh the power back to redirect the the decision-m uh power. Uh this is not the case in truly onchain permissionless governance. Right? We are building systems um where the the electorate has potentially at any point in time checks and balances over the entities that it does uh legitimize and it does empower to to conduct. uh daily operations. Uh and so the the underlying leg legitimization of of these decisions and those those entities comes from the uh from the token holder. I'm obviously referring to to to to very specific

onchain token holder based governance. There are different models but I think this is the one that we understand as as as the the DAO model. Um and so in that the the legitimization can remain and it it can uh enjoy certain checks and balances throughout the duration of that uh of that legitimized entity's existence. So that's a really good point because we're talking about not only a designs of governance but a technology that allows to experiment and to create new layers of governance at new ways to experiment and participate right so I will go with the following question it's

uh let's get into the map in the into the map right um what have we learned about like real dows let's talk about specific cases and what works what doesn't work And is there that talking like in real cases uh is there something that shouldn't be decentralized at all when we talk about current operations between different dows. Um I would say like like execution of specific tasks doesn't can't can't really be um meaningfully decentralized in in in a productive way.

Um, in a lot of cases like like and one of the one of the problems that we we ran into with with Dows is like there is a difference between like high level decision-m where potentially like like legitimacy plays a role in blah blah blah but like like like there is like oh you know somebody somebody proposes something and it's like we should do this um and like I can do it and then somebody else comes in and is like no I can I can do it and and you you you start mixing stuff which is like what should be done and who should do it right so so so and and that's not governance that's operations

>> right so so so I think on the operational level and and like sky is moving in this direction of like like what needs to be done is defined vendors bid you pick one vendor and that's like like that's it you don't try to like break everything and and there were like a lot of cases where like three teams are kind of competing. They're fe they're like you have like three teams that are funded for like growth and marketing. I'm like like like in most of your cases your product doesn't need more than three people. You have three teams that are funded in parallel just so that you appear more decentralized. So I would say like execution is something that uniquely probably doesn't need to be

decentralized on a granular level. Um and and also like on things that probably should be more decentralized. We we had a chat with retro like a few years back where it was like you know like like maker at the time they had this like very complex like like you know like delegate blah blah blah blah blah and at the end you arrive at an executable payload. There are like three people who can actually validate whether the proposal the delegates spend months arguing about matches the actual instructions that go on chain. And it's like the rest of it doesn't matter cuz you like you these three people are the breaking point and and and like like all the rest is like

okay yeah let's let's let's have like a a cool thing and like like talk about like how politics translates on channel like like no like you're not decentralized because you have a very clearly identified breaking point which which makes everything else sort of futile. I think it also depends a lot on what success is supposed to be. And this might be a bad take or a negative take. And I'm not saying it's my opinion or the opinion of my company or any of that, but Dows have been incredibly successful in the past four years when you measure it by

founders who went to jail for innovating, right? And when we think about it like that, Dow's yeah, kicked it out of the park, right? Like who went to jail, right? I went back here. How many people do that should have gone to jail went to jail? Cuz because like >> Yeah. Like like like it's it's a very good structure to avoid consequence. I don't think that's a net positive for the space or the world. >> No, no, but I think it's important to to think about like what DAO served in the interim four years under the Gensler administration, right? It allowed for continued human innovation

with some protection from a hostile state, right? Which is not I would argue that's not what Dows were supposed to be in 2014, 2015, 2016, right? It's what they became out of necessity, right? But it's not what it was really intended to be in my my opinion, right? It was just like, oh, people started to use this structure, use this concept, use this idea as a solution to a very particular problem at the time. And a lot of the, you know, a lot of the ills that did come out of the Dow space in the four years that we complain about are kind of a reflection of this like original sin or this original like seed of like

cancer in the application of this idea. Because what we originally were building I think was very very beautiful, right? Like at some point in time we would have like had green leaves all over here and we were picturing a future garden and like we're all happy, you know. Um but I think the application of this concept was used in such a way that sort of like dominated what we now think of as Dows being able to do. It's kind of corrupted in in my opinion the sort of like the vision and the integrity and the application.

>> So yeah. We sorry you were going to say something. >> I um No. >> Okay. Okay. Sorry for that. Uh I was going to say that we we we witnessed how uh this last year uh legal rappers were a thing right and it goes like by the hand what what you were saying right uh the the condition the market conditions and the regulatory environment are shaping governance or the way we do governance. So uh do you think that in the next few years we will see like more like and also bringing what what he mentioned a couple of minutes ago like splitting the execution layer from the

the decision making will we will we see like more ways uh different ways of organization such as the argument operations company the central and operations company how that's going to move forward in the next few years I believe this goes back to what Ivan said about um you know day-to-day operations not being necessarily the best to be covered by a DAO. Uh decision-m by a committee does not work when we're talking about you know a $3,000 grant to do YouTube videos. Uh which is an actual proposal that I saw in one of the leading L1s that was posted right after like a fundamental shift in the direction of the protocol that would impact it for like you know

10 years ahead. uh and it used the exact same governance mechanism, the exact same people voted on it and were involved in that decision-making which I believe is fundamentally flawed and that um as I was mentioning before you need to have mechanisms for legitimization that allow um you know companies and and and rappers to to take up day-to-day operations and execution and and execute on behalf of the actual um decision makers. So as long as it all rolls back to the actual you know onchain layer of permissionless onchain executable decentralized governance uh I think there is a ton of experimentation that Dows have been going through in the recent years and will be going through

in in in the coming years as well. >> The the thing that I think your question really hits at is Dows are trying to enter the real world. They've lived in cyerspace for a while. They've they focus on protocols, but to make that jump like people and pioneers have done and have been had their hands slapped for it's it it requires not negotiating or you know following what has been done but applying that innovation to the legal layer. So then the things that have worked in decentralizing an organization can be can be retained but still participate in the you know global financial markets or even uh you know any sort of business uh that could be

recreated. The the protocol piece I think is still important though. There there still needs to be some central mechanism that that isn't lost and how that applies to the legal layer. I think needs to be looked at a little bit more instead of just spinning up foundations. But the the next four years I think are going to be interesting where we're going to have some experiments that work and then some experiments that really don't work and get some people in trouble. But um that's that's the cost of innovation. >> Great. So we're coming to an end. We're running out of time. I know if you want to give some closing thoughts or just continue with this uh with this topic.

Uh but yeah uh I invite you if you have again let's go if possible with real cases what's the like the the case the DAO or the the governance that you see uh something that's going to be replicated or at least adapted uh for new governances or for the ones that want uh to have a revamp. And with that we close. Uh I think I think the one that's going to see a lot of adoption is unis swap's duna right like this is you know the the Hayden's proposal the dunitification I think it's called right I think this

is what we'll see you will be uh able to be a US corporate entity you will be able to have a protocol you will be able to have participants in the protocol who are not liable for the actions of the protocol and that's veryant important, right? Like we didn't really talk about the fact that just by voting means that you are responsible for everything everyone else who holds the token does which is an insane insane thing which has been an enormous setback for the industry right because the most capable participants were also the most rational say it's not worth participating the danger's too great um I think that we will see the unis swap model uh we we'll figure out if it's a good one right like it's it's definitely I I don't think there's a universal agreement on if this

is the way forward for everyone. But we're going to see that be used by a lot of entities going forward. Um where the token and the protocol is a product of uh a Delaware CC Corp and the operation of it is going to be governed to a certain degree by the participants and the market forces of whether or not you want to be there. I.e. Does your token return enough value um be this sort of deciding factor of like whether or not people participate or stay?

>> Yeah, I I'll give I think directionally I'm I'm in agreement. I'll just give another example. I think the example for like I would give is probably maker. Um I think the and and that's going to be driven by hopefully the market as they call it the the revenue meta which is called like business that makes sense. So, so in essence, structures that are able to create value, capture it, and return excess capital to those who've invested and and supported that entity's existence are the ones that are going to dominate in the coming years cuz like nothing else is going to get funding and like that's going to be a focal point of

like like like you know people should build something worth owning. I probably wouldn't be doing my job if I didn't say that the the the the future of I think Dows and what we're trying to build is alignment. So within the central governance, it's very simple, right? It a and locked down to your point about the cost of potentially switching or governance attacks. It's just going to be made impossible. So those problems are solved and then where we push the innovation and the chaos and the the the the experimentation is going to be in isolated pockets. So then again in terms of decentralization the core revenue the core machine the the core

monster is fine and then if you know part of the the ecosystem fails well at least everybody learns from that but at the end of the day it's it's yeah effectively one purpose within uh one segment of token holders which is yeah grow the revenue and allocate >> subdows >> stars or agents. Okay. Uh, I think we're good then. Thanks for thanks for the for the insights. Uh, I'm a huge fan of all of you. So, thanks for being here. And that's all for today, but for this panel.

>> Thank you all. Thanks. Thank you. Thank you. Thank you. What was it?