DAOs Professionalised - But at What Cost? - Oxytocin | Velora

Gosh, I didn't know I was going to be right before Lido. That's a big act to follow. But um hey everyone, I'm Oxy Oxytoen. I am the head of government strategy at Lighter Labs Balora. If the name is not familiar, we recently rebranded uh used to be known as Paris swap uh one of the largest dex aggregators.

So I wanted to come with this more stareyed question about Dows the idea that we professionalized but what was the cost of this professionalization um believe it or not even I don't believe it sometimes Dows have been out for a very long time and I've been in this very weird rabbit hole now for over four years and in this few years from the outside it looks like nothing really has changed people go the to the forums they fight each other on Twitter now and they have tokens and they vote on things. Um but at the beginning in my opinion if you now looking back things used to go much much more differently. Um we can look at some of the D5G protocols. Um just to clarify these have changed a lot since then but like we used to run what we call enthusiasm engines. The idea that people would participate in Dows because they believed in the protocol.

They would say okay I really want this protocol to succeed. I use it on a day-to-day basis and I want to have a direct say on what's going to be happening. Maybe because since I'm a token holder, by making good actions, price will go up. Who knows? But the idea was very simple.

You get token, you vote token, you participate, and things get better. And the reason why I called it enthusiasm engine opposed to a more boring D5 1.0 kind of protocol is because very crazy things happened on this period, what some people call the golden age of Dows. Um, we I always tell the story to people that are newer to web 3 and they never believed it. Did you know there was a DAO that raised $42 million to buy the US Constitution?

Yeah, shout out to Constitution DAO. Um, we had what I like to say, these are your dreams come true kind of Dows. The kind that people come together because they believe in something. They will pull resources together and they will try to accomplish a very specific goal. So the most egregious at the same time the most hilarious example is constitution DAO.

The idea of we are going to be getting together to get an auction buy an initial copy of the US constitution to show people that Dows work. Of course since it's an auction in the end it was a bit hard because everyone knew exactly how much you could bid. So it was easy to out bid but also you had people trying to um buy other kinds of intellectual property. So a DAO that came together to get a copy of the lost cancelled movie of Dune by Yodoski and try to get their own intellectual property going. What I'm trying to get across with these kind of more unusual dows is that back then with enthusiasm um people used to come up with absolutely bonkers ideas but they had this idea of innovation things that were only possible in the web 3 space.

How else could you get all these people together, all these strangers on the internet to pull resources together for something like buying the US Constitution? Would you just like get people on a Discord server and say like, "Send me money on PayPal and trust me, I'll actually buy this." No. But sadly, all good things come to an end and there were a lot of problems from this time. Um, we just had Kate talking about how Dows are very fragile and I have to agree with this.

Goodwill is not antifragile. The more pressure you get on a project, especially these kind of like purely communityrun and human organized projects, the more likely they are to break. Um, especially on these early dates when people used to participate on a purely volunteer basis with speculation maybe being the only compensation they received. Um, people start to work more and more on these things. You would have getting burnout.

You would have to start thinking about more longerterm things. Obviously, this wasn't a problem for things like um let's buy something and then be done. If it's an initiative that takes a few months and it doesn't work out, you dissolve the Dow and that's it. But what about the protocols that are supposed to be running for months, years, maybe decades, how are you supposed to be running these things? And the sad answer is that with this original system, you simply couldn't.

And usually I don't like to bring up this kind of negative examples but I feel the one that really gets the point across is what happened with compound. um they had a similar system where people would um participate in a traditional Dow governance but due to the nature of this kind of participation there was a lot of user burnout and despite compound being still very much alive and the protocol still being thriving actual people that were going voting talking on the forums defending the protocol ended up like really not participating as much so we had this problem where well I wasn't part of we as in we in DeFi I wasn't really a member of compound DAO But you had these whales that were forming because obviously at the end you have the centralizing bureaucratic forces in these kind of systems where the only people that care are the people whose size is really size. Um voting to basically move a huge part of the treasury to their own initiative. Everyone else was against it but the whale passed the proposal over three times. The first few times it was rejected but there's only so many times you can try to fight against this before it passes.

Uh this is one of the wakeup calls that the entire industry noticed that the current version of Dows, this idea of like we need to make it volunteer run by pure enthusiasm wasn't going to work out. It was a painful $25 million um dollar lesson. That's almost half a constitution dial. So what did we come up with? What was the next big stage?

Well, this shiny new product started taking form in many different dials, even the ones that I brought up before. um the idea of the professional meta delegate. This would solve all your all of your problems. It was everything you wished for. And it was only one neat package.

They had multi-person teams. It wasn't anymore that like okay, if you didn't like this one guy, this delegate wouldn't vote for you. It wasn't that like if they were on holidays or sick, you would start losing a big part of your voting power. You would have many different people running different projects. They would give you dashboards, reports more, which I know it sounds a bit silly, but it really helps knowing how much you actually have in your treasury, knowing who's voting on what and who you can actually delegate to.

Um, you have experience. Like some of these people, they participate in half a dozen dozen different dows. It's literally their job. So, let's say that you want to be running a grant system or you want to be running a new kind of delegation system. um you don't have to reinvent the wheel, which I know it feels a bit obvious, but in the early stages of the house, a lot of time was spent just like remaking everything from scratch.

Let's make yet another grant system. Let's make yet another treasure management pro process. And since it was just volunteers that have part participated for the first time, you would run into the same mistakes again. How do you make grant accountability? How do you make sure the proposals actually pass?

What's the best way to make a dow proposal happen? So by having this kind of professional delegates, you will be able to kind of like speedrun through this process. And this is the last in my opinion hidden benefit that people don't really talk about by not having just a pure um backholder, somebody who's just like voting for the project as a shortterm driven u token incentive. you had a separate kind of incentive which I'm not going to say one is better, one is worse, one is driven this way, one is the other, but by having different kinds of opinions all working together, you had more of a discourse happening. It wasn't just like what can we make to make token go up, but maybe what can we make to make sure the DO survives because these people were more interested are more interested in making the DA survive or making the protocol actually run in the long term.

So yeah, you might be thinking, hey Oxy, I joined this call to talk about the cost of delegates. What is the problem of professionalized Dows? You're here glazing the professional meta delegate. You're here telling me they're all so good. So, did you lie to me?

Was this a bait and switch? And the answer is no. There is actually a cost. I'm not talking about the salaries. That would be a very boring talk.

And talking about some of the things that we actually lost of the on the way. And in my opinion, what we lost most importantly, which is very difficult to notice, is how we lost the actual collective wisdom and accessibility of Dows because at the end of the day, and I know again I'm just basically stating the obvious, but Dows are a choice. You don't have to make a Dow if you want. There are some protocols that could perfectly just be a GitHub and just like keep it running like this. There are some protocols that like could just stay a simple traditional organization.

So what do you actually get out of becoming a DAO? What kind of benefits do you actually um facilitate when you start making this kind of traditional decentralized organizations happen? And obviously if you got like five different Dow people in a room, they would give you five different answers. But while I was trying to make this talk, I think I ended up coming up with these three key points. You allow the community to actually steer the protocol development and culture, which is very important because for example, by having the DAO be the one deciding the kind of communication, the kind of things that you're actually going to be pushing, you get to shape the culture outside of the usual core team, which are usually just devs that are coding 24/7.

um you allow people from the outside to be able to actually develop the protocol which again if you were just going to have the core team develop everything. If you were just going to have a single set of devs make the project 24/7 then why did you make a DAO just keep everything it's much faster to actually say I have an idea let's code it as opposed to making a proposal going through the stages discussing with all these people. So a big advantage of Dows is this idea of like maybe one team could be developing let's say a certain kind of front end another team could be making another kind of like grant distribution system or other incentive system and they could all be acting independently each one coming up with their own solutions and it's especially the case in my opinion when it comes to protocol versus UI separation. I feel a very good example of this that I really like is what you see with unis swap. you have the unis swap pools being developed on one hand and then you have front ends being developed by uni lababs oku um even def lama even though they they do a bit of everything.

Um but by having this kind of redundancy we see something that is simply not possible in web 2. This idea of like you have one protocol and you have a million different ways of selling it so the user can have their tailored experience and Dows can help coordinate this. Finally, you can actually get an edge over computing protocols by having these unique developments of the DAO. Um, again going into real life examples, I think the most obvious one that especially was the case a couple years ago was the example of arbitum versus optimism. Um, when they both first released, when they were both stage zero um, kind of roll-ups, they were essentially the same.

They were all layer twos that were general purpose. They all tried to get more or less the same thing, which was a general purpose from DeFi to NFTs kind of system. And Dows really helped them get an edge one over another be it in attention the people that were developing. Optimism took a bit more of a foundationrun u micromanage kind of system where they would have a very strong grant system and retrofunding to be able to drive people and arbitum at the beginning now things have been changing quite a while quite a bit but they try to go into like okay the community chooses the community is going to be voting on what things are going to be happening we're going to be able to um get everything rolling and this way at least we have a very different system compared to optimism um it's undeniable despite both of them being layer tools on the Ethereum ecosystem the culture of the Dows were like night and day, but professionalization um actually kind of limits all of these three in my opinion. It's true that it helps them um get developed, but it also helps limit a lot of things.

So, for example, by having a more limited set of delegates actually running the day-to-day business, you reduce the number of voices that are actually participating in the community. Um if you have the same metadlegate participating on like 10 different dows can you really form this unique culture in your protocol would you have this kind of separation between curve a compound if back then you you would have like the same people participating on each one of them delivering similar outcomes. Um open question I'm not going to say one way or another. In a similar way, if we wanted to um professionalize the the process, which I understand why we do it, we want to be able to make sure that everything is going to be flowing through smoothly. Um but this also makes it a lot more difficult to get outside people actually participating because they had to be jumping through a lot more hoops.

They have to be talking with specific stakeholders. Let's make sure, okay, we're still on time. um we have to make sure that everything goes as expected because obviously we don't want another compound um issue or we want people to actually know what they're voting for. But that can also limit what people are actually going to be participating in. Um, as somebody who tried to apply to grants from many different protocols, some Dows have gotten such a complicated grant system that at the end of the day only a couple people that were already part of the delegation system know how to apply for them because otherwise it's completely inaccessible and you don't know how you get rejected for different things.

And additionally, where's the edge? Where's the innovation if you just like get the same boxed procedure into one another? If you come up with the same like here is the nth different grant system that's going to be the same one that we deployed on all the different ones. What happened to the craziness that we saw on the enthusiasm engine days? Um will we still have ideas like Vomics happening if we had the same teams always running the same things?

Um in a way it's a bit of a stabilizing force. Um the idea of professionalizing DOS it's true that by introducing professional delegates we introduced the concept of stability. we suddenly are um a lot more resilient to governance attacks. We have somebody to keep out on things like how's the DA doing? How are these procedures going?

Um somebody's able to do the boring things. We can actually decide um no you you don't have a community member having to vote for the 11th time. Okay, this distribution looks fine. We have the delegates now um dealing with all these things. But on the other hand, we are now trending so much towards the stabilizing centralized forces that we are losing some of the edge that we had from the initial enthusiasm days where actual community members could have a say.

So I feel that going forward we need to start raising this bar a bit more to be able to make Dows actually a bit more than a fintech that is a lot more difficult to run. So to finish um I want to come up with some of the things that I've been observing. Um because as somebody who spends way too much time on X, I mean it's literally my job now. Um I want to point out some of the things that you might have missed about things that some people are developing to make Dows go forward. One of the most fun experiments that I got to play with was the um butter um buttery games experiment with optimism.

So I brought up earlier that one of the advantages of the optimism DAO is how they had the foundation really every season like trying new things and like revamping their grants. This this season they made something that was based on futur futuris futuri I only read it I don't know how it's actually pronounced but basically you would have people come with play tokens now in the future could be something else actually say I predict that this grant is going to be getting this much growth in the ecosystem I feel this is going to be the outcome and I put my quote unquote money I put my skin in the game for this to be happening I bring this as an example not saying that like everything is going to become like poly market but I believe in the future by having this kind of abstracted games where you make the community have say over specific outcomes you take away the sameiness that we used to have from old Dows that only professionals could be handling you can be abstracting it in a way that you can still be distilling the collective wisdom about what do you think this could be happening do you feel this is going to be good for the community without having them to go through a million different forms forum proposals or other kind of um resolutions Another thing again I didn't know lighter was right here before me so please if I say anything wrong if you're still in the audience raise your hand correct me ratio me but um I see this increased separation happening more and more between the executing branch and the governance branch um in future protocols. So in a similar way that we used to believe that everything had to be bundled together front end back end devril and then we noticed actually maybe we can have a team just doing smart contracts and another one doing something else. I feel something similar is going to be happening with Dows. you're going to have um I don't want to call it entity because that's a very legally defined name but you're going to have a group of people that are going to be working more in the Dow separation things and then the people that are actually doing the execution of stuff the ones that are actually working the protocol and actually holding the thing and making keeping the lights on are going to have a stronger say besides just playing the governance theater to say I don't like this I'm out of here.

So again, not a lighter DAO member, so I could be wrong on this one, but from what I understand, this dual governance parameters allowed um staked ETH holders to actually have a say after proposal passes, saying I don't like this. I'm going to be stepping out of of lighter if this comes up. And it introduced a way of saying like I don't have to be anymore just an LDO holder to be able to have a say in what's going to be happening. I don't have to be DMing um a million different delegates for something to be happening. Now we have these two separations and then the executing branch can say, "Okay, you know what?

I'm out of here." And then just to finish, I put these ones together because it's a bit more of an open question that I wanted to finish this up with. Maybe the problem is not with how we are coordinating the tokens and delegates. Maybe it's simply that the way that we're currently structuring our Dows is not really suited for these kind of like sudonmous um decentralized systems. So, what if what we have to do is escape this token delegate mousetrap where we have to have one token, one vote, and then delegate it to someone.

Could we have a proof of human participation system? Maybe for things that require more democratic participation, you have something that verifies you're actually a human and we do a more one person, one vote kind of system. Um, scan the orb, do whatever you want. Verify person to person do circles. But that could be a way going forward.

Or, I mean, AI is the buzz word now. So you knew it was going to come up very sooner or later. Um people think that AI is going to be running Dows like basically instead of delegates you're going to have a AI agent doing this kind of voting. Personally I strongly disagree with this because an AI cannot be held accountable. If an AI votes in a way that you really don't know why it voted this way you cannot blame the prompt.

You cannot say okay what happened. So instead of what I think is going to happen more in the future is that AI is going to become this kind of abstraction layer. So for example, AIS could be doing the summarization, could be doing the communication of what other people have been saying and could serve as a middleman. So maybe to take away this kind of politics of like I like this person, I cannot vote against them. You could have a middle layer with AI where they say, "Okay, some delegates said this and they really disagree with this.

So these delegates believe A, B, and C. Vote how you wish." And the final idea expanding on this kind of uh system of like um separation between execution and participation. Maybe the problem right now is that people that have token um delegations, delegation is only a reputation game. Um the worst thing that can happen if you make a bad proposal right now is you lose the position on uh Dow Council or you lose your delegation.

So perhaps having stricter participation or having more skin in the game um could actually help um reform the systems, come up with ways where like somebody um gets slashed if something happens, you have to be doing these kind of predictions. um some Dows back in the golden age tried to play with this. I know the first situation of Nosis did this kind of protection system, but maybe it's time we had a comeback of this. Um after all, that's kind of what we have on the layer one. Ethereum is not a DAO and it's been working just fine.

So, right on time. Thank you so much for coming and if you have any questions, corrections, please let me know to celebrate. Here's to winning in Rome like the Romans. Hey. Hey.

Um, really nice to see you talking on stage and I guess I wanted to just ask a bit more personal question because you talked about the consequences of this at the DAO level, but one thing interesting in your case is that you this was happening in the DAO. This happened at your personal level as well, right? You went from an enthusiastic community member in the Paris DA to know a professional delegate essentially. So, how does that look like uh more personally? Yeah, I mean um that's a very good point.

I should have brought up more the story so people know what um what how I ended up here. Uh in the audience I have basically my mentor, the one that made me finish up where I'm talking right now. But I started being part of Parasoft DAO literally um on day one when the airdrop happened and wanted to participate wanted to improve the protocol and wanted to um make the protocol that I like so much participate on a day-to-day basis and by having these kind of more strict systems where you have to be already part of a strict organization you have to have this kind of like close reputation to be joining in. It's true that we end up losing this kind of um onboarding method that maybe previously was not as accessible. We have this horrible reputation in web 3 in my opinion where people think that if you don't know how to code, if you don't know solidity, if you don't know how to be an auditor, you cannot do anything for web 3.

And in reality, we need more of these non-dev people to have a say on how these protocols are actually being shaped. Um so I should have added that point. Thank you, Bree. But by having this kind of open participation, we are also able to have this kind of on boarding of perhaps experts that come from other fields or people that were initially not web3 that are just like curious to double their um their feet into the world of Dows and actually be able to have a say and these fresh ideas can actually really help the system. I mean from ZK proofs to the way that AMMs are designed, we need anything from mathematicians to lawyers to be able to um define the future of Dows.

Thank you. Yeah. Um u regarding uh dual governance, I was actually the one who led the design of the system.

Oh my god. So yeah,

can I have your autograph?

Can I have your autograph?

Yeah. So um um you said everything correct. Uh let's say dual governance doesn't solve the problem of um like mediocrity, right? because like it by providing escape h for stakers and uh like a way to say that I disagree and want to leave it does protect from like really bad outcomes attacks but it doesn't actually solve the thing that like nonoptimal decisions can be uh can be taken or mediocre right uh because stakers usually don't care about like the the DAO being really like u competitive because they can just switch to a competitor if there is a better one, right? And um uh yeah, and futarch also I don't think it really solves this uh because like it it plays well in the short term when you can objectively assess uh the change, right?

But it doesn't uh it doesn't allow you to bet on I don't know 10 year ahead like outcomes. So yeah, this is a really really good question and like I I right now I don't have an answer how to solve it and uh if if any any has any ideas it will be great to chat because like I think it's one of the most most hard pressing problems. Uh yeah

yeah thank you for the insight. I try to get like it's always difficult to predict how the future is going to be going and which paths are going to be the golden path. So what I tried to get here was like some breadcrumbs of things that I can see developing and how we can be going forward from this. So for example the what you mentioned with the current way of future key it still has to be developed because you don't know what could be happening. So for example some of the limitations with the optimism experiment is that it was priced in dollars.

So people were speculating more about will the market go up and down in the next three months as opposed to the actual outcome of grants. But that could be a way of abstracting maybe not with future key but maybe these games that like take away the complicated part but still has an opinion and also with dual governance I agree that it's more of a I heard it called a rage quit kind of system as well but it's uh just a way that I saw the first steps of people saying okay maybe we can have systems that like people get to have a say without having to directly have a voice in the day-to-day operations because yeah you're right at the end of the day um these operators can always have other ways of signaling their dissatisfaction with the DA.

Thank you.

Um, hey there.

Hey. Um I want to thank you for making this talk because I've been a delegate in arbitum for a while a small delegate and I've been feeling this transition in arbitum from um you know more enthusiastic community maybe a couple years ago to something more um centralized now

I'm not sure if you're aware of the new vision that was posted by the foundation where the governance process is going to be different um and I was trying to articulate what you did in this talk and I think you did a great job at articulating what are the dangers of uh too many professionalized delegates. My question is what do you think about um incentivizing exclusivity of delegates? uh you talked about the metadlegate being you know um professionalized delegate across multiple dials and then how can you really um ensure that their interests are aligned if they're a delegate in arbitum and zk sync and optimism and so on right uh so my question is what do you think about having exclusive delegates to just one though

so well also to gain more context to be honest what's been happening in arbitum has also been a very big catalyst of mine um because also these big changes have made it more difficult for me to keep up with arbitrum dao. I was also one of the first people to make my deta platform on arbitum dao and I haven't been able to keep as much because of all the changes that have been happening. But going back to your question, I have a very biased answer to this because of um what I've been trying to help out with Laura Dao with the new delegates that have been happening. Um recently we also launched this kind of a professional uh delegates program and one of the outcomes that have been developing from this is this kind of growth committee. the idea of like having some key members of the DAO have a say of ways that we could be growing the protocol and be able to um get an edge and be able to acquire the most number of um volume users possible.

Um, one controversial thing that we pushed on our side in lighter labs that delegates were not really aware of was that we asked if you want to apply for this position, you're not able to participate in any other similar um positions in similar dows or have any conflict of interest

and it feels like I'm stating the obvious but so many people applied and then said like this is the first time that I see a DAO have this kind of requirement. So getting back to your question, I feel that for delegation itself, I'd say it's not much of a problem because it's out in the open in the forums and everyone can see how you're voting and everyone can have your final opinion whether to delegate to the guy that's in competing Dows or not. Um once something is out on the forums like uh you can have literally everyone step in and say anything and delegate. But for the more innermost operations like when it comes to running things like the grants, running the growth, running these kind of like things that your protocol actually depends on, I feel that exclusivity is going to be something that we're going to be seeing more and more people actually having to state their allegiance because this idea of like, oh, I'm just here trying to push paperwork is not going to be um helping in the long term um grow any protocols.

Thank you. Uh I have another question. I've heard you mentioned some external like checklist and frameworks that help delegate to do diligent work. But what about internal? Maybe you have some mental model like ethical rules for yourself or other delegates you can think of that may highlight the moment when okay I can't do great job as a delegate anymore for this D or like at all.

uh something like that.

So you're you're asking like uh internal frameworks when when you think

like you understand okay I'm burnt out for example or like I love this project so much I can't critique it anymore to the level it's needed to be. Yeah.

It's interesting because I feel that we have a very strong culture in web 3 where we push ourselves to the limits to the point that we end up hating and the things that we ended up loving so much. So it's very common for example that's why we see so much hopping from very similar projects. People go from like one um liquid staging project to one that's probably even technically worse just because it's new and it's something more exciting. Um when it comes to like how when to decide to get on boarded I'd say the bar to actually becoming a delegate is a lot lower than what people believe. Um people tend to have this opinion that to participate in a DA you need to know a lot more than what you do when in reality a good delegate just needs to ask the right questions.

They have to be just like, okay, do I understand what it's saying? Do I make a question that probably everyone is having but nobody's asking? That's already enough to start being a delegate. And going to your part of like internally deciding when you think it's enough or when it's time to step out. I feel that the biggest indicator for that is when forum discussions start being more about the state of internal management of DAO and other delegates as opposed to the actual growth of the protocol itself.

I've seen this happen in a million different Dows. You end up with you have at the beginning these questions about like let's grow this D let's do this initiative let's make this kind of staking system and when they get burnt out it starts being things like this delegate did this or this delegate did that or I'm upset with this and once you start having this kind of infighting you know you spend more time talking about the process than talking about the protocol I feel that's when you have to say time out okay everyone maybe let's gather a couple weeks think about what's going wrong and maybe try to get something out and final thing that I really enjoyed from some Dows it's starting to be normalized the the idea of standardizing voting periods and break periods. The idea of like most votes have to be happening at this time of the month and you have this two week breakout period where you can reflect and think about these things because the other thing that tends to happen a lot is the constant cycle of votes really ends up like getting into you. Um hopefully this helps.

Thank you. Uh do you think that delegate cartels represent as much of a risk to Dows in the one year ahead let's say than attackers such as Hampy on compound.

So I'd say the answer is hard to answer because Dows it really depends on how much power the DAO actually has. So if we're talking about these just like rubber stamping dows where you make a proposal and you do these kind of procedures just for the foundation to execute something then these kind of delegate um groups are not going to be doing much because then the foundation steps in says you know what I don't like this anymore and takes away the whole theater but then again was it really at the if they could do this in the first place the ones where you have actual treasury management and you have actual resources being spent I feel it's going to be a much bigger risk and that's why we are seeing this central like this counter weight right now happening as well. Um because you also have developers um that are currently worried about their Dows being captured by these kind of systems that are not really developing the protocol itself. And as we start developing the projects more and more we need to focus into I think I'm being called out. We need to focus more into active protocol growth as opposed to um just running the DAO as a normal community.

All right. Thank you.

All right. Thank you.