The following transcript is from a fireside chat that took place at Consensus 2022 featuring John deVadoss, head of NDG Enterprise, and Erhan Korhaliller, founder and CEO of EAK Digital. The session was part of a four-part exploration, “DAOs: Without Your Permission” and focused on emerging patterns and practices for Web3 development.
Erhan: All right. Thanks everyone for coming. It’s great to see so many people here. My name is Erhan, I’m the founder and CEO of EAK Digital, which is a blockchain PR and events company. We organize Istanbul blockchain week and BlockDown festival. I’m also a general partner at Matterblock, which is a Web 3 fund and accelerator. Really delighted to have John DeVadoss today with us.
John: Thank you, Erhan. Good morning. I’m John DeVadoss, I work on the developer tooling for Neo N3. So if you like our tools, I’m the guy, if you don’t like what we build, I’m the guy as well. I want feedback, we believe we’ve built the best Dev Tools ecosystem in the industry and are certainly very proud of what we have done.
Erhan: Yeah, so obviously the DAO space is still extremely nascent. We see two main styles of DAOs, investment DAOs and contribution DAOs. And I think right now, DAOs are particularly interesting, now that we’re in a bear market, especially on the contribution side, where people are earning tokens to do different tasks. Obviously we’ve seen that Coinbase is not hiring the people that they promised to hire, and then at Gemini, getting rid of 10% of their workforce. People are going to be able to get more jobs and incomes using new DAO systems, which I think is going to become more and more prevalent, especially as we may go into some kind of a recession.
So I think that from a future of work perspective, it’s quite interesting, because the pandemic really accelerated people working from home, which obviously gives them a bit more ownership of their job and their work. And then also now we have DAOs, people are able to jump from DAO to DAO, and not be locked into a labor contract which forces them to stay for a certain amount of time or do a certain amount of hours. So it’s really given more sovereignty to employees. And I think that’s a really interesting development to see where that’s going. John, what’s your current opinion on the DAO landscape? Where are we right now?
John: Thank you. I want to just give you a perspective. This morning there were a couple of sessions here, and I think there were comments that made us think of the DAO as the evolution of the corporation. I think that’s a fair perspective. There was another interesting point, that really DAOs are kind of the destination; the model that we’ve been pursuing for a long time. What we have today or over the last few centuries has been building up slowly towards this destination.
For me, I’m a design patterns guy. So what I really like, as I observe this, are the underlying design patterns, how do you build a DAO? How do you evolve the DAOs? And from a technology and a technical perspective, but also from a commercial perspective? What are the component patterns, how do you compose them? DAOs are not static things. You don’t just build one and say, I’m done. It’s something that’s living, it’s evolving. So I’m kind of obsessed with this notion of the design patterns, and how we can create the tooling on Neo N3 to help support people out there building DAOs, and how to evolve them eventually, in due course.
Erhan: You’ve been building tooling for products for decades now, and Neo has some of the best tooling for developers. Obviously DAO tooling at the moment is again, very nascent, it’s not really there yet in order to get adoption of DAOs, I don’t think. How is Neo helping that come together?
John: I want to maybe highlight one particular DAO recently that’s been built by some key developers in the community. It’s called GrantShares, grantshares.io is the URL. As you are probably aware, Neo N3 has been very ecosystem focused over the last couple of years, there have been efforts like EcoBoost and other similar initiatives to support, to grant, and to fund community initiatives.
What is very interesting about GrantShares for me, two things: one, the people who built it—amazingly cool developers. And so that design, the underlying design is something that I feel that we can learn, reuse, and build on top of. The second basis; it’s also a very natural portal, an entry point for people to go in and apply for a grant from the ecosystem. In some ways it is bootstrapping and self-referential. I recommend everyone to take a look at GrantShares in terms of how to design and build a DAO in keeping with the highest principles of software engineering.
Erhan: Yeah, and that’s available right now is it?
John: It’s out there right now. In fact, the guys, Guil is right here, the folks who built it are right here, please go check it out. It was released very recently, really good work. And again, an exemplar on how to design and build a DAO.
Erhan: Okay cool. Yeah, everyone do that. So again, looking at DAOs from a technical perspective. You know, obviously this is version 0.1, right. We’ve got a long way to go, how do you see them evolving as time goes by?
John: Certainly. And by the way I would love to get your perspective on this as well, because you’re an industry expert across multiple domains, I want to get your thoughts in terms of how you see the evolution. But let me start. What is interesting for me, as a guy who spent many, many years in the AI and machine learning space, I’m particularly keen on this notion of evolution of DAOs.
There was a field called, many years ago, genetic algorithms, in terms of how you fuse and evolve and create new generations of software artifacts. And particularly, I think a lot about this notion of how do we use machine-learning to take over the mechanics of a DAO over time. Of course, we are still figuring those things out for the most part, but just the core, in terms of the DAO, this notion of genetic algorithms and machine learning, and having micro DAOs, and how do you compose micro DAOs to create macro DAOs, if you will. I think it’s a space where we will see significant innovation and focus in the coming months and years. But now, I want to hear your perspective as well.
Erhan: Well just before we do that, you used a lot of terms there like genetic and AI and everything. So could you give an example of what that might look like, just for people that don’t have that technical understanding?
John: Certainly, I’ll give you a quick synopsis. The idea of machine learning you’re already familiar with, but this notion of genetic algorithms is a theme from the early 90s in terms of how do you evolve code, with minimal, if you will, human interaction. And the system is designed with intrinsic constraints and enablers. And you can think of it as parents and children if you will, where you’re able to combine the best and maybe worst sometimes of both sides to create the next generation, and the next, and so on.
So to me, the mental model of DAOs as we see them today is very primitive, I believe, and I apologize for being very candid, however we can all see the potential. And so to realize the potential, I believe we have to look at other similar fields and say, this has to be an evolutionary mechanism. Now, whether it is machine learning, maybe, maybe not, I don’t know, I just want to plant the seeds for you to think about it, not as human directed or human instead of mechanized, but really self-driven, autonomous if you will, organisms.
So ping me offline, I’m happy to share my thoughts if you’d like, it’s a topic very close to my heart. In fact, we’re working on similar developer tooling to support DAO builders out there, not just to build but also again, to take it forward, to evolve.
Erhan: Yeah. Obviously John, you see things from a technical standpoint, I see things more from a marketing and events standpoint. I’m quite interested to potentially incorporate a DAO into some of the events that I organize, it’s a decentralized way for people to vote and govern what they actually want to see on stage, rather than just one or two people behind it, thinking that we’re just making up what people want to see.
That way, it’d be a lot fairer to put together an agenda and make an event for the people, which I think is great. And then you know, from a sports perspective, I’m a big sports fan John, I think you are as well. I’m obviously English as you can tell from my accent, and our Premier League teams right now, as you know, some of them are owned by oligarchs and people that are purely in it for profit, don’t care about the teams. I would really love to see DAOs in some of the biggest sports teams.
I’m a big Manchester United fan, for example, we have our issues with our owners right now. And I’m a big supporter, you know our fans, if we were owning the club from a DAO, it would look after the club, after all, the whole club is based on the community and the fans and the memories. And I think that’s a perfect marriage of DAOs that I would love to see happen. Maybe that’s a romantic way of looking at it, but hopefully one day we can see some more sports teams do that.
John: Absolutely. In fact, again, at the risk of offending other fans, I’m a big Packers fan, and the Packers here in the US have had this model for a very long time of being owned and run by the fans themselves. This is long before DAOs, right, and so for me again, there’s a natural fit, like you said Erhan, of sports teams, franchises… this is the fans who make the club happen. So being able to bring the fans in, and again, designing probably the next wave of DAOs, I believe is inevitable.
Erhan: 100%. So going back to Neo, obviously you guys recently launched Neo N3. And it’s been a big focus. Neo is a blockchain with its own inbuilt oracle, you know it’s quite unique compared to some of the other blockchains out there. So, what’s kind of your focus area now with Neo N3? What does it enable builders to do?
John: Thank you. So I want to go back a few years ago, when Neo was designed in the beginning, and what really motivated me to work on and build tools for Neo, was this thesis that assets will be digitized, of course. And looking back today it seems kind of obvious, yeah of course. But we’ve seen over the last couple of years, gaming for example, we’re seeing more and more asset backed coins in those domains and their evolution. And I look at DAOs very similarly, in terms of, again, the transformation of assets from what you might call today “in real life”, to tomorrow.
So we are particularly focused, I would say maybe even obsessed with the notion of developers, and helping developers to be able to build what they dream very quickly. For example, again, not to tout, but, you know, you could pick up our Toolkit and in about three, four, or five minutes, build a smart contract, test and debug and deploy it, right? And our intention is to take this forward with the GrantShares initiative and other similar efforts to provide similar syntactic sugar as well as underlying framework support for developers to be able to quickly easily build, test, deploy, and evolve DAOs. Because we, I believe, certainly, that the notion of decentralized organizations, whether DAOs in the current incarnation, or perhaps something else, is inevitable. This is happening, whether we like it or not. And the question is, do you want to be on this train? And if you do, then now’s the time to begin to experiment and to play with it.
Erhan: And which industries do you think will be the first ones, I suppose, like venture capital’s already been disrupted with investment DAOs? Which industries do you think will be disrupted first?
John: So to your first point earlier, about 20 minutes ago, you talked about the current kind of wave and people and work the nature of work, right? I certainly want to believe that the notion of an individual being tied to a corporation or an entity to work, I believe, will disappear, that it’ll be loosely coupled. Whatever you want to call them, but loosely coupled structures where people are able to do multiple things. You don’t have to… this notion that I work for company X, and therefore I cannot do B or C or D, I think it’s just passé. It’s just it’s a sort of dumb, you know, artifact in my in my opinion. So this notion of a loosely coupled new world of work, if you will, will fundamentally replace everything we have today. And I think potentially what might be, shall we dare say a bear market, might be a catalyst to help us get there.
Erhan: So what about the security models of DAOs? There’s a famous first DAO in 2016, I think they had a hack, I read about it earlier. How do we, especially if we’re talking about people’s salaries, people working, people earning tokens, obviously, it’s pretty important that the security layer is clean. From a technical standpoint, how is that currently?
John: That’s a really good question, in fact that is something we didn’t discuss, we’re talking backstage about what we should talk about. This is a very interesting point Erhan makes. You know the way I see it, it’s very simple, right? To be secure, you want to have less surface area, which means less code, less design, less time involved in the preparation and deployment, less of everything. Because when you have more surface area, you open it up to risk, right. And so the first thing I would say is from a tooling and a platform perspective, in any tooling ecosystem, like of course N3, that minimizes the surface area for you, the frameworks are able to move up the stack. And the work on the part of the developer is less and less, is certainly the direction to go with respect to security.
And I don’t want to say it’s the only thing. Of course, we know we have to worry about threat models, right? When you design something, especially a DAO, you have to have a pretty robust threat model. So what is the threat model? You think about all the vectors in terms of protagonist, possible avenues of attack? And you say, okay, for each of these vectors, let’s understand first, and then analyze, what are the variables? Then how do you minimize each of those variables across those various vectors? So that’s something obviously, I would assume you do. And if you haven’t, I strongly advise you, especially as Erhan says, in the DAO space, because we are in new territory here. So two things, surface area, minimize it. And second, ensure that you have a very robust threat model, or even multiple threat models would be my strong advice.
Erhan: So how early do you think we are with DAOs? Where are we with DAOs right now?
John: Now? I’ll give you a perspective, and I want to get your perspective as well. And I see it as really very early, very nascent, in my perspective. That’s a good thing. And why is it good? Because we can experiment. Nobody is telling us “the DAO has to be like this”, which we don’t want. We’re going to learn, we’re going to experiment. And we will shape what the DAO’s of tomorrow look like. That is what we have as the opportunity. So I see it really as sort of a two sided coin, right. It’s very early. But that is the opportunity, because I don’t want to say, you know, somebody up there will tell me what a DAO should look like, right? No, we, us, you, me, together, we will define what this tomorrow looks like, for a DAO across the technology as well as the commercial dimensions. But what do you think?
Erhan: It’s beyond early. You know, it’s like, was it that 56k modem router? I think we’re about at that stage. Like you said, it’s a great opportunity for people to collaborate to get involved, especially with GrantShares DAO and N3 to kind of build the future of DAOs, you know, what do we want it to look like? It’s kind of up to all of us really, and how we integrate and use DAOs. So it’s going to be fascinating, isn’t it?
John: Absolutely. What a time to be alive. Right? I want to go back to your point. So again, if you’re thinking about building a DAO, if you’re maybe in the process of building one, do take a look at GrantShares.io, reuse the code, if you will, at the same time, if you want to sign up or make a proposal for a grant, it’s the same mechanism anyway. You can kind of get two birds in one stone, so give it a spin.
I know we have a couple of minutes, so if there’s any questions, happy to open it up?
Erhan: I think there’s someone over there with a hand up.
Audience member: So when you were talking about design patterns, are you using that term in kind of informally, or are you talking about more formalized design patterns, and if so, have we … [inaudible]?
John: Thank you for the wonderful question, I’m really happy you asked this question. Yes, I do mean this in the formal sense. And why do I say this? Because as we mature the space, the underlying patterns will have code instantiations. And there’ll be libraries that we can reuse; that can only happen, rather we can reify the patterns only if they are formal. That’s my position. Now to your second point. Where are we in this process? Again, very early, it’s something I’m driving with a few colleagues and friends. If you have an interest, I would love to partner with you. But I believe it’s inevitable. We’ve seen this before in terms of cloud design patterns, services, and earlier waves of computing errors, so it’s inevitable, but if we can make this, build this together, please ping me offline, I would love to connect with you.
Erhan: I think, do you have time for one more question?
John: Maybe a few seconds here, yeah anything else?
Erhan: Anyone else? I think there’s a hand over there.
Audience member: Could you talk a little more about the intersection between machine learning and DAOs, what do you envision it as, like a machine learning algorithm being used … [inaudible]?
John: Yeah, thank you again for a wonderful question. The question was about the intersection of machine learning and DAOs. So if we decouple, or at least asymptotically decouple, the human aspects from a DAO, then for me, the evolution of the DAO has to be a self-governing, autonomous if you will, mechanism. And that’s why this idea of machine learning, in particular around genetic algorithms as I referred to earlier, we have micro DAOs, composition of DAOs, perhaps pairing of DAOs, and then you can have a resulting kind of “child DAO”. This is still very early, but we can already see these discussions happening. People are talking about how do we merge them, how do you unify them? How do you have this conjunct or conflation? And what does the resulting child DAO look like, right? So that’s why I believe we will see significant work, and today these fields are somewhat of a diversion. I don’t think we have too many machine learning people in the DAO space. And we certainly don’t have DAO people in the machine learning space. And that is also an opportunity for us to be able to discover and say okay, look, let’s see where this intersection goes. Once again, I would love to follow up offline. Please ping me, I’m on Twitter, LinkedIn, the whole deal, so would love to follow up and discuss more.
Erhan: Cool. Well look, I think we’re out of time. So you can find John at the Neo booth, and we’ll probably just be backstage coming out of here. So if anyone has any quick questions for either of us, you know, we’d be happy to talk to you outside. But, thanks so much for coming to our session, and enjoy the rest of the event.
John: Thank you very much.