Ben Sheppard  00:23

Hello, everyone, welcome to another episode of the TX – Tomorrow Explored podcast show. I’m Ben Sheppard, Managing Director at TX. On today’s show, we’ve got a special guest, we’re going to be talking about DAOs. We’re going to be talking about API3, and we’re going to be talking about Björk. For those of you out there who remember who Björk is the famous Icelandic singer, round, still around. So, we’re gonna get into all three of those subjects on this pod. So, today’s guest is Sasa Milic, who’s one of the co-founders at API3. Welcome, Sasa, how are you?

Sasa Milic  01:09

Thank you so much. I’m good. How are you?

Ben Sheppard  01:11

I’m doing good. Did you know, I am actually in the happiest country in the world, for the fourth year running. Finland has done it again, we’re the happiest country.

Sasa Milic  01:22

I didn’t know that our one of the co-founders is finished and he just tells me a lot about there’s a lot of drinking in Finland. So, I assumed there wasn’t going to be a correlation between happiness, but I don’t know. I guess there is. I wasn’t aware that they were the happiest country. I’ll have to tell him that and I feel like he might disagree.

Ben Sheppard  01:44

Yeah, indeed. I think COVID and Finland, probably, you know, the lock downs because of COVID and the way that Finns are naturally goes sort of really well together because it’s self-isolate anyway, they’re very sort of keeping themselves to themselves sort of people so I wasn’t surprised when this year actually came number one but in previous years is coming number one as well.

Sasa Milic  02:09

Yeah, before any Fins get mad at me, I actually did a vacation in Finland I was in Helsinki and some smaller town. It was one of my favorite vacations, so I have nothing bad to say about Finland. I actually really liked it when I was there.

Ben Sheppard  02:22

As do I love it. [Inaudible 02:25].

Sasa Milic  02:25

Are you in Helsinki or somewhere else?

Ben Sheppard  02:27

Espoo actually, so it’s just outside of Helsinki. Yeah. Beautiful country. I’m looking out the window now. It is forest and snow. I mean, it’s just really gorgeous here and you’re in Canada?

Sasa Milic  02:40

Yeah, I was just commenting. I have my Tim Hortons cup. So, for any Canadians watching to make them feel a little bit more homey watching this, but yeah, I’m in Canada. If anyone’s wondering about my name, I was born in Bosnia. That’s like my background, but I grew up in Canada. That’s why I don’t think I have an accent. Some people say I do, but yeah, I grew up in Canada, and I’m here right now.

Ben Sheppard  03:04

Cool. Is it still snowy, where you are in Canada?

Sasa Milic  03:06

Yeah, it’s pretty bad. I’m in Calgary and actually, we got fresh snow, like two days ago. Probably like, I don’t know, a quarter of a meter of snow outside.

Ben Sheppard  03:17

Yeah. Well, I think we’ll do some mole, sometime next week. It’s been melting here. There’s just a little bit out there now. So, it’s kind of that wet slushy stuff that you just found on your backside. I think we’ll get another load soon. So, we have to talk about Björk first, because I was super interested by this in our prep call when you told me that basically, you’re quite a fan of Björk. In fact, you’re a blogger about Björk. So, come on, tell us all about this. Sasa, what’s this about?

Sasa Milic  03:49

Yeah. So, for people watching Ben asked me I think like, what are your hobbies or something along those lines, and I do have a lot of hobbies, or I should say I had a lot, but I’ve been pretty much like crypto full time for at least like a year like all my free time reading about crypto, but one thing I have maintained is that I have a couple of blogs actually about Björk and her music and I’ve maintained that. I don’t know how much [Inaudible 04:17], but I’ve been obsessed with her since I was like 15 and I even went a couple years ago, I started reading all of her interviews. There’s some database online. The thing about Björk is that people are either obsessed with her or they don’t really care. So, some other people that are obsessed with her had this like database of every single interview. She’s done since like, 1989 up until like, last year, and I started reading all of them and I found them so interesting, because a lot of people find her music kind of opaque like they don’t get it’s weird, but if you read her interview, she knows exactly what she’s doing. A lot of things like her work makes more sense in the context of an interview. So, I just started taking quotes from her interviews, and some of them are really profound just like quotes about life, quotes about dealing with balance, just very like kind of high-level themes about life she talks about a lot. So, I just started taking little quotes of that and posting it on a Tumblr and I got like, I mean, I normally don’t care about followers, but I thought it was cool that it was something I didn’t care that much about in terms of like, having a lot of followers that ended up having I think, like something like 5000 and then I started doing and then I had a fit oh, I still have a Facebook page where I look at tarot cards in the context of her music just because I find her music. So, like they talked about the themes in her music are very like high level archetypal, very all encompassing. So, I thought, like, tarot cards are the best context to kind of discuss for music. That’s got like 10,000 followers or something like that. I maintain it like once a week or something, just post a thought or a quote or something like that. I’m super obsessed with her and have said it, since I was like, 15.

Ben Sheppard  06:02

Is that when the big hit came out? It’s so quiet. Is that being the title of it?

Sasa Milic  06:06

Oh, no, like that came around when I was born actually. I think that was around 93. It’s also quite came out in 93. I can probably tell you exactly the exact year everything out.

Ben Sheppard  06:21

I remember when that came out. I mean, that was a good hit. That was really good.

Sasa Milic  06:27

Just my last comment about Björk, I can’t imagine actually living at a time where she was mainstream, and you could see her on TV because I got into her high schoolish to like 2007 when a lot of her albums came out. At that point, she was like, Avant Garde, only hipsters listened to her, but like in the 90s, she was like, you know, on the radio and on TV.

Ben Sheppard  06:46

Yeah, she was on top of the pops in the UK at number one for a while, I think with that hit. So, yeah, she was very famous for a while. Yeah. Cool. All right. So, world of Web3, what brought you into the world of Web3, Sasa?

Sasa Milic  07:04

I’ll try to keep this one short, too. I feel like a lot of people that do these interviews start off with you know, I’ve been interested in Bitcoin, like, since 2010, or something like that. I can’t say that I’ve been really interested in crypto for a little, almost two years and it started off, well, I actually was really I had bought some Bitcoin, Ethereum like, around like 2017 crash or like on the way down just because the crash actually made me pay attention. I read about the actual Bitcoin algorithm. I thought it was like the most interesting algorithm I’d ever seen. I took graduate courses in distributed consensus. So, I was familiar with, like, what the problem is, and trying to get different nodes of the network to agree. So, I was familiar with the general context, but I had studied algorithms from the 90s. They weren’t particularly interesting. So, when I heard about or when I read about the actual Bitcoin consensus algorithm, I was like, this is the coolest thing I’ve seen. I took a lot of advanced algorithm courses in university; I studied computer science. So, I bought some Bitcoin Ethereum, and then just forgot about it. Then I saw some crypto memes online, I think, May 2019. So, like, deep in the bear market, I see a bunch of crypto memes online for like projects I’ve never heard of. I was like, do people still care about crypto? Like, I was so surprised. I thought, yeah, bitcoins still around, but like, I had no idea, but people were like busy building things during the bear market and people are still, the people that were left out after the bubble in the crash are actually the people that are most obsessed about this stuff. So, that’s why you had all this, like really good projects coming out and I actually got interested in the Oracle problem.

Maybe it will probably touch on this after when I get into what API3 actually does, but it was this specific problem of how to get data on chain because I was like, wow, if you can figure that out, if you can do that securely. You can apply blockchain to you know, insurance to supply chain to all sorts of things, legal disputes, potentially just, there’s so many industries that can be streamlined with Blockchain. So, it was just a bunch of memes online in May that I was like, okay, crypto is still around. I started reading papers and just got really obsessed. It’s such a cliche to fall down the rabbit hole, but that’s kind of what happened. It was around the time that I moved to New York. So, I started reading about crypto and then I was like, are there any meetup groups in New York about crypto and there’s tons as you can imagine, so like all my friends that so I moved to this big city and then all my initial friends, and still most of them were people in crypto and so just my life began to revolve around it. I was actually working at Facebook at the time I moved to New York to work at Facebook, but then at work, I was literally reading white papers like you Dude work 50% of the time and kind of reading about crypto the rest of the time. So, I just got obsessed with it and then was just looking for how do I contribute to this space? I didn’t want to just jump to any project, because as anyone knows, most projects in crypto are not good. I don’t want to say it scams, but there’s those too, there’s only like a real handful of things of projects that are solving like really critical problems in a good way. So, how did I get into API3? Now I’m like, kind of rambling on. So, somehow managed to get myself in, like consulting for crypto companies, just through my friend network, and people like going to these meetups and realizing I kind of knew what I was talking about, because I come from an academic background. So, understanding white papers wasn’t an issue for me. So, I did a bit of consulting. Then I was looking at Oracle projects, and I had a lot of questions. I didn’t think they was solving the problem in the right way. So, I contacted the people that are my co-founders now and was like, I know you guys also work in Oracle space I can get into that after but I’m trying to kind of keep it short. As I know, you guys work in Oracle space, are you noticing the same issues that I’ve noticed from consulting with some of these companies? The answer was yes. Then pretty much like two days after having this conversation of Brock, the first author on the white paper, and also one of the co-founders, he reached out to me and was like, do you want to work on this project we’ve already kind of brainstorm this idea for API3, we just got to start writing the white paper. So, I had a conversation with Brock and then like two days later, we were working on the white paper together. So, it happened very, very quickly.

Ben Sheppard  11:50

Super interesting. I mean, it seem a little bit of similar to me then in that case, crypto was huge in the Philippines, and it has been for a few years now. You got devices out there, like coins.ph is 6 million users and that’s been in use since…? Well, I was first buying Bitcoin and Ethereum, from seven elevens stores in the Philippines back in 2016 I think, and you can put it into this coin, [Inaudible 12:19], there’s quite a strong crypto community. Now I didn’t understand really what crypto was then I just thought I should probably buy some of these it feels like something I should buy a few of and just see what happens. Then before you know it through meetup groups and other things, you find yourself talking to different people and I guess like you, I found some interesting projects online, and I wrote to one of them was like, hey, I’ve been looking at your project, just kind of curious if you’re coming up against these issues, here’s some use cases that you might want to think about. One thing led to another had a conversation and found myself in the world of crypto and I think that’s what’s so of wonderful about this industry that is very open in that way, you know, these projects are very open that if you spot something, can you write to these people then actually, they will respond sometimes or a lot of the time and listen to you and actually, you know, take you on board and it’s that sort of decentralized openness that the world of these things.

Sasa Milic  13:17

I completely agree with you I was actually talking to someone about this recently and the word I used was porous. Like, you know, hierarchical, hierarchical, crypto world where you know, or it’s not really like closed off like a cool, you know, art scene, like I encountered in New York, for example, like people are really like, oh, you’re interested in crypto, like, you know, talk to us or join our team or like people are so open and I love the fact that the idea behind the technology like decentralization a transparency, actually. How do I say it kind of permeates the whole community in the culture of blockchain. So, it’s not the technology has these features, but it’s also like the community that’s built around it has these features, which I find amazing and I think that’s what people say, they fall down the rabbit hole, it’s because, yeah, there’s the tech, but then there’s all these layers, the community, and then the kind of political implications and all this stuff that has to deal with decentralization and transparency and stuff like that.

Ben Sheppard  14:21

It’s fascinating to see where people have come from in their previous sort of careers as well before they come into sort of the world of crypto as well. There’s a guy called Jay that from Avalanche that we’re interviewing on the pod in a couple of weeks’ time and his backgrounds, photography, and music and like it’s super interesting, like the different backgrounds that people have had before they come into this industry and then the different knowledge and skills that all of a sudden is very relevant as well and what they can bring to it. So, I do think that’s a wonderful thing about this space that people from all walks of life can actually find a place in this sort of ecosystem and add value to you in different ways.

Sasa Milic  15:01

I just want to add something before we kind of move on to the other topic, but you know, I agree with everything you’re saying and I find it also necessary that you have all these people from different backgrounds because crypto, you have to think from, like, first principles a lot of the time, like, yes, I know, computer science and I also study statistics, but I don’t have a finance background. So, Defi, it’s like you can’t understand Defi, unless you have some basic understanding of finance and having all these people around that have these different skill sets and also understand their field from a very fundamental level, because you have to like rethink all these things that you think are obvious in your field, because kind of crypto re does everything from the ground up, even like to give an example. So, I’m not talking high level, but like something like Uniswap, like, all trading was order book trading, where like someone has a buy order, someone has a sell order, and you match them up. Uniswap is a completely different model and in order to come up with it, I mean, I should also clarify, I’m not entirely sure if they’re the first people that come up with autumn AMM’s although I think they are like that requires some really in like understanding things at a fundamental level and being really creative and to touch back on what you said like that requires a group, a community of people that have a very diverse skill set.

Ben Sheppard  16:18

The innovation in this industry is outstanding. I mean, I often joke, you know, just when I think I’ve caught up with all my reading, and I feel I’m now on top of a topic, it moves again and then it’s like, oh, man, you just have to keep going with this thing and just keep reading whichever facet of Web3 you’re involved with, whether it be you know, AI or a specific type of blockchain or a particular solution, such as yours around Oracle’s or whatever it is, it’s always evolving, and it’s doing it at a very fast pace, which is great, but it means you’re constantly learning. If you’re someone that seeks to sort of develop your knowledge, I mean, it’s fantastic flex, it just never ends. It’s a constant churn of new things to learn so tell us about API3, then what is API3?

Sasa Milic  17:09

So, before I ever start talking about technical things, I want to know, like, who’s the audience like, how familiar would they be with the Oracle problem, for example?

Ben Sheppard  17:20

Not at all.

Sasa Milic  17:21

Okay. So, I’ll start there. So, there’s something called the Oracle problem, although I want to put an asterix next to that, because we one of our theses at API3 is that it’s actually a poorly specified problem, but the general idea is that, the blockchain doesn’t have access to outside data and the reason is, and I’ll give like a short technical reason is, the way that you reach consensus, the way that you can get a decentralized network to agree on anything to agree on like, you know, the state of the network, which is who owns what, is every single node in the network has to replicate a complex computation locally, and then somehow agree with all the other nodes that yes, this is the right outcome of this computation. That means that all these nodes are using data on the network, like, and what that means practically, is usually just ownership data. Like, like when we say the state of the network, the state of the network, essentially, I’ll be speaking about Ethereum, specifically, but it applies to pretty much all blockchains. The state of the network is what is in everyone’s account. So, in Ethereum, yeah, you have an account, you have Ether, you have some other tokens, and sometimes an account is mapped to a smart contract, but that’s getting a little complicated. So, essentially, like what’s in everyone’s account? That is why I think that cryptocurrency is still the main application, because it immediately translates to money, right? Like, who owns what, that’s the data on the blockchain is what is in everyone’s account. Yeah, so essentially, ownership data is the only data that natively lives on the blockchain, but if you want to do something interesting, like write a contract, an insurance contract, like a flight insurance, you know, if my flight gets delayed, I want my money back. That can be really easily coded. That’s like in terms of code. That’s a simple if else statement, if the flight is delayed, return my money else, like in terms of code, that’s extremely simple. The hard part is how do you actually get that data securely on the blockchain? Like how do you know the flight was delayed?

So, you can get one single node in the network to transmit that information? You could say this node is going to tell us if so you have a network of 1000s of nodes and one of them is like, okay, I’m gonna say whether or not the flight was delayed, but that is now the weakest link in your smart contract because you’re trusting a single node you’re building on top of this amazing networking, this amazing algorithm that allows people to agree without any centralized point of failure, but as soon as you trust one person to deliver data that’s it’s like you’ve kind of diminished the whole point of the blockchain. So, the Oracle problem is, you know, how do you know, I think the way the Oracle problem is formulated is like, you don’t know what’s happening outside of the blockchain and then it’s a question of how do we approach this? How do we get data safely on the blockchain? And usually, the solution has something to do with redundancy, like you don’t just trust one node, you somehow get many nodes to agree on what the data is outside of the network and put it on the network. Many projects solve this in many different ways. So, this is the space, we’re working in API3, this is the problem that we’re addressing, although I should be clear, I did mention actually, I made a tweet about this on the API3, Twitter just the other day that explained that the Oracle problem is ill, I think it’s ill formed, there’s actually a mathematical definition of what an ill formed problem is. The Oracle problem fits that it’s too broad. There’s so many different types of data that you can get on the blockchain and the best solution for different types of data is different. I’m not explaining that very well. I’m just trying to avoid technical details, but we focus specifically on API data. If someone has a Web API that you can query with, you know, a get request. How do we get that on the blockchain? So, we’re not talking about all possible data that could possibly get on the blockchain? Because I mean, there’s some things in siloed industries that are not accessible by API’s. So, the Oracle problem is extremely broad, but we’re focusing specifically on if someone has an API, a Web API that can be accessed through the internet. How do we get that on the blockchain?

Ben Sheppard  21:49

Yeah, so another issue with the whole sort of data value chain is whether the data was actually any good in the first instance and if it’s been verified by a third party. So, you’re not dealing with those two segments, what you’re dealing with is, it’s that sort of validation that yes, this is the data from the source that hasn’t been tampered with in any way that is now going on to the blockchain. So, you give that reassurance of that, that is the case, but it could still be corrupt data that’s going to the chain, you don’t solve that part of the problem, but you do make sure that it hasn’t been tampered with in the process of getting it to the blockchain. Is that right?

Sasa Milic  22:31

Yes, thank you. I sometimes do interviews, and I feel like the people don’t even understand why the project is but you have like the main essence of the idea? Yes, that’s correct. So, I explained the problem we’re solving. Let me just quickly add on to that about how we’re doing it differently. So, most Oracle solutions, they say they solve the Oracle problem, but essentially most of the data on the blockchain right now is Web API data, like the majority of data feeds on the blockchain have to do with price. That has to do with Defi like, if you want to write any kind of financial derivative, which is essentially what Defi does in one way or another. You need to know the price data. So, it’s kind of ironic that on the blockchain doesn’t actually know what the price of Eath is that’s decided outside of the blockchain and just to be precise, some trading does exist on chain. So, some price data can be decided on chain, but most trading happens off chain. So, ironically, even though you’re passing around Eath, on the blockchain or some other token, the blockchain doesn’t actually know what that price is that the information from outside of the blockchain. So, you get that information from Web APIs from people that specialize in crypto price data, which is actually really hard. I can get into that after but it’s actually really hard to get accurate crypto price data, just because there’s always exchanges popping up. Sometimes the information isn’t correct, and whatnot. So, the way that a lot of solutions, a lot of Oracle projects solve the Oracle problem is they get nodes, which they call Oracle’s. So, there are nodes in the network, but they also have a technical component that can interact with like, HTTP, so like, the internet.

So, it’s kind of like this bridge between on chain and off chain and these projects, they’ll get like a bunch of nodes to query an API, or a couple of APIs, and then they’ll take the average, or something like that, or take the medium. That’s essentially the current solution are what we call the third-party solution, but we realize that you’re implicitly trusting the data source anyways, like if you ask a bunch of Oracle’s to query some API for you, you’re implicitly trusting the data source. So, whether or not the data source is correct most of the time, you’re implicitly trusting it. So, might as well get information directly from the data source. That’s the crux of that meaning, our solution is a bit more complex than that, but that’s the crux of the solution. Is that like you said, if the data comes directly from the data source, so an API provider is running an Oracle and transmitting that data on chain, you’re removing a bunch of middlemen and that means you’re removing an attack surface area, you’re also reducing costs, because you’re not asking like five Oracle’s to get information from coin Gecko or something like that, that information is coming from coin Gecko. Also, this isn’t exactly quantifiable, but it does reduce risk, because a third party, no name, that you don’t know, they have incentives to manipulate data, but like if you’re getting data from a source, I’m not saying the source would never manipulate it, but they have a business, they have an off-chain reputation. Like the risk, it’s hard to quantify, but it’s obviously there, the decrease of risk, when you’re getting it straight from the data source, like their incentives to lie, are much less than some third party that’s just running a node to get money. That’s just a quick note. For your audience, the reason why people would want to manipulate data, here’s a simple example. Like, imagine some really simple derivative contract, like…

Ben Sheppard  26:08

I was just about to say this exactly my mind was gonna, you carry on Sasa, I think this is a great example.

Sasa Milic  26:15

Yeah, I don’t come from a finance background. So, sometimes I get this stuff wrong, but think of a simple derivative contract, like, you know, if the price of Bitcoin moves up, whatever, some percentage, I get this payout. If it doesn’t, I get this other payout. So, almost like a futures contract. Again, I don’t know if your audience will be familiar with that, but like, you’re basically betting a derivative. I hope no finance people get mad at me when I say this, but derivatives are essentially bets on the price of assets. Like they’re just like people betting where the price is gonna go and if you have a large money on a particular derivative, and you’re running an Oracle, you have the incentive to manipulate the price in the way in the direction that would pay out the derivative money to you. So, if you’re betting the price will go up, you’re gonna lie about that. And, then you pay out. If you have a better example. Go ahead.

Ben Sheppard  27:12

No, definitely. I mean, I recorded another pod earlier today with Opium Protocol, we were talking about derivatives and as you were explaining exactly what API3 sort of does, I was thinking this is the exact use case where you know, someone’s bought Bitcoin, they’ve then bought a call option, sort of derivative on how that price might fluctuate and it would actually not saying they would, but it could be in their best interest to somehow manipulate that price in order for that, that call option to be valuable of them to take the return on it. So, by having this infrastructure that API3 sort of providers, in addition to the blockchain, it helps to solve them at some of those problems, and, you know, get around it. So, I think it’s a brilliant use case. We have a project that we’ve been working on for a while now called Tracy, which involves fisherfolk sharing data on their catch into a blockchain enabled app, that data is then sent to a bank, who then uses that information to ascertain credit worthiness and often microfinance back to these fisherfolk. So, basically, there’s a good reason why that data might want to be manipulated to make the credit worthiness of these individuals look better than what it is and again, it’s about how can you protect at least no third party trying to somehow manipulate that data in that value chain, so that it is actually correct from source as good as that source might be? You can rest assured that it’s as accurate as it can be. So, yeah, I find this quite interesting, actually.

Sasa Milic  28:56

Yeah, I just have two quick comments on that. So, like a high-level way of thinking about this and I know, you know, this, I’m just saying, you know, for the audience, a high-level way of thinking about this is you have a contract, and it gets triggered. So, it gets some input to trigger an output and that output is money. The input is data. So, like if you can control the data, you control the output of the contract, regardless of what that contract is. So, that’s like the high-level attack. Also, what we’re talking about isn’t theoretical, like people might have heard of flash on attacks and I think it’s a subset of flash on attacks, but something called sandwich attacks, but all these kinds of…

Ben Sheppard  29:34

What sandwich attacks is a new one on me?

Sasa Milic  29:37

Yeah, it’s very similar. So, I’ll explain it high level and I might get it wrong. So, just be aware of the audience but a flash on attacks people, basically. Yeah, again, for your audience. There’s some lending platforms where you can borrow money, a lot of money that you don’t have as long as you return it within this same transaction. So, it allows you to borrow a bunch of money, you don’t have to do something and give the money back and whether or not this is good or bad is up for debate. It means it allows for arbitrage which can, you know, is arbitrage good or bad? That’s a philosophical question kind of in the world of finance, but flash on attacks are kind of poorly named, because essentially, as far as I know, all flash on attacks have to do with manipulating Oracle prices. So, some people use on chain on to determine the price, which is good in some ways, but thing is someone can manipulate a price by just making a giant trade. So, like on a swap, you make a giant buy order, and you push up the price. So, the way flash on attacks work is someone borrows a bunch of money manipulates a price on chain, and then they get some payout because they have some derivative on the movement of the price. So, flashing on attacks are Oracle attacks, essentially.

Ben Sheppard  31:04

Wow, I’ve never thought of that angle. So, what’s a sandwich attack?

Sasa Milic  31:10

Again, Asterix, I might get this wrong, but it’s kind of like what it sounds like you manipulate the price one way, do something, and then manipulate the price the other way. So, it’s like a flash on attack with a flash on attack, you borrow money, you do something, you get some payout, and you repay it. So, sandwich attacks is like you manipulate the price one way do something and then manipulated the other way and it’s based off of the way you profit is I can’t think of an example but there are some derivatives that work on like, are not necessarily derivatives, but there’s ways that you can earn yield if like price moves in two directions, or something like that. So, it’s like profiting off of making the price move in two directions. Then I might get that wrong, but it’s the reason it’s called Sandwich attack is because you manipulate the price, you do something and then you manipulate the price the other way. Alright, I think that’s the essence of the attack.

Ben Sheppard  32:16

You’re manipulating both ends of the price.

Sasa Milic  32:19

Like I said, it’s like a subset of a flash on attack. So, if you get like the general gist of a flash on attack, it’s very similar in concept.

Ben Sheppard  32:26

Okay, interesting. I’ll be Googling that one afterwards to learn a bit more about it. So, let’s move on to DAO’s then, now that we’ve covered derivatives, and Oracle’s and Björk, so DAOs for today, so decentralized autonomous organizations, this is really fascinating for me, my background is management consulting, business at university. So, I find anything to do with new business models absolutely fascinating and just always want to experiment and see how you can run businesses differently. So, let’s talk about DAOs. What is DAO Sasa? How would you define a DAO?

Sasa Milic  33:12

Since I come from an engineering background I’m going to give a quick like technical explanation of what allows, how DAOs come to existence so normally like if you want to make a transaction on the blockchain so whether send money on Ethereum and that can be like a transaction could be like what’s the word putting a contract on the blockchain there’s a word for this so on Ethereum, saying a transaction is more complicated on Bitcoin transaction is sending Bitcoin and that’s it whereas Ethereum of a transaction is you can send Ethereum you can what’s the word Alicia contract on the blockchain trigger a contract so there’s so anything that gets the state to change this giant computer is a transaction and normally you just need your private key to do that. So, you sign a transaction with your private key however, there’s technology called multi signatures, which means that you in order to get a transaction executed you need five people to sign it or however many numbers and there’s also something called threshold signatures, which is like at least five people need to sign this transaction before it’s executed on the blockchain. So, the reason why I start with this low-level tech stuff is because I’m sure your audience I imagine they’re pretty smart, they can immediately see the implications. Like we can, you know, write a contract that or we can write a Treasury for example, this is probably a good example, like let’s say, we have some organization, whatever it may be, and we don’t want one person to control the Treasury when we move money from the treasury, we do not want one person. So, back to that is the general idea of like, no single points of failure. So, how do we do that? Okay, well, we create a multi signature threshold signature such that anytime money moves from the treasury, at least X number of people from our organization have to sign it.

So, there’s no one person that decides where money from a treasury goes and that’s just a treasury example, but this can be any contract change. So, that could be like Uniswap. If Uniswap wants to change there I don’t think you know, so I actually think they are a DAO, they recently became a DAO, but if they want to change their algorithm, or they want to change some parameter of their system, or if they want to whatever, like major code change, if they’re a DAO, that can’t be done by one person like that has to be transparent. First of all, like it has to be lit. Like this is our proposal to change the code. At least, a certain number of people have to agree to that proposal before it’s executed. So, again, so it’s taking that idea of no single points of failure and moving it kind of off chain. So, like no single person can change code because you have that now, even though the code on the blockchain is just decentralized and distributed, anyone can submit a contract on chain, but once an organization is a DAO, moving money from a treasury changing code, none of that can be done by a single person off chain, it has to be agreed upon by a certain number of people, depending on how the DAO is implemented. There’s many different implementations, but essentially, it’s not one person with their one private key that can make any change, it has to be by a small group of people.

Ben Sheppard  36:55

I’d imagine that must have to be quite fundamental changes because otherwise, it’d be a very inefficient model. If every decision had to go to a wide-open governance, you needed voting on it, and you need people to be available to vote and to participate. So, does that happen when it’s a material change to the technology product itself or is it for everything that it goes to the sort of voting platform?

Sasa Milic  37:22

You’re asking a very good question. So, yes, that’s pretty much the essential question when it comes to DAOs. So, I explained the default behavior. So, the way that DAOs work by default, like kind of extrapolating from this multi signature, technical concept is a proposal and a vote, a proposal, and a vote, but obviously, this is inefficient, and this goes back to political science debates of 1000s of years ever. Since, the Greeks came up with democracy, this is the big question, right? Does everyone vote on everything? Just think about it like a government. Like should everyone vote on everything and there’s good reason that no system really works this way. So, now, the big question of DAOs. Like, DAOs are sort of like, I don’t want to see reinventing the wheel. That’s a negative thing to say, but DAOs are addressing all these like really fundamental questions about political science just like Defi, just like Bitcoin addresses, like really fundamental questions about money, Defi starts, not intentionally, but you start having this conversation about like, what is a derivative? What is finance? What is trading? What is arbitrage? Like DAOs, we’re getting to ask these fundamental questions about like, how do you organize people? Like that’s a question that’s been around for 1000s of years, at least. So, now directly to your question DAOs are structured, there’s very few patterns, in terms of like, this is a good way to create a DAO this isn’t, but a lot of people are working on like, how do we make it so that we have sort of like representative governments? It’s like how do we implement representative governance like, everyone votes on like major things and some of them being like sub teams and sub DAOs that are that focus on specific tasks?

That’s a major kind of solution. That’s what we’re going to for API3. I guess it’s a tradeoff between fairness and getting in participation and just efficiency. Like if someone needs to fix a bug, we don’t need to make a proposal. We don’t need to make a proposal and get 1000s of people to vote on someone fixing a bug. Just so not talking to high level, a proposed solution, what we’re probably going to implement with API3, our DAO isn’t live so I can’t speak of what’s going to happen because you know, it’s up to the DAO, but the idea is that we’re going to vote on a team that works on specifically like bug fixes and that team gets a multi signature. So, that’s like three, four or five people, they have a multi signature. So, like, none of them can change the code, but we as a whole, we, as the entire DAO have voted them as having a multi signature that can change certain aspects of the code. Another thing that’s really cool is that you can actually program what people have access to. So, if these five people have a multi signature, we can program which parts of the code they actually have access to, they can just go on and change everything or have access to the Treasury, we can actually program that you have a multi-sig, for this little aspect of code, if anything goes wrong, you guys can take care of that. I haven’t seen it work too long in practice, but that’s kind of the approach that we’re probably going to take something like representative governance.

Ben Sheppard  40:53

So, you might have like, represented governance for different functions of the organization. Those people, presumably, they’re in those roles for a period of time and then there’s another election and it could be a different group of people come in, because otherwise you’ve just built the same again, with those people basically, in that role and that’s, you know, they’re there for good. I guess, you have to keep re-electing and re-voting and moving around, because this is wide open, anyone can come and participate in it. Now, how do you ensure consistency in service level quality, when you’re constantly bringing different people into these different elected teams? How do you ensure that there isn’t kind of like if you have a traditional company, and you bring a developer in it takes them time to get up to speed with the projects you’re involved with what the business is about to learn about the culture, all of those things like you give them a three month window to get into that and you guide them you give them mentors and buddies and that sort of thing. In a downtime situation, how do you deal with that side of things so that you’re getting that consistent sort of level of quality? And do people know where to reach out to other people for support and that kind of part of the business?

Sasa Milic  42:19

That’s a really good question again, and I want to clarify, our DAO isn’t live yet, like we have a version one DAO which is just the original team members, but the official DAO or like any API3 token holder is a DAO member if they stake that isn’t live yet. So, everything I’m talking about is theoretical and things we plan to do. So, just to get that out of the way, but I imagine that the original, the initial team members, especially people dealing anything with tech, are gonna be the people that have already worked on the project thus far, like I just don’t see people voting on someone that’s just new to the project. Like I feel like someone has to be entrenched, someone has already provided code. So, this would be like me, and the initial founders, probably seeding. At least seeding the initial team because we have like a track record. I guess it’s similar to like, why would someone elect a politician like I mean, that’s kind of you can get really messy discussion there. [Inaudible 43:23]. Yeah, hopefully, I mean, I don’t want to be naive and maybe we’ll face the same things with DAOs, but you should show some sort of a track record. So, I imagine the initial tech team is going to be at least seated with a couple of people that have already been working on the code base. So, people like me, and some of the other developers.

Ben Sheppard  43:47

The thing that’s really cool about this, and I interviewed, I forgotten her surname, but Irina from Polkadot the other day. So, she’s one of the head ambassadors over at Polkadot more on the marketing and sales side, but she’s involved in three different DAOs. She’s able to do that, because you know, you can work in any of these projects, you can be involved in all of them and it occurred to me there must be brilliant sort of cross fertilization of knowledge, by being able to work in one company, and then bring what you’ve learned to these different projects. So, from that perspective, as an organization, or you sort of looking great, this person has been involved in Avalanche, and Polkadot, Ethereum, and now they’re helping me in whatever my chain is, and there’s so much that these people can bring because they’ve sort of been subjected to these other projects as well. I think that is super powerful, because everyone that you know, every leader in an organization wants their team to constantly be learning and developing so they can bring innovation, but actually the DAO does that naturally because you can work in these different organizations very easily and I think that is super interesting, and it will drive innovation in these projects as well.

Sasa Milic  45:04

That touches on something I was gonna say in my previous comment then forgot, but another thing that aligns people is, it’s really kind of basic, its incentives, its financial incentives, like the API3 token omics are a bit complex. Now, I don’t want to get into that, but if you’re a DAO member, that means you’ve staked your tokens and your tokens are influenced by both the risk and the reward of the system. So, you’re financially aligned. It’s not like we’re getting you know, some sort of consultant that doesn’t really care about our project working on things. It’s like, if you’re a DAO member, that means your tokens are staked, and you’re earning both the profits, but you’re taking on the risk, so the system just too quickly, we have ensured data feeds. So, if our data feeds mess up, we pay out claims, and there’s a lot of technical details there, but I won’t get into but that’s what I mean by people that are API3 DAO members, they take on the risk and the reward of the system, they’re directly incentivized. So, that’s another nice thing. Because, again, I’m talking theoretically, and you’d know more about this, because you’ve actually worked with companies and management, but the direct financial incentive, I feel is going to alleviate a lot of that, like top down. What’s the word like, the things that companies have to do to force their employees to be good employees. Like, the companies don’t say this explicitly, but I mean, there’s things you have to do to actually make people care about the company.

Like one thing, like I worked at Facebook, and a common thing for tech companies is to get tech stocks. It’s actually very similar. It works for them for many financial and tax reasons, but it is also like, if a good chunk of my money is Facebook stock, you know, you think I mean, I kind of slacked off a bit. Like I said, I was reading about crypto work, the incentive didn’t work for me, but I imagine for a lot of people just that direct financial incentive. It’s similar to a DAO. The reason why I brought it up is after what you said about this really fertile knowledge transfer. I completely agree and it can also be even more, you can do that financially as well. Like there’s something that I’m seeing more of and something that we’ve been participating in. It’s called DAO to DAO. It has a fancy name, but it’s essentially like two DAOs, giving each other their tokens. So, is it like you agree and not immediately sell them? Like what’s cool about smart contracts is that you can actually lock tokens like my API3 tokens are locked for over three years, invested over three years. Do you know how to give people trust in the project? So yeah, DAOs can exchange tokens, and that immediately gives them an incentive alignment, like I mean, you should care about things, not just for monetary reasons, but it does give this kind of it’s nice having the monetary incentives as well. I mean, I knew there was some word for it.

Ben Sheppard  48:23

Yeah, now, I find it super interesting. My first startup, which was a law and advisory firm in Vietnam. When we launched that company, we wanted to do things very differently because myself and my business partner had come from very top-down corporate type structures. We just desperately wanted something different. So, we employed a bunch of staff in Vietnam and basically said, unlimited leave, work wherever you want flexible work in there is an office you don’t have to come their work wherever you want. It was like we tried to make it as flexible and as open as we are possibly good. We pay just above market rate salaries. So, it was good incentivization and the behaviors that we saw as a result of it was fascinating. So, the Vietnamese didn’t take any leave. They were scared to take leave because they thought it was a trick. They thought actually, it was a trick and if they went on leave, there wouldn’t be a job when they came back.

Sasa Milic  49:25

It makes me sad.

Ben Sheppard  49:26

In the Philippines, we also tried it and in the Philippines. They waited until I took leave and then they wanted to take leave and I was like no, you can take leave anytime. The British that we had employed, went on leave for three months, didn’t tell anyone and then came back. We’re like dude, where you have been, he was like, I was on leave. I was okay.

Sasa Milic  49:54

I think British people have a lot of vacation days. I mean, this is off topic, but I know when I travel, especially to like the kind of places around the Mediterranean, there’s Brits everywhere and like, yeah, we have like six weeks’ vacation or something. That’s true, right? That like six or seven, like actual, I think, all the holiday and everything.

Ben Sheppard  50:12

Yeah, I think it depends on the company. I wasn’t lucky to work in one with that company.

Sasa Milic  50:18

You guys do have better than two weeks, right?

Ben Sheppard  50:20

Yeah. I mean, Finland is fantastic. You gotta think six weeks here. So, yeah, it’s good, but I think where I was going with it was, I think different cultures will react to DAOs in different ways as well, and for me, culture is a really important thing. I’ve worked in a few different countries now and sort of just those habits and what’s ingrained to people in and how they perceive the rules that have been laid out in front of them can be interpreted in so many different ways. So, I think it will be interesting to see where DAOs are really successful, and where they’re not so successful and what was the reasons for that, and was some of that down to culture, and just sort of understanding the principles of what the DAO is, and also security. So, you know, a big thing when we had the companies in Southeast Asia was security, because the government’s not really there to provide security, things go wrong. So, having that employment contract and being part of a company and knowing that, okay, if things go badly, I’m at least got this to protect me. Whereas in perhaps more, well, Finland, for example, were very strong government gives you lots of support, if you’re out of work, you know, not so reliant on a good employment contract, happy to take risks working as an independent contractor going into different things. It’s a very different sort of playing field for those people in countries like that. So, I think it’d be very interesting to see the dynamics of where DAO has worked really well, the countries that take it on, and just how that evolves as well.

Sasa Milic  51:55

Yeah, that’s a really good point, but I guess one thing about DAO, is that it, I mean, it exists off chain, but it’s represented on chain, and that’s completely distributed. So, with API3, it is like our core team is all over the place. I think, East of me, for this East is India and Turkey, someone in South Africa, couple of Europeans all over Europe, a couple of Americans, one person in Venezuela. So, like I mean, maybe certain DAO will surely have like, overrepresented geographically one way or another, but yeah, I agree culture being an influence, but since DAOs, aren’t really geographically located. I feel like maybe some of that difference will kind of dissipate. If that makes sense.

Ben Sheppard  52:57

Yeah. I’m just curious to see how it sort of evolves.

Sasa Milic  53:02

But you do bring a really good point about culture. That’s actually one thing that I found so interesting about how much I’m learning about other people’s cultures through crypto just because crypto is so global. Yeah, there are the ways certain, even just the way certain tech gravitates towards some regions and not some other reason. Like I’m just from the Balkans. As I mentioned, I was born there and there’s this project Idina, have you ever heard of it?

Ben Sheppard  53:29

No. What’s Idina?

Sasa Milic  53:31

It’s really interesting. I haven’t read enough about the tech to say like, you know, to the point where I can show the project, but it is interesting. So, it’s the way you do these tasks. In order to mind tokens, you have to do these tasks that are called Ai hard. They’re things that AI apparently can’t solve. So, that means that if anyone is mining and they’re doing these tasks, it means they’re an individual. So, the thing with again, you know, this, I’m just saying this for the audience with a network, like Bitcoin, or Ethereum and proof of work, you don’t know if individual nodes are actually individual human beings. It’s just, you assume that like, no single person can have that much computer power. So, like, you have this assumption. It’s relatively decentralized, because there’s way too much compute power in the network to be represented by a couple of individuals, but Idina tries to make it so that every single node in the network is an actual human being, but off topic, the reason why I’m mentioning it is because I went to their telegram group and it’s like, 50% people from the Balkans and I was like, why did Idina? Like, I’m just so confused. Why did this project like there’s an Idina Balkan group too, and I’m just I even asked them one time I was like, why is everyone hear from the Balkans people are like, yeah, because we’re poor, but I’m like, I don’t think that’s a good enough explanation, but just one example of like one project, one technology gravitating towards one geographical reason, and I can’t think of a good logical reason for why.

Ben Sheppard  55:04

Yeah, well in Axie Infinity, the game has taken off in Southeast Asia as this sort of play to earn. You know, Filipinos and Vietnamese that have been out of work because of lockdowns have been playing these games, basically earning crypto by playing them. Now I think there’s some Filipinos are actually earning more playing the game than they were in their jobs prior to going into the lockdown. So, that particular game, for example, is really taken off there and it’s got a really strong following now. So, I think there’s definitely certain products in our sort of world of Web3 that are very well suited to certain geographies, and some of them are just global as well and open to everyone. So, Sasa, it’s been super having you on the pod today. I hope you’ve enjoyed being I’ve really enjoyed chatting to you.

Sasa Milic  55:54

Yeah, this is super fun. While we were recording, I was just having a fun conversation.

Ben Sheppard  56:01

Indeed, me too. Okay. Well, great. Thanks for coming on and I hope everyone out there listening has enjoyed the pod as well. Thanks very much.

Sasa Milic  56:10

Thanks, everyone.

56:13

If you enjoyed listening, don’t forget to rate, and subscribe to our podcasts and please share your feedback. Thanks for joining the TX podcast.