Money 3.0: Tom Kysar from Augur talks about prediction markets

In this episode of Abra Money 3.0, Abra founder and CEO Bill Barhydt is in conversation with Tom Kysar, the head of operations at Augur. Augur is a decentralized prediction market that’s built on Ethereum. Bill and Tom talk about some of the ways that people are using prediction markets, but the conversation also touches on the potential impacts of the technology that the team at Augur is building.

Check out the episode wherever you get podcasts or find a full transcript of the conversation below.

Bill Barhydt: Hello everyone, Bill Barhydt here, welcome to another exciting episode of Abra’s Money 3.0. So with me today I have Tom from Augur. Tom, you there?

Tom Kysar: Yes, hey, thanks for having me.

Bill Barhydt: Well, welcome to Money 3.0. I’d love it if you could just give us some background on yourself, and how you got into crypto, and then I really want to get into this whole idea of a decentralized prediction market. It’s something that I think our audience is going to be super excited about. So let’s go through this methodically, because I think there’s a lot to break down here. So tell us how you got into crypto, and your role at Augur.

Tom Kysar: Yeah, absolutely. Maybe five or six years ago, I was studying computer science in college for a year and had discovered bitcoin. It ended up taking up the majority of my time over studies as I was learning more stuff about bitcoin and playing with it, and trying to figure out what I can do with it and what other things I could do with it. And long story short, I ended up finding a company in Silicon Valley maybe four-and-a-half, five years ago, that had just raised a very small, seed amount of money for wanting to do something in the crypto crowdfunding space. And thankfully, I got a job offer to go move out there and work on that about five … yeah, it was about five years ago now. And it became something called Coinify, which was a crowdfunding platform prior to Ethereum and a lot of those token sale-type booms.

Tom Kysar: So we did a couple of early crowd sales, one of them being Factum, I think that’s pretty much the only one that’s still around. But through that, I ended up meeting Joey and Jack because Augur was one of the earlier cryptocurrency crowd sale projects that were potentially in the pipeline, and that was kind of the stuff we were working on. So we ended up not actually doing the Augur crowd sale for Augur under Coinify, but I had met Joey and Jack, stayed in touch with Joey for a little bit, and I don’t know, maybe about a year and a half later Joey gave me the opportunity to come to join him over at the Forecast Foundation, and I’ve been there for about … just about three years now. And director of operations, pretty much one of the few engineers on the … one of the few non-engineers on the team, so anything non-engineering, more or less, is stuff that me and another guy, Peter, who works with us, handle.

Tom Kysar: So it’s been a fun ride. Very interesting. But yeah, through the kind of San Francisco crypto network, I got my first job out there and made my way to Augur and pretty much have never left.

Bill Barhydt: Wow, fantastic. So what is a prediction market, and what is Augur, and what is Augur’s vision for prediction markets?

Tom Kysar: Absolutely, yeah. So Augur is … on a technical level, Augur is a peer-to-peer prediction market, and oracle network, a decentralized peer-to-peer prediction market, and an oracle network. Now what that actually means, I guess in simple terms, is these are no limit, censorship-resistant markets. Any market that has a definitive event or definitive outcome can be a market that is traded. And people trade these markets with their confidence of the likelihood of whatever outcome happening is. So you’re essentially trading event outcomes. And it’s kind of parallel to how a lot of traditional financial markets work today, but it’s very similar to like traditional betting exchanges and betting platforms that we have today. But the difference between those and what Augur can provide today is that Augur is a set of 90 contracts that live on the Ethereum blockchain, and you get all the inherent kind of properties that came along with bitcoin, but now for markets of any type. These are limitless, censorship-resistant, anybody can access them. Nobody can modify them, nobody has any sort of extra proprietary power or control over the software than anybody else participating in the system.

Bill Barhydt: As a prediction market, the idea is that I can basically say I want to facilitate or participate in some type of bet on the outcome of an event, right? So let’s say the presidential … the upcoming primaries, let’s say, the Democratic primary leading up to the presidential election. There are 15 candidates, and I basically want to facilitate a prediction or betting process to see who is going to win the nomination, who might come in second, who might come in third. And I can basically then open up a new type of prediction … I don’t know if you call it a contract, to determine what the market thinks about who is going to win the nomination. Would that be a good example?

Tom Kysar: Yes, absolutely. So you could either come and create your own market on your own terms for that event. Or you can go and trade on somebody else’s market that, presuming there is a market that meets what you’re trying to trade, but yeah, there are a few markets on Augur today surrounding some of the upcoming political events. And yeah, you can buy a share of whatever you think the outcome winner is going to be. Say you buy it at 60 cents, or 60% in this case, if that outcome ends up being the correct outcome, those shares will be redeemable to one. So the price at which you pay for the share is inherently the price that you also believe is like the probabilistic chance of that event outcome being the winning one or the occurring one.

Bill Barhydt: Right, exactly, so you also mentioned this idea of a decentralized oracle, so in the example you gave, you have to determine either … well, in my example, who won the primary, or in your example the price of the share for another example. So you need some external data source that everybody agrees is valid at the beginning of the bet, I would assume, to determine who wins the bet based upon some external data. I assume that’s what you meant by a decentralized oracle, in this case.

Tom Kysar: Yeah, absolutely. So every market has … the market creator is allowed to designate either a specific resolution source, which would be saying something along the lines of, “I want this baseball game to resolve to whatever ESPN says is the score of this baseball game.” Or alternatively, you can make your resolution source general knowledge, in which you are not specifying a specific place you want this information pulled from. You want the kind of general consensus.

Tom Kysar: And so a market creator specifies which one of the two ways they would like that market to resolve. There are kind of pros and cons of each different way, depending on what you’re going for. But then that gets into essentially the second part of Augur. Augur’s really a decentralized exchange on the first half of it, and then the second half of it, that majority of traders and people using Augur, in the long run, will presumably never really see or touch is the oracle system and the distributed reporting. And those are people who hold the REP token that we created. They essentially have a financial incentive to keep that oracle always honest, and/or reflecting reality. Reflecting, presumably, what traders are expecting those outcomes to be. As the REP reporting token holders essentially work for the protocol and they work for the traders. They have their incentives aligned for traders to continue to trade on Augur, they will continue to Report on those markets, and extract the fees from the market for doing their work there.

Tom Kysar: But it’s essentially a big kind of distributed backstop reporting game, where it’s designed in a way that the most profitable move should always be being honest, versus attempting to collude.

Bill Barhydt: Right, so the Litmus test is something decentralized to me has always been is there a central off switch? Meaning can I as the FBI walk into a room, shut off the lights, shut off the power, and the system is now off. So what you’re saying is, is that in theory, you can’t shut Augur off because even with the oracle, even with the ESPN score source, there’s no place for that FBI to go in this case to say, “Okay, here’s where I shut the oracle down, or I shut the prediction market down.” Is that accurate?

Tom Kysar: Correct, yeah. So the logic for the trading network, for the trading and matching exchange, and the logic for the oracle network exists on-chain, it’s all on-chain on Ethereum. And then there are front ends that are used to interact with Augur, but those are primarily used and shared upon using IPFS today. So the client UI isn’t really actually hosted on a … we absolutely don’t host or serve any of it, but it seems that most people have been accessing it through a gateway in their browser.

Tom Kysar: But yeah, everything is hosted in … quote, unquote, “hosted”, but everything is served in a way where there’s no central point of failure, the code’s freely available on GitHub, you can download it and always run it as a desktop client if you would like as well. But it’s free open-source code, and anybody can access it, and all the parts, the entire system of Augur is designed in a way where yeah, it views every participant equally, and the Forecast Foundation, we have no sort of either control over what’s in these contracts currently deployed, we can’t update them, we can’t modify them, we can’t stop them. We have no escape hatch or like a centralized kill switch key like a lot of some other projects do. It’s out there, and we can’t even modify the contracts if we want to. So Augur v.2 is going to be an entirely new contract deployment with a manual migration over to it.

Tom Kysar: But yeah, there’s nothing to shut down if you would like to shut Augur down, you’d essentially have to shut Ethereum down with it.

Bill Barhydt: Yep, yep. What are the kind of contracts that are most popular today? If you look at all of the different prediction markets that are out there, or the predictions within Augur, which ones are getting the most attention?

Tom Kysar: Yeah, absolutely. In the short term right now, so political betting markets definitely get some interest, I think particularly due to the … the other places where you can bet on similar types of political markets have … like PredictIt has an $850 bet limit, and withdrawal fees, and etc. And a lot of these offshore sports exchange and sportsbooks don’t trade political markets. So we had the who will win the House vote market at some point last year that is still to-date the largest trading market on Augur. I think it settled with like a million and a half dollars in open interest and three of four million in total volume. But the Canadian election market two weeks ago got good volume. I think Augur matched like $125,000 in bets, and Betfair matched like 20k in bets on the same market. So political ones have been showing recently to definitely be something of interest.

Tom Kysar: Same with sports betting. Particularly just because sports betting is currently today, the operational process of actually sports betting is very inefficient for a lot of the sports betters. Whether that’s because they don’t have an exchange or a place that they can bet based on where they live, so then they’ll go to offshore exchanges, it’s like a 50/50 crapshoot if they’ll actually send you your money back with a withdraw, and there’s no protections, no liabilities, they charge very big fees, they charge very big spreads. So people have been using the NFL markets and some of these other sports markets have been getting … not anything crazy in terms of volume, but the consistent trading volume of decent and sizable bets that people are actually using it for.

Tom Kysar: And I think lastly, some of the things we were … at least I was more surprised in but then ended up being super interesting is like these … people have been kind of little niche markets, and they don’t do crazy amounts of volume, but they’re cool because they’re markets that we really haven’t seen before. Like we had a guy who wanted to make … this was before Augur had launched, but he wrote out a big proposal, he wanted to make a market on where the next Amazon headquarters is going to be, because he was looking to purchase a home in one of the potential locations and wanted an avenue to potentially hedge against housing pricing going up if he was going to go live in one of these places.

Tom Kysar: So it was like kind of a niche little use case like people have been using … they’ll be bounties sometimes for teams, which is really cool. So like you want an open-source project to implement some feature, you could create a market on Augur that integrates support for so-and-so wallet by a certain date? And put 2,000 bucks on it. And in theory, the opposing team could purchase those shares, make that event happen, and it’s kind of being used as like a bounty funding mechanism. So these ones don’t do big amounts in volume or anything like that, but I think it highlights and illustrates that people are going to find ways to use this technology and these markets in ways that we definitely haven’t thought of or can think of right now.

Bill Barhydt: Yep. Given that it’s decentralized, how do you prevent… let’s say in air quotes nefarious usage of this with things that may be unethical, like let’s say somebody wanted to create a prediction market for who’s going to assassinate the president, or on what day will the president assassinated? That’s probably something that the … I would assume that the creators of Augur probably didn’t want out there. So how do you police something like that in a decentralized environment? Or do you just throw up your arms and say, “We can’t police it.”

Tom Kysar: Well, on a fundamental level, we can’t police it. On a protocol level, we can’t police anything. And that’s kind of some of the trade-off of a truly decentralized technology. When you’re creating anything new, there’s going to always be some level of inherent … bad actors, and people using it for nefarious reasons with anything that’s created. I think at least in terms of Augur and some of these markets, it’s something that at a protocol level, it’s a human curation problem — like you can’t have code enforce what ethics is because it’s all very subjective depending on who you speak with. Obviously there are clear examples of what would be yes and no, or acceptable and not acceptable if somebody was policing it, but still, it’s very difficult to put that sort of thing into protocol-level code that you then go and deploy and that’s the part of the logic that lives on Ethereum.

Tom Kysar: But in terms of actually combating it being a problem in the real world, I mean, none of these markets I think will actually ever be traded on. I mean, I think we’ve seen that cryptocurrency isn’t as anonymous as everybody would like to think it is if you use it in the default manner. But yeah, this provides market-creating, you can create a market, anybody can create a market for a couple of dollars. So with that will necessarily come with some bad actors doing some bad things, but I think one, as more companies start coming in and building for profits on top of Augur, which we’re starting to see, they will all have their own kind of curation methods, and their own ways of highlighting what is good and what is bad on Augur. Because outside of just like nefarious markets, there’s also nothing stopping somebody from creating 20 of the same markets if you want. And currently in the UI, it’s weighted by volume, so ideally you will see the ones that actually have the trading volume, those would be the ones that people trade on, and the 19 other ones will just kind of be dead markets.

Tom Kysar: But there’s no way to actually control or police that on a protocol level. So UIs coming out and people that will be providing access to this protocol I think will play a large role in what types of markets are curated, how they’re curated, and how you essentially kind of … it’s like how do you score the quality of this market in a distributed network in a way where you’re confident that it’s a valid market, it’s the one that has the most trading volume. You want to give … there’s always going to be a level of human curation that will make the experience better, and I think we’re going to… come v.2 there are a couple more projects that will hopefully be opening up their doors. But we have things like Guesser now today that are doing it.

Tom Kysar: And so yeah, I think in the long run it’ll kind of be like self-solved in the sense of that majority of people who probably use Augur are going to go through some person that’s built a for-profit relayer on top of it, or they’re going … there’s going to be some other interface channel that the user is using Augur with where they’re only going to be shown like the top hundred markets for whatever they want to look at or whatnot, and hopefully the decentralized UI can get to that level at some point, too. But yeah, that would kind of be the general take on that.

Bill Barhydt: Are there applications of Augur that you think are going to be really exciting but just haven’t taken off yet? Meaning I’m just kind of surprised that this type of market isn’t there yet, or that type of market isn’t there yet?

Tom Kysar: Well so in the early days, we saw a lot of markets on like … essentially just cryptocurrency speculation.

Bill Barhydt: Right.

Tom Kysar: But some of the largest markets were just will ETH close at a certain price by the end of the month, or the end of the year, or some altcoin they were doing with REP as well. And those seemed to take off originally, and then have started to kind of slowly fade away. And I would imagine it’s … it was one of the initial interesting things for crypto people to bet on, and nowadays too I think we’re attempting to attract a somewhat different audience in the realm of more traditional betters, and people who are more familiar with betting, and try to get them into the crypto ecosystem, versus trying to convince people in the crypto ecosystem that they’re betters.

Tom Kysar: But, so I think some of these financial and speculation markets, in the long run, will probably be quite big. However, in the short term, like right now we have like a cutoff date for Augur v.2 as well, there’s kind of some bumps as we’re in the growing process. But once this kind of matures a bit, I think some of these financial speculation markets will end up taking off based on the way that you can do … you can set up pretty fancy and sophisticated scaler markets that give you essentially synthetic leverage on your position without necessarily putting up the additional capital. There’s a bunch of I think interesting things that could be done in the realm of just like … not even cryptocurrency, but kind of just general market speculation, and derivatives based on that and stuff like that. So I think that’ll come when Augur matures a bit more, but I definitely think that’s going to be a large use case in the future.

Bill Barhydt: Do you have any way of knowing how many people have used Augur? Meaning I’ve actually bought or sold on a market?

Tom Kysar: Yeah, well, we have unique Ethereum addresses, which isn’t the best way, but it’s a relatively decent proxy. I believe it’s in the ballpark of three to four thousand unique people who have placed bets on Augur over the past year. Yeah, it’s like three to four thousand. A large portion of … Augur v.1 was essentially: Look, we need to get this deployed, we need to prove that this works, that this whole kind of incentive system, reporting, resolving markets, that this all actually works. And I think we did, in a sense, at least to ourselves. We’re confident at least now that this entire system can hold up. And come v.2, and come in the future, now I think it’s a lot a matter of fixing that experience, polishing up that experience, and getting this product out in front of more users.

Bill Barhydt: Why create a REP token? What purpose does the token actually serve versus Ethereum itself in the system?

Tom Kysar: Absolutely. Yeah, so one of the reasons, I would like to think that Augur’s one of the few projects that can answer the question of why do you need a token quite well. The fundamental thing that protects Augur’s oracle, and the reporting system, is that in the event of two different sides really disagreeing on the outcome of an event, Augur can fork. And this isn’t a fork in the sense of Ethereum forking, this is in a sense a direct token can fork itself. So reporting has rounds, and essentially people can crowdfund a dispute bond, if that bond fills then it switches outcome and it allows the other side to potentially post a dispute bond and it can keep going all the way up and the sizes get larger and the bond sizes get larger.

Tom Kysar: And at the final dispute round, I believe it’s right now, by design, it’s somewhere in the ballpark of about two and a half percent of the total REP supply for the final round. The final dispute round will be a forking round in Augur. Meaning if two sides sufficiently disagree with what the outcome of the event was and stake enough capital asserting that, the REP token will fork into two separate universes. And essentially there will be REP token one where that exists in the Augur universe that says, say, Hillary Clinton won the presidential election. And then there’s Augur two which is the outcome that says Donald Trump won the election. And if you’re a REP token holder, come v.2 you’re going to be forced to actually participate in these network-wide forks.

Tom Kysar: But you have to take your REP and decide which universe you want to exist in. Do you want your REP to exist in the Hillary Clinton universe, or the Donald Trump universe, and you’re forced to migrate your REP to one of those two individual worlds. And this was kind of the underlying principle of the oracle that this is one of the main things that different implementations of people working on a decentralized oracle change and play around with. Like a master oracle is nearly the identical oracle design of Augur’s reporting system. However, the difference is is that in that event of a fork, they would fork Ether. Or they would fork Ethereum, versus forking their own token, they wouldn’t use their own token as a part of the reporting system.

Tom Kysar: So one, we needed to get REP out into … we needed to have some value out in the hands of people that wanted to participate in Augur in the long run, in order to … and just know that we have participants that will hold this token and actually come in and do this work weekly. And then additionally on top of that, it’s one of the fundamental underlying security properties is the fact that the protocol has the network to fork the token. And without that, it’s like … the oracle can’t ever define itself what is true and what is not true. And if two people really don’t agree and they stake enough money to say so, the protocol essentially allows them to each go exist in their own universe, and everyone else gets to decide where they want to exist and participate within it as well. So-

Bill Barhydt: Has such a fork ever happened? Or happened yet within the REP token?

Tom Kysar: No, no, and it would be a … it would be a very large event in Augur’s history if that were to happen.

Bill Barhydt: Sure, right.

Tom Kysar: The thresholds are quite high and whatnot, but it is there, and it can happen. So the threat of that happening is generally enough to keep the oracle honest for the vast majority of the time.

Bill Barhydt: Yep. And so basically, if I understand you correctly, what you’re saying is is that in theory, you could use Ether, or any other asset, digital asset, as the incentive mechanism for the oracle function, but the fact that you have this forking feature that needs to be unique to your token is what actually makes the token necessary in order to basically have a decentralized oracle, as opposed to a centralized party for dispute resolutions. Is that accurate?

Tom Kysar: Yeah. Rep doubles as a way for us to … as a way for the protocol to transmit value from essentially market creators and traders to reporters. So it is used as like a means of transferring value. I think that a token that is just used for the means of transferring value is a pretty low-value proposition for that token alone. So yeah, the fact that the protocol, the Augur protocol, has access to the underlying functionality of that token or asset that the players have a financial stake in, because that allows the protocol to modify it or change it, in this event, force a fork.

Tom Kysar: And yeah, that’s kind of our REP thesis.

Bill Barhydt: Do you see … well, first of all, I’ve been thinking a lot about this whole idea of decentralized oracles. I mean, in any kind of smart contract, it seems like the trust issue is always going to be there, right, so … the idea of not having to trust a third party oracle to determine the outcome of a smart contract is very, very compelling. But as you’ve demonstrated with this necessity to fork in the case of a dispute is very complex. Do you foresee a model where your oracle system could be used outside of Augur, or outside of your own prediction markets? Meaning let’s say somebody was building their own Ethereum based contracts, or EOS for that matter, based contracts, could I use your decentralized oracle model in my own contracts, or would I have to basically take the code and re-implement it in my own contract system?

Tom Kysar: There’s a yes and a no answer to that. So like you could, in theory, take the outcome from events that resolve on Augur, and use it for your own project, or some other market that you have to use that as the outcome to resolve it. The issue with this is something we call, it’s the parasite attack, which is a nearly, relatively unsolvable attack. Which is in the sense of that if there’s a third party that is using Augur’s resolutions to resolve their own amount of value outside of the Augur system, if that amount of value outside of the Augur system is larger than the equal amount of value that is locked in the Augur system for that same outcome that they’re using, there now becomes a potential profitable attack on the oracle, which would be buying the majority of the REP tokens and forcing a fork, and whatnot. But it might be profitable in that sense.

Tom Kysar: So it’s in the Augur white paper too, it’s the parasite attack. Essentially, if someone parasites the outcomes out of Augur without participating in the market and paying reporting fees in the market, if that ever grows too large, there’s a chance for both that external market and Augur to both kind of implode and kill themselves. Because it could, in theory, be profitable to attack Augur to get the value on the third party or outside exchange paid out to them in a certain way, and it screws the incentives, and potentially can kill the Augur oracle, which would also thus kill the third party market that is now trading outside of Augur as well.

Tom Kysar: So it’s kind of like the suicide bomber attack on Augur if somebody wants to come in and do that. There’s a way to solve that, which is just participating in the Augur market for an equal amount of open interest just to pay reporting fees. And so you don’t even need to participate in the market, you just need to hold a complete set of shares in the market and participate in the market for an equal amount of value that you’re going to resolve outside of the market. And then, in theory, that’s just kind of plugging into the Augur markets, and you could do that with external chains as well. But anyone who wants to potentially use a decentralized oracle is going to have to decide if they want to participate in that oracle’s system that makes that oracle secure, and they realize that if you parasite from that oracle, that you are deeming is secure, and you’re parasite volume grows to a level larger than the volume in Augur, you can inherently kill the secureness of the Augur oracle, as well as the integrity of your own thing that you’re trading outside of there.

Tom Kysar: And that’s the whole kind of Augur security model, is that Augur is secure under certain conditions. And the conditions right now are like the market cap of REP needs to be … I believe it’s five times larger than the current open interest escrowed in Augur. Essentially it’s as long as an attack on the Augur oracle is unprofitable, we deem it secure. But there is a potential world where an attack on the Augur oracle could be profitable, and then we would deem, at that point, the Augur oracle is non-secure. If somebody would presumably be attacking it, and then we would eventually get to the point of a fork, fork in the REP token like we talked about earlier. And in the event somebody actually did do a hostile attack and attacked the oracle and even if it was profitable, the token would still fork in two different universes, and hopefully, REP holders would be able to realize that there was just a hostile takeover and still migrate to not the attacker’s universe.

Tom Kysar: So and that’s kind of another reason for the whole forking thing, is that there’s always optionality to not agree with what somebody else says on Augur at the end of the day, if it gets to the final step.

Bill Barhydt: Gotcha. I’m thinking a lot about collateralized stablecoins, ala Dai, that rely on collateral priced in dollars in order to create a stablecoin. And ultimately, they do have to trust an outside source for determining how much collateral needs to go in that contract to create that Dai-like stablecoin. I’m wondering if there’s a great overlap between what you’re doing and how to basically solve that trust model inherent in those collateral stablecoins that actually does potentially create an off switch, right? Because if you can’t determine to some degree of trust what the price is, of whatever ether or bitcoin or whatever you’re collateralizing the stablecoin in, then you have a failed system. Is that true? And then could you solve that problem here?

Tom Kysar: Absolutely. So yeah, that train of thought and way of thinking is valid and absolutely correct. The one, I think, hurdle that I’ve spoken with one or two people roughly about the idea, and I think the hurdle that we’ve faced with this such far is that … and it gets better in v.2, so in Augur v.1 right now, markets take … the reporting cycles are a week. And at best right now, a market will resolve in like five days if undisputed. Come v.2, that’s going to get to like 24 to 36 hours. So within 24 to 36 hours, a dispute round will have been completed and you’ll know either if this market’s being disputed or if the outcome was accepted as truth.

Tom Kysar: So I think the issue with some of that stuff is that there is a need for a more realtime data feed of some of this pricing. And that’s something that Augur is not necessarily great for right now, which is fast reporting. Because essentially, if somebody says what the outcome is on Augur, and then we give a 24 hour period for anybody who holds REP to come in and stake some REP and say, “No, that’s wrong,” and if they’re correct, they will make money. But we still have that natural period of where we have a distributed network of reporters that need to come into some software and validate claims, essentially. And there’s always a financial incentive to validate that claim in a way that you believe is going to best reflect reality.

Tom Kysar: But so in the event of say you’re using the Augur reporting system to attempt to provide a price feed for what bitcoin is, because you want to value what the collateral is, if that price ends up being incorrect for some reason, or if somebody disputes that price, you have a 24-hour period now where you have to like field the real information in. So the hurdle that we’ve always kind of got caught up on is how would you get that in realtime? And actually, Augur faces this problem inherently within the protocol itself. The Augur protocol needs to know what the outside USD market value or REP is in order to properly value reporting fees being paid to the reporter.

Tom Kysar: So in Augur v.1, this was a multi-sig. Very similar to what like Maker does today for their price feeds as well. There was ironically enough a bug in Augur v.1 at the protocol level that made reporting fees almost … no Augur needed, at least at this level and what the Augur volume is doing. But, come v.2, we explored a bunch of ideas on how would you get a somewhat … or how would you get a distributed price feed in a somewhat realtime type way? And I believe what we’ve decided on for v.2 is using Uniswap’s v.2 new deployment. They’re going to have a new deployment where they allow access to that realtime market value, slightly easier than the way that they’re doing it right now, and some other stuff.

Tom Kysar: But our leading thought for that was to use Uniswap, and I believe that’s what’s going to be powering the REP price feed come Augur v.2. Because that’s kind of the one … the thought years ago was always, “We’ll use DEX price feeds. We’ll be able to get price feeds on-chain from something like Ether Delta,” or whatnot. But I think at this point it’s been somewhat proven that these DEX’s don’t have sufficient liquidity, at least to a level where you would want to confidently take a price feed from it to price some other asset, just because they’re just not mature enough books to where they … they could be very easily manipulated and whatnot. But Uniswap’s quite good. So I believe that’s actually our solution, but yeah, there are quite a bit of teams that are working with the how do we get a price feed in realtime type problem. And I’m pretty sure Maker still today is running, it’s just a fancy multi-sig setup feeding in a bunch of the prices.

Bill Barhydt: Yeah, that’s right.

Tom Kysar: Yeah, pretty much anybody I think who is going to be dealing with DAPS this year faces that in one way or another. Most of these things need a price feed, for whatever asset it is. And there’s still not a great way to do that, and I would still think there is stuff to explore in the realm of something like having a Uniswap price feed for your realtime data feed, and then default, and then using Augur as a backstop to that. So essentially take Uniswap’s decentralized real-time price feed, and if anybody really has a problem with it, kick it over to Augur and let Augur use the distributed reporting setup to sufficiently assert what has happened here.

Tom Kysar: But I don’t know how that would work in the realm of who’s using the price, and what you would do in the meantime when those claims are being either validated or invalidated. But-

Bill Barhydt: Right.

Tom Kysar: Yeah, super interesting topic.

Bill Barhydt: Yeah, lining incentives there for bad actors is very, very hard, because you could easily see a model where people would lock up other people’s funds to extract some portion of their losses back, right? I mean, it sounds a little scary, but I think it’s like the Holy Grail of smart contracts to me is exactly that, right, we talk about that all the time at Abra.

Bill Barhydt: So that’s a good segue into what you think is next here. What are your predictions for what happens next with the Augur protocol, with REP, with the Augur network in general, and how do you see 2020 evolving for you guys?

Tom Kysar: Yeah, absolutely. I think it’s going to be super exciting. We’re at the point of development where it’s like you wake up every day and get in the chat group and catch up with everybody from what happened over the night, and it’s like every day at Augur just gets continuously more exciting now. Because I think we’ve kind of at least pinpointed what our kind of clear value props today to some direct audiences that we want to target. We’re confident now that the system can work, not it’s kind of how do you make the best actual product and start onboarding users to not just Augur alone, but if you’re going to onboard somebody to Augur, they’re onboarded to the entire crypto ecosystem. And there are many parallels that you need. To be onboarded to Augur you’re going to have to presumably buy crypto, whether that’s through wire or whatnot, whatever … but you’re going to be introducing a whole new user into crypto. But come v.2, it’s entirely … most of our development and focus is entirely around the product itself, the user experience and the product itself, and then user acquisition.

Tom Kysar: So Augur v.2 is actually going to have two new UIs being released. One is essentially a revamped trading UI, which relatively has the same look as the current UI. The UI is actually much better, but the experience of trading is still very similar, it’s much more familiar to trading on GDAX. And then additionally, we’re going to be releasing a new betting UI, which is something we haven’t had before, which is a UI that looks much more like Betfair with back and lay and odds, and you can toggle odds to whatever odds you’re familiar with.

Tom Kysar: So I thin come v.2, it’s going to be very heavily focused around building a good and smooth product that we can actually onboard users to. And just essentially continuing to make the protocol hardened to a level where there are people using us, there’s a community, an ecosystem that cares about it, we have users, people are coming in, they’re resolving markets. But I think, generally speaking, we have out v.2 deployment, Maker has its v.2 deployment coming up next month as well. Uniswap has a v.2 deployment. I think, generally speaking, a lot of the industry is slowly but starting to begin to actually mature. At least on protocol levels, too, I think a lot of us have kind of hardened where we live on the spectrum of like what are you building on-chain, and what is your actual protocol? So I think the protocols are starting to really mature, and now what this next year I think for a lot of us is going to be the maturity of just the experience, and getting the experience of using these applications to an equal, if not better level than current web apps today.

Tom Kysar: And I do think it can be done to an extent and level, and there’s always going to be some limitations on dealing with the contracts versus a central server somewhere. But even with 0x trading, it gets super quick. We’ve segmented the trading engine from the contracts in a way that you could essentially … we could deploy a new trading engine if we would like, and then Augur users can decide to point to the new trading engine. So after solely Augur v.2 launch, we want to deploy some sort of off chair scaling trading contracts, which in theory can get trades down to a sub a second. And when you start removing some of these limitations, a lot of those limitations were a lot of the cause for just terrible user experience I think across in cryptocurrency. So the like middleware and infrastructure tools for all this stuff is getting a lot better. Like we can sync the part of the node in a browser through Tor if you wanted to, peer to peer, like much faster and we know how to do that now, and Ether PC handles it.

Tom Kysar: There’s a lot of these inherent problems that in the early days of crypto I think everyone focused on making sure their protocol was okay, which is probably the right way to go about it, too.

Bill Barhydt: Yeah, right.

Tom Kysar: That’s a part we can’t change, or edit, or modify or anything like that. And user experience I think was always kind of just … I don’t think anyone wrote it off, but I don’t think it was optimized for, at least for a lot of projects in their early days. And I think everyone’s kind of getting there now, and I do think by the end of next year we’re going to have … I hope Augur can be one of these, but I definitely think we’re going to have a handful of applications where it is a truly decentralized application running on Ethereum, and the web experience is identical to you nearly using Twitter or Gmail, or something like that.

Tom Kysar: So I think it’s going to be a very exciting time for people who are building all this stuff.

Bill Barhydt: That’s awesome. Why don’t we stop there, this has been fascinating. So where can people go to find out more information and learn more about the details and innards of Augur?

Tom Kysar: Yeah, Augur.net, Twitter @AugurProject, those are our two pretty much primary communication channels. We have a Discord, you can go to Augur.net or invite.Augur.net and that’s where all of the people who work for the foundation, all the engineers, and the whole kind of little Augur community exists. So it’s a pretty good place to come ask us questions or talk about any of this stuff if anybody would ever like to. But yeah, come find us, Augur.net. Hopefully aiming for January 2020 you should see the v.2 launch, so it’ll be exciting.

Bill Barhydt: Fantastic. Well, this has been a really interesting discussion, I’m a huge fan of the Augur concept. Thank you so much for your time. I hope people will dig in and you can actually deposit, withdraw, exchange Augur REP tokens in the Abra system, and so we hope you’ll use Abra for that. And thank you again for your time, I really enjoyed it.

Tom Kysar: Thanks for having me, this was great, I really appreciate it guys.

Bill Barhydt: Yeah, thank you. And so that’s it for another episode of Money 3.0. Thanks for joining us.

 

---