Joey Krug Explains the Augur Prediction Market

Joey Krug Explains the Augur Prediction Market

Augur is a decentralized prediction marketplace bulit with Ethereum, and sidechains. The Augur interface will match bettors to oracles, and incentivize its users ...


What's up party people? Chris DeRose here, community director of The Counter Party Foundation. And I'm here at the Inside Bitcoin's Conference 2015 in New York City. And I'm here with Joey of the Augur project. Hi, thanks for having me on. So, I've been watching this project. I know a little bit about it.

It's a prediction market. I think it's built entirely on the BItcoin blockchain. Or are you using some Ethereum parts? Tell us about that. Yeah. So, we're using Ethereum. So what we do is we have smart contracts on Ethereum.

And these contracts alow you to create any sort of predictive market place. And you earn trading fees from markets you create. And the way Bitcoin comes in is it will be sidechained to the Bitcoin blockchain. So, you'll be able to use Bitcoin on these markets instead of just, say, Ether. Gotcha. So, when you say sidechained to the Bitcoin blockchain, how does that work? You're going to be using the actual block streams sidechain project to create a Bitcoin sidechain that is in one way or another controlled by Ethereum contracts? Does that sound about right? Yeah.

So, there's a few ways to create sidechains. There's two main ones. There is federated pegs, which they only have a few people using them, so they're pretty centralized, which we don't want. And there's the block-stream two-way peg. And you can actually do a two-way peg in Ethereum pretty easily. Probably even easier than using one with Bitcoin, because in Ethereum you can write the contract that allows you to do these sidechain proofs, and it's pretty simple.

Like, you could write in a few hours as soon as Bitcoin has made their soft-fork to support sidechains. If that comes out sometime soon, we'll do that. If not, I have some ideas for a, basically, a distributed federated peg. Or instead of five people, you'd have a few thousand verifying transactions. And I'll be making a blog post about that soon. Interesting.

What is your blog? You're all handy, is it easy? Yeah. So, it's Okay, that was much easier than I expected. Actually, no.

No, it's not. I'm wrong. Okay, here we go. I knew it's harder. It's .

It's pretty easy, but... Only by a little bit. So, tell me a little bit about how development is proceeding then. How many languages are you using for Augur? Tell me about that.

So, we're using this language called Serpent. And it's a smart contracting language. It's pretty similar to Python, except it has less features. So everybody at Counterparty's kind of very familiar with Serpents. We at Counterparty know Serpent very well, so, that makes sense. Are you using Serpent exclusively, or are do you not need some GUI works, or interface work, command line stuff? So, our GUI is in JavaScript, and it's written by our UI developer named Scott.

And it uses the latest the framework which is like, React and Flux. Cool. And how many people are on the team right now? Is it just you and he or is there others? So, there's six of us. There's four dev's, a marketing guy, and a business development guy. Got you. And how did the idea kind of come about? How long have you been working on Augur for what have you been thinking about it? What led to you finally committing to this full-time.

So, the base idea is a project that came about by guy named Paul Sztorc. He published a white paper, but he didn't have any intentions of implementing it himself. And one of my friends basically called me up and told me about this. And he'd had some ideas in the past that weren't that interesting, so I kinda ignored it. But then, I read into it about three days later, and there's about 100 pages or so of documentation by this guy name Paul. And I was immediately enveloped, because I realized that prediction markets are like a great tool that you can use to forecast pretty much anything with more accuracy than pretty much any forecasting tool that's ever existed.

So, when I realized that we could actually build it and implement it using blockchain tech, I had to do it. Okay. You had a calling. How do the oracles work in this environment? How do we know what the truth is for any given forecast? There's two ways you can do this. One is you just trust a third party, which we obviously don't want to do, because if you did that, you'd defeat the purpose of blockchain technology. So, what we do instead is we have people reporting on the outcomes of events.

And your report is weighted by a thing called reputation. And so, the more reputation you have, and the more your reports weighted, and people who report are rewarded with trading fees in the markets. So, I'd mentioned earlier, that the trading fees go to the market creator. Half of them go to him or her, and the other half go to the reporters. And so these reporters decide what happened. Got you.

And then all these will be done through the react JavaScript interface, and that will relay through a running Ethereum node then, or does it go into one of your federation servers? Explain how that part of the process will work. So, there's no federation servers. So, what happens is you'll install the UI with basically a double click install. And in the background it'll run an Ethereum node. The end users do not need to know about the Ethereum node. It's just running in the background.

And it will basically, communicate with the node, and you'll input things to the UI, and it will send it to the blockchain. And then, it will respond back with information from the chain. And since Ethereum's block times are 12 seconds, it's pretty quick for blockchain technology versus say, 10 minutes. Let's talk about Ethereum a little bit. Now, the Ethereum project, it seems like it's taking a little while to come out, it's sort of neutering some features. Does that concern you in terms of tying your projects to that, or are you just in no rush really to get yourself out? So, there's a few things.

So, they have been taking a while to come out, but their latest version has said, basically, they're going to release when their GitHub issues are at zero left, and right now is 11, and that's down from like 150. So, they're making progress, and they should be releasing in the next month. And then, about the features, well, what we aim to do is, we aim to basically, launch Augur as an alpha with fake money on Ethereum, and do that for a few months, then move to a Beta, do that for a few months, and then, we'll launch with real money in the fall. So, by then, Ethereum at a really good position, because they aim to have their first released version in the next month. And then after that, it can only improve from there. Okay.

What made you choose Ethereum say, over obviously, I like the Counterparty Project? Was it the 12 second block times? Was that what the clinching issue for you, or is the platform more? Tell us more about that decision. Obviously, we like Counterparty. We also like to know why you're preferring Ethereum over Counterparty. So, there's a few things. One's, block times are a huge boost. Two, some of the features that we wanted to do from our contracts on Counterparty.

.. I talked to some of the dev's, and they weren't sure if they were going to allow them or not. With Ethereum, I just went to Vitalik, and he said, yes. It's a more ease of use thing for us. Do you remember what features those were? Yeah.

The ones where...basically, with Ethereum,...

Do you want to come on camera? can issue an asset from a contract. Jimmy Carter's on, off camera over here. I'll come in if there's anything meaningful I can contribute, I will.

I want you to wave to the camera, and we'll get back to what we're doing over here. Okay, sorry about the distraction. Okay. So, on Ethereum you can issue assets from within a contract. And if we were going to able to use Counterparty, what I really wanted to be able to do was issue Counterparty assets from a surface smart contract. And I don't know if that's changed or not, but when we started it was impossible.

Got you. And so, that was the main thing. And another thing is we really love Bitcoin just like the Counterparty guys, and since we're able to sidechain to Bitcoin, I don't see it as that big of an issue that we're not using Bitcoin. And there's one final thing, which is with a prediction market platform, you have so much data that it'd be really expensive for us to store it in a Bitcoin blockchain. That makes sense. You know, smaller smart contracts would be great for Counterparty, because you can store in small amounts in blockchain cheap.

But with Augur, you're going to have Megabytes of data and that's just not feasible with Bitcoin at the moment. Very fair points. I could probably dice up some of those things, but it's all made a lot of sense, actually. I get that. In terms of the Blockstream team, in terms of sidechains, that's another project that's probably going to take a while to really be mainstream. How much do you think that will impact your project success as we wait for that to roll out? So, I talked to Adam back in October and I asked him will there be two-way pegs, the full deal by October of this year.

And he said, yes. I have no idea if that's going to happen or not... That sounds very ambitious. I'd loved to see it.

That sounds very ambitious. Yes, it does sound ambitious. So, our backup plan is just basically to use a distributed federated peg, which I mentioned earlier. I'll have a blog post on that coming up soon, because I really want to get community feedback. Because, I think that's a way we can have sidechains now, and still maintain 99.999% of decentralization, and not have the mining issues you have with regular sidechains.

Are you just hacking up all day long at this point? You've got a lot of stuff on your plate. Are you a little overwhelmed, or you're making it work? You're always a little overwhelmed doing these things, right? But I pretty much either hack all the time or go to these events sometimes, but mainly just hacking from morning to late at night. Did you guys fundraise too, or are you going to fundraise, or did you fundraise? I can't remember. So, we have some...

we're self-funded a bit, and had a little bit angel funding to make it to the crowd sale point, in which we'll basically, sell this token called reputation. And basically, the more reputation you have, the more your report is weighted. And the reason we have that is to prevent civil attacks. And then the other thing is reporters are rewarded in proportion to how much rep they have with trading fees on our system. And do you know when that crowd sales going to be? It should be in June. Okay, and who are you working with on that? Quantify, somebody else? Yeah, Quantify.

Awesome. And can you us who your angel investor is? Is that public information? I can tell you one of them. There's three. Actually I can tell you two. One of them is Vitalik Buterin's gave us some funding. And then, another one is Joe Castello.

Okay cool, real cool. Well, I appreciate your time here. Are you enjoying the Inside Bitcoin Conference? It is a pretty good one by your standards? Yes, it's been one of the best one I've been to in quite a long time. It's been...

I'm sorry? It's been the best one I've been to in quite a while. So, you went to the last one. Do you think things are a little bit more professional this year or do you think it's about the same, less professional? So, I wasn't at this one last year. But I've just been to a bunch of conferences over the past few months, and this has been the best the most well attended in a while. Yeah. I had a really good time.

And I think that's it for our interview. Is there anything else you want to tell the audience? Nope. Where can they find out more? Wonderful. Thank you again. And I guess that's it for this interview.

If you like this video and you want to see some more interviews here at the NABC. There's a whole bunch on my channel. Check them out. Subscribe to the channel if you like them, and stick around, later, party people.