Zeitgeist On The Relay Chain Podcast
Co-founders Logan Saether and David Perry were invited onto Relay Chain last month, the premier podcast in the Polkadot ecosystem. The episode was published July 23rd, 2021, and you would have noticed that we've used a lot of the material as key Twitter content. The episode revealed some great insights into Prediction Markets and Futarchy, and gives intriguing indications about where we are headed as a company and what our vision is.
The podcast episode was hosted by Jorrin Bruns, an experienced support engineer at Parity Tech (the organisation behind the Substrate framework upon which Polkadot and Kusama are built).
Here is the full transcription from the episode!
JORRIN:
Welcome everyone, thanks for tuning into another episode of The Relay Chain. I’m super excited to be with Logan and David from Zeitgeist, a prediction market on Polkadot and Kusama. We’re going to dig deep into Zeitgeist and to find out what it’s all about… but first, welcome gentlemen, thanks for being here.
LOGAN:
Good, thanks for having us.
DAVID:
Hey Jorrin. Great to be here.
JORRIN:
Zeitgeist has been a little under the radar recently. But you just closed a nice seed round, but not a lot of folks really know what Zeitgeist is… so let’s talk really high level... What is Zeitgeist?
LOGAN:
Zeitgeist is an evolving blockchain for prediction markets and futarchy.
Futarchy is the application of prediction markets in a governance aspect.
We’re building a layer one parachain with the core functionality of permissionlessly creating, trading, and resolving prediction markets. Then we take this core functionality of prediction markets and apply it to our on-chain governance as well. That’s the futarchy aspect.
We’re building the entire tech stack of a prediction market… from the parachain layer - the primitives of prediction markets - up to the application. This includes the middleware layer (we’re building an SDK for that). There’s also a lot of focus on the user experience and the UI of prediction markets.
JORRIN:
Cool. So much to dig into there. But before we do, I want to know who you guys are a little bit. So Logan, I remember when I was researching Parity and Polkadot before I started working with Parity… and I remember watching a bunch of videos of you because you worked with the Web 3 Foundation and you were in the education team. David I’ve just met you, and I know you’ve been doing a bunch of work in the Prediction market space. So how did you guys come together to create Zeitgeist?
LOGAN:
Yeah, that’s a good question… so I’ll talk a bit about my background, and how I ended up at the Web 3 Foundation and then how that became Zeitgeist...
I got started in the blockchain space in late 2016. I had heard about Bitcoin at some point before then but what kicked off my interest was when I discovered Ethereum.
I started learning solidity and researching what smart contracts were. I remember having this moment where I thought, “You can create money!” and this kind of blew my mind as something that had so much power and potential to do really cool things with. I started initially building some dApps (decentralized applications) with friends and was part of the computer science club at my university, so I spent some time with friends I knew there. I just started building dApps and talking about Ethereum, and just jumped down the rabbit hole.
After I graduated from university, I started working with a company called ChronoLogic on a project called The Ethereum Alarm Clock, which was a decentralized scheduling protocol. Essentially, it solved this very practical need of setting up the transactions beforehand, and letting them execute for you at a later time.
We had a whole ecosystem built around that, involving these decentralized actors who would execute those transactions. I worked on that for a little while. Going forward to late 2018 I was working on a new project called Convergent. We were building a personal tokens platform using bonding curves as the token emission mechanism. It was during this time that I started to hit my head against the limits of the EVM (Ethereum Virtual Machine) - let’s say, the constraints. Even while working on the Ethereum Alarm Clock we had this issue where it was really expensive gas-wise to use. There were ways to optimize it but they were always more “hacks” than good engineering practice.
So, once again with Convergent, I started to get into this situation where I felt like a lot of the things that would make the product better were really just hacking what you could do inside of this constrained environment.
That was when I started to look around.
I discovered Substrate and again had one of these big moments where it felt to me like this was the next step of blockchain tech and building blockchains. I built some early substrate modules and then noticed that the Web 3 Foundation was hiring for some positions - I think I completed a bounty for them... After that, I got hired in the technical education department.
Initially, I worked on building out the Polkadot Wiki, which at that time didn’t exist. I wrote a bunch of content for that plus worked a bit on Substrate Docs. Then, as things moved on, I started to get pulled more into tooling and building some of the operational stuff around the Kusama and Polkadot launches.
After that, I also got involved with the validator ecosystem, helping to set up and run the Thousand Validators Program with Will from Parity for a while. Around the beginning of this year (2021), I got to the point where I had been thinking about prediction markets for a while - ever since I’d got into this space. Augur was one of my favorite projects at the time. I had been a fan and part of their community over the years, although in the background.
I was eagerly waiting for a really good prediction market to pop up. After not seeing this and going through this experience of building different things at different times and progressing through blockchain tech, I saw this fit of prediction markets as an application or a protocol. I saw Substrate as a technology to bring this protocol to fruition. I started to work on Zeitgeist during nights and weekends, then earlier this year left the Web 3 Foundation to work on it full time.
JORRIN:
Sweet, and David, how did you and Logan link up with this project (Zeitgeist)?”
DAVID:
Sure… I got involved in prediction markets back in 2004. I had been interested in politics for most of my life. After the consternation and uncertainty around the 2000 election, I figured that there had to be a better way. I thought about that for a period, then in 2004 I read the book The Wisdom of Crowds*. It talked about prediction markets and how the many are smarter than the few, and how if you can harness this information, it’s very powerful.
That led me to Robin Hanson, who is the originator of Prediction Markets. I was very interested in setting up a political project so I asked Robin if he had any software available. He said, “You should meet this guy Ken Kittlitz.” Ken and Robin built the first prediction market in existence back in the early nineties.
Ken and I met at a conference at Rutgers University and hit it off. We decided to start a company together called Consensus Point. The first project we started was a political project called The Washington Stock Exchange where we ran markets on elections, policy, and different events in the political space that ended up being very effective and successful. That led to companies reaching out to us. GE became our first customer, and we worked with several dozen Fortune 500-sized enterprises and other different-sized organizations.
I’ve spent the bulk of my career in Prediction Markets. After Consensus Point, I spent some time on a next-gen prediction market system for the US government, around the time that Bitcoin was emerging. I was very interested in that but didn’t become professionally involved in the space until Ethereum came along.
I was fascinated to see what Augur was doing and what Gnosis was doing, and I reached out to Gnosis to partner with them on some ideas and they were very supportive. They introduced me to ConsenSys where I went to work for a few years on product strategy with the teams there.
I met Chris Hutchinson (“Hutch”) who is the CMO here at Zeitgeist several years ago. Of course, he was at the Web 3 foundation. In the last year or so I’d been telling him of my interest in developing a prediction market product. We discussed that on and off and then he mentioned that he knew this guy, Logan, who just left the foundation to work on a prediction market platform. The three of us got together and the rest is history.
JORRIN:
Awesome! Like all great stories, it’s not always what you know - of course, that’s a strong part of it - it’s who you know and the connectors that bring everything together. That’s a great story to hear.
You guys mentioned Futarchy, and I think we went (too) quickly past that. That was a new topic for me, a new word for me, and I think it’s going to be new for a bunch of people. So, let’s try and clarify what Futarchy is and what does it mean within the context of Zeitgeist?
LOGAN:
With Zeitgeist, what we want to do is push prediction markets further. What we see as one way to apply prediction markets further than they’ve been taken before is to apply them to on-chain governance in a binding way.
In other words, to not only use the signals to inform governance decisions but to actually use prediction market signals to implement rules, apply them to what proposals can get passed, when proposals don’t get passed and when proposals get halted for some time.
Futarchy is just using prediction markets for governance. We have a few different ways of how we’re thinking this will look but maybe I’ll let David talk a little bit more about it before we go into the details.
DAVID:
Futarchy allows you to aggregate information for the purpose of decision-making. You decide some outcome of success that you want - whether it’s a development milestone, or security uptime, or whatever this outcome is. Then, you have a variety of ways to get there…
If you’re a development team working on a new product and you’re concerned with the impact that product is going to have on the network’s uptime, you can make a bet on how we go about implementing this product and its probability of affecting the network negatively.
It allows you to get decision-contingent outcomes. It basically asks the question, “What would happen if we did this?” If we have a problem that we’re solving; some sort of knowledge problem and different ways of going about solving that problem, what’s the most effective way?
What we’ve seen is that futarchy and prediction markets are the best mechanisms that we know of when you compare all other approaches of asking that question - and that’s been pored over many times.
JORRIN:
I find that super-fascinating because when I think of traditional prediction markets, they’re very much a form of gambling. It’s betting what outcome’s going to happen and you’re putting your money where your mouth is but, in this case, it seems like it’s betting on things that really matter. It’s potentially having real-world implications on the outcome of this. I think that’s super-fascinating.
You guys mentioned that you’re building on Substrate, and maybe for some of the folks reading, it’s the first time they’re learning about this. Substrate is a way for you to build bespoke blockchains that are specific to your use case.
For the use case of Zeitgeist, what is your recipe for your blockchain? There are different pallets that you can put together. What kind of pallets are you guys using and are you developing your own pallet?
LOGAN:
Yes, we are. With Substate, you have the ability to both go into the internals of the chain and modify very low-level pieces, like the transaction pool, the way blocks are produced, and so on. You also have the runtime code, which is what’s being run inside of a block.
So, we are indeed building our own pallet. The pallets are these modules that you put together to create your full runtime. The pallets we’re building are centered around prediction markets, of course. We have a few pallets that we’re working on: We have the core prediction market pallet, and this has the logic of creating markets; it contains the interface of resolving them and disputing them. We’re also creating separate pallets for the dispute mechanism.
We take an approach where we have a bit of flexibility like how markets are reported - how the oracle first sources the market result and also how the oracle is secured. This means, for instance, under a dispute how does the chain come to a consensus on what the result will be? How can everyone agree on what the result will be? We’re building some pallets for this.
We have one that we’re calling Simple Disputes: It’s really just a game of putting more and more stake behind a result until the other side gives up.
We’re also building something a bit more robust called the Court Pallet. The idea of the court pallet is that we want to economically incentivize actors of the protocol. Anyone can join the court by putting some stake there. When someone creates a dispute, they put a bond and they create a court case. When they create a course case, we draw a random subset of these staked users and ask them to come to a consensus among themselves.
In case that’s disputed again, we have an escalation mechanism. We allow further court cases, each time taking a larger subset of these staked users. At some point we get to a cap; we say that there’ve been enough court cases. It’s a parameter in the runtime; I think it’s set to six so after six court cases, we pull that into a chain-like vote. It becomes available for anyone to say what they think is the correct outcome.
That’s part of it but we’re also building pallets for trading. We’re building AMM pallets (automated market makers). One is based on the balancer formula and another is based on the LMSR formula which was invented by Robin Hanson as well. We’ve been doing some research on how we can improve this specifically for our own use case.
As far as I know, we’re the first to be implementing it in a blockchain context. I know that some projects have tried this before, but there have always been major challenges that have made it impractical on the EVM. We’ve been working on solving these challenges and implementing it as a substrate pallet. We’re also using some of the frame pallets. Substrate comes with a collection of modules and we’re using a number of those on our chain as well.
JORRIN:
Awesome. It seems as if you guys are deep into the Substrate weeds there and you're hacking away and building the parachute as you're going. That sounds awesome.
You talked about a couple of things there... First of all, I was surprised to see automated market-makers working in this space. Maybe you can talk about how that fits into a prediction market framework? The way I’ve traditionally seen automated market-makers is more of a use for fungible tokens, and trading them and creating bonding curves to derive the value of certain things. How does that work in the context of a prediction market?
LOGAN:
The core concepts are the same. With prediction markets, you create tokens that represent the outcome. We're able to use these tokens to provide liquidity in the automated market-maker, and for other people to come and trade using that provided liquidity. The core concept is the same. There's a difference in the risks of providing liquidity in prediction markets versus the risk of providing liquidity of just some coin pair (like ETH and DAI, let's say). The risk is what's called impermanent loss. This exists in any coin pair but in prediction markets, it's a bigger problem.
The reason for that is because at some point the outcomes become known and for one of the tokens the value will essentially go to zero, which means that for the people who initially provided the liquidity, some part of the value that they put in will essentially disappear and people will pull as much value as they can out.
It's a problem that all prediction markets that use AMMs face, and it's the reason we're doing so much research into new AMM models. Specifically, on the LMSR AMM or the LMSR-based AMM that we're working on that we've codenamed Rikiddo. We're continuing to do further research on it.
We're currently putting together a best practices guide on how to provide liquidity to prediction markets to reduce those risks. We're also looking at how we can modify the AMM to make it harder for liquidity providers to shoot themselves in the foot when providing liquidity.
JORRIN:
That makes sense. I think it might be useful here to go through a test example.
I know you guys recently did a testnet campaign called The Kusama Derby. You were playing a game to predict who was going to get the first Kusama Parachain Slots. That might be a good example to walk through. Maybe we can even make it simpler and devise a more straightforward kind of example?
Maybe the flow of how a prediction market starts? People interact with a contract, then it gets resolved, and then maybe how it could be potentially challenged afterward. Does that make sense? I guess it makes sense to use the Derby as an example here because it's something we can relate to. I think… Hopefully, everyone can relate to the Parachain Auctions.
LOGAN:
With the Kusama Derby... Just quickly, our Testnet itself is actually called Battery Park and the Derby was our Testnet campaign. What we were doing with the Derby was that we wanted to preview the tech, while at the same time build out a community and get people to come and use it.
We could ideally then solve any bugs, practice with deployment, and so on. We decided to set up three markets, one for each of the first three Parachain Slots, and make this our campaign.
To explain how a prediction market works - when you create a market, you attach some metadata to it. This metadata is essentially what this market's about. For the Derby market, let's say the first Parachain auction markets, the question inside this metadata was who would win the first Kusama parachain slots.
Each of the outcomes was one of the teams that were one of the first ten projects that connected to Rococo. You had Karura, Moon River, Shiden, and so on. With the prediction market, you're able to generate a full set of these outcome tokens by exchanging one base unit token. The reason that works is because, at the end of the market, the winning outcome is redeemable for one base unit.
In the beginning, you don't know who's going to win. You can create one of each for one base unit because later just one of them will be worth that and the rest will be worth zero. What's interesting about a prediction market is what happens between that; between not knowing which one is going to win and then knowing. This is where the dynamics of the market mechanism come into play.
People will trade the markets and the outcome tokens’ price will reflect the probability of that outcome being the one that happens. In the Derby at different times, we had Karura as an early leader but there were other ones like Shiden that changed a lot throughout the auction. It changed from 0.10 base units, which would be 10% probability, to 30.3, which would be 30% probability.
There are a lot of ways for traders to analyze the markets and move the probabilities to be closer to what they think will happen. The power of the prediction market is that you're aggregating all of these viewpoints and providing an incentive for them to provide their knowledge in a public arena.
Then there are various ways to arbitrage different outcomes and it's essentially a freely traded market. Usually, you specify when the market will end; when it expires. When the market expires, you also specify who's going to be submitting the results; who the Oracle of this market will be. When the market ends, the Oracle should submit its results.
There's a 24-hour window where the Oracle can submit the results. The reason it's a 24-hour window is that it's possible that the Oracle won't show up. In the case that the Oracle doesn't show up, the market creator has placed a bond. The market creator gets part of their bond slashed because they picked a bad Oracle. Then, it's open to anyone to submit a result at that point. Even someone who wants to volunteer to say what their result is can submit a result.
After that, we have this dispute resolution mechanism. If someone else says, “Hey, that's not the result. I don't think it's A, I think it's B,” they can put some bond down and say that it's B. Depending on which dispute resolution mechanism it is, we either enter into the court or we enter into simple disputes or potentially another mechanism that we’ll develop a bit later on.
JORRIN:
You mentioned earlier that this can go for several rounds of increasing jurors and that then eventually if you push it to its limits, you evoke everybody to be a juror. Then for sure, we'll figure out the true wisdom of the crowd.
Okay. That makes sense. I guess my next question is that works in the perfect world very well for empirical things… For example, we know for sure that Karura won the first slot; it’s unequivocal. Are prediction markets useful for things that are subjective? Can I open up a market and say, “Relay Chain is the best blockchain podcast of all time”? Can we come to a reasonable conclusion of whether or not that’s true using a prediction market?
DAVID:
Yes, prediction markets do that very well. I’ve seen a lot of applications over the years on prediction markets applied to product development and market research. As you’re thinking about which products to build, you’re doing a survey of the market and trying to find that product-market fit. You’re saying, “We could build this and we could build that.”
(In the past) I’ve worked with a few mobile device companies where you have a lot of decisions to make on things like the size of the screen, the processor, the battery size, the enclosure. There’s no right or wrong answer but what you want to do is - based on the objectives that you have on, say revenue or your development timeline - use the prediction market to help decide these metrics.
A prediction market can, therefore, be used for aggregating opinions about subjective data. Ultimately, you do have an objective measure to judge the subjective. At some point, we’re going to end this market, and then we’re going to use a way of analyzing the market. The subjective trades in an objective way. So yes, you definitely can use markets in subjective contexts.
JORRIN:
Nested layers of game theory where people participating are incentivized to provide an answer, but knowing that other people are going to provide an answer, you’re incentivized to pick the answer that you think everybody else is going to pick. How does that play out? Are there any mechanisms or levers in Zeitgeist that mitigate these things, or is that a true prediction market? Is that healthy for a prediction market?
DAVID:
Great question. What I described a moment ago was a Keynesian beauty contest for building a phone: Should we have a large screen or a large battery? Helping you decide those trades. That’s why you want to have an objective measure like a completion date or revenue.
You can ask the question in a market, “Which is cooler? Red shirts or blue shirts?” And you can run a market on that for, let’s say, a month. At the end of the month, you’re going to do a payoff. Let’s say, the blue shirts are higher based on some volume-weighted average over the last week. Just as you would do for a survey or an opinion focus group where you ask that question. You could aggregate the same information in a prediction market.
What’s different about a prediction market is: If the reason you’re doing that is if you want to make the most money from whatever the shirt color is, or whatever you’re deciding, then prediction markets are going to help you answer that question most effectively. There are other mechanisms that we can bring to bear; there’s something called a “Bayesian truth serum” that is very effective at helping measure and reduce these biases. As long as you think about it up front, you can design your incentives accordingly.
JORRIN:
Let’s talk about some of the other use cases for Zeitgeist, maybe getting into some more real world cases. You mentioned you’re very interested in politics. We know in recent times the election process (has been challenging) - how do we find out the true wisdom of the crowd, the true will of the people? So how do you see prediction markets in general, and Zeitgeist playing into this kind of problem?
DAVID:
Politics is a great domain. A political market has several gates that you get through, like an election. There’s a nomination process, there are fundraising milestones and then there’s a general election. At the presidential level, there are different state contests.
What you want to do as a candidate or a party is pick your candidate that will ultimately win the election. That candidate may not be the obvious choice when you’re talking about, say, at the primary level. We’ve seen examples where there’s been a really popular candidate... But if you were running a sort of Keynesian Beauty contest of who’s popular right now, you’d have one answer, and then if you looked at it in the light of a broader general action, it would be a different answer.
That’s a good contrast to how prediction markets compare with something like opinion polling, for instance, where a poll asks you, “If the election were held today, who would you vote for?” But a prediction market is asking, “When the election is held, who do you think will win?”
To go back to our own Kusama Derby example, if we’d run a survey about that we would’ve just asked, “Who do you want to win the slot?” versus “Who do you think’s going to win?”
Incentives matter. When money is on the line, people have an incentive to pick what they think is the actual outcome versus the outcome that they want. It’s got a lot of relevance in the political domain, for sure.
JORRIN:
I think that's super-attractive to me and blockchain in general. I'm a huge fan of governance over government.
Can something like Zeitgeist or a prediction market be used on a more granular level to determine what our government's policy should be like? IE. Should we have higher taxes, lower taxes; any kind of decision that a government would make? Is it feasible to do that with a prediction market on a more granular level, where it's less about an individual making these decisions and it's more about the consistent will of the people?
DAVID:
Absolutely. Yes. One of our key goals for Zeitgeist in regards to politics is to be a mechanism for generating policy advice. As we do that, and we build up a track record of providing signals about which policy would yield the most advantageous outcome, we expect that savvy politicians will look at that and follow suit. If they don't, there's a very robust mechanism providing advice and we can ask them, “Hey, why are you not following this policy bias when it's consistently yielding accurate intelligence?” We think that's very interesting.
In politics, we're less interested in making the existing system more efficient than we are in helping more people get engaged in politics and just having the whole process be more transparent.
JORRIN:
Absolutely. Transparency is key and blockchain aids that perfectly.
Let's talk about the types of prediction markets. I know you guys support at least three types - categorical, scalar, and combination. What do those mean and what are their use cases?
LOGAN:
With Zeitgeist, we're aiming to build the primitives of prediction markets. We want to cover as many types of prediction markets as we can. With categorical markets, you have a number of separate outcomes; categories of outcomes. This can range from, in the simplest form, two outcomes. Yes or no. This is commonly known as a binary market. It can either happen or it doesn't. Or we can have up to ten or more potential outcomes. This is what we used for the Derby, for instance. Each outcome was a different parachain team. That's one type of market.
Another market is a scalar market. For some markets, you don't want to pick between outcomes because you could potentially have a whole range of outcomes. It could be zero; it could be a hundred thousand. It could be any point between that. For instance, with financial markets. Predicting the price of an asset in the future would be on a range. You would set a minimum and a maximum then the market would create a prediction between those two points.
I think the most interesting one is the combinatorial market, which I'll let David talk about.
DAVID:
Combinatorial markets basically allow for composable prediction markets. This is something we haven't seen in the crypto space yet, and we're excited to be supporting that. You can imagine working on a project on their platform and it has markets on deciding things like, “What features should we build? Is the impact on the security of the network based on these features? We want to increase the transaction throughput, the number of addresses, and other dimensions.” In a typical prediction market, you'd have all of these as individual markets. It’s the same thing with Zeitgeist but with combinatorics, you can link all these things together to have a composable prediction market.
You can get an answer about optimizing over these dimensions. It'll tell you, for example, what features will have the best security with the best transaction throughput. It's a very powerful mechanism that is vastly more valuable than traditional prediction markets. We're really excited about it.
JORRIN:
Very interesting. Could you give an example of where combinatorial markets would be used?
DAVID:
Sure. Let’s go back to politics - let's say you're trying to pick a candidate… there's a market on the primary election. There's a market in different States, like in the New Hampshire or South Carolina primary. What you want to do is work out if we pick Candidate A, what's the probability that they will make it through the primary, as well as both of these states?
Then for Candidate B, what's the probability that they will make it through both South Carolina and New Hampshire? Based on those prices, you could see that we had this front runner, who’s going to primary well, but they're not going to make it through these other early contests. We can have the security of picking maybe the less obvious choice who’s actually going to do better in the general election.
You could also think about it in terms of, let's say a basketball bracket, where you have the final four. In addition to picking which team is going to win each bracket, you're picking who's going to win at the semifinal, final, and championship match all in a series.
JORRIN:
I see. So it allows more, more granular, more sophisticated kind of knowledge?
DAVID:
Right. And composability.
JORRIN:
That makes more sense. You guys said that you're building an SDK and you're going to allow apps essentially to plug into Zeitgeist. Does that mean that anybody can have their own front end, ask their own questions, and use Zeitgeist as the backend for that?
LOGAN:
Exactly. That's the goal. Zeitgeist is a permissionless protocol. We encourage people to build their own front ends. The purpose of building the SDK is to make this as easy as possible for entrepreneurs or builders to come in who might have a really good idea and just to quickly set up their prediction market-based application.
By no means is the SDK required. You could always access the chain directly using the Polkadot.JS API. The SDK just aims to provide a guided way to do that and a way to do it quicker with more safety measures turned on.
JORRIN:
Is there a way by which an organization, a startup or a community can introduce futarchy into their own governance structure?
LOGAN:
Definitely. There are two different sides to this. There's the side where it's futarchy by signal only. This involves honoring the prediction markets and the signals that you're getting from them, but there's no kind of enforceability there.
And then the other side is that futarchy is enforced. Since we're building prediction markets in the native layer, we can use prediction markets and enforce in different ways. Proposals will take place, etc.
What we want to do with Zeitgeist is first to demonstrate this on our own chain. Then ideally, as it's theorized, this is an augmentation to governance so other projects would come and use it as well.
Lately, we've been talking to some people, just brainstorming, how this could work. The cool part about building in the Polkadot ecosystem and deploying to Kusama is that with XCM we could do this for other chains too. So, if it is something that we prove on Zeitgeist to be useful, then potentially any parachains, if they opt into it, could have enforceable futarchy as well. Further on, we would support DAOs that could do that as well on Zeitgeist.
JORRIN:
That's awesome. A very attractive piece of Substrate in the Polkadot ecosystem is that it's so composable with XCM. That's the magic of it.
I can imagine how parachains would be able to leverage Zeitgeist in the future. That's super-awesome.
What else are you guys excited about? What's on the horizon for you? How do you see Zeitgeist changing the world?
LOGAN:
Most of our effort at the moment is spent on building an application that's going to have really good UI and UX, and is going to be something that people will come and use and will want to use. I think we've seen prediction markets in the past, prediction markets on blockchain, being plagued by this curse of bad usability. We want to make sure that the Zeitgeist application is dead simple to use, super simple. Just come in, if you have some information, you can use that information in a market. We're excited to ship that soon. We're still actively working on it. It's the next thing we want to push out the door while still in this Testnet phase that we're in.
Then of course we're excited about the Mainnet launch later this year when we're planning to deploy as a parachain to Kusama.
JORRIN:
So, you're planning on doing an initial launch on Kusama. Are there any ideas of whether or not you will go for a slot on Polkadot as well? Or do you think Kusama is particularly catered to your needs?
LOGAN:
We're big fans of Kusama. We like the “move fast and break things” and “embrace the chaos” ethos of Kusama. We think it's a good fit for Zeitgeist.
We're not ruling out the potential of Zeitgeists’ Polkadot deployment in the future but we're only planning to deploy Zeitgeist Mainnet to Kusama.
We're going to focus on our Kusama deployment and then see how things might change in the future. We're thinking now that with the Kusama Polkadot bridge when that's available, we'll have interoperability in the wider Polkadot ecosystem. It's not something we're rolling out but it's also not something that we're seeing as super-critical for Zeitgeist.
JORRIN:
Makes sense. You mentioned that you've got Battery Park running and that's your Testnet and the Kusama Derby recently concluded. What's the next step?
Are you going to do a couple more campaigns on the Testnet before launching your main project on Kusama? What do folks reading this have to look forward to there?
LOGAN:
We are planning on doing another campaign. Not in the very immediate term but not too far out either. Before Mainnet we'll hold another campaign. Essentially, it gives the community a chance to get to Zeitgeist.
Before we do that, we do want to do the main application that we're working on. With the Derby, since the main application is still being worked on, we deployed a preview of the application. The next step for us is to build a generalized application where people can not only trade the markets but also create the markets. Once we have that, once we ship that on the Testnet, we're going to hold another campaign and go through the rounds of getting people to use it, break it and solve bugs in preparation for our Mainnet deployment.
JORRIN:
Perfect. I know you guys are working on a whitepaper as well. Do you have any hints on when that will be released and we will be able to sink our teeth into it?
LOGAN:
We don't have a date on that. We do have most of the meat drafted out there. The only reason we haven't published it yet is that we still have some more research aspects that we want to work into it - some of the recent stuff we've published on Rikkido, as well as some of the stuff we want to publish on combinatorial markets as well. Expected soon but there's no date!
JORRIN:
“Soon ™”, very classic. Definitely common in the blockchain space. We'll be waiting and watching.
Where would be the best place to keep up to date with the project? Where will all the announcements be released?
LOGAN:
We have all the traditional social media channels where you can find us - Twitter is @ZeitgeistPM. Our website is www.zeitgiest.pm - PM is short for prediction market. From the website, you can find our telegram room and our discord server, which is our main community. That's where they are.
Then, of course, our newsletter is probably the best place to sign up and to make sure you stay up to date on any new events we might be having. This is found from the home page of the website.
JORRIN:
Great. You guys are a small but growing organization. I noticed that you are hiring right now. Do you want to put a call out to the community for the type of folks that you're looking for?
LOGAN:
Sure. We have a couple of job posts open. We're looking for a really solid designer who likes to do the branding and marketing sides of things, as well as take bites out of the UI / UX part. We're also on the search for a really good typescript developer who likes to work more on the back end of things; not really on the front end. This is to help us architect and build out the SDK, and other tooling that we'll need.
Then, one that we are looking for (we haven't put a post up yet) - we haven't fully defined it but something like a “Head of Prediction”. This is someone who would be taking ownership over when we have the prediction markets live, taking ownership over moving them forward. Not only doing that but also coming up with interesting market ideas, creating new markets, and promoting these to the communities in different ways.
JORRIN:
That sounds like a very interesting role. I feel like a prerequisite to apply for that is needing to have a crystal ball or something! Might be a magic eight-ball to consult with, but Head of Prediction is awesome. I love the sound of that.
Alright, gentlemen, thank you so much for being here today. Is there anything else you wanted to mention before we head off for the day?
LOGAN:
It's all good from my end.
JORRIN:
Anything for you, David?
DAVID:
No, I was just gonna say thanks very much. I really enjoyed the conversation, Jorrin.
JORRIN:
Thank you so much, gentlemen, for joining us today. I’ve enjoyed the conversation and I'm excited to follow Zeitgeist as it continues to grow. I think prediction markets are a fascinating topic and I can't wait to see it all come to life. Thank you very much again, have a good day!
LOGAN:
Thank you. You too.
DAVID:
Thank you.