🔩 Augur v2 Launches — Issue No. 105

The team behind Augur has launched the next version of their decentralized betting market protocol. The launch of Augur v2 comes almost exactly two years after the first version shipped. Link.

Augur was an early high profile ICO. Unlike many such offerings, Augur succeeded in delivering a working product, though it took them a few years. Despite excitement around Augur v1's launch, usage of the protocol eventually stalled.

Now, the team behind Augur is back with the next version, replete with UX improvements they hope will help drive adoption. Let's review the basic premise of Augur before discussing these improvements, and what we might expect to see from Augur v2 in the coming months.

Decentralized Oracles

The basic idea behind Augur is rather elegant. The system of smart contracts allows anyone to create a prediction market on chain. Users can buy or sell stakes of the outcome of a future event. If the outcome they predicted comes true, they receive a payout, at the expense of the users who bought the other side. Centralized versions of such prediction markets are used for political elections and sports betting.

Astute readers will have an immediate question: how does the on chain smart contract system "know" who won the election or the game? After all, smart contracts don't natively "know" about anything except the state on chain, nor can they call out to third party web 2.0 APIs. How are the outcomes of these markets determined, such that funds can be released to the winning parties?

The challenge of getting off chain information on chain is generally known as the "oracle problem," and it's one of the most important research topics in the crypto world. Certainly there are plenty of quick and dirty engineering solutions— such as semi-trusted multi-signature schemes— where some group of trusted entities simply feeds the data on chain. For many applications, this is probably good enough.

Augur, though, has a more interesting solution. Users of platform stake REP tokens to participate in market "resolution." When it comes time to end a market— say, for example, after the election has occurred— the users must assign their stake to the outcome they believe is correct.

If a user asserts the outcome the majority of other stakers choose, they receive their share of the reward. If they disagree with the majority of stakers, they lose their stake. In this way, Augur not only decentralizes the resolution of markets, it becomes a decentralized oracle itself. Other smart contract developers can hook into the results of these votes to bring data about off chain events into their own systems. In fact, one way to look at Augur is as a decentralized oracle solution that uses betting markets to incentivize accurate on chain event reporting.

You might wonder: what happen if the majority of stakers lie or collude? For contested outcomes, there are multiple rounds of disputes possible, with exponentially increasing rewards and penalties at each level. At the limit, the protocol itself can fork, with two alternate versions of the on chain universe proceeding forward, each with their own outcome for the contested case. The assumption is that, in the rare case where such a fork occurs, the market would ultimately choose the version of the protocol that best reflected reality, and ignore the one that has results perceived to be untrue.

UX Improvements

The first version of Augur launched in July 2018. It quickly rocketed up to over $1 million at stake in its betting markets, but later slowed. Usage remained relatively flat, with both the money at stake and number of liquid markets plateauing. Why did Augur usage stall?

The project's creators think UX and onboarding challenges are what held the first version back. In Augur v2, they've focused on solving these problems. In particular, the protocol now uses the Dai stablecoin for betting markets, instead of Ether, removing the volatility risk of the underlying asset. Additionally, the new version uses 0x for off-chain order books, greatly reducing the fees to participate. Markets also resolve much faster in the new version, and there are a host of other small improvements. Link.

Crossing the Chasm

Augur is an interesting project. It provides on chain oracle data to developers in an elegant, decentralized way. It offers a service to users— namely, betting markets— that one would assume there would be interest in. There's been enthusiasm for it from the crypto community, but that enthusiasm hasn't translated into heavy usage.

As mentioned, the team behind Augur is betting that the lack of growth is due to UX issues. In the next few months, we'll get to see if this theory is correct. Certainly, with a heated election cycle happening in the USA, it's hard to imagine a better time for open betting markets.

I can't help but to be a little skeptical of the "bad UX" explanation, though. It's not that the UX of Augur v1 wasn't difficult. Certainly it was. But so many other DeFi products, with most of the same UX challenges, have nonetheless attracted thousands of users and billions of dollars in liquidity. What makes Augur uniquely unappealing to the gamblers (or to use the polite term, "speculative investors") that populate the DeFi ecosystem?

I'm honestly unsure. Is it possible not that many people care to participate in betting markets? That for all the blustering, over-confident predictions people like to make, they're actually just unwilling to actually put money on their predictions? This seems unlikely. After all, sports betting is a huge market! But I don't have a better explanation. I suppose the next few months will yield some answers.

A trust-minimized oracle system, even a slow and expensive one, would be a useful primitive for the ecosystem to have. I hope Augur's creators are right, and that Augur v2 crosses the chasm where the first version failed.