Supra's CEO: Oracle nodes that never rotate will eventually collude

Ashton Addison: What is Supra Oracles?▶ 00:23

Joshua Tobkin: Supra is a next-generation oracle. Many people have heard of blockchains and smart contracts, but they don't realize that what a smart contract can do is limited to on-chain data. Once you can bring external data from the real world in, the utility is much greater. A common pattern is that developers come from Web2 with a great idea, jump into Web3, and halfway through realize they can't build it without getting external data into a blockchain. We have a mechanism, a strategy and a technology that brings cryptographic guarantees to any data entering a blockchain, and we do it at scale, with a lot of throughput. We're also positioned as a middleware network in the middle of multiple blockchains, serving that external data to multiple ecosystems.

Ashton: For people who hold Bitcoin and Ethereum and are just getting into DeFi, what does it mean to say DeFi wouldn't exist without oracles?▶ 01:50

Joshua: Use cases for smart contract platforms are limited in scope because the contracts have to act on some information. If that information is purely on-chain or historical data, the range of what can be done is limited. With oracles you can create synthetic products, such as tokens from other ecosystems on chain, or digital assets represented on chain. You can build insurance products, and you can get analytics and performance data on NFTs. There are a lot of things you can do with oracles that you otherwise simply cannot do with blockchains.

Ashton: Are price feeds the main use case, or is Supra working in other industries too?▶ 03:23

Joshua: We're working across multiple industries. I have a background in IoT myself. We built early prototypes for a publicly traded company in Taiwan called Avontec, and we entered a food safety challenge run by Walmart China, where we were one of the teams chosen to present our solution. That solution included both IoT and blockchains. DeFi is the low-hanging fruit. It has clear product market fit. But whether it's DeFi, IoT or supply chains, the quality and security of the data is the whole game. Data integrity is the whole game. We apply more rigor than our competition. For cryptocurrency pricing, our methodology has a security guarantee of 100 times more data correctness guarantees than the incumbents. Some solutions use a small cluster of nodes that never change, and given enough time and incentive, those nodes will collude. We saw this in the real world with the LIBOR scandal, where trusted institutions like banks were colluding on interest rates. If it can happen there, it's definitely happening among anonymous node operators. So we bring blockchain principles to the oracle layer. We use more nodes, with random selection and random rotation. We're also a proof-of-stake network, so nodes have to bond our token. That gives us a cryptographically secure network, layered with economic security through staking. It's basically the inability to cheat.

Ashton: Decentralized exchanges rely on price feeds that come from the real world. How do you make sure that data isn't manipulated?▶ 07:16

Joshua: Something less aggressive than changing the data is reading it first and trading against it before you publish it. That's front-running, or delaying publication if you see it isn't beneficial to your position. That's not as obvious as actually changing the value. It's also hard to prove whether some solutions adjust the reported values by a couple of basis points in their favor. So you want to make that behavior virtually impossible in the first place. You also want to organize the economics so there's a race to be the first node to publish the correct information, and that's where the reward is.

Ashton: Where is the platform today?▶ 08:47

Joshua: We're still in the early stages. We have a proof of concept that's effectively a private testnet, and we're working with select partners. We're moving from this alpha into a test phase in the next few months. Then we'll go through a series of rigorous audits, which could take three or more months, and we'll invite multiple audit firms from a variety of ecosystems. We'll also create a large bounty program that invites anyone to attack our system so we can fortify it before launch. We take security very seriously. This isn't the real world, where you can roll back if bad data enters a system. Smart contracts compose and interact with each other, so the consequences cascade. The data correctness guarantees need to be as strong as the blockchains underlying these ecosystems. We want to move fast, and we are moving fast, but without breaking things.

Ashton: Who are you partnering with, and how will end users actually touch the oracle?▶ 10:19

Joshua: We have 400 partnerships already in the works. We've had a lot of early discussions with projects. Many are DeFi, many are GameFi, and most have some financial component. Some are sports-related and some are prediction markets. Protocols call our data periodically. We're also building an automation service into Supra Oracles. It will take some time to roll out, but it's a decentralized if-this-then-that. If the price drops below a certain threshold, it triggers a certain behavior. Because we're a cross-chain oracle sitting in the middle of multiple ecosystems, behavior in one ecosystem can trigger a reaction or transaction on another. That doesn't exist yet in the oracle space as an all-in-one solution to facilitate cross-chain DeFi.

Ashton: Where do you see Supra fitting into DeFi over the next few years?▶ 12:37

Joshua: I want to mention constant function market makers, or AMMs, like the Uniswaps of the world. They're a really interesting innovation and very much responsible for a large amount of this boom. Most of them, like Uniswap, don't use a traditional oracle like ours or any other. What we're interested in bringing to market is the next generation of AMMs that do use oracle data. That can mitigate impermanent loss, where there's some value loss in these trades. It can also reduce slippage, where you might not get the best deal because of the way these products are priced. We think there will be new innovations in that area using oracle data, and ultimately we can get better accuracy. I have a tremendous amount of respect for that innovation, but I think the space will evolve, and using oracles in unique ways will prove to do some very interesting things.

Ashton: What are you aiming to achieve in the first half of 2022?▶ 14:34

Joshua: We're trying to get to market as quickly as possible, in a sustainable and secure manner. We have a few initiatives in the works: the oracle service, an automation network, and a verifiable random function, or VRF. That's a very important service. Things like generative art and NFTs use randomness to make fair selections. We have a very unique approach. We use cryptographic primitives and threshold signature schemes, and we also bring in the idea of chaos theory. Chaotic systems like weather data, or the price of Bitcoin, which nobody really knows in advance, add entropy to our randomness. On timeline, we're moving from alpha to testnet fairly soon, maybe in a couple of months. We're already talking to node operators, and they've been quite open to working with us, since node operators are willing to run nodes across multiple projects. We'll slowly but surely decentralize during the test period, which will take three to six months or so, plus the audits. If we're lucky, we're talking about maybe summertime.