Hello DeFiers,
Nicholas Fett is the CTO of Tellor, a decentralized PoW oracle on Ethereum. We sat down with Nick and talked about his project and the awareness of oracles in DeFi.
Enjoy reading this (project-specific) #5 edition of My Two Gwei by dex.blue!
A Chat with Nick Fett
Hi Nick, what’s your background and how did you get into crypto?
Right after college, I interned at a libertarian think tank and in the context of the Ron Paul 2012 campaign and the End the Fed rallies, I got into Bitcoin. I started off with some shitcoin trading and in 2014/15 I began building some market making bots. My day job by then was at the U.S. CFTC which was a good educational phase for me. In 2017 I then left my job to start a derivatives platform myself and since then I’ve been here full time.
So you started off as a trader. Do you still trade at all?
That’s just how I got into crypto back then since trading was quite straightforward. Now you’re competing against the big trading firms, so, I only hold some positions, that’s all.
I think it’s important to find out where your expertise lies. I’m very technical and, on that level, have a good understanding of what projects are more real than others. It’s always nice to find a cool project early on, but if we are being honest, it’s really hard to get in early enough to make it worth it, even if you like the tech.
You are working on an oracle project right now. What are oracles all about and what makes them so essential to the ecosystem?
I think what explains it best, is just telling you how we stumbled upon the problem ourselves:
We were building a derivative platform and the information we needed were things like the price of Bitcoin. The problem with Ethereum is that it doesn’t know this information and therefore relies on someone to enter it externally. The most straightforward way to do this would be to get a company to just tell you this information, but the problem is that you rely on them. If they’re not there, you don’t settle. If they lie, it could be to your disadvantage. And that’s what the oracle problem is.
When we were building this platform, there was a lot of good research, but no real solutions that were decentralized and ready to use, so, we built our own and that’s how we came up with Tellor.
As you mentioned there are different approaches to solving this “oracle problem”. Why is the way you’re doing it the most promising for you?
A lot of other oracles solutions provide off-chain information by fetching a whole lot of queries at a very high speed and to do this in a secure way is really hard and this is where Tellor works a little bit different.
To really do this in a secure way we went back and looked at Bitcoin 101 and its mining concept. We decided to also do some meaningless Proof of Work computation (PoW) but instead of letting the winner say what the order of transactions is, he gets to say what the price of Bitcoin was. And that’s the base of our system: A proof of Work oracle.
And if the winning miner provides false information?
We require miners to stake 1000 tokens if they want to compete in our PoW challenge. If the winner lied though and someone sees this, we put it up for a vote. If he’s found guilty his tokens are slashed. That’s how the system is secured.
So, at the end of the day, the community validates the data?
This the hard part about oracles, you can’t necessarily validate it. With on-chain information, the system itself can validate whether it’s right or wrong. The system, however, knows nothing about off-chain information - you have to put it up to this subjective truth of the community.
It was really important for us from the very beginning to be as decentralized as possible and since the token holders vote, we started with a zero token supply and minted with PoW going forward – so we think we have a really good distribution of tokens and maintain an active community that knows what good and what bad values are and knows what Tellor is trying to build an honest oracle. That’s what really gives it value.
Can you elaborate on this “subjective truth”?
Generally, it's important to know what exactly an oracles give back. There are different types of data: there’s the price of Bitcoin, the price of Bitcoin you can buy OTC right now and the price of Bitcoin according to the Coinbase API. Now, say Coinbase goes down and their API says the Bitcoin price goes to zero, then that’s a valid oracle query – but its not what you were expecting. You have to understand what you are asking for, so we can provide the right data. And this is what we are working on moving forward.
And the awareness toward decentralized oracles in the space generally?
Just building secure projects is one thing but educating the users is another. Whenever you go from a traditional safe database to a blockchain there is a learning curve and we just have to start teaching people and manage their expectations.
You see these people coming into the space trying to build DeFi applications that they see in the real world and often they find out it doesn't work that way. You know, it’s a lot harder to put things on-chain. A lot of things in the traditional financial world don’t make sense when you move them to a trustless architecture. So, I am hoping we can start building new things, things that actually make sense. And that’s what I am excited for.
As a last question, what’s your sentiment toward the future of decentralized oracles and where is this realistically going?
In the next year, I think we will see people actually starting to care about oracles more. It’s really nice that now whenever you see a stablecoin or derivates project people instantly ask what oracles they are using. And that’s a refreshing toss because the whole point of the project comes down to whether your project is decentralized. So, I think we are going to see more communities pushing people to actually decentralizing their oracles.