Freedom Money Episode 22 The Oracle

Today on the podcast, we're doing a very special issue. We have Siller here, uh, to talk about the oracle. Welcome to the podcast.

Hey everyone. Hey Kevin, thanks for having me again.

Um, so the oracle is kind of an important piece, uh, to the discrete log contract protocol. And I always like to describe the oracle as a fancy way of saying Bitcoin's price. And there's some, I guess, question as to whether everything is trustless or trust minimized. And I would say that there— what we've built is a system that is trust minimized, and the trust is in the oracle itself. And so it's worth doing a bit of a deep dive into the oracle. Maybe just, you know, I'd like you to add to that, but for listeners who are not deep into Bitcoin or discrete log contracts, can you explain what the oracle is doing?

Oh yeah, oh yeah, certainly. So hi again, thanks for having me. This is now the geeky part speaking, the geeky part of mine. Obviously what you said about the oracle is like perfectly well-formulated summary. For people who are interested in trading, in executing transactions, and like who wants to know what happens under the hood when it comes to contracts. This whole oracle thing is like, I'm a Bitcoin geek to the core, so when people ask me about oracle, I get started like Satoshi Nakamoto. Yeah, certainly. Satoshi Nakamoto solved an issue back then which everybody thought was not solvable, and that was the Byzantine Generals' Problem, which after he solved it was renamed Nakamoto Consensus, because what he did in this consensus part was so groundbreaking, so over the top, it was so unexpected that it was renamed after him. And people just gloss over it that just as a side note, just as a footnote, this guy solved another problem, which is an even bigger problem in computers and information technology, which is the oracle problem. Like, the consensus, what he solved, was big enough because we never figured that it was that we would be able to have consensus in a decentralized manner between unknown, like anonymous parties. Now, there is another issue in computer technology, like how do you handle an information that is needed inside a system if the information comes from outside the system? And just to shortly touch upon what Satoshi Nakamoto did there is His proof of work itself is based on some spent energy in the real world. So he built a system which is self— how do you say this in English— self-concluded. It's like a, it's a closed system. However, it confidentially, it really knows how much energy in the outside world was spent in order to get to the next block, if you will. So this is, again, this is a huge thing because it is a specific solution of the oracle problem, which again is an issue that how do you handle an information which comes from the outside world. So that's why we call this, this oracle that you said is basically for your intents and purposes the Bitcoin price. It is a number that comes from the outside world because Bitcoin as a system knows lots of things but it doesn't have a clue what happens outside the system. So if you want to make, like, if you will, a smart contract, or if you will, any kind of a calculation based on a Bitcoin price related to USD, euros, whatever, or for that matter, or a classical outcome, Barcelona-Real Madrid, or whatever bet you want to make on outside data, you need an oracle. And that's— that oracle is just an entity who tells you what the outside event has been or what it is at the current state. So that's, that's pretty much an oracle for, for a really geeky guy like me.

All right, I'm glad I've got you wound up now. I think for me, one of the, you know, as we were building this, there were all kinds of cognitive unlocks, and this was really kind of foundational, and I don't think it's intuitive to people. Like, there is— on the Bitcoin blockchain, there's transactions, and those show UTXOs moving between addresses, but there's no price attached on the blockchain to those events. And so all you really see is UTXOs moving and there's no price. The price happens outside. And again, to me, that, that wasn't intuitive initially. It was, uh, something I discovered. And so, I mean, we're talking about decentralized and how the Bitcoin network doesn't know the external price. So can you get into some detail as to why we need have this price for US dollars and for Bitcoin and for the Real Madrid game?

Yeah, by the way, if somebody is looking for a price, you could tell them to take a look at the current proof of work that the system did, and then thanks to Satoshi's solution, you could in fact kind of tell what the price of a given block in energy term was, because that's pretty much something that the price might be. But we as humans, we don't like to calculate in energy. I mean, and it's counterintuitive, it's just way too— it's just, that's, that's not the everyday experience we have. So if you want to have something that we use more to indicate value with, then you obviously want to have some fiat currency or even a gold price expressed in something different than Bitcoin, which is from the Bitcoin system point of view just a number. So you just want to assign a number to a specific amount of Bitcoins, and then you says— then you say that this is the price of it. But again, in order to have this number interact with the system, this is why you need to use an oracle at the end of the day. And this whole Real Madrid-Barcelona thing, again, it's just another way of showing that value is— and in my opinion, value is so subjective. What do you think? What do you mean by value? You value something, somebody else might value something different. And at the end of the day, The whole point of currency and money is that we come to, how do you say that, to at least like to common denominator on determining what the value is, which if everything goes well will be Bitcoin itself at the end of the day. But that's probably a topic for another discussion.

So, I mean, I guess where some people maybe get confused is when they hear Oracle. They're thinking it's kind of some centralized power, some custodian. And so our oracle, again, is not involved in holding funds at any point in time or custody. What is the oracle's role?

Oh yes, okay. So this is, I mean, this is crucial. We need to understand what you expressed at the beginning is What we already touched upon in our last conversation, what Nick Szabo said, that obviously this will be a trust-minimized system, because if there is an outside oracle telling you something about the real world, you obviously will have to trust it on some level. So—

Can I interrupt you here?

Sure, sure.

To me, it's very basic. The level on which they're trusting users of Kadena versus other potential potential centralized. They're trusting Kadena— or not Kadena, they're trusting the oracle to be able to provide a price for Bitcoin. The alternative with the old model is to trust a custodian with their Bitcoin. And so that's the level of trust we're talking about— the ability to produce a price for Bitcoin versus the ability to give somebody your Bitcoin.

And this is absolutely crucial. And this is, yeah, I mean, if there is a real value proposition on Kadena's side, that's it. Like, in the old model, you gave your money to somebody and then the execution of whatever outcome was resulted by a given price was then supervised and executed by this centralized party. No, this is— no, this is—

are there any conflicts there?

Wow, I wonder. I mean, I wish there was another solution. I really— I mean, this is, I mean, this is crazy. I mean, I'm not an expert in financial stuff, but, but there are like very strict regulations against this, but the system wasn't even, even able to solve it at the very fundamentals because the technology, the knowledge wasn't there. So I mean Yes, you trust your money to a centralized party and then you just, I mean, you just think it over. Like, what can go wrong? Jesus, really? So what Bitcoin proposes—

In this space, has anything ever gone wrong before?

Never. I mean, I haven't ever heard about anything fishy right there. So instead, this whole oracle thing means that you hold on to your money. It's you and your peer. And if you guys both trust this entity— and when we talk about trust minimization, we will come to this, but it's like, it's really an order of magnitude below whatever you had to trust in an old system before— well, then you are on your own. What, what I remember when we discussed it at the very beginning of our co-working is like, Just imagine yourself being a pirate, and then you are meeting another pirate on an uninhabited island, and it only comes down to what you two discuss, what you two agree upon, and that's all. And through magic, through cryptography, and through Bitcoin, the system enforces that what you agreed upon in advance at the beginning, none of you can screw the other one. If the oracle produces the price. And the point is that the oracle, like, the main point again is that the oracle is, he is not taking part in your contract. The oracle does not know that there is a contract that you're making. It doesn't sign anything inside your contract. The oracle is never contacted, not even in the beginning, not even at the end when it comes due. It It's just an entity shouting prices into the air that anyone can listen to, but it's not just shouting the prices into the air at one time when it attests to an outcome, but it previously announces that it will shout it out later at some time. And this announcement of a later shout-out is what you can then with your peer pirates— and I'm, I'm using this pirate analogy because the point is pirates do not trust even each other. So this oracle setup enables you to have a contract with somebody who don't— who you absolutely don't trust, don't know. Because this Bitcoin setup with the oracle stops anyone from screwing the other party and doesn't need a centralized party. So that's basically it.

There's pieces I want to touch on here, but I want to kind of move on, and I think— All right, come on, because Within Kadena Bitcoin, one of the things we're saying is that the oracle signs the outcome, and so trying to understand what that means in practice, and we're touching around that. What is the event that's defined in advance and what gets signed later? How is this happening?

Right. So just imagine you and me wanted to make a contract and we know that in a year from today, we want to know the exact price of Bitcoin in USD. So we go to an oracle who right now tells— again, everybody on the planet, anyone— tells that in a year from now, it will attest to a USD Bitcoin price with a certain signature. So we know in advance that in a year we will have an outcome from this oracle. So when we both go to this oracle right now, we fetch an announcement of an upcoming event. An upcoming event, what you are asking, is, is the price, is a number that we expect to get from the oracle in a year. But again, I really have to be careful about my formulation. It's not that the oracle— I mean, first of all, we're not asking the oracle to give us anything. We're fetching public data to implement it into our contract right now. Then in a year, the event, the price then the US price then of Bitcoin is again, it's not that the oracle whispers it to us. It's also not that we send our contracts to the oracle to sign it. The oracle will not interact with our contract in any way, shape, or form. We just know that in a year we will have an outcome with using the signature that was announced right now in advance. So when we talk about trust minimization, This whole trust in the oracle, actually it means that we trust that the oracle will produce an outcome. That's the first trust. And then obviously, because we are not totally stupid, we ensure ourselves that if the oracle is down, if the oracle doesn't tell us anything, we still have a backup plan, namely whatever we agree upon, we just redistribute the funds if necessary. But if everything goes well according to the plan, then in 1 year's time there is going to be an event that was announced, that we use the announcement for, and then this event will be attested to by the oracle with the cryptographic signature. Meaning, which is also very, very, very important, that the oracle cannot attest to 2 different outcomes. And if you think about how 2 pirates could screw the system is Like, one of us goes to the oracle and tells the oracle, well, hey buddy, just for me, for me privately, can you please produce a different outcome? And in a normal system, why wouldn't you? You can blackmail, you can force the oracle, whatever. But the solution that this DSC and the oracle provides is the oracle needs to reveal its private key in a way. So if it makes two different attestations, then the oracle itself is compromised. So we have— the oracle has a way to, to prove that it is attested, that an outcome was attested to, and also prove or make you really believe that it's not going to produce two different outcomes. So much for the pirates.

The, uh, one thing I want to touch on a little bit is we talked about the price of Bitcoin, and I, I think, you know, as a market participant for a long time, there's no price for anything, right?

Yeah, that's, that's true.

There's a bid offer, you know, depending on the amount you're doing. And so price is a very, very difficult concept to get to. So I guess the question is, how are we getting— how is the oracle determining what the price of Bitcoin is? And to me, this is kind of infinitely fascinating and like a regulatory market structure question that, you know, hopefully one time in my life I'll have time to dig deep into. And then secondly, kind of related to that, you know, in the white paper we've talked about some of these issues that the oracle doesn't know what's going on. Like, again, it's just essentially a price feed. But for me, one of the key features, the key feature of the oracle is reliability, right? Because if there isn't a price then that creates some issues. And so maybe you can talk a bit about the construction of the oracle and how we're able to make a reliable price. And then, like, you know, the things we think about are like, hey, if there's no electricity in the world for years, what happens to the oracle and our ability to produce a price? Um, and these kind of really distant situations that, you know, could potentially exist? How do we give our users comfort around those things?

Right. So first, uh, yeah, I mean, it's absolutely true that there is probably no price for anything, even if it's centralized. Like, you can say that's the gold price, but you probably get some, some gold somewhere else for a different price. Or like, for anything, there is probably all the time a local price, even more so for an asset like Bitcoin, which isn't even centrally regulated. So yes, the question of price is an interesting one, and like I touched upon it in our last conversation, if we can kind of make it, make our clients comfortable to just think about a DRC of, in terms of you are in fact betting on what the oracle will tell you. And this is, on the mathematical level, this is— that's, that's the absolute truth. Like, what the oracle will tell you, that's actually what you are betting on. You're not betting on a price because, A, we discussed it, there is no such thing probably that has a Bitcoin price, and B, the oracle itself has to come up with a way to calculate it from different sources. So for the concrete case, we have an Oracle site, and on this site you can actually take a look at how the USD price is calculated in any given time, any given point in time, which is also like— it is very important because at the end of the day, what our users can check again anytime is that was the price that we predicted actually based on the formula that we said we would use this time, 10 minutes ago, 20 minutes ago, 50 minutes ago? Because that can build trust in our users that the oracle actually does what it says it should do. Now, we are not talking about technical issues yet. We're just talking about if we have those price sources, we aggregate them, We use some mathematical formula to arrive at this price. And the good thing is that because a DRC can— like the two pirates, if they agree, they can all the time, any time, execute a transaction even without the oracle. So if they know that the formula that is announced by the oracle is used to get at the price, then they can do the calculation by themselves. And if they got the same number, Well, they just settle on that number. So the oracle calculates a given formula and, well, right now at this current point in time, as far as I can remember, we're using 5 or 6 different exchanges. Then we do some mathematical tricks, which is again fully transparently displayed on our oracle site, and then this is the price that we consider to be the price, because this is what the oracle will attest to at the end of the day, or at the end of every 10 minutes, or whenever we say the oracle produces a price. This is the way the oracle derives this price from outside sources.

So two things I want to touch on related to this. We initially had open-sourced the oracle. And I think we have a desire to open source as much as possible. And we made a decision to not open source that after a period of time. Do we want to talk about that?

Yes, I would. I would really, because this is as well something that we discussed before. Just so our listeners don't get confused about what we mean by open sourcing. The source code itself is not open for the oracle, so the oracle itself is not open source. But again, uh, the, the main point that you like, like the, the absolute truth that we have to keep in mind every time we talk about the oracle is the way a price is calculated is totally transparent and again verifiable. You can check it, and if you take every little price point the oracle ever produced and compare it with the formula, and you substitute the inputs into this formula, and you see that for the last 1.5 years, every, every time we produced the price with this oracle, it was the same outcome that you would arrive at if you did it by yourself, then you build trust. So this is a very important point. The oracle itself, the code itself is not open source. That's a technology. That's also a technology which is, in my opinion, like, we would love to have outside oracles. And we're also offering this technology to someone who might want to use it, because we obviously had to build it and it is working quite well. So our oracle is, first of all, right now the primary oracle that we are using, but as a second aspect, this is also a proof of concept for anyone who wants to run an oracle. And just to touch upon the last point that you were mentioning, this issue about the oracle going down or technical problems. Yes, we are handling it actively. It sounds more scary than in reality it is, because if you are thinking about what the oracle does, the oracle signs outcomes. Even if the oracle itself is down, Even if I right now cannot go to your oracle and fetch the data, well, maybe they're live in 3, 5, 6, 7 hours. And once they go, once the oracle is live again, the data is still there. So it just means that if you guys have a debate and if you cannot execute the contract, well, then you will have to wait until the oracle comes online again. Then you fetch the signatures. The signatures enable you to spend your money. So it's not that if the oracle is offline for a period of time, then disaster sucks. Nothing really happens. Your funds are secured by Bitcoin, and they're not going to be stolen, they're not going to be spent. Maybe you have a small window of loss of access until the oracle comes back online, provided you are in disagreement with your peer, because if you agree, you can still fulfill the whole contract without the oracle itself.

So as the architect now, there's a lot of customization, or I guess things that can change within the Kadena structure, from currencies to to tenors, to different things related to these contracts on Kadena. One of the features that can be changed, kind of with the dropdown menu, would be having a different oracle. So as of today, we have a single oracle. Can you talk about the roadmap and how we can potentially support additional oracles over time? Yes, you know, again, I'll say this question and I'll give you the floor, but you know, Kadena is really not— I mean, I'm interested kind of intellectually, but I'm not really interested in the Oracle business itself, right? It's a different thing than Kadena.

Exactly, exactly. I mean, that's, that's, that's true. I mean, uh, On a conceptual level, the further away the oracle is from us who are— well, we are basically providing the technology and the marketplace. So we are not active participants in the contracts anyway, but it's still even better for us if the oracle is a bit more distant. Now, with this in mind, The original concept, like, uh, the guideline along which we built Kadena was that we are Oracle agnostic. We assume syntax, timing, content of the Oracle data that we use for announcement and for attestation But if those are there, we can use any oracle provided by anyone. And it's not us, I mean, we're not the clever guys here. It's the two pirates, it's our peers who come to the conclusion that they want to have a contract using a given oracle. So yes, our plan for the future, for the near future is to enable our users to pick from different oracles, even if it's for different symbols. Like, I mean, even if you are using different currency pairs or God knows, again, football outcomes or whatever. There can be oracle for this and our technology is at the level. Like, we did everything on our side to enable whatever type of oracle wants to come in and provide data. We can give, uh, we can, we can give the floor, we can give, uh, the option to anyone to bet on outcomes from different oracles. And this is in fact actively now what we are trying to achieve, that we have multiple oracles. And, uh, well, I think, uh, our user interface right now is really intuitive and it's slick and it's elegant. And I think if you just pick your oracle and make your wishes for whatever DLC you want to sign, then the other side hopefully will find you easily. So I don't see a problem there from a UX point of view. Yes, we want to use as many oracles, and we can also use and serve as many oracles as there are out there.

All right, I think we've touched on a lot. The Oracle's role, how the Oracle works, some of the issues, some of the security, some of the fallbacks. You know, I don't know that I can add anything to this discussion without muddying the waters further. I think we're always available for clients who want to talk about the Oracle or have specific questions. Is there anything you want to say about the Oracle in wrapping up?

I would just wrap up with a single— I would urge participants in our ecosystem that to be an oracle for certain businesses might be of great advantage. Because if you are serving people with an exchange option, like that you can exchange Bitcoin to any kind of fiat. That means that you on some level do have a price, even if it's notional. Anyway, you have a much better notion of what the price of— what the so-called price of Bitcoin is than anyone else. Plus, a big plus would be that if you already give the price, then my user would confidence— would be confident to know that the price that I'm reading from an exchange is the actual price that I can buy my Bitcoin or sell my Bitcoin at. So what we are actually looking out for is, is, uh, some sort of exchanges that might provide us with data which then could be used for the actual exchange before the transaction or even after transaction comes to.

Uh, yes, the holy grail. There's always other things to work towards.

Oh yes.

We have lots to do and get a lot of good suggestions, and mostly we're already working on them. So thank you again. I think you've made a lot of— answered a lot of questions and made things much clearer for a lot of people, hopefully. So thanks again, and I'll just wrap it up with that.

Thank you, Kevin Bell. Thanks to our listeners. See you.

Goodbye.