Search
  • Bianca Chen

Exclusive interview with Emin Gün Sirer:We're going to disrupt the world

In the fourth episode of Next: Blockchain, we interviewed associate professor Emin Gün Sirer from Cornell University. As a professor and a computer scientist, he is known for his contributions to peer-to-peer systems, operating systems and computer networking. As a blockchain expert, he created the first crypto based on proof-of-work - Karma System. Now he is launching his own cryptocurrency and blockchain network - Ava -, which has attracted $6 million from major investors such as Andreessen Horowitz, Polychain and MetaStable. Is Ava a bitcoin killer? How Ava is going to disrupt the world?




Bianca:I heard you just launched a new token. Can you tell us more about it?


Emin Gün Sirer: We're building a new system called Ava. And the core idea behind it is a new consensus for the goal that completely revolutionizes how cryptocurrencies are built. So, within a second or so the protocol is able to say this is accomplished. You do not have to wait for, in the case of Bitcoin, six confirmations in 60 minutes. You do not have to go through the very expensive process of mining. So, it is a good store of value because there is no leak of a value from the system. There is no mining being performed. None of the miners have to pay enormous power bills. It is a very efficient and it does have very high throughput. We just demonstrated six thousand transactions per second. And that's works for thousands of minutes. This has never been done before and people thought that achieving this level of throughput was not possible. It just wasn't possible with what we knew.


But this is possible when you change what you are doing. So that is what Ava does. And the final special feature of this protocol is that it allows enormous numbers of participants. You and I and anybody else can be the massive miner of Ava.


We can participate first hand in the protocol. You do not have to be advanced with power companies. You don't have to have special relationships with mining rig manufacturers or with hydroelectric power providers. You do not have to have these weird and bizarre things. The value of the cryptocurrencies comes from this incredibly decentralized and the incredibly distributive ledger.


If you look at the existing systems they are not distributed or decentralized sufficiently. Bitcoin has only 19 mining pools in theory, but, in fact, it is only 11. But in Ava, you can have thousands, maybe tens of thousands ,maybe millions of participants. So it's inclusive to a degree that nobody else has ever done. It's fast to a degree that nobody else is able to reach. So we're very proud of what we have just built. It's not public yet as of this moment. We are in the private testing stage, we plan to make it public as soon as this is ready to do so, and it should be soon.


Bianca: Sounds like it is more inclusive and it is faster. So how can you achieve that goal? What are the mechanics behind it?


Emin Gün Sirer: It works differently from everybody else's protocols. That entire mode of operation is different. So, people who sell you the shovels will tell you that the shovel was apparently limited in what you could do. But they will say ,look ,this is a shovel there and there always a tradeoff between how long you make the handle and you can make the handle longer but you can't really lift it as much as etc.. That's all true as long as you're within the shovel family, right? There are people with short shovels and long shovels, wide shovels and thin shovels. And then suddenly somebody comes in with the steam shovel and that's what we really do. So we have this proof of work protocol. It's very robust but it doesn't scale. And Bitcoin is stuck. It's intellectually stuck. And in fact it's so bad that the people in charge have given up on it. They're trying to develop layer two. They are saying, look layer one, we have to freeze layer one, and we don't know how to make it better, we can tweak the values. It's not going to take us to where we want to go. It's not possible to reach people in these numbers. So they're saying , look, we're giving up. We're going to go to Level 2. And so then along comes the Avalanche paper which is the basis of this new token called Ava. And it shows the world, you don't need mining, mining was a kludge. It's an afterthought. It's sort of an evil necessity. Like it's something Satoshi came up with to achieve its goals, but now 10 years after Satoshi came up with it. So that's a devil's edge that you brought into the picture, a completely different way of doing consensus.


What level of detail on the technical front if anybody wants it, but very roughly speaking, there has been 45 years of distributed systems research. There have only been two families of consensus. Now I should define what the consensus protocols, this is the system of messages or the way of messaging so that your network remains in sync.


You want all of the loans inside the network to agree on how much money every participant has. This is the crucial point. If they disagree your system fractures and you know you've got some people saying oh you've got five dollars and somebody else saying you got fifty dollars. Which one is it?This should never happen. We knew of two approaches to this problem.


One approach is you designate certain nodes, you can call them anything, you like you could call them “Validators”. But they're specially designated nodes and these protocols work by having every validator talk to every other validators. And making sure that we all agree on how much money you've got. This is an okay idea. We call this approach classical. And the two people who came up with this approach both have Turing Awards. It's a brilliant idea. It's great in many ways. It's very efficient. You just have to have about 100 of these guys and they talk to each other. Then that's great. Now if you try to have more than 100 of them, then the system doesn't work. If you have a thousand of them and they're trying to talk to a thousand other people. That's a million messages being sent. It's not going to scale, though. Everybody talks to everybody else. So, when Satoshi came along, we were only aware of the Classical approach. He looked at them and he said, look, these systems are completely unsuitable for an open currency. They require everybody to agree on a set of designated record keepers. And it's very fragile to agree on this. So he said don't be doing this instead of doing this mining trip. So this is what he says, is spend a lot of energy coming up with some kind of a lottery and somebody every now and every 10 minutes on Bitcoin. Somebody says, hey guys I have the winning block. I did this hard work, and I will now transcribe what's happened in the financial history of the world in the last 10 minutes. And for this I will get the reward.



This is an OK idea with at least to an arms race. Everybody's trying. All the winners are trying to mine. They're competing with each other. The ones would have lower energy costs tend to win. Certain miners have access to the latest hardware and they attempt to collect all the power. So, these systems have been incredibly centralized, Satoshi that put that discipline point. So Avalanche works differently from these two for the protocols. It is not based on everybody talks to everybody. It is not based on the permitted selection. It doesn't have specially designated nodes. It's just that it has everybody participate. But it doesn't have meaning. So how does this work. Essentially what happens is it's a probabilistic protocol where every participant talks to a subset of other nodes and tries to measure the center.


So imagine I'm in a very very big stadium. I randomly select a couple of people from the audience and I say hey guys what do you think about this transaction. This is a good idea. And if the answer is positive. That's good it makes me feel positive about the transaction. And they repeat this process and so does everybody else at the same time as me. So. You can easily see that in a good case, this protocol is incredibly efficient. Some of the issues are transaction. There are no conflicts. And at some point you've asked so many people that you have a good statistical basis for saying, look. I've sampled enough. And it looks to me like this network is ready to accept this transaction. And there are some sub-corner cases that this is where it gets interesting. In some kind of cases like you might have to make a decision. And what this protocol guarantees is that all of the corner cases are handled properly. That is there is never any likelihood of any two nodes and it to correct nodes that are disagreeing with each other with high probability.


So what does that mean? That means that we can build really large networks where everybody can participate. All of the participants are asking each other, hey guys you know did Alex pay Bob or did Alex pay Charlie, or you know what do you think. And after a modest number of rounds, they were able to come to a consensus decision. The math works out and that in the very very large stadium, let's say hundred thousand participants. But every node is asking about 10 nodes. After about 20 rounds or so, after about 200 message, people are able to start a committee. So this is in great contrast with other techniques where one hundred thousand people would each have to ask a hundred thousand other people.


And that's completely untenable. So I think with a hundred you'll get the message, and that is outstanding. So this is a very big breakthrough I think. It doesn't happen all that often historically obviously, it's happened only three times in the history of mankind. So that's kind of a nice feature.

And it's only changes the game, so we can have really large networks. They don't have to burn electricity and they're very efficient at this process after only a modest number of rounds, which only takes about a second or so. We're able to achieve finality and we're able to be more decentralized than bitcoin and much faster.



Bianca: But is this Safe? Do you think your way is efficient but also safe?


Emin Gün Sirer: Absolutely. The safety requirements for this process for this protocol can be set. It's under your control in the same fashion that bitcoins safety is under your control. Bitcoin. You have to wait for a certain number of transactions for a certain number of confirmations. So typically people wait for six confirmations, so that after that, your transaction is buried under six blocks. It's incredibly unlikely to change. So just like that, in this protocol we called Ava, what the participant can do is wait a little longer. So you can set the likelihood of a safety failure to any value you like. So typically when we run experiments we set this value to something really really high, something like we would like to see a problem once every 20 thousand years.


Bianca:Do you think Ava is better than Bitcoin? And do you think of Ava will replace Bitcoin? And what kind of application do you think Ava can bring to the world?


Emin Gün Sirer:I don't think anybody is going to replace Bitcoin. We all worked really hard to make bitcoin what it is. I know a number of people I briefed personally. I know the number of politicos I personally agreed with. it's a huge number. It's got a lot of goodwill built up over the years, and despite the stagnancy of the coin, it's not going to go anywhere, it's fine. It's not going away. It will be where it is. It will always be around. Its price will go up and then down and then up and down.


As I said, my competition is the other coins. We're not trying to be a bitcoin killer. We're going to all rise together. We will all stay where we are. So we approach the one trillion dollars with absolutely crappy technology. Now that we have much better technology I see no reason why we can't break a trillion. And that's going to happen with an influx of new money. And it's not going to happen with speculation, it's going to happen with real use cases. And the use cases we couldn't even fathom building on top of a block chain we now can't. So this is a brand new world. So yes I don't see bitcoin as competition. It is what it is and I'm just building my own stuff with my team as well as other people who have partnered with us. It's going to be fantastic. We're going to open up new areas that are untouchable for the incumbents.


Bianca: You attract a lot of investment so far. I read something about you having raised almost six million already.


Emin Gün Sirer: Yeah I did. I wouldn't say that's a lot of investment. That's sort of, it's our seed round. But as far a seed around goes, you know in the old days it used to raise one hundred fifty thousand dollars or two hundred thousand dollars. So we ended up raising a sizable seed round from some very good investors. We're very happy with our investors and with the signal that it sends to everybody that all these people who know what we’re doing, and we know what we're doing. So it's time really a chance to build a community. To get the influx of people who were building systems. We had the first explorer, we had Satoshi and his brilliant creation followed by a big wave of craziness that we don't believe, now with something like eighteen hundred different kinds of tokens in existence. Most of these coins have got nothing going for them. They're just copies of Satoshi's version with just small changes here and there. There are lots and lots of noise around them, you know, you change one little thing it's a big deal. These people bring nothing to the table. These are all substitute goods. They're all the same damn thing. None of them actually work. And in some cases like we've even had, completely centralized, what I would call embarrassment things like they were alive for many years. These are laughable laughable systems. It's time for the for the people who know what they're doing to come in with actual innovation and change the system.



Bianca: Do you think of this is complying with SEC rules?


Emin Gün Sirer:Luckily we are as aggressive technically as we are conservatively. Everything we have done has been double or triple checked by all sorts of consultants. And it's of course by the rules of traditional standards, risky financing all around. We are an American company headquartered in Delaware and Singapore. We're not in the funny jurisdiction where laws on the island go no farther. I'm not Fred Siegel in the Mongolian steppes in search of my jurisdiction over you. You know, Like an area that is kind of mark with dashed lines on the map. Nobody knows what laws apply. We're not doing any of that and we are standard American companies that we're complying with the world's best business interests.


Bianca: Interesting and I am also curious about once Ava goes public. How are you going to launch it? Are you going to have a mechanic like mining for people to just mine out their own token or you are going to issue a token like ICO kind of thing?


Emin Gün Sirer: We have no notion of a mining protocol. We don't want to. We need to get these tokens in the hands of as many people as possible. So they can participate in the system and they can participate as takers. They can use the system. I need to do justice to the technical innovation. I want to get these tokens in the hands of as many people as possible. We're looking for the legally safest and soundest methods of doing this. Sadly the state of regulation in the US is such that there are no bright line rules. This is a technical term that I just recently learned is a legal term. And it means that the regulators haven't said if you do this, you shall be okay. And so instead what they prefer is they like to look at the facts and circumstances, and so they want you to sort of chart the path for yourself. And if you stray from what they consider to be a good path and they might come in and punish you. This creates a lot of uncertainty. So, I don't believe there is a single person in the US including the FCC commissioner himself who is willing to tell you that if you do X it's okay. So it's a bit of a murky situation. We're not the only ones in this situation but we're in talks with all of the regulators.


Bianca: And how you're going to use the money by selling the token to the users.


Emin Gün Sirer: So the money raised from the sale is going to be used for building the ecosystem. And the platform is step one for us and it's amazing. It has great features as I talked about. But it's only step one. So our goal here is our ultimate eyes are on the end goal. And here is the end goal. It is to build these decentralized applications that will bring trillions of dollars on board into a digital ecosystem. That's not happening right now. It's not happening partly because the current technology performance is scale. And we talked about that at length. I go to repeat that one. Well it's also not happening because the current technology is functionally limited. You can't build a good exchange rate on the market on the blockchain. All of the decentralized exchanges are absolutely horrible to use. There have no liquidity. They are open to fraud trading attacks left and right. That's not a good foundation to build a new world. And I can name a number of other areas as well.


So the money raised will be used to go into these different areas. And show the world how it's done. We have a track record now of innovating at the hardest level and the consensus level. We have a track record of building a new kind of network that I didn't really talk about during this conversation. We plan to also build really interesting what will be called verticals that show the world how these new financial systems can work. Some of these vticles are highly capital intensive. So for compliance reasons or for stability reasons or for liquidity reasons one might need large sums of money. Time will tell if they can raise those kinds of amounts. We want end up partnering with people where there are existing credible innovators and we would love to partner with them. There are some on top of the ecosystem. I would love them to build on top of us. But in many different areas I just see less art. I see a lot of people opening up the coin market cap and looking at the price movements. I see a lot of people making noise and not building. And But we are aware driven team and we are the aggressive team and technically competent. And we have a lot of tricks up our sleeves so that is where Ava came from. There are many many other tricks and so I just can't wait to get to the polls. If you ask me What are you going to do with that money. My short answer to you is we're going to disrupt the world.