(Interview English Transcript) Matthew Spoke: AION Talks Blockchain 3.0

Interview Link

See Eun Ha Minseok Matthew Kim Matthew Spoke

This is your host See Eun and Matthew. We are the Blockchainers and we come from the heart of Seoul, Korea bringing you all that goodness about blockchain cryptocurrency news and interviews. We’ve got what you need. Today, on the Blockchainers hot seat we have a very special guest he is the greatest player of all time. Please introduce yourself the greatest player of all time.

Hey guys, it’s good to be here I’m Matthew Spoke from the AION project I’m dialing from Toronto Canada. It’s good to see you guys again.

It seems like your reputation precedes you. What kind of things have you done at all these like different conferences, Matthew.

We’ve been spending a lot of time over the last few months. Just getting around. We’re trying to spread the word. So I’ve been traveling a lot going to different cities and countries talking about what we’re working on. Obviously back home the team is developing and getting the first releases of our roadmap out the door. Our testnet is launching today which is pretty exciting for us. Big milestone, so we’ve got a lot more coming up over the next six months. We’re just doing a lot of conferences and meetups. When we go to cities we’re not only doing conferences and speaking but then we’re also trying to host community meetups, so we get a chance to talk to people and get feedback and spend more and more time with developers because we’re really trying to shift the conversation towards the tech rather than the coin which is a big transition.

Let’s talk about that a little bit. I think for all the projects that is one of the the problems that they’ll face soon enough once the marketing hype dies down. How are you going to have a healthy ecosystem of developers and how do you have an healthy ecosystem of developers continually flowing into a particular project? What have you guys planned for that?

I think the big thing this year is like making sure that you’re actually delivering on the promises of your roadmap. There’s a lot of projects that had really ambitious roadmaps, us included. Now there are objective measures: will you be able to build. It’s a challenge we’re going through daily. We’re trying to hire a lot of people really quickly and train them as quickly as we can. But it’s not that obvious because it’s really complicated problems so that’s going to be a big differentiator between projects this year. Who’s actually living up to the expectation.

In terms of attracting developers, the good thing is that developers just get attracted to cool new ideas so I think if you’re genuine about what you’re building and you’re actually delivering. They’ll be organic adoption and following just because people are curious. That said we are doing a lot more meetups and a lot more developer events. We may do some hackathons this year. We’ll host the first AION focused full-day conference all focused on talking about our research talking about our tech and then increasingly we’re getting a lot of projects that are asking us if they can build stuff on top of the AION protocol. So we’ll be looking at some third-party projects and companies that are launching their own apps or launching their own tools on top of our network. So that’ll kind of kickstart that which is pretty helpful and then we’ve maintained a pretty good balance of tokens to use to incentivize the community to pick up what we’re doing. So we’ll have code bounty programs and DApp development for bounties and if people can build visualization tools and the network will have a mechanism to continuously reward them for doing that so we keep attracting people’s attention.

Based on your interviews, I heard that AION has a very unique vision and very unique process for hiring developers. Could you describe the person you want to hire?

I think this is part of the fun and part of the challenge. We’re trying to keep our quality and expectations really high when we hire. Most of our engineers, they come through a process of probably four or five steps by the time we hire them including a pretty in-depth, full-day technical challenge where they’re in the office coding some sort of challenge that we put in front of them, so that we can see the quality of their work when they join us. But there is this expectation that they’re starting for the most part with very little background in the blockchain space. So that means we have to train them on a whole bunch of new concepts. It might be engineers that have done work it at the operating system level in different contexts, building on top of building VMs and compilers and but not necessarily work directly in a blockchain environment.

There is some learning curve so that means that although we’d love to quadruple the size of the team in a month. We know that we have to grow according to a plan because if you grow too quickly there’s not enough people there to train the new people and then we end up just becoming unproductive. So we do have a big focus on that quality. Now we’ve got two parts of our development team, the one that’s focused really on our kernel, the core of the operating system. This is everything from like the peer-to-peer protocol to the consensus algorithm to the VM to our bridge design. We’ve got people building tools and applications on top of that kernel. So we’ve got some projects that we’re working with that are in the process of building their own apps and potentially will run their own ICOs on top of AION and then we’ve got tools that are necessary for any one of these ecosystems such as wallets and explorers and dashboards and that type of thing. Depending on what type of skill set the developers come to us with, they end up falling into one of these two teams. Generally and obviously around that there’s QA (Quality Assurance) and user acceptance testing.

In another interview, you said it’s quite challenging to be where you are and trying to hire developers because Toronto’s well known for a being an AI city. The blockchain space has to compete with the AI space in order to get talent. How’s that working out?

I think Toronto’s a great city to be building a company. But, definitely it has some limiting factors and not just because of Toronto, but because this is just the reality of building in a single place in a single city. You’re always going to run into the constraints of how many people are on the market. Toronto happens to be a big hub for AI which is great it also happens to have a lot of big offices of tech companies so there’s a big Google office, Amazon office and Facebook office that sucks up a lot of engineering talent out of our top schools. So it’s a very interesting problem right now for engineers that are curious about cryptocurrencies more and more. We’re naturally getting people interested because they hear about it, they’re starting to trade, they’re starting to buy Ether and Bitcoin so they want to get involved. They want to apply to a kind of newer crypto company so we’re getting better at that. The second thing though is we’re looking at how we grow outside of Toronto. So this year there’s gonna be developing teams in other cities around the world to kind of augment what we’re doing here so we’ll have like a part of our core team in Toronto but we’ll also end up having a part of our core team likely in Europe and in Asia. We’re still working on some options in terms of what cities we place those teams in. But we can’t grow this to scale in a single city so we’re gonna end up being in a few different places.

We talked about AION where we assume that the listeners know exactly what AION is. A lot of our listeners I think are actually the semi-technical crowd who have some idea about AION. Can you give us a high-level summary of AION project and your elevator pitch that you’ll make to persuade developers to join AION?

The high level pitch is we focus on a problem in the market that we think is not super evident yet, but that will become very critical. That’s the problem of multiple blockchains communicating with each other and so right now we live in like the early stages of an industry where we can get away with poor performance and poor scale and low transaction. All these things are not the end of the world because there’s not very many people using cryptocurrencies but we’re trying to envision what does the future look like when the world is using this technology so then at that stage the big layer of infrastructure that we focus on is how you scale across multiple blockchains and that means being able to build these communication protocols that move value and data and information between blockchains. If you wanted to go build a solution or an application today you’re building it on top of a single network: on top of Ethereum or on top of Qtum or whatever. We want to get to the point we’re building apps in this environment not restricted to a single network where you can have a token that moves from one network to another. You can have a logic that moves from one network to another and you just open it up a lot more for the creativity of app developers. Whether they are app developers in the enterprise market that are working on business process transformation or app developers that are just trying to build new cool business model.

In both cases we think that this layer of interoperability is going to be a big obstacle. More and more we’re seeing that there’s quite a few projects that kind of followed us into this space because they realize that there is a big gap here. We’re pretty optimistic that this is going to become an important layer and now there’s just a couple of different approaches on how you solve it. So we’re trying to make sure that we maintain compatibility at the app layer for our first release. It’ll be completely compatible with Solidity programs that would run on top of Ethereum and that you can move over to the AION Network, but more and more we want to get people thinking about where our bridges fit into their app designs. Why would you want to build an application that ties into two different blockchains or three or more different blockchains and how do you build that concept directly in your application is kind of our long-term thinking.

One of the things that you mentioned is the possibility that is unlocked by building on AION’s blockchain. Because AION’s blockchain offers so many avenues into all the different bridges and the different functions that you can tap into all the different blockchain ecosystem when you’re building a truly decentralized application. Do you see AION-1 project being the place to start building DApps? Can people build DApps that use the functionality of multiple blockchains but it doesn’t necessarily have to be on the AION-a blockchain it can just be somewhere else.

I think both are possible. We’ve got kind of two types of projects that we’re talking to today. One, that’s looking to build something directly on AION-1 so they’re looking for a starting point and they say hey I’m gonna build my public application on top of this network. The other type of project we’re talking to is projects that already have applications built on Ethereum that are looking at how do you use this concept to be able to open up multiple blockchains. I want to expand its reach to more and more networks so that would be the other way to look at it.

It doesn’t really matter to us. We’re just trying to open up the idea in people’s minds that when you build applications you don’t need to think about it as where I can only build applications for users of one network. I can build applications for users of any network at some point. Obviously there’s a lot of work that needs to get done to get there. We need to build bridges that connect all these networks to each other.

AION intends to do that but then after those bridges are in place then all of a sudden I can build an app that’s accessible to users of different networks rather than only an app that accessible to users of one network. Because right now if you want to become a user of an app you have to convert into that coin to use the app and so we want to remove that step so you want to use an app on Ethereum you don’t need to convert Bitcoin into a Ethereum to use the app on it, you can go directly. That’s kind of the long-term goal but there’s a lot of work that needs to get put in place. We’re building our first release and we’re also in the process of building our first bridge and our first bridge is AION to Ethereum. You can start to have apps that are made on both of those networks at the same time and we want to start demonstrating what that looks like as well.

Some audiences might wonder what bridges are and I understand that it’s one of the core components in the AION Network. How do you think about bridges and how bridges will bring the value to the AION network?

Sure, if you’re familiar with the industry you’ve probably heard the term, oracle. Oracles were designed and they’re still being worked on by some cool companies like SmartContract.com, ChainLink, and Oraclize. They’re working on this concept of how you put data into a blockchain and as you put it into a blockchain how do you make sure that it’s valid and authentic. Generally, those designs were paid or focused on pulling data from non-blockchains into blockchains. You’re pulling it from a database or some sort of off-chain system. We looked at it the same way. A bridge is essentially how do you create a decentralized oracle that pulls proofs or data from one blockchain into another blockchain. When you see the intersection of two blockchain networks, you need a way to validate that something happened on the other side. You can’t rely on a single party to do that because then that party becomes vulnerable. They become the point that you might hack or you might go in and change the data. The bridge became this decentralized protocol, so a group of nodes would sit on a bridge and they would essentially watch for transactions. Let’s say on Ethereum and if they all agree that a transaction happened on Ethereum then they could pull that transaction in and transmit it over to AION. Because they’re playing that function, they’re getting rewarded by the network for validating cross-chain information.

They have to go and witness events on one side of the bridge and they have to transmit it to the other side of the bridge. If they do that according to the rules, then they get rewarded for doing that. Those rules are focused on incentivizing decentralization so the bigger the bridge the more decentralized it is, more people have confidence in the bridge so that’s kind of the goal. At every layer of communication, it’s as decentralized as possible. When we originally thought about it if I have a whole bunch of different blockchains, blockchain number one two three and four traditionally the only way to go between them was a centralized structure. You can only go through an exchange to connect all these decentralized systems. Actually we want to connect blockchains with a decentralized system. Logically in our mind if you want to connect blockchains with a decentralized intermediary then that intermediary needs to be a blockchain. That’s why we built AION. We built AION to be a blockchain that connects blockchains so that any point of intersection is always decentralized. There’s going to be different types of decentralization like massively decentralized networks or the less decentralized bridges. But at least you reduce that single point of failure concept and you try to incentivize decentralization as often as you can. People can have confidence that when data is moving from one chain to another. It hasn’t been changed or tampered with.

You just said big bridges and in your whitepaper you mention that between Ethereum and AION-1 there’s a possibility that multiple bridges will exist and naturally some sort of market emerges depending on the reputation or different factors people will choose one bridge over another multiple bridges connecting AION-1 to Ethereum. Do multiple bridges for the same blockchain networks help in scalability at all?

Another bottleneck in scale is the bridge. Let’s say I have a million transactions that I want to move between AION-1 and Ethereum. If all those million transactions are using the same bridge then the bridge becomes one of the limiting factors. Opening parallel bridges actually could offer alternative paths for that transaction to get validated. The original concept in multiple bridges was not that we want multiple bridges but the market may offer different structures of bridges. The market would offer those with different pricing essentially. Somebody who builds a bridge does that on the AION blockchain by creating a bridge registry contract and essentially defining the rules of the bridge. The rules of the bridge primarily are what’s the fee on the bridge how do you pay for access to the bridge and the other nodes can then join the bridge by staking tokens into that contract. But somebody else might say well that bridge is too expensive and because it’s so expensive not a lot of nodes joined it. Now it’s relative centralized and very expensive so I want to build a cheaper bridge and I want it to be more decentralized so then I go build an alternative essentially and I could go and design kind of a separate economic system. We wanted that free market opportunity but we also want to make sure that there’s a scaling solution that relies on multiple bridges so you can remove some of the bottleneck on a single bridge if that makes sense.

When it comes to bridges, what is the difference between a bigger bridge and a smaller bridge?

The number of nodes that are active on it. I think the nodes will be attracted to bridges if there’s enough transaction volume to justify significant fees on the bridge so the fees on the bridge are created every time a transaction crosses the bridge. If I connect AION to some empty blockchain that nobody uses, then there’s not very many transactions going back and forth so there’ll be very little incentive for bridge nodes to participate in those bridges. Because there’s not a lot of rewards so I think the bigger bridges will come as a result of the more transaction volume so the more transaction volume the more fees and the more fees the bigger the reward for these nodes. So that’s kind of the basic premise.

As full disclosure, I have my money in AION so I’m not trying to shill or anything but I just wanted to get that out there.

To be clear, I mean there’s nothing in our design that restricts a bridge from being centralized. The design could allow for a centralized bridge and just the market might demand a decentralized bridge. If somebody really trusted it and decided to build a bridge and made it really cheap, maybe people will use that bridge and they might say hey I trust it enough to use it. But in another context you might say ‘No, I don’t trust anybody. I want to use the more decentralized one.’ And that’s where you get like a market opportunity where some people are gonna be confident enough to say this one is someone centralized, but maybe it’s faster. I’m gonna go on this somewhat centralized faster bridge or I can do the trade off and say this one might be a little slower but it’s more decentralized I can go that way. We don’t want to dictate that in our design we’re keeping the design kind of as flexible as possible so people can pass their transactions wherever they want.

The Tencent, the really cheap, literal and figurative, ten cent bridge versus the TOR bridge which is really slow.

I understand that the AION network provides opportunities with various incentives as well as penalties. Especially, there are two types of consensus algorithms: PoS and proof of intelligence. Could you explain both of these?

We got two things about consensus: one, our first release which is the one we’re releasing this week is a proof of work. We’re releasing a form of version one of the network that functions like a modified version of the Equihash mining algorithm. That’s what’s being released this week. Our long term design and consensus is a delegated consensus model and the delegation means that we have a core group of validating nodes that that actually provide the infrastructure for block creation and block propagation. That core group of several hundred nodes provides that core function and then behind those validating notes is a group of backers and those backers are nodes that support them so they’re also active nodes in the system.

They support a validator and they share in the reward of the validator and they can pick and support them in two ways. They can support them by staking tokens or staking coins into the system or they can support them by providing a computing resource to the system. The computing resources that we’re designing we call them proof of intelligence because we’re actually starting to think about the long term implications of being able to run artificially intelligent applications on top of decentralized networks. And you need an on-chain resource that has AI capability essentially. We’re trying to incentivize nodes to create that machine learning functionality so that they can actually be a called as a function inside of a smart contract. And this is where we start playing around with terms where are we talking about smart contracts or are we shifting towards intelligent contracts where you can start building like AI at the features inside of contracts and that’s part of our design.

So, the goal at some point is to say well the long term convergence of artificial intelligence and decentralized networks is how we continue to have confidence in increasingly sophisticated systems. The big risk that people see in the AI industry is that you’re putting so much confidence in centralized institutions to act honestly. You’re putting it into Google and IBM and other massive institutions that are going to be running AI infrastructure but you have to trust that they’re gonna act honestly and in your best interest. So long term is if we prove the scale of decentralized infrastructure we think that we can move a lot of these AI characteristics onto decentralized networks and really make sure that trust is fundamental to all of these systems regardless of whether they’re smart contracts, dumb contracts, or intelligent contracts. Trust is still the base layer of everything.

Ethereum is doing a lot of cryptoeconomic research and recently Vitalik had his famous trilemma which I’m still struggling to wrap my head around. But for you it seems even more complicated than what Ethereum is suggesting. You’ve written that you’ll come up with some research and some papers that’ll help us wrap our heads around AION’s cryptoeconomic model, can you give us a teaser or what are some tradeoffs that you guys are balancing in your head for choosing both proof of stake or proof of work or just even in the way how you have multiple incentivization schemes to reward people?

The trilemma concept is something that we’ve been thinking a lot about. Because there is a tradeoff in a lot of context in the industry around decentralization versus efficiency and I think what we’ve kind of opted for in our design is that decentralization can happen in many ways. You can have decentralization of a single Network they can also have decentralization across a number of networks. As contrary as this sounds centralization to the same decentralized network causes less and less efficiency so if we all use the same decentralized network then we’re actually depreciating its value because we’re slowing it down and we’re damaging its performance essentially. Not only decentralizing the networks but decentralizing the number of networks that we use and that’s the core of our design. Now the cryptoeconomic layer we’re really fortunate to follow projects like Ethereum. We look up to Ethereum with an enormous amount of respect and we follow a lot of their leadership and the research fields. Because they’re leading a lot of really interesting cryptoeconomic research and a lot is coming out of the Ethereum foundation, the Ethereum community. I think a lot of those lessons are trickle down very much to us. There’s nothing all that unique about our cryptoeconomic design versus another multi-blockchain interoperable system. Everybody realizes that individual networks need to have their own stable concept of cryptoeconomics and then we also realize that for us to connect these networks, those connections need to have a layer of cryptoeconomics. You need to figure out why people would participate in the connection. Whether that connection is being done using the Polkadot model with the relays and the parachains and/or it’s using COSMOS and the zones/hubs or you’re talking about ICON and the Nexus and the different the different blockchains off of the Nexus or Wanchain. Everybody has a similar concept where at the connection you need an economic model you need a reason for people to participate in the connection so I think we’re gonna learn a lot from each other. Most of our research is gonna get put out in the way of requests for comments. We want to start conversations. We don’t know all the perfect and final answers, but luckily as we progress in our design we can start to collect more and more feedback and the reason we’re really excited about starting to grow the impact of our community is because we actually want more input from people. We don’t have all of the answers internally. We’re 30 to 35 person team right now and we could use way more brains on this problem. We could use way more researchers, way more academics. Our objective is to just start opening the dialogue so that we can get better input but we are going to be publishing some of our early thoughts on our economic design. We do expect that we’re going to get some of it wrong and we need people to put up flags and give us comments when they disagree with something that we put forward but I spend a lot of time reading what’s coming on the Ethereum foundation - a lot of our early thinking that we do comes out of Ethereum.

Recently a paper published by IC3 talked about how these decentralized networks namely Bitcoin and Ethereum are not as decentralized as we’d like to think. Actually we can substitute them with a better system in terms of decentralization. I think a lot of public blockchain protocols are in fact more centralized than we’d like to admit and you touched on how to keep AION decentralized. The question is if the interoperability protocol is decentralized but the blockchain connected to it is not as decentralized or they’re not, do you think there’s any value to the interoperability protocol?

We’re going into Inception here. We’re like six levels into the dream right now.

I’m just wondering because you’ve mentioned just now that you will have bridges that are centralized that can be more centralized than others. It’s very interesting because in your whitepaper you explicitly say that they are “decentralized to a certain extent” I think that’s the wording that you use.

This is like a spectrum. The perfect decentralized network does not exist because the perfect decentralized network is inefficient. That’s where we run into challenges like the reason bitcoin has centralized around massive mining pools is because there’s economies of scale. Everybody organizes itself around efficient use of resources. You have to assume that’s gonna be true always. Economic principle will dictate that if we will be more efficient. If organizing ourselves as a collective is more efficient than organizing ourselves as individuals then we will always opt to organize as a collective. We have to have that kind of in our thinking. When we designed our delegated consensus model, we call them tiers in our whitepaper and these tiers are essentially like how do you participate at different levels of reward in the consensus algorithm so that you can get a larger or smaller reward and the purpose of defining the size of the tier was to make sure that you were not incentivizing centralization behind a single validating node. If everybody said hey I want to back the one validating node that is most likely to get the biggest reward all the time, then you’d have a massive centralization behind that node.

We end up structuring these tiers so that at some point there’s a diminishing return to centralization. The cryptoeconomics is all about thinking about the behaviors you like and reinforcing them and thinking about the behaviors you don’t like and penalizing them essentially. If you don’t like centralization then think about how to economically incentivize people away from centralization. For example, if you were to take the Bitcoin protocol and say at a certain scale a pool would start to receive less reward at some point there’s a diminishing return to centralization. That’s not how the Bitcoin protocol got designed, but it could have got designed that way. You could say a large pool, a large centralized pool of miners has a diminishing return the larger it gets. Now every new member I join into the pool, I reduce the overall percentage of reward that people get. Then you find the equilibrium of the balance between efficiency and decentralization. But you have to design that into the protocol because if you just leave it up to the economic motivation of the individual they will always opt for the most efficient not the most decentralized and often the most decentralized is the less efficient. I think you have to set these rules into the protocol as much as you can.

Let’s talk about your recent releases. You have the testnet up now and I think people should jump on that and try it out for themselves. What were the improvements you made on the existing EVM with AION’s fast VM - what are the specs of it?

If you look at our roadmap, you’ll notice that our long-term design focuses on something we call the AION VM. As an interim we ended up building a modified or enhanced version of the Ethereum virtual machine that focused on one major characteristic difference. It changed it from a 256-bit data word size in the EVM to a 128-bit data word size in the AION’s fast VM. Completely compatible at the smart contract layer, so it still supports the same solidity but allows you to run applications significantly more efficiently. Less usage of gas and faster execution through the VM, so that’s the major difference that we did in this change over. We rewrote the VM from a 256 bit to a 128 bit and then we also ended up essentially designing in Ethereum VM just-in-time JIT as the execution engine so we swapped out the execution engine to make sure that we could process transactions more efficiently and more quickly essentially. So this is like an incremental improvement. This is not an order of magnitude improvement. This gives us like tens of percentages improvement over the EVM and performance but the long term design is still to move towards a completely parallelized VM with the AION VM design. But it’s a significantly longer roadmap to make that happen so that’s why we ended up starting with a small improvement as we work towards our long term design.

Why did you choose PoW like equihash? Vitalik and Nick Szabo recently got into a little tussle on twitter over PoW versus PoS and which one has better social scalability. Some might says proof-of-work is outdated but why’d you guys choose equihash?

Two reasons, one it’s relatively well understood by the market and it’s been proven economically by the market whether it’s the long term design or not. We’re unconvinced it’s not part of our long term design. This was really kind of a starting point in the system. When you get something out there that the ecosystem understands, that miners could support, that comes out of the box with a relative amount of certainty on its security. If we were to start from scratch going into a completely new model, we’d have to question the security premise from day one. We wanted to make sure that we at least had some solid understanding of what we can expect in terms of security based on the mining algorithm that we decided to support. What we then did on top of that, we looked at all the different kind of PoW algorithms. We landed on equihash and then we made modifications to the equihash mining algorithm to essentially make it more ASIC resistant. We to memory hardened the algorithm. ASIC resistance really comes down to how much memory is required to run the mining algorithm. Because ASICs are pretty inefficient at storing large quantities of memory so the more memory you require at the mining then the less likely it is that an ASIC could mine that Network essentially.

So we wanted it to be somewhat easy for a GPU based miner to adjust to mine this and somewhat resistant to the ASIC world stepping in and over-mining or dominating the hash power with ASICs. That’s our early design. We’re working in parallel on proof of intelligence. It has some similar characteristics of the proof of work because you’re still proving some sort of compute to the network in anticipation of a reward. But the compute thing that you’re working on is actually some sort of functional compute because a big criticism of proof of work is that you’re just doing math. You’re not really solving a useful problem to prove that you can solve a problem and that’s where we ended up starting to question. What if we can make the problem useful so that the output of the problem was useful. That’s the kind of the hope that we have behind our proof of intelligence design but we’ve still got a little bit of research and it’s kind of in development stage right now. It’s not perfectly ready to release but we’re getting some pretty cool results in proof of intelligence and we’ll hopefully have something to show you guys.

You touched on how proof-of-work naturally evolves or changes into proof of intelligence. Can you elaborate a little more on that? What role does your partnership with SOMN play here?

I mean two things. One of the designs that we’re working on is to make sure that the components that we build are modular enough that if you wanted to launch a new blockchain connected to AION, you could pick up any one of our modules and use it to launch that new blockchain. If you wanted to have a blockchain that only relies on proof of intelligence, its core consensus algorithm, you can do that. In our design, proof of intelligence becomes one subcomponent of this delegated consensus model. The cool thing about working on a interoperability project is that when we think about updating our system we don’t have to think about it as forking our system. We’ve got version 1 of our network that’s a live operating blockchain and we want to launch version 2. We can actually launch it as a completely parallel blockchain and they could operate beside each other and after they operate beside each other then you can build a bridge between them. After you build a bridge between them, you can move value over from one to the other so you don’t have to go through these contentious forks every time you have any proposal. Because you have that as an option you can actually see two networks continuing to survive. You could say hey I’ve got AION version one and AION version two operating in parallel. Our bridge design is such that the supply of our coin is stable regardless of whether it’s on one blockchain or two blockchains. Today we’ve got roughly 465 million coins and many of them locked off but that supply is all on the Ethereum Networt. Then we want to migrate that supply over to the AION blockchain and we want to migrate that supply over to the next version of AION blockchain. We actually plan for the fact that some of the supply stays behind. You might have one blockchain with one quarter of our supply and another blockchain with three-quarters of our supply and the bridge is efficient enough to move the supply back and forth.

As long as supply is stable and it falls into the same monetary policy that we that we’ve created, the monetary policy doesn’t shift regardless of how many blockchains. Today if you think of a single supply of a coin it’s always locked to a single blockchain. We’re trying to get a single supply to be able to exist across multiple blockchains. As we update the system we envision that there’ll be people that like version one better than version two or better than version 3 and they may stay behind and continue to use an older version versus a newer version. We’re not anticipating that everything has to be an all-or-nothing update. You can have parts of the community that stays behind. As it relates to SOMN, the relationship we’re developing with SOMN is focused on how we make mining of the AION Network in its first release available to people more efficiently through a decentralized marketplace that’s on this building. SOMN is building this marketplace where I can put forward GPU power and somebody else can come by that GPU power and allocate the GPU power to the AION mining. For example, we can prepackage an AION miner on top of the SOMN marketplace so that you can buy GPU and have a AION miner running on top of it in a single step. That’s kind of the intent of that relationship.

Something that wasn’t clear to me was you would have multiple versions of these blockchains but how would each of these blockchains and the different versions be secured? Is the concept similar to merged mining or is it all done by staking?

It depends on the version. Version 1 of the blockchain is based on proof of work, so proof of work will imply that they’re all going through the equihash mining process providing security to network. As we progressed, the final version of the network will have a staking function as well as a validating function as well as a proof of intelligence function. Moving between these networks the amount of coins that sit on top of a single Network, our current hypothesis - and I’ll say hypothesis because we’re asking the community for comments - we’re releasing in the next week or so our intended monetary policy and our monetary policy implies that the reward on a blockchain is a percentage of the supply on that blockchain. Let’s say we have the 465 million coins across two blockchains. The reward is stable as a flat percentage and it’s stable based on how much of the supply is on blockchain number one vs blockchain number two. If enough of the coins move over this then the reward on the original blockchain goes down so much that there’s less incentive for miners to be there. More and more incentive for miners to shift over to the new version but if it stays kind of balanced, then you may have miners that stay on the old version and miners that move on to the new version. The reward follows the supply and that’s what we’re trying to make a stable rule. We can calculate monetary inflation on the basis of supply regardless of where the supply is.

Is there a possibility for a particular version of a blockchain to be completely abandoned by either the stakers or by the miners?

There absolutely would be a possibility for that to happen. If one miner continues to run the mining algorithm, in theory the blockchain still exists. But at a certain point you get below a certain threshold of hash power and for all intents and purposes it would stop becoming a trustable system. You probably would slowly die off. What we’ll do as a foundation is be very vocal about which networks work and we’re still in the business of maintaining. But the community could go maintain whatever they want. You can go and have a network that is being maintained by another party and that’s fine. If people still see value in it they’ll keep running applications on that.

It’s very cool. Because it seems like you are empowering and giving choice back to the users. They can choose which version of the blockchain they prefer but at the same time it’s not forceful like a soft fork nor is it contentious like a hard fork.

This is a design that actually having two networks operating in parallels is a cool and desired outcome rather than it being a negative outcome. When you go into the traditional design of a decentralized blockchain, a fork looks like an undesirable outcome. But we’re saying actually having two parallel block chains that run the same monetary supply is cool as long as you’re not duplicating the monetary supply. You don’t want to have a Bitcoin Cash or an Ethereum classic situation where you’ve doubled the monetary supply. You want to have two parallel networks that still maintain a single monetary system.

Other interchain project, including COSMOS and ICON, describe the decentralized exchange in their white paper. Do you have decentralized exchanges planned in the AION network?

We’ve always envisioned that’s an application that would get built on top of AION. But it’s not something that we’re prioritizing as a foundation. We may work hand in hand with third party projects to do that. But we do think that this is a cool protocol that could be used for decentralized exchanges. The idea of something like EtherDelta that sits on top of AION. Not only deals with tokens but also deals in currencies - I think it’s a really cool concept. We’re not prioritizing product development right now. We’re prioritizing the network and the protocol. For now it’s not on the roadmap but it might be something that somebody decides to do with our support. But nothing in the pipeline right now.

This is gonna be a little complaint, but let’s hope it’s not like EtherDelta.

I used it as an example but it’s not the best example. It’s the one everyone’s familiar with.

It’s true. Okay, questions about partnerships. Let’s jump into EEA because you’re a founding member/board member. We’ll ask the nitty gritty, harder questions now. Do any members of the EEA actually contribute? As a founding member in the EEA, what have you seen in the past months what kind of developments have been there?

Biggest development at the Enterprise Ethereum Alliance is we just hired an executive director his name’s Ron Resnick, solid guy who’s going to start putting a lot more structure into the priorities of the Enterprise Ethereum Alliance. I was on the phone with Ron today talking about this. The original focus of the EEA started off as how do we take Ethereum style code - I say Ethereum style because people are constantly modifying it - and optimize it for private networks. That was the original mandate of the EEA because the hypothesis was that enterprises were only comfortable working with private blockchains. So you have projects like Quorum; our company when we started Nuco was only doing private networks; our friends over at BlockApps; companies like Monax and the list goes on. What I’m trying to push the EEA to do is to start recognizing that public networks become a really important part of every business whether they’re large enterprises or small startups. Public networks are going to become fundamental to all aspects of society in the near future. The forward-thinking companies are gonna want to think about how do they integrate their systems into public networks. But we’ve also held the position that private networks and public networks don’t need to be mutually exclusive. I think that the long term solution is that some applications and some business processes are going to be better suited for private networks. Not necessarily because of privacy requirements which might be one reason, but also because of efficiency requirements. If you close off a network to a smaller group of participants, you could actually achieve significantly higher throughput and efficiency. You might have parts of a business process in a segregated private network and then part of that business process in a public network and then we come back to the same problem of how do you connect those networks to each other. That’s where we started designing on the basis of private and public and public in public how do you build those connections. The bridge concept is still the same whether you’re doing a private Ethereum network to public AION or public Ethereum to public AION.

We also have this more contentious view that other protocols, other than Ethereum, are gonna have long lives. It wasn’t just going to be Ethereum and Bitcoin at the end of the day. We were gonna see enterprise adoption around R3 and their Corda protocol and Hyperledger and then we’re gonna see public adoption around networks like Qtum and NEO and Dfinity and others. We wanted to make sure that whatever we were designing was generic.

Now to your other questions about members of the Enterprise Ethereum Alliance there are a few members that are extremely high-value contributing members and increasingly now there’s a very robust group of working teams that are working on different aspects of the spec. The goal of the Enterprise Ethereum Alliance is to publish a spec on how you run an enterprise network. Off the top of my head, we get a lot of involvement from JP Morgan we get a lot of involvement from Microsoft and Intel. Accenture is really active. Deloitte is increasingly active. The list goes on. There’s 400 plus companies in there. They’re definitely not all contributing on a daily basis, but I’d say over a quarter of those companies are actively participating in working groups and actively contributing code and actively feeding requirements into the end of the structure. The Enterprise Ethereum Alliance now has not only an executive director but it also has some full-time staff that are working on writing this spec and designing the spec. There’s some full-time technical staff at the EEA that are spearheading that with input from members. We’re going through a little bit of a board shuffle pretty soon but we’re still on the board of directors along with ten other companies and they mostly are large companies. There’s only three startups on the board and the other eight are big companies like Microsoft, Intel, and others.

The last two minutes of what you just said really made EEA less enigmatic for me. It sounds more like a standardization body.

Yeah. We’ve been having this conversation about what’s our mission in life at the EEA and there’s definitely been different perspectives. One of the big players inside the Enterprise Ethereum Alliance is Consensys. Consensys has a huge vested interest in seeing the success of Ethereum. Obviously there’s a mandate for Consensys that might be different than a mandate for a bank that’s got an internal innovation team. But we’re kind of finding our common ground. Our common ground tends to be that there’s going to be a need for a set of specifications on how large enterprises integrate or interact with networks - whether they’re private networks or public networks. Some of that is focusing on privacy requirements - that’s a big focus of the Quorum project, how do you do private transactions. Some of it is focused on just doing more efficient supply chain management or whatever but I think increasingly crypto is gonna become part of that conversation because now you’re gonna get investment banks and large financial institutions that are interested enough to get access to these markets.

Do you expect that more and more companies including banks or financial institution will join the EEA?

I’ve got two perspectives on this. One I think it’s really important that these large incumbents take this seriously. I also think when we started enterprise in our business since we started a few years ago, everybody assumed that you’re talking about existing enterprises. But I think more exciting is what’s the next generation of companies that’s coming out of this industry. There’s some massive analogies to the Internet. The earliest websites built by banks were like online brochures. They were really shitty and they didn’t really do anything. They just were like static pages. But in parallel to that PayPal was building a completely new payment system. I’m more interested in the analogy to PayPal than the analogy to the bank. Who will be the next PayPal who will be the next Amazon on top of this system. Because the enterprises of the Internet era are not the big companies before the internet era. Companies were created during the internet era. That’s why it’s tough to realize sometimes but we look around. We’re all friends in the industry and we know each other pretty well. When we go to conferences, among us is the next Jeff Bezos and the next Mark Zuckerberg. What’s cool is to think about that next generation of tech companies. It’s different because we’re not building these massive centralized kingdoms like Facebook but we are building very relevant long-term technologies. To be fair 95% of us are gonna fail in our attempt, but some of us will make it through and be really relevant to the to the long term adoption of this technology.

I wish that AION will be one of the pioneers for the future PayPal. In the white paper you talk about scalability and interoperability but you do not address privacy, how do you plan to do this? Is the partnership with Enigma a part of it?

Yeah it is. Privacy I see kind of two ways. One there’s the layer of privacy that says I want to segregate part of my business application or my process into its own network and decide who’s entitled to be on that network. This goes back to the concept of permissioned or private blockchains. And then there’s the concept of at a transactional level I want to maintain privacy. I’d say firstly we don’t want to go solve for transactional privacy because other people are solving for transactional privacy. So we’re looking at the models of like zk-SNARKs and Monero and Quorum on the enterprise side. There’s a lot of cool models for how you maintain transactional privacy. We’re more likely to integrate or adopt one of those standards and to go create our own standard. We’re also more likely to say well if we want access to privacy as a feature we can just build a bridge between AION and one of those private networks like Zcash and build that function to allow for private transactions on one network to connect into AION on the other’s.

Private networks is still something that we intend to encompass into our design so if people want to use more a voting consensus protocol - a lot of these private blockchain protocols are based on some derivative of a Byzantine Fault Tolerant protocol so PBFT or Honey Badger or Tendermint or we built something called Nuco BFT and all of these are more communication based like voting protocols - they allow for you to restrict who is entitled to be part of that voting. You can say you need a restrict based on unique node ID and only these unique node IDs are allowed to be part of the consensus. They’re the only ones who see the consensus data. So that’s one layer of privacy that we’re absolutely going to support. It’s not part of the AION design because it really comes down to the individual network inside this network of many blockchains. The individual network could decide to be as private or as public as it wants to be. We want to make sure that our design is generic enough to encompass both.

COSMOS has a similar thing. I remember in a podcast interview they were asking how you would shift coins from Zcash to Bitcoin and the answer was that it really depends on the network, it’s not really our responsibility.

Because we don’t control the protocols on the other side of the bridge, we just control how our protocol works and how the bridge works. But at the end of the day we’re still limited by the functionality of the other side of the bridge. Simple example is connecting to Ethereum is a significantly different challenge than connecting to Bitcoin because Bitcoin doesn’t have this second layer kind of smart contract logic layer on top of it. So we have to think of a very different approach on connecting to Bitcoin. That’s a limitation based on the design of Bitcoin. So there’s not much we can do about that we’re just gonna have to optimize where we can to fit those different characteristics.

How much can Bridge builders make?

Three factors in there. What’s the fee when you build a bridge, how many nodes are sharing that fee and how many transactions are crossing the bridge. So it’s some factors of those three things. This is something that when we design a transaction format what we call an interchange transaction. The interchange transaction has to tag a fee in AION coins that will be then shared by the bridge validators. The bridge that they’re crossing will essentially publish its fee structure and then when you send a transaction that is intended to cross that bridge you need to attach the appropriate fee, very similar to attaching a fee to an Ethereum transaction to get picked up by a miner or to Bitcoin transaction for that matter. Then it really depends on how heavily the bridge is used because if it’s really low volume then the fee might be very inexpensive just like how gas prices change on a day to day basis or transaction fees change on a day to day basis.

It depends on how many nodes are on the bridge. If there’s a hundred nodes on the bridge and the fee is getting shared among the hundred nodes. If there’s a thousand nodes and it’s being shared among a thousand nodes so I don’t know that there’s a quantified dollar figure I can give because it really depends on those factors.

Your two revenue streams that people can do by contributing work on AION is either validating AION bridges or by mining on AION-1. It is the trade-off where on AION-1 you get more of a consistent return but by validating AION bridges it’s more or less very dynamic and depending on what bridge is set up you can get a better return.

The difference is that the consensus on the AION-1 blockchain has a predictable reward. The reward on the bridge is a fee. The reward on the bridge is paid by somebody like they’re paying a transaction fees across the bridge. But participating in consensus, you’re getting new minted coins from the protocol that is always consistent. The monetary policy that defines how that works never changes so the same amount of reward is going to be created on a block by block basis or day by day basis or year by year basis. But the bridge if nobody’s using it there’s no fees if everybody’s using it there’s lots of these turns. So it’s a lot more dynamic.

What are the key differences between AION and Ark?

I haven’t spent enough time looking at Ark other than kind of surface level to give a well defined answer. I know that there’s been a few commentaries that have been written online about the comparison between the two. Honestly I’ll probably pass on the question just because I don’t think I’ll do them a fair representation. I met the Ark team in Miami a couple weeks ago and they were phenomenal guys and hopefully we’re gonna start a conversation and work a little bit more closely together. But I don’t know enough about their design to comment.

Another question that we actually got from our audience is actually from our Korean audience. This guy typed in a very long response about like AION the game. I think he thought we are interviewing the founder. It was a very heart-wrenching, I love your game and I don’t know what you guys are doing with the game but I really want to play it. I was gonna read that question but I figured you’ve never played that game.

It comes up all the time every time anybody searches AION they’ll probably see some cool knights with swords.

I heard it’s also a problem in the Korean community. Your community manager was telling me that because the other name on the Kakao group or the telegram group was just “AION” people would come in and start asking about the game. So they’d had to change it to AION cryptocurrency and then people stopped coming.

How’s your presence in Korea? How’s it going in Korea for you?

We could do better and that’s why we’re coming in April. We want to do a big event in April. We’re working on lining up all the major pieces of being a relevant part of the community ecosystem so wallets and exchanges and then meetups and partnerships. We’ve got a really good relationship with ICON. We’ve known those guys for a number of years and we continue to work really closely together, but we do want to come. We’re gonna be there at the beginning of April. We’re hoping to get as many people out as we can. It’s on the priority list but it’s not something that has gotten as much emphasis as it probably should have.

You recently announced the Alliance with ICON and Wanchain. What does this alliance mean to AION?

A lot of people have asked us why those two and but not other projects of interoperability. A lot of it stems back to relationships that we’ve had for a number of years. I’ve known Min Kim, one the founders of ICON, for several years, Jack Lu at Wanchain and Dustin Byington we’ve known for a long time. We just realized in talking to each other that we were approaching very different problems but we were kind of approaching them with similar solutions. So the more we could share and open our notes and share research, the better we felt about the output. One of the things that I’m constantly trying to reinforce internally is whenever we can leverage work that’s already being done somewhere else it’s more efficient. We don’t want to go and reinvent the wheel on every layer of our infrastructure. There’s some layers of our infrastructure we’re going to be better off working with other companies. So the first layer was cross chain communication: ICON and Wanchain both have very interesting designs that we want to learn from. The second layer that we’re starting to talk about is a collaboration on our VM design because there’s a lot of projects working on new types of virtual machines and if we can get to some sort of acceptable standard on high-performance virtual machines and decentralized networks we’re better off doing that as a group rather than everybody going and designing their own and having no compatibility between them. So the general premise is if we can collaborate with others we will thus it saves us some effort, makes us more efficient, and getting to market faster.

Where’s the partnership with Bancor going? Are there other partnerships?

Two things. One we’re really interested in their relay design. We’re likely going to be doing an AION to BNT relay with the Bancor team pretty soon. I think it’s a cool concept to have this self managed liquidity pool of coins to trade. But the other thing that Bancor came to us with was the fact that they wanted the BNT token to be freed from only one blockchain. They wanted BNT tokens to be able to move across multiple block chains. So we’re looking at how do we leverage our bridge design to help the BNT token move across blockchains. They can rebuild the Bancor relay on more than one network. Right now they’re restricted, and the only thing they can do with the BNT or with the Bancor relay is only do ERC-20 tokens on top of Ethereum. They want to broaden that so that this BNT pairing they can do with any coin across any blockchain. So we’re hoping that the design that we’re working on is something that they can leverage for that. But in the short term, you can expect to see an AION BNT relay pair.

What your thoughts on the current BTC market crashing. How it’s impacting altcoins?

Yeah that’s a good question. Every project in this industry has a lot of money caught up in the market. So obviously we’re paying close attention. But it doesn’t change what we have to do on a day to day basis. So we have this conversation almost daily with the team that if we’re trading at 10 cents or $100, our roadmap doesn’t change. So we need to continue hiring engineers, we need to keep building the roadmap, we need to keep researching the unanswered questions of our design and that will always be true. It’s sometimes more encouraging when the price is higher and discouraging when the price is lower but fundamentally it doesn’t change what needs to be accomplished. We’re building something that is really relevant five to ten years from now. So that patience. There’s gonna be a lot of ups and downs in the next 5 to 10 years. I’m constantly reiterating this to people that follow our project. There are some apps out there that their token might show value very quickly and the use case might show value very quickly. There are some protocols that need time to develop and evolve.

We’re gonna have versions of our protocol in market used by people very quickly. We’ve got our network going out this week and we’ll have it live in the next few months. But the long term design of what we’re doing is a very long road map. So have patience. It’s how we’re trying to operate. It’s sometimes difficult to disconnect from the market but at the end of the day going back to that internet analogy I was using earlier - the best companies that came out the other end of the dotcom bubble are the ones that had that long term view. Amazon got caught up in the dotcom bubble just like everybody else did but they still stood at the end of the day. Google got caught up in the dotcom bubble and still stood at the end of day. So I don’t think it’s a matter of where’s the price in the market. It’s really a matter of are you staying true to what your goal is and our goals farther out than day by day movements of Bitcoin. The other thing that we have to work on pretty consistently is I want to redefine what it means to be a coin in this market. Right now you’ve got Bitcoin, Ethereum, and altcoins and everybody gets bucketed in there. There’s one category. But the reality is there’s a lot of differences in these projects, so we’re trying to make sure that we’re defining well enough what it is we’re doing and what are our KPIs. What should people be measuring us on so that you don’t just measure the market as one giant lump sum you measure individual projects on their individual merits. I think over time we’ll get better at that. Right now obviously we get stuck in these ups and downs of the Bitcoin price and the Ethereum price but we’ll stand on our own two feet at some point.

It’s like you said every single coin has a different proposition. It’s kind of like dating girls. You have girls who really stand out and then you have the ones that are dime a dozen.

I think that’s all the time that we have for today. Thank you for joining us. Every time I get schooled so much whenever I talk to you.

Thank you so much. I especially agree with your opinion that there is no KPI for evaluating each project and there will be more and more research and people who are interested in developing appropriate evaluation for evaluating the real value for each project.

Yeah totally agree. I think it’s important for us to be able to tell our own story and not just get caught up in the crypto market. Because the crypto market is a very broad story but individual projects are very specific stories and I think we all need to get better at that. Right now an outsider sees one massive bubble or one massive, very difficult to understand thing but if you look project by project you find really interesting concepts. You find really interesting technical designs but you have to you have to look deeply. If you just look at the surface, there’s no difference between coin number one and coin number ten they all look the same to the average outsider.

Give us your social media handles. How can people get more information about AION?

I want to make sure I don’t screw this up because there’s a couple of scam accounts you should be aware of. Website pretty simple aion.network, we’ve got a telegram chat for most people to participate an english speaking channel. We also have a Korean telegram channel and a Korean kakao channel that you can get access to. You can follow me on Twitter @MattSpoke or @AION_Network. There’s a whole bunch of other channels. We’ve got an active blog, we’ve got the AION Network sub-reddit, we’ve got a forum published now at forum.aion.network and more and more we’re trying to push people to our github. We just started publishing on our public github over the course of the last few weeks, so we’d love to get some feedback and kind of reactions from the community.

Yes, sir. Thank you once again, Matt for joining us.

Thanks so much guys looking forward to seeing you in Seoul.

Thank you everyone who’s tuned it and once again, this is a your man of the hour See Eun and Money Makin’ Matthew. Our special guest Matthew Spoke representing AION.

If you like this video, please subscribe, like, comment below. If you have any questions either please go over to AION and ask or we’ll make sure to relay them along. We hope to see all of our audiences in the future and also Matt, we hope to see you in April at Seoul.