Sharon Goldberg, CEO of Arwen and Boston University computer science professor, describes how the protocol enables people to trade on crypto exchanges without making their coins vulnerable to exchange hacks, how this differs from trading on a decentralized exchange, and how Arwen would function if the exchange you were using on was hacked during a trade. She also discusses how Arwen works with any custody solution, how they choose which coins to support and what other features they are considering adding in the future.
Read the full show notes on Forbes: http://www.forbes.com/sites/laurashin/2019/05/21/how-to-trade-on-crypto-exchanges-without-your-coins-getting-hacked/
On May 27th I will be hosting a conversation about the future of finance and human rights at the Oslo Freedom Forum in Norway. As the world continues to move toward a cashless society, paper currency is disappearing. Companies like Facebook, Apple, and TenCent are becoming increasingly influential in the digital payment space. We’ll discuss how individuals and companies can preserve and protect financial freedoms in the digital age. I’ll be joined by Bitcoin author and educator Jimmy Song; Casa chief technology officer Alena Vranova; and CoinCenter founder Jerry Brito. To register to attend, you can visit oslofreedomforum.com today — use a discount code “unchained25” to get 25% off your ticket price!
Interested in the crypto workshop I’m teaching with Meltem Demirors and Jalak Jobanputra at Omega Institute in September? Sign up here!
Thank you to our sponsor:
CipherTrace: http://ciphertrace.com/unchained
Episode links:
Arwen: https://www.arwen.io
Sharon Goldberg: https://twitter.com/goldbe
How Arwen prevents exit scams: https://medium.com/arwensecure/how-arwen-prevents-exit-scams-55d1a9de32b
How Arwen protects traders from a hacked exchange: https://medium.com/arwensecure/unleashing-mt-fox-4e8b20a11dea
Forbes article on Arwen: https://www.forbes.com/sites/ktorpey/2019/01/28/this-startup-is-fixing-the-biggest-security-hole-in-bitcoin-exchanges/#5f68d5905812
The Block on Arwen: https://www.theblockcrypto.com/2019/02/05/a-startup-is-launching-a-solution-that-protects-users-from-exchange-hacks/ https://www.coindesk.com/crypto-startup-wants-you-to-trade-on-exchanges-without-trusting-them
Kucoin launching Arwen: https://www.coindesk.com/kucoin-exchange-traders-can-now-self-custody-their-crypto-assets
Transcript:
Laura Shin:
Hi, everyone. Welcome to Unchained, your no-hype resource for all things crypto. I’m your host, Laura Shin.
Interested in the crypto begin retreat I’m teaching with Melton Damaris of Coin Shares and Jalak Jobanputra of Future Perfect Ventures? If so, be sure to check out the show notes for the link to sign up.
Also, Unchained is now on YouTube. You can find the most recent episodes there every week on the Unchained podcast channel, and if you’re not yet subscribed to my weekly newsletter, go now to unchainedpodcast.com to sign up.
Cipher TraceCipherTrace makes it easy for exchanges and crypto businesses to comply with cryptocurrency anti-money laundering laws. Avoid illegal sources of funds, and maintain healthy banking relationships. Ciphertrace is helping you grow the crypto economy by keeping it safe and secure.
Oslo Freedom ForumOn May 27th I will be hosting a conversation about the future of finance and human rights at the Oslo Freedom Forum in Norway. As the world continues to move toward a cashless society, paper currency is disappearing. Companies like Facebook, Apple, and TenCent are becoming increasingly influential in the digital payment space. We’ll discuss how individuals and companies can preserve and protect financial freedoms in the digital age. I’ll be joined by Bitcoin author and educator Jimmy Song; Casa chief technology officer Alena Vranova; and CoinCenter founder Jerry Brito. To register to attend, you can visit oslofreedomforum.com today — use a discount code “unchained25” to get 25% off your ticket price!
Laura Shin:
My guest today is Sharon Goldberg, co-founder and CEO of Arwen and computer science professor at Boston University. Welcome, Sharon.
Sharon Goldberg:
Hi. Thanks for having me.
Laura Shin:
So, let’s talk about Arwen, your new project. What problem are you trying to solve with Arwen?
Sharon Goldberg:
Right. So, Arwen is a new way to trade at centralized exchanges. The idea is that you could trade on specialized crypto exchanges order book without trusting the exchange to custody your coins, and our notion of trustlessness is really strong. We guarantee that, even if the exchange is hacked in the middle of the trade, your coins should not be at risk.
So, what Arwen does is it’s a cryptographic protocol that allows you to transact with the exchange and settle your transactions without ever having to put your coins in the custody of the exchange.
Laura Shin:
And how did you come up with the idea for this protocol?
Sharon Goldberg:
Yeah, so, that’s kind of like a long story, and I can sort of start at the beginning, which is my background and how I sort of got into this space. I have a PhD in computer science focused on cryptography and networking, and so, really, my most fundamental background and how I got into my whole career is through cryptography, which I started in 2006.
So, when you are sort of basically trained as a cryptographer, the kind of problems that you want to solve are you have two parties, like A and B, that which to transact, but they don’t trust each other, so it’s usually Alice and Bob, and they want to do something, but Alice doesn’t trust Bob, and Bob doesn’t trust Alice.
And so, what cryptographers do is they build protocols to allow these two parties to communicate with each other without trusting each other. So that’s kind of like the fundamental problem that I’ve been trained to solve for the last, you know, 13, 14 years.
And so, we came into the blockchain space about…let’s say about four or five years ago. We started looking at the security of the coins, security Ethereum and then getting into designing also protocols that build on top of blockchains.
And so, for me, with the background that I have, the very first question that I always ask is like who you need to trust in order to accomplish this particular goal.
So, when you look at the market right now, it’s centralized exchanges. You regularly see, unfortunately, you see compromises of centralized exchanges where people’s deposits on those exchanges are getting stolen, you know, starting with the Mt Gox hack, which, you know, is famous from a long time ago.
We’re still seeing that, and recently saw Binance get hacked, and basically, what happened is that the coins that traders are depositing at the exchange are being stolen by hackers. So, essentially what’s happening here is you have the trader being required to trust the exchange in order to trade, so the cryptographer, the first question you ask is how can you allow the trader to trade on the exchange without having to trust the exchange with their coins, and that’s really how we came up Arwen, like the idea of transacting on a centralized exchange without needing to trust the exchange.
Laura Shin:
I want to go back to another piece of your bio that I thought was interesting, which is I believe that you actually started in electrical engineering, and so, how did you make the switch to cryptography?
Sharon Goldberg:
Oh, okay, yeah. So, when I started, it was like 1999, when I started school, so at the time, like I wasn’t really…computer science wasn’t so much on the map, so I studied electrical engineering. I went through this program called Engineering Science where you could choose what stream of engineering you wanted to do, and they were like civil engineering and nanotechnology, biotechnology, so I ended up choosing electrical engineering.
And what I liked about electrical engineering was the precision. You know, you can use formulas to characterize, you know, activities in the physical world or building systems and things like that. What happened was I discovered kind of like as I got further up in my education, so when I started my PhD, I was working on optics, like the communication using light.
I found that this was all very imprecise and very analog, and there were a lot of sort of fudge factors that were preventing you from like really characterizing the systems you were building using math, and so, when I took cryptography for the first time in 2005, I was just super excited about the ability to like represent the physical world and relationships and trust using math. That was, you know, fully capturing everything that was going on in the situation, and for me, that was like the big driver to sort of drop everything that I was doing and stop really being an electrical engineer and start becoming a computer scientist and studying cryptography and all of that.
So, it’s really the precision and the ability to characterize trust relationships in a mathematical way that really drew me into the field at the beginning.
Laura Shin:
And then, when you were also describing how you came to found Arwen, you kept saying we had gotten into the space four or five years ago. So, who were you referring to and what happened at that time?
Sharon Goldberg:
Yeah, so what happened is that…so, I’m a computer science professor at Boston University. I’ve been there for…I think this is nine years now. In 2013, I had a new PhD student whose name was Ethan Heilman, and he joined my lab. I was working on internet routing security and cryptography at that time, and I was real excited about that stuff, and so, I was…you know, at that point, like five or six years into this long research program on internet routing security and basically all these protocols that formed the guts of the internet communications network.
So, Ethan showed up, and he was obsessed with Bitcoin, and at the time, there was only Bitcoin. There was really nothing else. He just would not stop talking about Bitcoin, and basically, we figured out the only way to get like productive research out of this person was to do Bitcoin research, and so we did, and it ended up being amazing, and my most-cited research paper is actually with Ethan that we wrote in 2015, which is the first paper that we did in the blockchain space.
And what we did there was we were basically the first to consider how Bitcoin nodes actually communicate and find each other because prior to our work, everyone sort of assumed that like any node that’s participating in a blockchain knows what the blockchain is, so has the same data about what the blockchain is, and that was sort of an underlying assumption in all of the analysis of blockchain up to that point.
And we had realized that like there was a communication layer here. There’s a network here. What is this network? How can you attack it? And we did one of the first attacks on the actual communication layer of Bitcoin that resulted in changes in the Bitcoin protocol in 2015.
So, that was me and Ethan, and Ethan really pulled me into this space over the last, you know, like six years that I’ve been working with him, sort of culminating in us founding Arwen in 2017 and basically working on it together for the last couple of years.
Laura Shin:
And so, you started to describe kind of briefly how Arwen works, but let’s go into it in a more detailed fashion. Why don’t you just walk me through what it looks like when someone trades using Arwen?
Sharon Goldberg:
Right. So, with Arwen, the idea is that you wouldn’t deposit your coins into exchange custody. You would use your own custodian to custody your coins while you trade. So, the way that this plays out is we use this concept called escrows. So, if you’re very deep into the blockchain space, you’re probably familiar with Layer 2 protocols, like the Lightning Network, and there’s various other Layer 2 projects out there, like Plasma and others that are built on Ethereum.
So, with Arwen, what we do is we have these escrows where, if you’re going to trade, let’s say, Bitcoin into BitQuick cash, you’re going to have escrows on the Bitcoin blockchain and escrows on the Bitcoin cash blockchain. The idea is that every coin’s blockchain acts as the agent of escrow for trading that coin, so, if I’m trading Bitcoin, the Bitcoin blockchain will be the agent of escrow.
And so, what you do is, instead of taking your coins and depositing them in the exchange’s wallet, what you do is you deposit them into an escrow on the blockchain, so Bitcoins go into a Bitcoin escrow in the Bitcoin blockchain, and then that escrow backs your trades with the exchange.
So, trades themselves are happening instantly in the sense…the same way that you would place a trade today, you know, for instance, through the API of an exchange. You would just place a trade through the API. That’s exactly what Arwen does.
You’re placing trades through the API of the exchange, but what we have in front of that API is basically the Arwen protocol that translates your trades into a cryptographic atomic swap messages that they happen instantly.
So, at the end of the day, what Arwen looks like is you take your coins, you put them in escrows. You do trades backed by those escrows, and then, when you’re finished trading, you would close your escrows, and your coins would return to your wallet. So, it’s sort of functionally similar to trading today where you would deposit coins traded on the exchange and withdraw coins, but the difference is that, instead of depositing into the exchange, you’re depositing into these escrows, and Arwen guarantees that, even if the exchange gets hacked or goes offline, you’ll still be able to close your escrows and claim all the coins that you’ve traded.
Laura Shin:
So, there’s something that I don’t know if I fully understand here. So, I put my coins that I want to trade into an escrow, and then, that smart contract, the escrow, talks to the exchange. So then, at what point does the atomic swap happen because, as far as I understand, that can just happen directly between two blockchains, so now I don’t understand where the exchange comes in.
Sharon Goldberg:
Yeah, so the escrows are not the atomic swaps, so the escrows are a smart contract on whatever blockchain it is that you’re using. So, for instance, Bitcoin versus Ethereum, you might have two escrows, one on Bitcoin blockchain, one on Ethereum blockchain, and then, you’re doing atomic swaps that are backed by those two escrows.
So, there would be a message that adjusts the balance in your Bitcoin escrow and also a message that adjusts the balance in your Ethereum escrow, and the adjustment would be exactly the value you traded. So, for instance, if you’re selling one Bitcoin, your Bitcoin escrow would be minus one, and if you’re buying, you know, ETH, then your Ethereum escrow would be plus two, and those messages essentially are sent back and forth between you and the exchange, and they adjust the balance in your escrow.
So now, when you have those messages, you have the ability to close your escrow at any time and claim the actual coins that you’ve traded. This is the idea of a Layer 2 protocol where you don’t have to actually go to the blockchain every time you do something. You just create transactions that, if you took them to the blockchain, they would do something for you, right?
So, what we’re doing is we do a trade. We create these messages that could close the escrows for us with the balance after our trade, but we decide not to close the escrow. We decide to do another trade, and then, when we get these new transactions that could close the escrow with the new balance of our trades, and we keep doing that until we actually decide to really close the escrow and take those messages and post them to the blockchain.
Laura Shin:
Okay, I think where I’m getting confused is when you say if I want to trade BTC for ETH, I’m not creating both escrows, am I? I’m getting one, and maybe if you…
Sharon Goldberg:
Yes.
Laura Shin:
Oh, I am. Okay. So, I have to put Bitcoin into one escrow, and then I put ether up into the other escrow?
Sharon Goldberg:
Yeah, so let me rewind a little bit. So, here’s what would happen.
If you were to trade with Arwen…let’s say we’re trading from Bitcoin to Ethereum, so the first step that you would do is you would take your Bitcoin, and you would put them in an escrow where the agent of escrow is the Bitcoin blockchain, right? So, that will collateralize your trades of Bitcoin.
The second thing you need to do is you need to get an escrow of Ethereum that will actually collateralize the purchase of Ethereum that you’re about to do.
So, the first step of Arwen is setting up the escrow that you funded, which is the Bitcoin escrow, and setting up this additional escrow, which we call an exchange escrow, that’s going to be funded by the exchange. So, in this particular example, when we’re doing Bitcoins-to-Ethereum trading, you would fund a Bitcoin escrow, and the exchange would fund an Ethereum escrow.
Now, once you have those two escrows set up, you can do atomic swaps that are backed by those two escrows.
Laura Shin:
Oh, now I get it.
Sharon Goldberg:
So, what’s happening is that the exchange is putting in Ethereum to collateralize your trades, to collateralize your purchase of Eve, and you are putting in Bitcoin to collateralize your sale of Bitcoin, and then you’re doing a swap across those two escrows and across those two blockchains.
Laura Shin:
Oh, wow, that’s super fascinating. So, something then that I want to understand here is, is there any central point of failure in this process?
Sharon Goldberg:
So, each party is sort of protecting itself, so in the sense that the guarantee is that, as long as the user’s machine is not compromised, or their custodian is not compromised, they are guaranteed that they can close their escrow with the correct balance, regardless of what’s going on, on the exchange side.
So, let’s say the exchange goes down, stops talking. You won’t be able to do any more trades with that exchange, right, because it’s just not there. It’s not accepting any trades, but you will have a particular balance in your escrow, so for instance, you know, you sold one Bitcoin, and you bought two Es. So, let’s say the exchange just disappears or loses access to its wallet or does an exit scan or something, you would be able to close your Bitcoin escrow and your Ethereum escrow with the correct balance after you trade. That’s what Arwen guarantees.
So, in terms of set single point of failure, there is a point of failure in the sense that, if the exchange went down, you can’t trade there anymore, but it’s not going to cause you to lose your coins.
Laura Shin:
And then, there was another aspect of how this works that you described to me before where you talked about…so, the Arwen protocol is what enables this to happen, but as far as I understand, there’s also something that Arwen runs, which is the Arwen hub, so what role does that play?
Sharon Goldberg:
Right. So, I just want to emphasize that Arwen provides sort of the same, similar value proposition to a decentralized exchange. In a decentralized exchange, a lot of the goal is to have self-custody of your coins, not to trust a third party with custody of your coins. That’s the same with Arwen.
The difference about Arwen from most decentralized exchanges is that Arwen does not have its own separate order book and its own separate ecosystem, like you might have, for instance, in 0x protocol or in the EtherDelta. Arwen is going to be plugging into the order books of the exchange.
So, that’s the big difference, and because we plug into the order books of the exchange, the trades have to happen necessarily as quickly as trades would happen in a normal, you know, regular exchange that was custodial. So, our trades are fast because they’re actually just regular trades on the order book. So, that the big difference.
You asked me about the Arwen hub. So, what the Arwen hub is, is it’s basically a cloud system that speaks Arwen on behalf of the exchange, so it translates Arwen messages into messages that can be placed on the exchange’s order book. So, rather than having the exchange like implement an entire Arwen protocol themselves, which is not very realistic, you know, my company writes the Arwen hub, which will take Arwen protocol messages from traders and translate them just into regular API messages that will go into the order book of the exchange.
Laura Shin:
And is there any chance that those messages could be hacked along the way?
Sharon Goldberg:
If they were, then that would affect only the exchange but not the users themselves. The guarantee of Arwen is that each party takes care of themselves. The traders are going to be secure, even if the exchange is hacked, and the exchange is going to be secure, even if the traders are hacked.
Laura Shin:
Oh, interesting. Okay. And then, is Arwen limited in the types of exchanges that it works with? Or I guess is it only centralized exchanges that you work with?
Sharon Goldberg:
Yeah, it’s currently focused on centralized exchanges. We haven’t looked at integration with decentralized exchanges yet. The reason for that is because we’re focused on getting into the places where there’s the highest amount of liquidity, which is currently the centralized exchanges.
Laura Shin:
All right. And then, we started to draw the contrast with trading on a DEX, but can you flesh that our even further? Like trading on Arwen versus trading on a DEX?
Sharon Goldberg:
Yeah. Yeah, so there’s one key difference, which is sort of the fact that we’re plugging into liquidity of the centralized exchange. If you’re trading on Arwen, you could be the only Arwen user on that exchange, and that’s fine. You don’t need to have your counterparty be another Arwen user. It’s just basically a trade that’s going on the order book and being filled and on the way that a trade would be sold on order, so that’s really the key differentiator is that we’re not as dependent as a DEX is on the growth of the network effects of that specific DEX because we just go into the exchange and use that, the liquidity from the exchange itself. So, that’s the key difference.
There are other sorts of important differences, which are the speed. So, Arwen trades are not actually executed on the blockchain, and so, that means a couple of things. It means that we’re not subject to slow blockchain confirmation times. Any sort of confirmations that have to hit the blockchain are for opening escrows and closing escrows but not for trading. So, the trades can happen fast. That’s the first thing.
The second thing is that, when you do a trade with Arwen, that trade is not visible to anyone except to you and the exchange that you’re placing the trade on, and so, that saves you from a lot of issues that are affecting DEXs right now, which have to do with frontrunning.
So, in the DEX…typical DEXs you execute your trade on the actual blockchain, so your trade will go to like an Ethereum smart contract, and the Ethereum smart contract will be the one to actually execute the trade. So, that means for your trade to actually happen, it needs to go through an Ethereum miner.
And so, that creates frontrunning risks because if the Ethereum miner doesn’t like your trade or wants to front run your trade, they have the power to do that before the trade has actually executed. And so, we see a lot of interesting frontrunning tricks that are happening today, like even now on a lot of DEXs.
We sort of sidestep all of this because there is no on blockchain execution of individual trades, so no miners are really involved in the execution of the trade, so you lose all of these risks.
So, the sum up, really it’s the liquidity of the exchange’s order book. It’s the speed because we don’t go to the blockchain, and it’s the lack of frontrunning because we don’t have to involve miners in actually executing trades.
Laura Shin:
Yeah. I was going to bring up that paper, that Phil Daian at Cornell wrote recently with some other academics about how arbitrage bots in DEXs are doing certain things, like paying high transaction fees and taking advantage of network latency to front run the ordinary user trades on DEXs. So, it sounds like, with Arwen…
Sharon Goldberg:
Yeah, that’s exactly what I was getting into there.
Laura Shin:
Yeah, and is there any situation in the future where it would ever make sense to try to implement something like Arwen on a DEX, or no? Is it only really for centralized exchanges?
Sharon Goldberg:
So, Arwen on a DEX is tricky because the model that we’ve taken with Arwen is…well, I have to be careful about that actually. Let me answer like the level one answer, and then, I’ll do the level two answer.
The level one answer is that Arwen on a DEX is a little tricky because Arwen swaps. The movement of clients is from the user’s wallet to the exchange’s wallet or to some sort of collateral party wallet. You know, if Alice is trading with counter party Bob, where Alice is an Arwen user, and Bob is not an Arwen user, Bob is just a regular custodial user of the exchange, there’s really no direct movement of funds from Alice’s wallet to Bob’s wallet.
It’s more that there’s a movement of funds from Alice’s wallet to the exchange’s wallet, and even though that sounds strange, that’s exactly what happens today. Like if you trade at…you know, pick your favorite exchange. The first thing you’re going to do is you’re going to move your coins into that exchange’s wallet, and then you’ll do some trades, and then you’ll pull your coins out of that exchange’s wallet.
So, it’s really trader-to-exchange wallet coin movement, and that’s what we have in Arwen, right? And so, that looks kind of different from what you have in a DEX, where DEX is the movement of coins as peer-to-peer. So, if Alice is trading with counterparty Bob, the actual swap will be from Alice’s wallet to Bob’s wallet.
The reason we took the Arwen approach was because we really wanted to integrate directly into centralized exchanges and not have to build kind of our own ecosystem because we didn’t think that was going to take off as quickly, and you can see that now with the liquidity indexes.
We think that, you know, that’s coming from the fact that you really need like the growth inside the DEX, and there’s all these speed and frontrunning issues. So, that’s kind why…the fact that it’s from the user’s wallet to the exchange’s wallet makes it really tricky to have it like fit directly into the DEX model. That’s level one answer.
Level two answer is that there’s a lot of interesting things that are coming out now, like Binance DEX and Hobie DEX, and these are like, you know, very different from what someone has as a DEX when you look at something like EtherDelta, and the difference is that these DEXs are really blockchains, separate blockchains that are supposed to allow you to trade like a large number of assets.
So, for instance, the Binance chain, you can trade Bitcoin on the Binance chain, right? And that’s kind of weird because Bitcoin is not issued by the Binance chain. Bitcoin is issued by the Bitcoin blockchain, right? So, how do you actually trade Bitcoin on the Binance chain? And the way you do that is through a peg.
So, Binance change currently uses this thing called BTCB, which is basically bitcoin.bpeg to real Bitcoin. It’s very similar to the way that Tether is pegged to US dollars, and so, you know, when you trade Tether, you’re not trading US dollars, but you’re trading an asset that’s pegged to US dollars. BTCB is similarly an asset that’s pegged to Bitcoin.
So, to make a long story short, that idea of like going from the asset that’s pegged, BTCB, to the actual real Bitcoin or going, you know, from real Bitcoin into an asset that’s issued on the Binance blockchain, that’s kind of a place where Arwen can come in and facilitate some of that swapping and some of that security.
The idea is that Arwen is designed to swap coins from their native blockchain, and so, rather than having to kind of, you know, deposit your Bitcoin at a custodian and withdraw BTCB, you could use Arwen to swap directly from the Bitcoin into whatever the asset that’s natively issued on the Binance blockchain would be.
So, that’s kind of one way that we could fit into a DEX, but it’s not really kind of plugging into EtherDelta the way you would think about it sort of at first glance.
Laura Shin:
Oh, super interesting. Yeah, so I guess it could be used in that fashion that is actually really fascinating. And then, one other thing I want to ask you about was when you kind of talked about, you know, why it is that you’re more focused now on centralized exchanges and how you get a certain level of speed, does that mean that high-frequency traders could also use Arwen?
Sharon Goldberg:
So, we are paddling in that direction right now, so if you have high-frequency traders use Arwen, we need basically bidirectional escrows that you can sell and buy on the same escrow. So, if I put in Bitcoin in an escrow, right now you can sell the Bitcoin out of that escrow, but you can’t buy it back with the current like product that we’ve launched as of today.
But you know, if you read our white paper and on our roadmap it’s to support both buy and sell, and so, one could use that que very quickly, you know, like buyer-Bitcoin, buyer-Bitcoin, buyer-Bitcoin, buyer-Bitcoin, which is what I think high-frequency traders are really looking for.
We’re still doing a bunch of research on like what the high-frequency trading market would look like in the crypto world. What I’m hearing is that there’s starting to be interest in building HST data centers and using these strategies, but I’m not hearing that the demand is that high for like HST at this specific moment.
So, we’re sort of…that’s a little further on our roadmap, but we have the capability of doing that, and if you look at our white paper, like the protocols for doing that are all laid out already in there.
Laura Shin:
And earlier, when you were talking about why it is that Arwen works with centralized exchanges, I think something that’s interesting is that you went in that direction while, at the same time, in the industry, it sort of feels like everybody’s working on decentralized exchanges. So, do you feel that somehow Arwen’s going against the tide in the industry, and why do you think that is, and how do you kind of plan to navigate that?
Sharon Goldberg:
I think we definitely went against the tide, and we did it very inadvertently because, you know, in 2017, when we started thinking about what kind of company we wanted to build, we were never of this opinion that decentralization is the key value proposition of blockchains. We were of the opinion that trust and trustlessness and managing trust was the key value proposition of blockchains.
So, let me explain what that means. One thing that has been sort of this constant current through my career as a security researcher and engineer is that everywhere in the internet, you have these things called trusted third parties. For instance, if you want to use HTTPS to securely access a website, at the end of the day, the security of that entire ecosystem comes down to a few centralized parties, which are called certificate authorities.
And these guys are…you know, they get hacked, and bad things happen, and there’s all sorts of attacks that you can do because of this like trusted centralized authority, and it’s sort of like at the root of a lot of internet systems, like, you know, web encryption or DNS, the Domain Name System. You find this stuff everywhere.
And so, for us, like blockchain was this amazing thing that is an entity that you can trust, but it’s not a single entity. It’s not a single trusted third party. The blockchain itself provides a root of trust that you can do cryptography with, but it’s not a single party that can get hacked, so it’s very hard to sort of roll back blockchains. That’s the whole point, but there’s not just like a single place you can go and attack and like kind of skew the whole blockchain, and that’s really fascinating.
So, that’s always been my understanding of what blockchains are and what they’re for, and so, when we started this, we said, okay, look, there’s a whole bunch of things called centralized exchanges. They’re getting attacked. They’re getting hacked. Money’s getting stolen. How can we continue to trade on these things but not have to trust them?
And then, we said, okay, let’s continue trading on them but just use the blockchain as a root of trust, and that’s where Arwen came from. The idea is you want to trade on centralized exchanges. Everybody wants to trade on centralized exchanges. Look at all this volume, right. Twenty seventeen, look at all this volume. How do we trade on them without having to trust them? And that’s where the whole thing started.
So, when the whole DEX, you know, movement took off, we were just going in a completely different direction, saying like everybody wants to use centralized exchanges. Why don’t we just figure out a way for them to use them without having to trust them?
Laura Shin:
And from the exchange’s perspective, how does working with Arwen benefit them, and you know, what do you have to do to convince them to work with you? Like, I don’t know how much work it takes to integrate Arwen.
Sharon Goldberg:
So, working with an exchange is like your general enterprise SASS sales cycle, so you have to convince basically the executive at these exchanges this is a good idea. So, you know, that takes some time.
In terms of why they would do it, I think that there are a couple of reasons. There are a lot of actual leadership at exchanges that really believes in the promise of blockchain and the idea that blockchain provides sort of trustless…all this ability to do things trustlessly, that things are decentralized, and we have found actual leadership at a number of exchanges that this is like something that they truly believe.
And so, they’re bothered by the fact that they’re like, therefore, acting as this custodian…basically, acting like a traditional bank for everyone’s crypto. Like, they just feel like that’s against the ethos of the space. So, that’s like one thing that you sometimes find, and that helps us get to a deal faster.
The other thing is that, you know, providing these types of features puts them basically in a leadership position in the market. So, we launched on KuCoin four weeks ago, and there was bunch of press about, you know, this is like a bold move that they’re leading the space, and they’re kind of doing things that other exchanges haven’t bothered doing. So, at the moment, what Arwen is, is really a differentiator to draw traders to an exchange that would allow them to transact from an exchange in a way that they can’t really do in many other exchanges.
And so, that’s really the main value proposition that we’re offering. These exchanges like being able to provide the service to their users and kind of lead the industry in that direction.
The other piece that I think is really important is that, if you look at what’s been going on in the crypto space in 2018, you know, we can sit here and just start naming the number of start-ups that are trying to solve the custody problem. So, let me just start.
You know, Casa, Ledger, Trezor, BitGo, Anchorage…like let’s just keep going…Xapo. There’s tons of them, right? And they’re all super interesting, different coin type. Everyone’s doing something different, and they’re all really interesting, but at the end of the day, like you have your coins in this custodian, and as soon as you go to trade, they’re not in your custodian anymore.
And so, you know, what is driving me personally to kind of like push this into the market is that there are going to be users. They’re going to be using these custodians. They’re going to be invested in their custodian, and they’re going to say this is my secure custodian. I don’t want to take my coins off my custodian, but I still want to do things with them, and protocols like Arwen allow them to do that.
And so, I think that as a custodian gains market traction, there’s going to be more market traction for protocols like ours, and not having to like really change what they already do, because people are used to trading on centralized exchanges, but they just want to get to those exchanges in a way that keeps their coins protected by their custodians.
Laura Shin:
Great, and actually, just that last piece about what it takes to integrate with Arwen, is it just a kind of writing some code to interact with the smart contracts, with the escrow smart contracts?
Sharon Goldberg:
No. No, no. So, we take care of all that, so really, the integration has to do with us coming into the exchange, looking at their ATI endpoints for trading, and terminating against those ATI endpoints. So, we need certain endpoints. So, for instance, some exchanges have the ability to programmatically withdraw coins from the exchange, right? There’s like a withdraw endpoint in the API.
So, if that’s present, we can use that to do an Arwen integration. If that’s not present, then we may need that to be exposed. So, that’s sometimes where the latency comes in. Like if you look at the big exchanges like Binance, and I think, Piloniax, Bitstamp, they all support this type of programmatic withdrawal from accounts.
We need that, but some of the smaller ones may not have that, so we have to wait for that to be added. That’s kind of the integration. They don’t have to do anything Arwen-specific. It’s just that we need certain like basic functionalities from the API that some of them have, and some of them don’t have.
Laura Shin:
All right. So, we’re going to talk more about the custody stuff that you mentioned and also more about smart contracts after the break, but first, a quick word from our fabulous sponsors.
Oslo Freedom ForumOn May 27th I will be hosting a conversation about the future of finance and human rights at the Oslo Freedom Forum in Norway. As the world continues to move toward a cashless society, paper currency is disappearing. Companies like Facebook, Apple, and TenCent are becoming increasingly influential in the digital payment space. We’ll discuss how individuals and companies can preserve and protect financial freedoms in the digital age. I’ll be joined by Bitcoin author and educator Jimmy Song; Casa chief technology officer Alena Vranova; and CoinCenter founder Jerry Brito. To register to attend, you can visit oslofreedomforum.com today — use a discount code “unchained25” to get 25% off your ticket price!
Cipher TraceDid you know that if money laundering were an economy, its GDP would be the size of Canada’s? Large volumes of tainted crypto assets move through financial networks, often below the radar of banks. Cyber-criminals use unregulated crypto exchanges to avoid detection. No wonder governments around the world are rolling out tough new anti-money laundering laws for cryptocurrencies. Complying with those laws isn’t easy. Banks and exchanges need the best cryptocurrency intelligence available, to avoid penalties. Now you can use the same powerful AML and compliance monitoring tools used by regulators. CipherTrace is Securing the Crypto Economy. To learn more, visit Ciphertrace.com/unchained.
Laura Shin:
Back to my conversation with Sharon Goldberg of Arwen.
So, when you were mentioning custody earlier, do you have to sign on specific custody solutions, or can people use any other custody solutions with Arwen?
Sharon Goldberg:
Yeah, so, currently, you can use Arwen with any custody solution that you have, and again, this is a two-part answer to your question. So, right now, what we do with Arwen is that there is a piece of code called the Arwen Demon that speaks Arwen on the trader’s behalf.
What the Arwen Demon really has inside it is keys that are associated with the Arwen escrows, so when you take your coins and you put them in an Arwen escrow, there are keys associated with that escrow that allow you to move coins around and do trades.
So, the ideal integration from a security perspective for us would be to have the keys associated with the escrow actually housed inside your custodian, right? So, for instance, if you’re trading with a ledger, then those keys would be inside the ledger.
And so, that type of integration requires, you know, a bunch of work on our side and on the custodian’s side to actually execute upon, and you know, we’re interested in doing some of that with like a variety of custodians, but at the moment, we have sort of a different way of doing it, which is that keys associated with a particular escrow would live on the user’s local machine.
And so, those coins are actually never going into the custody of any third party. The coins in the escrows are actually being moved around by the Arwen Demon, which is an executable that the user would locally download. So, that means that you can basically use any custodian you want to transfer coins into the Arwen escrows. The Demon will move the coins around in the escrows, and then, when you close your Arwen escrows, those will go back into your custodian directly.
So, at the moment, we can support any custodian, but as we go into sort of like higher levels of security, we’d have more deep integrations with specific custodians, and that’s kind of what we’re working on right now.
Laura Shin:
And when you say any kind of custodian, do you also include things like Xapo and coin-based custody…stuff like that?
Sharon Goldberg:
Yeah.
Laura Shin:
Okay.
Sharon Goldberg:
Yeah, yeah, yeah, because today, what you need to do to actually move a coin into an Arwen escrow is that the Arwen Demon will tell you an address. So, for Bitcoin, you’ll be given an address that is the hash of the Arwen escrow smart contract.
So, you have the Arwen escrow smart contract, and you hash it. That gives you an address, and that’s an address like any other address, and so, what you would do is you would fund that address, and that locks your coins in escrow.
The address itself is the hash of a smart contract, and that smart contract is associated with escrow keys that allow you to adjust the balance in the escrow.
Laura Shin:
And so then, if Arwen were widely adopted, the general vision would sort of be like people could, if they want to, outsource their custody and also transact with a centralized exchange, which would enable them to, number one, not to have to worry about managing their own private keys necessarily and then also to not worry about their funds on any exchange being hacked. Is that right?
Sharon Goldberg:
That’s right. Yeah, that’s right.
Laura Shin:
And so, how does Arwen make money?
Sharon Goldberg:
So, we work with the centralized exchanges, and we share part of the revenue with them that comes through the trades that come on Arwen.
Laura Shin:
Like from each transaction?
Sharon Goldberg:
That depends on the particular exchange, but it could be done per transaction. It can be done sort of as a fee at the start of the year. There’s different arrangements that we’ve considered with different players, but it has to do basically with, you know, the traffic that comes in through the Arwen platform to the exchange.
Laura Shin:
And what coins do you support? And how do you decide which ones to support?
Sharon Goldberg:
Yeah, so currently, we have three coins because we’re still in beta, and we’re moving towards like the full set of features that we plan to support by the end of the year, but right now, we have Bitcoin, Bitcoin Cash, and Litecoin. So, we’re probably the only protocol right now that can support swaps from Bitcoin into Bitcoin Cash actually, and so, those are the three that we have right now, and the team is currently working on Ethereum and ERC-20.
And so, when we have that part, we’re going to have the ability to basically swap from all the ERC-20 tokens plus the Stablecoins into Bitcoin, Bitcoin Cash and Litecoin. So, that’s where we’re starting from.
We started there because we’re looking for the highest-volume coins at the moment, and that was always Bitcoin, so we started there first. The other reason to start with Bitcoin is because Bitcoin is actually harder to write protocols for, and so, we started with basically the most difficult part of the problem first, and when we have solved that problem, we can basically migrate that onto other coins that are easier to write smart contracts and protocols for, and that’s, for example, Ethereum. So, that’s why we did Ethereum second.
I mean, if you can actually nail down all the protocol and specificities in Bitcoin, it’s a lot easier to port that over to Ethereum, which is completely different and a lot easier to program than Bitcoin.
Laura Shin:
And then, earlier, I asked you so how do you decide which coins to support?
Sharon Goldberg:
Yeah, so, we looked at the highest-volume coins, which were…
Laura Shin:
Oh, no, but I mean like going forward, how will you decide?
Sharon Goldberg:
Oh, going forward. Yeah. Sorry. Going forward, you know, I’m very interested in a lot of these different new coins that are coming out, like we have Cosmos, and we have Binance chain, and we have Zcash. There’s just a huge number of super interesting new blockchains that are out there.
I think that, right now, we have our hands full with the Ethereum and ERC-20 leg, and I think that like, at the end of the 2019, we’re going to have an interesting view on like what are the next set of really key blockchains, and I think, at that point, we’re going to look at, you know, what are those blockchains…what type of scripting functionality do they have, and pick on the basis of both, you know, like the popularity, the importance, and the success of the coin, and also, like what we can actually do with it in terms of scripting and protocol design.
Laura Shin:
Okay, but generally, you sort of envision that eventually Arwen will support as many coins as possible.
Sharon Goldberg:
That’s right. Yeah, yeah. I mean, we just have to pick and choose in terms of the size of our team and what we can execute on, but yeah. Our goal is really to have coins that are not designed to interoperate with each other, to allow those to interoperate.
So, we’re looking at the type of coins that aren’t supposed to really be able to do cross-blockchain protocols and focusing on actually executing the cross-blockchain protocols for coins that aren’t supposed to be able to communicate.
So, like the classic right now example of that is, you know, Bitcoin and Bitcoin Cash. When they fork, you know, protocols that are designed for Bitcoin typically cannot work with Bitcoin Cash, but we, on purpose, took the point of view that we need to be able to support something like Bitcoin Cash even though they’re not really designed to support each other because people are actually trading, you know, these pairs in practice.
And you know, that’s where the difficult problem is, and that’s the one we’re solving.
Laura Shin:
So, just a minute ago, you said that Ethereum is easier to program on, but I know from our pre-interview that you personally actually find it the other way around, that Bitcoin’s easier to program on. So, can you talk a little bit about what it’s like working with smart contracts in Bitcoin versus Ethereum, and you know, what the differences are and why maybe your point of view is a little bit different from others?
Sharon Goldberg:
Yeah. So, the definition of the word easier, like easier can be interpreted in so many different ways, so I have to be like precise about the meaning of easier. So, the way I was using easier two minutes ago was, you know, Ethereum smart contracts can support a massive amount of functionality. They are Turing complete, which is something that everyone says, but you know, something that’s like more meaningful, apart from the Turing completeness of Ethereum smart contracts is like you can write all sorts of code in Ethereum. You’re not super restricted into what actual, you know, functions you can execute in Ethereum.
With Bitcoin, you can like basically do three things. You can find transactions, not messages. You can’t find whatever message you want. You can only find transactions, and you can compute hashes and check that the hash matches, like the pre-image of the hash matches the hash value. So, those are the two things you can basically do in Bitcoin.
So, given that you can only do two things in Bitcoin, it’s not that easy to come up with protocols that have rich functionality given that you can really only do two things in the scripting language. And so, that’s why I say it’s, you know, harder to write protocols in Bitcoin than it is to write protocols in Ethereum because, in Ethereum, you can put a lot of the complexity of the protocol into the actual smart contracts because the scripting language supports doing all sorts of things that are not possible to do on Bitcoin. So, that’s one definition of easier.
Okay, so, now let’s go to the next definition, which is when you’re actually building a smart contract, right, so now, you’ve figured out your protocol. What is a protocol? A protocol is what your smart contract is plus what messages you send back and forth. That’s the definition of a protocol.
So, in Bitcoin, your smart contract is really simple, right, because you can barely do anything with the scripting language, but the messages you send back and forth may be more complicated, right?
In Ethereum, your smart contract is much more complicated, but the messages you send back and forth may be simpler.
So, what’s easier is like when you’re writing the Bitcoin smart contract, it’s really easy because it’s just a little, dumb thing that can do very few things, and actually, programming that and making sure it doesn’t have any bugs is not that hard, right, because it just doesn’t really do much, so there’s very few places where you can go wrong.
For instance, the Arwen escrow smart contract has 16 op codes in it. Bitcoin script is like assembly language, and we only have 16 op codes in our escrow smart contract.
If you try to take the exact same smart contract from Bitcoin and try to port it into Ethereum, you end up with this complex, you know, much more complex contract that has many more lines of code, many more function calls, many more, you know, things that are being done just to execute the exact same functionality, and that comes because the Ethereum scripting language is more rich than the Bitcoin one.
So, when I say it’s easier to write Bitcoin smart contracts, it is because they’re so simple, once you figure out that like this smart contract lets you actually do something, then writing it is easy. The hard part with Bitcoin is figuring out like how you can actually do anything with this very crippled scripting language that’s available to you in Bitcoin. So, that’s a very long answer to the question, and the reason that I find like Ethereum more difficult is because actually writing the smart contract itself, once you’ve decided what it needs to do, is trickier because there’s more instructions, and there’s all sorts of strange things that can happen in Ethereum because of the way the Ethereum virtual machine works that is very different from like just writing regular like Java code or C Sharp code that programmers are used to.
Laura Shin:
And so, I actually don’t know what an op code is. So, when you said there are only 16 of those in Bitcoin, is that like lines of code?
Sharon Goldberg:
It’s actually an instruction, so it might be like…like an if could be an op code, or sign something could be an op code, or check the hash could be an op code, check the time could be an op code. Those are examples of op codes, so there’s just a very small number of instructions that you do. It’s not a function call, but it’s like an instruction, almost like a machine instruction.
I don’t know if that clarifies. It’s a very engineer answer, but it’s just that there’s a few like commands you can use in Bitcoin, and each op code is like one of those commands.
Laura Shin:
And then, in Ethereum, for the same functionality, how many lines is that? You said it was a lot more.
Sharon Goldberg:
Yeah, it’s a lot more. I don’t want to tell you the number because we’re still working on the exact details of the smart contract right now, but the difference is that you don’t have specific instructions, but you have, you know, like function calls. You have like a rich, you know, programming language. You’re not writing code like machine code, like assembly language like you are in Bitcoin. That’s the real difference.
It’s like if you think about normal programming languages, when you’re programming your CPU on your computer, you’re using a language called assembly language, which is very simple, like move something here, multiply something, hash something. That’s what you would do when you’re programming like a CPU, and then, like a normal developer would be writing like Java script code that’s many, many layers above what you would be doing when you’re programming a CPU with assembly language.
So, the difference is, when you’re programming Bitcoin, you’re really doing like assembly language programming, like one layer above the CPU essentially, and when you’re doing Ethereum programming, you’re using a richer programming language that has a lot more functionality but also a lot more complexity.
Laura Shin:
All right, super interesting. So, let’s move on to some thought experiments. You did mention before Mt. Gox and Binance, and just in general, you know, obviously, there’s this whole history of exchanges being hacked in crypto, and here it is that you are working on a project that, you know, could potentially, I think, prevent situations like that. So, what do you think would have happened if users had been using Arwen in the Mt. Gox situation?
Sharon Goldberg:
So, if an exchange is hacked, basically what happens is the exchange would be disabling the ability to deposit and withdraw coins from that exchange, and so, in Arwen, what that looks like is that the exchange would refuse to cooperate in closing any escrows that you have with that exchange.
So, remember in Arwen what you do is, you take your client’s deposits in an escrow, do a bunch of trades, and when you’re finished, you close the escrow, and the coins are returned back to your wallet and can be used for other purposes.
So, if something bad happens to the exchange, when you go to close your escrow, the exchange will basically not respond in the way that you expect or just refuse to respond at all. So, you’d see that something’s wrong, that this escrow is not closing because the exchange is not cooperating with you to close this escrow.
At this point, what Arwen would do is go into a coin-recovery procedure, and what that is, is that the Arwen Demon, the piece of software that speaks Arwen on your behalf, will detect that the exchange is not properly responding to your request to close an escrow, and so, what it’ll do is it’ll say to you, you know, this escrow is frozen, the exchange is not responding. You’ll be able to recover your coins and get the coins out of this escrow on your own without the assistance of the exchange after a particular time.
And so, Arwen escrows are associated with an expiry time, so when you set up an escrow, you might say, you know, this escrow is for three days. What that expiry time does is that, when it’s over, it allows you to unilaterally close the escrow without the exchange’s participation, so even if they’re hacked or they’re behaving badly or something else has gone wrong on the exchange, you can still, on your own, basically withdraw the coins from the escrow and use them for something else, even if the exchange is not there.
So, the bottom line is, if the exchange is hacked, you’d just be having your coins frozen in escrow for some period of time, at which point you’ll be able to recover them according to the balance of the trades that you’ve done up to that point.
Laura Shin:
So, that sounds like it would apply really to any of kind of these high-profile situations on exchanges like Mt. Gox, Quadriga, Binance. Is that true that it would apply to all those?
Sharon Goldberg:
That’s right, yeah. So, essentially what happens is, if something goes wrong on the exchange, and they stop processing withdrawals, which is essentially what you see in all these cases, right? For instance, Quadriga, you know, the founder disappeared. You know, they lost access to their wallets, so if they don’t have access to their wallets, they can’t process withdrawals, right?
Another example would be like if an exchange is hacked, they can’t process withdrawals because there’s no longer coins in their wallets to support the withdrawals. So, it all comes down to just basically withdrawals not happening.
So, what Arwen looks like in all those cases is, you know, the exchange is not participating in closing this escrow, so we’re going to close it on our own.
Laura Shin:
Yeah, and just to clarify, the Quadriga owner died apparently, but there is some suspicion about whether or not that was true. But then, I actually also wanted to ask…so, I know we don’t have all the details about Binance just yet, and for listeners, we’re actually recording very shortly after the Binance hack was announced, so potentially, between the time we’ve recorded this and the time we’re releasing it, there could be more details that are revealed, but Sharon, for you, knowing what we know so far, is there any…because like as far as I understand from Binance’s announcement, it looks like individual-user accounts were compromised.
So, I don’t know if an exchange were using Arwen, and if like, you know…I don’t know how that would affect the ability to do something like that.
Sharon Goldberg:
Right. So, with the Binance hack, right, what we know, and we’re only like a couple days from the hack, so I’m sure more details will emerge in the coming days…what we know about this was that, according to the information that’s out there, is that several user accounts were compromised, and it sounds like those accounts were sort of compromised at whatever rate they were compromised, slowly accumulating compromised accounts.
And then, at one point, the attacker sort of took all those accounts, and it sounds like it told them all to withdraw at the same time, and to withdraw coins to a particular place which is controlled by the attacker or places that are controlled by the attacker. And so, all those accounts kind of just, in one transaction…all those accounts requested a withdrawal kind of at the same time, and that withdrawal was processed at the same transaction all in one shot, and then, we had 7 thousand bitcoins from the Binance hot wallet just removed into the attacker’s control.
So, that’s what it sounds like, so individual accounts probably were compromised at some very slow rate, and then, the attacker like waited until the moment where he or she actually did the withdrawal and processed at the same time.
So, that’s, you know, that’s a pretty aggressive and precise attack, and that’s what we know right now, but you know, I’m sure more details will emerge in the future. So, the effect of this was that, essentially, the hot wallet was emptied, and there were no more coins in the hot wallet.
So, consider another user that wasn’t compromised that’s just coming in and wants to withdraw their bitcoins from Binance. Well, there’s no more Bitcoin in the hot wallet anymore, so you can’t really do withdrawals unless they take them out of their cold wallet, which of course, they weren’t going to do at the moment that they discovered that something went wrong, right, so the exchange is not going to like, you know, be like, okay, now let’s pull out our cold wallet and expose that and risk that being hacked.
So, at that point, you know, withdrawals are frozen because that’s the only like natural thing to do at that point. So, there’s nothing, you know, strange about that, and at the end of the day you know, Alice, who has nothing to do with any of this, can’t withdraw her Bitcoin even though they’re her Bitcoin, right?
So, the difference with Arwen would be we’d be back into this case where like the user wants to close an escrow. She never actually gave her Bitcoin to the custody of Binance, so even if Binance’s hot wallet was empty, she still has whatever her escrows are. Those are still locked on the blockchain, even if Binance is like I’m going offline, I’m not going to talk to anyone for a week while I figure out what happened here.
She could unilaterally close those escrows without the participation of the exchange, so really, the whole technical contribution and all the innovation in Arwen is really just dealing with the case where the other side of the escrow…you know, I’m escrowing with Binance…just disappears and doesn’t want to participate anymore, and so, I have to be able to recover on my own without the exchange’s participation.
Laura Shin:
Right, but actually, so my question was ever so slightly different, which is that it sounds to me like an attacker wouldn’t be able to perpetrate this type of hack because…and sorry, if you know, all the users are using Arwen, because essentially what would just happen is that then individual users might lose their coins, but the exchange…I mean, granted then it gets a little weird because then we’re sort of presuming…anyway, I think you see where I’m going because then, it’s sort of like the exchanges don’t have any coins, but essentially, it would…so if an individual user gets compromised, it kind of like localizes the damage to that individual user, rather than the exchange suffering. Is that sort of right?
Sharon Goldberg:
Right.
Laura Shin:
Okay.
Sharon Goldberg:
Yeah, so I mean, there’s two answers. One is that the user really doesn’t care what happens at the exchange, so even if the exchange is like out to get her the entire time she’s trading on that exchange, she doesn’t care because the exchange’s job is to put up escrows properly, you know, and when you’re setting up the escrows, if the exchange is not doing that properly, you won’t set up escrows, and so, those coins never go anywhere, and you’re never interacting with this exchange.
Once you set up the escrow, even if the exchange is attacked, to you it doesn’t matter because you always know you’re going to get the coins you expect. So, that’s really the first piece, right?
Second piece is that, when you have self-custody, you kind of make it harder to attack because the coins are housed, you know, in a distributed way, and so, that’s really like the value proposition of a lot of these self-custody solutions is that you have to really go after the custodian in order to attack the user, and that’s harder to do because it’s like more places that you have to attack.
Laura Shin:
All right. So, let’s talk about kind of where you’re going. So, you guys only recently launched in beta, and as far as I know, there’s pretty restrictive limits on the trading, so you know, probably because you are in beta…first of all, can you actually state what those limits are right now?
Sharon Goldberg:
Yeah, it’s about $100 right now for the next little while, but we’re going to be raising it soon.
Laura Shin:
And so then, going on to the future, it sounds like right now, you’re working on adding Ethereum and ERC-20 tokens, but what are some other features you’d like to add on Arwen in the future? I kind of wondered if something like margin trading would work on Arwen.
Sharon Goldberg:
Um hmm. Yeah, you know, I can say this right now, but I don’t want to like commit to anything I’m saying because we’re still evaluating specifically what’s next.
There’s a lot of different things that we like now how to do and are interested in doing, so one of them is a high-frequency trading support that you mentioned, which we think is really exciting, but you know, we’re looking at what exactly the demand is for that particular feature right now. There’s also, you know, the notion of options, so you know, talking to traders, there’s a lot of interest in having options that are not fully collateralized, so finding a way to do that could be really interesting. It’s not something that we’ve started building, but something that we’ve talked about and thought about.
Other pieces are, you know, providing you know fast movement of coins between exchanges, so you know, you’re on one exchange, and now you can really quickly move your coins to different exchanges in order to exploit whatever price opportunity you have at different exchanges.
So, you know, all of a sudden, just kind of throwing out in the air because we haven’t committed to exactly which one of these steps is next. Potentially, integrating this with like banks and providing Fiat support as well through atomic swaps is another thing that we’re interested in, so there’s a lot of different opportunities that we’re looking at right now, and you know, we’ll have like more details on exactly what that is in like the coming months, but right now, our real focus is just like providing this support for Ethereum and basically reducing some of the friction in using Arwen.
Laura Shin:
And sorry…just want to understand what you meant when you said that you would be interested in integrating with banks for doing atomic swaps with Fiat. Do you mean with like Stablecoins that are pegged or something like that?
Sharon Goldberg:
No. So, Stablecoins are going to come through Ethereum, so that’s already on the roadmap, right, so when we have the ERC-20, we will have Stablecoins, and then, that’s done basically.
No, there’s also a question like real fiats, through real banks, and how can you work with a bank to actually have them support…you know, if there’s movement of Fiat in the bank, how can you actually reflect that movement of Fiat on the blockchain? You know, so if you move, let’s say, 100 dollars in the bank, then maybe you move like part of a Bitcoin on the blockchain at the same time.
So, some of those ideas are things that we’re thinking about, too.
Laura Shin:
Okay, okay. So, I feel like I’m a little bit confused. So, it’s like, for a bank to use Arwen for their internal ledger…
Sharon Goldberg:
No, no. So, let me back up. So, people are very interested in crypto-to-Fiat trading, right? So, they want to trade dollars for Bitcoins. That’s perhaps the most popular market you have in Korea, you have Bitcoin for won and things like that. So, these are very popular markets, and this concept of atomic flops, which is something that’s very particular to cryptocurrency, can be expended, so when you do an atomic swap, you typically have, you know, two cryptocurrencies, and you’re swapping from one to the other.
So again, an atomic swap means Alice has item A, Bob has item B, so either Alice gets B and Bob gets A, or Alice keeps A, and Bob keeps B, so it can’t be the case that Bob gets both A and B or Alice gets A and B. That cannot happen. That’s what an atomic swap is.
And you can do that on the blockchain without trusting anyone because that’s the whole point of having a blockchain. It’s a way of, you know, having mutually untrusted parties agree on some form of truth, and so you can use that to build these protocols. That’s like an atomic swap when people in the blockchain space say that. That’s what they mean.
But you can also extend that concept, right? So, if you’re actually moving Fiat like real money, you know paper money or money inside a bank, you know, there’s no blockchain there, so it’s hard to kind of use the blockchain as a route of trust, but when people move Fiat around, they use banks, right? When you’re moving Fiat back and forth, you might be using a bank to facilitate this transfer.
So, if you’re transferring, you know, Fiat in the bank world, maybe you can reflect the movement of Fiat in the bank world actually for the movement of coin on the blockchain, right? So, you don’t really have to trust anyone to move the coin, although you may have to trust the banks to move the Fiat, so tying those two pieces together is something that we think is pretty interesting.
Laura Shin:
Hmm. Something about this sort of reminds me of like colored coins or something, but anyway, okay. Well, it sounds very interesting, and I guess, you know, we’ll sort of see if that the directions you go in.
Well, it’s been so great having you on Unchained. Where can people learn more about you and Arlen?
Sharon Goldberg:
So, we have our website, Arwen.io. If you want to try Arwen right now, you can go to Kucoin.com which is the first exchange that we launched with, which is one of the top exchanges in Asia, and just click on trade from your wallet, and you’ll be able to use Arwen to trade directly on Kucoin.
Laura Shin:
Okay, actually, there was one piece of that I think I saw, which is this is restricted to non-US residents.
Sharon Goldberg:
That’s true. Yeah.
Laura Shin:
Okay.
Sharon Goldberg:
So, just if you want to talk a little bit about the whole regulatory…
Laura Shin:
Actually, yeah.
Sharon Goldberg:
…space…
Laura Shin:
Yeah, can you?
Sharon Goldberg:
Let’s do that, yeah. Okay, so, you know, we’ve been in this business for a year and a half, and we’ve been talking to a whole bunch of exchanges, and we discovered a bunch of things.
So, the first thing we discovered is that, in Asia, crypto-to-crypto trading is a really big thing. It’s a much bigger thing than it is here in the US. So, that’s really interesting.
Second, is that in Asia, sort of the regulatory situation looks very different than it does here. So, let’s take the example of Japan, for instance. In Japan, you have the Financial Service Agency of Japan, the FSA that regulates all crypto trading and transactions and payments and so on. So, there’s this one body called the FSA.
They regulate everything. They’re a very strong regulatory, and they’re very tough, and they’ve very involved, but there’s just one of them, and you know who they are, and so, if the FSA licenses you you’re done.
And so, you know, that’s sort of like the picture in Asia. Either the regulators are very friendly, and they’re just like, you know, try what you’re doing, and we’ll regulate you later, after we figure out what it is that you’re doing here, or it’ll be, you know, we’re the FSA. We’re very tough, but you know, you just know that it’s us, and you can work with us, and that’s the end of it.
So, that’s kind of the picture in Asia, and so what happens for us is, because what we’re doing is so new, it’s hard for the regulators to really know what we’re doing because it’s new, and so, that makes it a little bit harder to kind of like launch with a US exchange or with US users because we have to kind of make an argument based on laws that are very old, that involve multiple regulators…like you have the SEC, the CFTC. You have the states. You have the FinCEN. You have just a large number of different regulations that apply to cryptocurrency.
You have to figure out how your new protocol fits into all of these different rules and what that means for taking US users. So, that’s really tricky when you’re a start-up, and so, what we’ve done is essentially like just worked outside of US because the regulatory landscape here…people say that it’s unclear. I think a more correct term is that it’s complex and that there’s a lot of regulations that you have to comply with, and not complying with them results in very bad things.
So, kind of sidestepping all of that is what we’ve chosen to do at this point because we’re really kind of breaking new ground from a technology perspective, and understanding how that fits in with the regulations, you know, is something that we’re working on, but it takes time.
So, at this point, right, we’ve launched with an exchange that is abroad and doesn’t really take users that are KYC’ed from the US and as a result. You know, Arwen is not available to US users right now.
I think, broadly speaking, I personally would like to see less of this happening because I’m sure that you’ve encountered a lot of new projects, coming from teams here, actually launching outside of the US, and it really comes from this. When you’re doing some of the really innovative things that are coming out of this country, understanding how the regulations apply to those things is really hard and expensive and involves lots of lawyers, and it’s not just one set of rules, but it’s like 10 of them, and that’s what you see.
You know, you see launches outside of the US. That said, you know, we are working towards getting launched in the US, but it’s like a longer and slower process that’s involved…you know, figuring out exactly how we fit into many more regulatory frameworks that are old and not really designed for this space.
Laura Shin:
Yeah, I’m so glad that this came up at the end in a very roundabout way because I did ask SEC Commissioner Hester Peirce about this when I interviewed her, and I think she seemed to say that she was kind of sympathetic or did understand that this was something…or actually that she was just concerned about what you said, that kind of innovation was happening here, but people were bringing their innovation elsewhere for that very reason, and she did say she listens to the show, so perhaps she’ll hear this or I think other people that the SEC listen to the show as well, so maybe they’ll hear…but yeah, I do find it’s a pretty wide spread sentiment actually amongst a lot of entrepreneurs in this space, at least in the US, so you’re definitely not the only one.
Sharon Goldberg:
Yeah, and I just want to say that like if I can move a conversation just little bit through this podcast is that like you always hear like there’s no regulatory clarity. There’s no regularity clarity. Like that’s not the point, right? Like so we don’t know if ETH is a security or not a security. Like that’s an important question, a huge question, right? But there’s so much more going on, and there’s so much more innovating happening.
So, for instance, what’s, and what does that imply? When you have a 2 of 2 Multisig, who’s controlling the asset? Who’s transmitted the asset? How does that fit into, you know, money transmitter laws, and how does that transmit into like the state money transmitter laws, and how does that affect the CFTC’s view of the object, right?
Like you’ve just got to think about so many different regimes and so many different rules, and you know, this country has very well-developed regulations for financial services, which makes sense, right, given like the history of all sorts of things happening when there aren’t regulations? But you know, what this is, is like a new way to transact, and what we’re doing is, is we’re having to fit this into these old rules, and you know what? That would even be fine if it was just one set of old rules, like you have in the Japan FSA approach, but here, you just have lots of them, so like you really don’t know.
You really don’t know how many of these rules can affect you, right? You have money services businesses in each state, and you have money transmission like the FinCEN regulations. You have CFDC. You have SEC, so it’s not just that it’s sort of like lack of clarity, but it’s the fact that there are a lot of rules. You have to do some really creative lawyering, which I personally actually really like, like I enjoy thinking about all these old laws and how exactly like our escrows affect these specific laws and how they fit in.
I find that like intellectually fascinating, like I am a professor at the end of the day, so I like love this stuff, but it’s complicated, and if you make a mistake, like the penalties are high, right? And so, it really sort of prevents you from even taking the risk, right, because, you know, like I think this is the interpretation, but like who knows what the regulator’s going to think because, you know, no one’s really done this before, and so, how much risk am I willing to take to actually deploy this technology in the US, when I don’t know exactly how the regulator might apply this, or maybe there’s like an attorney general in some state that doesn’t like the way I did this and comes after me.
So, like, it’s just so complicated, and the penalties are so high that it really is stifling our ability to launch products in the country.
Laura Shin:
Hmm. Okay, well, that’s a grim place to end this but hopefully, that’s an important statement that some regulators will hear.
Well, it has been so great having you on the show. This was just…yeah, this was an amazing episode, so thanks again for coming on Unchained.
Sharon Goldberg:
Thank you. I’m sorry to end it on a sad note, but let me try to end this on a happier note, which is that I do think that there’s just like a massive number of technological innovations happening right now in this space. You know, people kind of trying to do things that don’t really make sense in the physical world or in the traditional world but do make sense and are possible when we use blockchains, so like I’m really excited about like pushing this type of field forward, and like I just am looking for kind of places where we can get this out there and get people using it, and I do think that things will kind of fall into line, and this innovation will be accepted, but you know, we’re really like at a place where we’re experimenting, and I’m sort of looking forward to seeing all of the new things that different people in the space are working on.
Laura Shin:
Yeah, me, too. All right, great. Well, thanks for coming on the show.
Sharon Goldberg:
Thank you.
Laura Shin:
Thanks so much for joining us today. To learn more about Sharon and Arwen, check out the show notes inside your podcast player.
If you are not yet signed up for my email newsletter, go to unchainedpodcast.com right now to get my thoughts on the top crypto stories of the week, and be sure to check out our new channel on YouTube.
Unchained is produced by me, Laura Shin, with help from Fractal Recording, Anthony Yoon, Daniel Nuss and Rich Stroffolino. Thanks for listening.
