Elastum’s Deep Dive Interview with Loopring
This interview was originally published on Elastum’s blog on Feb 27th; https://medium.com/@elastum/elastum-insiders-so-what-is-loopring-66f28a021922. It provides a great summary of our past, and a look at where we’re headed.
Speaking about ring miners — have you ever wondered what is behind Loopring, how the name was created and most importantly how this amazing company has started? This week we are happy to share an interview with Matthew Finestone from Loopring!
1. How would you describe the main vision of the project?
The Loopring project is a framework that seeks to underpin the trustless, non-custodial exchange of value in a tokenized future. The project most basically consists of two parts:
1. Loopring protocol — an open source set of rules and a common ‘language or logic’ that handles trade settlement on-chain (for example, of ERC20 tokens on the Ethereum blockchain),
2. Loopring relayer — an off-chain system of servers concerned with communicating, sharing, and matching orders that ultimately pass through to the protocol for settlement.
So, essentially, you have a set of smart contracts for the swap/settlement of assets, and you have a relay system for messaging/matching these orders off-chain.
Combined, these are the components required for anyone to build a non-custodial exchange, or what is often referred to as a decentralized exchange (DEX). So Loopring falls into the bucket of a DEX protocol.
Loopring simply drastically lowers the barriers — technical and economic — for a team or project to incorporate non-custodial trading functionality into their DEX/app/product. The simplest case would be for someone to put up a nice GUI, and work on the branding, UX, customer acquisition, legal, etc., and run their product. In this case, they would be using the protocol, and also our relay implementation as Matching-as-a-Service, which we think is really cool. Or, they can just use the protocol, and build their own relay — which is a higher degree of difficulty — in addition to the UI.
The Loopring Foundation is a blockchain research organization that perpetuates this project and vision. In reality, the foundation is working on a few different major releases and versions, some of which may exist in parallel and serve different use cases. So you can say that the overarching vision is to build decentralized trading protocols. We really believe that if crypto is here to stay — which we believe first and foremost — then the different assets will be traded using the technology itself, and not in companies’ centralized databases as is today (or at least not solely in that manner).
2. How was the Loopring name created?
The name was derived from our order matching format, called ‘ring matching’. In Loopring, multiple orders can be combinatorially matched — settling in a circular manner in what we call order rings. So, trading doesn’t necessarily occur by matching two sides of one trading pair — ETH/DAI for example. Instead, it can match across ETH/DAI, DAI/LRC, LRC/OMG, OMG/KNC, KNC/ETH and so on…with each leg of the trade going to the next trader to find matches where there may not have been on a traditional matching scheme.
3. About the team and supporters, what do you believe are your strengths?
Loopring consists of a very strong technical team. Of the 18 current team members, only 4 are non-engineers. So I’d say our strengths lie on the development and technical side. Our founder/CEO, Daniel Wang is a very experienced engineer across a range of technologies and large multinational tech companies. You could say the team is kind of modeled in his image — very hard working, heads down, build first and talk later type of attitude.
In terms of supporters, I’d say our strengths lie in our Chinese communities. Given that’s where most of the team is based, we have more mindshare and visibility over there, and it is reflected in the community. We have a tremendous community of supporters in the West as well, just smaller.
4. Share how core team members entered the crypto space, maybe overcome some interesting issues.
I personally came from a finance background. I was working at an investment bank that specialized in fixed income (bonds), and left that back in mid 2017. At first I was just on my own, trading, speculating, learning as much as I can. I started writing for a large crypto exchange, again mostly just to learn. I came across Loopring early on, and was just a supporter who got very interested in the project as I understood the importance of the trading protocol given my involvement with traditional securities and market structure (especially something like bonds, which is still rather archaic, trading over the phone and on Bloomberg chats!). I started communicating with the founding team, helping out, and eventually just came on board. Was super organic.
As far as someone like Daniel, he has a very fitting and interesting background. He had an engineering career spanning companies like Google and JD.com. He had been involved in crypto for a while, and in 2014 he founded a centralized exchange startup, allowing people to buy and sell BTC/LTC/XRP etc. After some time, he closed it and joined ZhongAn Insurance, but in his spare time kept focusing and building on digital asset trading. That was the beginning of Loopring, and he became obsessed with it, finally quitting and starting Loopring in earnest in early/mid 2017.
5. You are a decentralized project — how is the team distributed, how do you manage all the tasks, work? What challenges you are facing and how you are solving it?
The team is distributed, but definitely has the home base in Shanghai, as well as some in Shenzhen. So most of the team is in China. I am located in Montreal for the most part, and we have an engineer in Belgium as well as San Jose. Daniel splits his time between Shanghai and New York. I think it has challenges, but also some advantages. Lots of different timezones and perhaps some weird hours — definitely some late nights. But we can cover more space and always have someone awake!
As far as managing workload and tasks, the team is broadly split into protocol development, relay development, and then the business development/marketing/community building, etc.
One of our challenges on the team front is certainly that we can’t necessarily build as fast as we’d like. If it were perfectly feasible, I’m sure we’d increase the team size. That’s both to speed up current development, but also to take on spinoff ideas that we can’t pursue at the moment.
6. What’s the added value of Loopring in comparison with other competitors?
As mentioned above, Loopring is a hybrid on-chain / off-chain decentralized exchange protocol. We use the Ethereum blockchain to settle trades, but order management happens off-chain. This hybrid architecture is similar to the 0x protocol.
The differences — and in my opinion, advantages — from other hybrid trading protocols are the following:
1. Ring matching: Every Loopring order is modeled as <giving-X-tokenSell-receiving-Y-tokenBuy>, or, more concisely, <X tokenS->Y tokenB>. This is akin to an exchange rate, as indeed, every order is just an exchange rate between two tokens, not necessarily a sell or a buy. We call this aunidirectional order model (UDOM), and it enables order rings: <10 X->10Y> <10Y->10Z> and <10Z->10X> can be traded in one transaction. This cannot be done using other protocols and is a built in arbitrage mechanism. Further, this is able to improve liquidity, and find matches across disparate order books.
2. Dual authoring: There is a pernicious problem with DEXs called front-running, which occurs when malicious middlemen, or, even more dangerously, Ethereum miners, insert their own orders in front of users’ to take advantage of an attractive trade, or effectively steal orders. Loopring can prevent this through a mechanism called Dual Authoring.
3. Fee model abstraction: In Loopring 2.0, DEXs (relayers) can accept anytoken as fee payment, not just LRC (Loopring’s token). When a relayer gets 5 DAI as a fee for matching a trade, for example, 1 DAI out of the 5 will be sent into a fee holding contract. That 1 DAI, (and the other tokens in the holding contract) will be auctioned off for LRC, and those purchased LRC will be burned automatically. For DAI, in this example, the “burn rate” was 20%, but if relayers accept LRC as fee, the rate may only be 2.5% percent, meaning they keep more of the fee, with less being burned. Either way, this guarantees LRC’s total supply will decrease over time with network usage, but doesn’t cause friction for users.
4. Division of duties: In the Loopring architecture, many different types of participants can work together, focusing on what each does best, and sharing in the success/fees. As opposed to operating a DEX completely by yourself, a project can, for example, focus on building the best wallet/UI, while collaborating with a third-party relayer for backend services (Matching-as-a-Service). This backend can be the Loopring relay implementation, or their friend’s relay, or 5 relays with whom they have partnerships. Users of the wallet would be able to trade directly from the wallet, and the fees would be split by the wallet + relayer. Conversely, you can build the best relayer out there, no UI, and offer your services to all sorts of exchanges or applications that may need to manage multiple tokens, earning fees in the process. The ability to be the master of your own business is there.
7. What are the use cases you are looking to cover?
Loopring has thus far focused exclusively on order-based trading systems, meaning traders specify their order intentions and exchange rates (make limit orders, and take limit orders) and all the action happens on order books. This is in contrast to systems like Kyber or Uniswap that are purely on-chain liquidity providers. I call them the token store model: a user goes, requests a swap, and is shown the level at which she can execute — take it or leave it, like at a store. That is very useful, but we believe the vast majority of value and asset exchanging will happen on order-based markets, as that’s how most traditional markets operate today, and for good reason: the largest traders (professionals, institutions) are price sensitive, and they need to specify their intentions, or not execute. So building the best order-book based trading paradigm for tokens is our main focus.
However, in the course of our research, we have thought up a few non-order book based systems, which approximate a novel type of auction market. We’ve alluded to that recently and have codenamed it “Oedax”. We haven’t begun developing that, but have designed it, and may build it if we have the resources, or put out a grant to the community to have it built.
So in that sense, our future and ultimate vision may be broader: to build global, non-custodial, high performance trading systems. Further, any blockchain application that needs to manage multiple tokens is in our realm of teams we want to support.
8. Tell more about 2.0 protocol.
We deployed Loopring protocol 2.0 to Ethereum mainnet in December 2018. It is broadly similar to 1.0, but more extensible, powerful, and efficient. For example, multiple rings (which contain multiple orders) can be settled in a single Ethereum transaction. Also, now any token can be registered on the protocol, and it is fully up to different exchanges/relays which token they’d like to list/support. Furthermore, orders can now be created by a different party than the ‘order owner’ itself, by giving permission to a broker, for example.
In addition, it affords a few new specific capabilities. Most notable perhaps is the flexible fee model alluded to above — in which fees can be paid in any token. That was implemented in 2.0.
Also 2.0 supports other token types besides ERC20, specifically ERC1400 — which is a security token standard, and ST20, which is Polymath’s specific implementation of ERC1400. With this, exchanges that plan to trade security tokens can more easily be built atop Loopring. ERC1400 allows extra logic to be coded into tokens by the issuer, such as only being tradable amongst whitelisted addresses for regulatory compliance, and checking vesting schedules and ownership limits before transfers.
9. What do you think are the key factors affecting Loopring price?
I know everyone says it, but we truly don’t analyze or fret over price. Mainly because we simply don’t have the luxury of time to do so as we try and build the best protocol possible. However, we are cognizant of LRC’s utility and its place in the ecosystem, and we believe it’s a very strong one. Specifically, we believe the fee model in 2.0 is very robust, as it does not add friction, and because overall Loopring ecosystem usage translates into a deflationary supply of LRC. Ultimately, it will be Loopring network-wide usage that is the catalyst for price, or future expectations of usage, discounted to the present. In my opinion, it will only take a few high-performing DEXs to satisfy this.
In terms of what’s affecting price from a ‘crypto-macro’ perspective, I am generally of the opinion that the market is not very rational at this stage. The distillation of strong projects and mechanics into strong price is not something that happens with any certitude yet. And the inverse is equally true: many ‘less-than-honest’ coins/projects are valued more highly than genuinely useful ones. As it goes, the market is a voting machine in the short term and a weighing machine in the long run; we’d much rather focus and let the tech speak — and be weighed — for itself. I’d say one of the strongest price related signals is our team members, especially at the founding level, having tremendous faith in the project/token, and actually using dips as an opportunity to buy back the token in the open market. My guess is you can probably count on two hands other projects that are doing that.
10. How do you rate current ‘bear market’?
It’s definitely been a big bear. 90%+ drawdowns for many altcoins is serious, and may mean the end of certain projects. That is unfortunate for some great teams and talent. At the same time, it’s needed: irrational exuberance of 2017 had to be digested. Crypto enthusiasts have just witnessed in 18 months what traditional market participants see over 18 years. The swings from greed to fear will always accompany humans in markets. Knowledge may progress and compound, but our emotions won’t.
What’s changed in my eyes is that the barrier for a token to be worth something has been raised significantly. We must keep in mind that this whole space is so young. Everyone is making this up as they go along, which is both good and bad: innovation and destruction. There will be many more up-cycles, but it will accrue to fewer, more quality projects. There are still very high positive correlations across tokens — near +1. I suspect and hope this changes as idiosyncratic factors are evaluated instead of just binary risk on/off. Overall, I’m still super-bullish on crypto, just much more discerning at the asset level. Personally, as I’m answering this in mid/late February, I feel like the market put in a bottom (but my view doesn’t mean much :).
11. Thinking about the future where is your main focus now and on further steps in a roadmap, what will bring the highest value to the community and ecosystem in general?
There is tons going on for us right now. 2019 is quite frankly a huge year.
Our top priority at the moment is ‘Relay 2.0’ — the second iteration of our own off-chain order messaging and matching engine. We refer to this internally as ‘Lightcone Relay’. As mentioned, this is the backend that does all the order handling and then submits it to the Loopring protocol smart contracts for settlement on-chain. We are building our own version because it is an extremely hard problem to solve. It is not just a typical matching engine. At a high level, normally, a centralized exchange backend just has to deal with one ‘state’ or data source, which only changes as traders interact with their API/UI. With a decentralized exchange ecosystem like Loopring, there is that similar API for communicating orders to keep track of, but also the state of the Ethereum blockchain itself! Remember, users keep custody of their own assets and their balances stay on chain, so we must constantly monitor those while not controlling it, and also keep aware of the blockchain state for many other things, including re-orgs, etc. Building a high performance relay while not sacrificing across any custody dimensions is very difficult. Daniel has been leading and building this part for about 6 months. It is nearly code complete, but will soon undergo tremendous testing for the next 6–8 weeks before we launch it.
When we do release it, it will be very significant not just because it emulates centralized exchange functionality and performance, but because we will open this up to some partners, and effectively provide a cloud-based API service for exchange backends. This could be a game-changer, as new DEXs/apps could spin up and use this backend for Matching-as-a-Service (MaaS), and have non-custodial exchange functionality out of the box.
We’ve also begun building Loopring Protocol 3.0. This is somewhat of a departure from previous releases, and uses some experimental and powerful technologies. The main benefit is huge potential throughput gains, which, of course is an important feature. Scaling is a big issue across any blockchain endeavor. Protocol 3.0 uses some awesome technologies to scale up settlement ability. Unfortunately I won’t say more about it right now, but in the next few months we will release more info. Interestingly, 3.0’s design lends itself to a very natural additional utility for LRC.
Also of note is the Loopring-specific sidechain we have been building. This sidechain will be used by different relayers to communicate, share, and match orders with each other. Because it will be application-specific, it is optimized for DEX considerations such as quicker blocks, lower fees, etc.
Of course, bootstrapping liquidity is of prime importance for us, and indeed for any DEX. We have been working on or planning this strategy internally as well as with some specialist partners for some time now, but have not deployed the resources until we are ready to put our best foot forward. Liquidity is paramount, and we are taking it seriously. I believe you need to work with some professional market makers, as well as open up the system to as many ‘hobbyist’ market makers or large asset holders as possible.
Finally, as all of this comes together, making it a lot easier for developers and third party projects to build on us is what it’s all about. We are working with some very strong teams such as Dolomite and UP Blockchain, and will be working with many more. Any team that needs to build a DEX or incorporate non-custodial token exchange in their app will have an amazing toolkit to work with if building a top Loopring. Whether that’s just using the protocol itself, or using our relay implementation for MaaS, ecosystem partners will take center stage.
You can purchase LRC directly with euros on elastum.io, an intuitive and easy to use cryptocurrency exchange platform.
Thanks to Gabriele and the Elastum team for a fun interview and continued collaboration!
To stay up-to-date with Loopring, please sign up for Loopring’s Bi-Weekly Update, and find us at:
⭑ Twitter: twitter.com/loopringorg
⭑ Reddit: reddit.com/r/loopringorg
⭑ Telegram: t.me/loopring_en & t.me/loopringfans (Chinese)
⭑ Discord: discord.gg/KkYccYp
⭑ GitHub: https://github.com/Loopring
⭑ Kakao: open.kakao.com/o/gJbSZdF (Korean)
