Ethfinex Telegram AMA with Loopring’s Daniel Wang

We appreciate Ethfinex for listing LRC on their platform and organizing the Loopring AMA on the Ethfinex Telegram on Thursday, Feb 14th, 2019. This post sums up the questions and answers.

1. An introduction would be great, perhaps we can kick things off with by telling us a bit about the Loopring genesis story; how you guys came to be and what key problems you are solving?

Daniel: Alright, good idea. First of all, English is not my native language, forgive my typos please. I used to work for Google and in 2014 I founded a startup to offer centralized exchange service to help people to buy/sell BTC/LTC/XRP etc. The business didn’t go well so I closed it and joined an insurance company and secretly built a similar product, but still focusing on digital asset trading.

The idea of Loopring came up to me while I’m with my previous employer, but I’m obsessed with it, so I quit and did an ICO. The ICO is very successfully but ended up refunding most ether raised (close to 100K) to participants due to the regulation in China. We decided to go with what left with us to continue the development though. The Loopring protocol adapts the idea of “managing, sharing, and matching orders off chain and settling trade on chain”, the on-chain part is the protocol, the off-chain part is the relayer, which is very similar to the 0x protocol.

The differences from other trading protocol is the following, and I believe some of them are our advantages:

  • Loopring protocol is order-based. I personal believe most of the values are generated by order-based trading because that’s what most people are used to.
  • Loopring features “ring matching”. Each loopring’s order is modelled as <giving-out-X-tokenS-for-Y-tokenB>, or <X tokenS->Y tokenB> for short ,regardless whether the order is a sell or a buy. With such an unidirectional order modelling, we can enable <10 X->10Y> <10Y->10Z> and <10Z->10X> order to trade in one transaction. This cannot be done using other protocols and is a built in arbitrage mechanism.
  • Loopring can prevent orders to be stolen by any middleman or Ethereum miners, which is what 0x is lack of and will be suffering.
  • But the most existing thing about Loopring 2.0 is the fact that relayers can accept any token as the fee token, not LRC. When a relayer get 1000 GTO as fee, for example, 200 GTO will be “taxed” into a fee contract, and those 200 GTO and other tokens, will be auctioned off for LRC, those purchased LRC will be burned automatically. For GTO, the “burn rate” can be 10%, but if relayers accept LRC as fee, the rate is only 5 percentage. This guarantees LRC’s total supply will decrease over time.

2. Hi Daniel, thanks for being here to answer questions today, much appreciated! I’ve been a holder of LRC but, aside from general market conditions, I feel underwhelmed by the growth of LRC. For a project that seems to tick a lot of boxes, the interest feels low. Do you have any plans to ham up your outreach efforts to try build some hype around it?

Daniel: We are mostly engineers, actually only 4 people are not engineers. We want to focus on coding and delivering the protocol and a new relayer implementation, we don’t do a lot of marketing as of now. But this may change in the future — once we deployed the new relayer backend and get liquidity, we’ll do more promotions. The plan is to deploy the relayer in about 2 months. We are close to code complete but need to do a lot of testing internally.

3. Hi Daniel! I’m interested in your liquidity strategy as your model seems to be dependent on having high volume. Are you engaging market makers?

Daniel: We’ll open our relayer backend API for people to make marketing on our relayer. But you are right, given the current Ethereum throughput, it is hard to generate comparable trading volumes on DEX. We are also working on a new protocol upgrade, v3, to migrate some on-chain verification logic off-chain to reduce gas cost and can potentially settle about 100 trades per Ethereum TX. This is helpful for liquidity generation, but it has some other drawbacks we are trying to mitigate.

4. I heard elsewhere that you are introducing a new token (LRX) resulting in two Loopring tokens. Whats the point of this? How will it work? And will it impact the price of LRC?

Daniel: There are LRC (for Ethereum) and LRN (for NEO). NO more LRx Tokens. We are working with the Dora team to build a cross chain solution, on the up-coming DEX-chain, LRN will be used as the native coin, and people who hold LRN can transfer their LRN from NEO to the DEX-chain and even convert to WLRN on Ethereum. The DEX chain will host only DEX protocols and DApp. LRN will be used as the fee token and governance token. We’ll release more details regarding the DEX chain and LRN in Q2.

5. Any limitations to the orders people can execute? Is it only limit orders or will there be functionality for market orders and possibly even margin?

Daniel: Currently we still focus on limit price orders. The limited price order guarantees once traded, the price meets the owner’s expectation. But the issues is the order may never be filled; we have a new (a pretty innovative ) trading model to be released in about a month, witch guarantees the settlement but not price, similar to Dutch Auction.

6. Have you considered communal pooled liquidity similar to Uniswap to bootstrap 1) a community 2) liquidity?

Daniel: We did. But as I said, pooled liquidity mechanism is not the silver bullet, we believe that order-based trading will still prevail. Don’t take me wrong. I like Uniswap, but I cannot image a crypto-world without a well designed order-based trading protocol. Loopring wants to be the protocol.

7. I am thinking more so adding on that feature along with ring matching. Both benefits can be had. DEX’s are plagued with liquidity issues. How is LRC fixing this? the LRC protocol is great in a world where there is at least $15m in total liquidity. How will it get to that point?

Daniel: There is no easy fix for the lack of liquidity, even for CEX. What Loopring can help DEX is sharing of orders across multiple DEXs. This will certainly be helpful. For tokens of very small total supply, pooled liquidity may help, but don’t forget that a certainly amount of those tokens will be kept in a smart contract in order to make the pooled liquidity works, which is bad for the token as its supply is small already.

Order-based protocols are best for tokens of larger supplies or market cap and a lot of holders. For smaller tokens, our new model, mentioned above, called “Oedax” will help.

8. How about a Totle like approach. Creating a contract that aggregates Kyber, 0x, Uniswap orders INTO Loopring. Have LRC be the dex aggregator?

Daniel: We don’t want to be an aggregator, we want to be one of those fundamentally important protocols.

9. It makes sense to create a smart contract that pools ETH/ERC20’s. The liquidity pool can then be used to market make, the benefit for liquidity providers will be 20% of fees from trades using the liquidity pool. I don’t see how this isn’t a win win?

Daniel: Those numbers above is for illustration purpose only, the actual burn rates are different. We also support WETH as fee as you can imagine. It depends on how the pool is implemented. Ideally it should not be just one pool, it should be many of them and anyone can create a new one or join an existing one.

10. Loopring is a protocol that works well at scale, we both agree there. Where we disagree is that liquidity will just come into DEX’s. You must understand that this can take 5+ years. What is the strategy to get more liquidity — if there is one?

Daniel: Or if it is the only one, decentralized governance should be applied. I don’t think I disagree. Getting liquidity to DEX is hard and it’s not going to happen over night.

11. Will you stay on the Ethereum network or do you have plans to migrate to your own mainnet?

Daniel: We focus on the protocol layer, which means we can move to other networks. But for the near term, we’ll stick to Ethereum as I see it as the best ecosystem for dApps. The DEX chain we mentioned is NOT a Loopring mainnet, we co-develop it with another team, we focus on protocol, they focus on cross-chain. There will never be a Loopring mainnet.

BTW, I don’t think any existing DEX protocols are good enough for large scale adoption. Counting on the underlying blockchain to scale up is a bad idea. Here at Loopring, we are trying to scale the protocol using other means. We plan to go to Paris to attend a couple of Ethereum Dev Conference to show some of our work in March and April. So stay tuned.

— Thanks for the questions everyone and to Ethfinex!


To stay up-to-date with Loopring, please sign up for Loopring’s Bi-Weekly Update, and find us here:

⭑ Twitter: twitter.com/loopringorg
⭑ Reddit: reddit.com/r/loopringorg
⭑ Telegram: t.me/loopring_en & t.me/loopringfans (Chinese)
⭑ Discord: discord.gg/KkYccYp
⭑ StackOverflow: stackoverflow.com/c/loopring
⭑ Kakao: open.kakao.com/o/gJbSZdF (Korean)