As each year draws to a close, Loopring releases the R&D roadmap for the coming year. This supports transparency and allows the community to obtain a greater understanding of the plan. Equally important, it forces us to think hard about the future and to reach consensus on the plan so that subsequent actions will be highly consistent and effective.
2020 Development Roadmap at a glance:
- Hebao Wallet App
- Hebao DEX
- Learn from v3 in Production
- Improve Composability
- Make Loopring Customizable & Upgradeable
- R&D into other ZKP systems
Relayer & Backend
- Lightcone 3.0
- DEX Browser
- Hebao Backend
- Volume Goals
- LRC Staking
Plan all you want, but the rapid development of blockchain technology — especially Ethereum — and its evolving ecosystem overpowers even the best and most concrete of roadmaps.
Take our 2019 roadmap for example: we ended up investing heavily in the R&D of a small part of the original plan, zero-knowledge proofs (ZKP). This small part has become the strongest competitive advantage of Loopring, and will likely continue to be for years to come.
While scalability was indeed the priority for v3 for the past year, and our ZKP work was already underway at the end of 2018, we didn’t know how our specific solution would take shape. From the 2019 Roadmap:
Our main goal with 3.0 is to reduce settlement cost by an order of magnitude, while also increasing the number of order rings per settlement by an order of magnitude. Significantly increasing the scalability of the protocol is front and center with 3.0.
We are incredibly proud that with protocol v3, we beat our goal by more than an extra order of magnitude, or 700x total, and without sacrificing Ethereum-level security guarantees. We believe our recently deployed v3 is the best high-performance orderbook DEX infrastructure available today.
This is due to the team’s continuous learning and improvement, and the ability to make flexible adjustments to the plan, so long as it maintains our principles.
Principles for Planning
In blockchain land, there is no lack of tempting ideas to make the world a better place. But if there are no principles for planning, decisions risk becoming unconstrained, and there will be endless and aimless paths. Jack Ma’s dream, for instance, is to “make it easy to do business”, and Alibaba will not engage in unrelated pursuits if it doesn’t support this guiding goal.
Loopring’s goal has been and will remain building infrastructure to power the secure, non-custodial exchange of cryptoassets. We will not deviate from our plan to make DEXes the obvious option for trading on Ethereum.
With this goal in mind, we relied on the following guiding principles when making the 2020 plan.
1. Focus on doing one thing well
We believe that the various protocols and dApps on Ethereum will continue to become increasingly composable, gradually connecting to each other and forming an economic organism. We don’t believe that any team is capable of solving all the pain points or potential opportunities in the Ethereum ecosystem. Therefore, in the process of productizing the Loopring protocol, when some deficiencies make our efforts less perfect (such as cross-chain (in)capabilities), we need to be cautious in our judgment and wait for other teams to implement the best solution instead of rushing. We recognize the fact that we have limited capabilities, resources and time. Doing one thing and doing it right is imperative.
2. Do not reinvent the wheel
Highly related to the above. For example: Uniswap has achieved great success this year, which will entice some teams to do similar things, or even simply copy-paste. But an ‘upgraded’ version of Uniswap will not make a substantial contribution to the Ethereum ecosystem as much as Uniswap’s own progression may. Indeed, often it is degraded versions. We believe that doing such technology R&D is to reinvent the wheel. Loopring has always been at the forefront of our respective track, and will never be a laggard or copycat in terms of R&D. That is not to say that we don’t learn from and collaborate with our peers, which we most certainly do, and value. This point is more nuanced because, of course, following another’s track can be a source of inspiration to improve upon. We just believe there are some scenarios — wheel scenarios — where the ‘thing’ itself is simply the best solution for its intended purpose.
3. Believe in the value of users
We are slowly and thoroughly realizing the value and importance of users, especially for a DEX protocol. Technology alone won’t reflect our accomplishments; we need users. The value of the Loopring protocol, and that of LRC, depends on the core engineering team at the beginning; but in the middle and later stages, it will depend on our userbase, and those of our DEX operators. We have just reached that middle stage. Therefore, it’s time for us to plan for products, user acquisition, and the commercialization of our leading technology.
Steve Guo, Loopring’s CTO, will lead the design and development of our products and liquidity provision tools.
1. Hebao App
We are developing an Ethereum smart-wallet mobile app for non-technical users in the mainland of China to solve the user onboarding challenge. Most people in China don’t have access to Chrome and MetaMask, so this is a critical piece. We call it “Hebao”, which means wallet or purse in ancient Chinese (荷包).
The overall contract design has actually already been completed and is being reviewed. Hebao’s smart contracts are open-source. Smart contract testing and app development will start simultaneously in the first quarter of 2020.
The Hebao app will have the following functionality:
- Support multiple guardians including hardware wallets, allow resetting of lost owner keys through guardians; support asset inheritance based on time locks/dead-man switch — when a wallet owner has not performed any on-chain behavior for a long time, the specified heir can claim the wallet as its new owner. Through these features and ENS integration, we hope that users will be completely free from managing exposed private keys and mnemonics.
- Support meta-transactions; wallets can perform various Ethereum transactions and dApp activities without any Ether.
- Provide in-depth integrations with Loopring protocol and mainstream DeFi protocols to enable users to securely trade and earn interest.
- Integrate LRC staking to enable users to benefit from Loopring protocol fees.
- Support WalletConnect standard, multi-signature wallets, address whitelist, one-click asset migration…
We have high hopes for Hebao’s ability to become a leading smart wallet in the country, and serve as an excellent entry & auxiliary to Loopring protocol.
2. Hebao DEX
The Hebao exchange will be built on top of Loopring 3.0. We will launch a beta in the first quarter of 2020. The web version is planned to be launched first and will support Loopring 3.0’s core functionalities as a high-performance orderbook DEX. It will be accessible via MetaMask and other wallet extensions.
Note: the Hebao App and DEX, as commercial, for-profit products, will be independently operated by Lightcone Technology Limited. The Loopring Foundation is one of the non-controlling shareholders of Lightcone Technology.
Brecht Devos, Loopring’s Chief Architect, will lead the development of the Loopring protocol.
1. Learn from Production
Protocol v3 is live on mainnet and will be used in production environments in late December 2019 with the WeDEX beta, a strong partner in Shenzhen. Having real exchanges on top of v3 running on Ethereum mainnet gives us extremely valuable data. We’ll discover if there are any bottlenecks that can be avoided, and learn if there are any useful features that are missing. We’ll learn how the protocol behaves on Ethereum mainnet in realistic scenarios. All this data will ensure that the protocol (and the relayer) will be even better for future releases.
2. Improve Composability
Protocol v3 has brought a massive increase in scalability and a massive decrease in costs for non-custodial trading on Ethereum, to the tune of 700x improvements from v2. As explained in multiple articles, this is achieved by us using the zkRollup design, a layer 2 solution. However, layer 2 solutions necessarily make it more difficult to interact directly on-chain. The more operations you do off-chain, let alone in a zkSNARK, the harder it is to interact with other smart contracts on-chain. From a UX point of view, this is great (apparent immediate finality), but this can have less desirable consequences in that it limits the composability of Loopring exchanges with other smart contracts on Ethereum. Our aim is to improve this significantly by making it possible to do additional on-chain logic for user requests when needed.
Here are a few concrete examples of what this will make possible:
- Allow exchanges to implement custom on-chain functionality for users (like tapping into on-chain liquidity pools, lending pools)
- Allow exchanges to do faster withdrawals.
- Allow exchanges to build exchange wallets that act similarly to a standard Ethereum wallet. Anything you can do on a normal Ethereum wallet you will be able to do on your exchange wallet.
- Support general meta transactions so users need not interact with Ethereum themselves if they don’t want to.
3. Make Loopring Customizable and Upgradeable
All exchange functionality in v3 is currently bundled in a single contract. This contract can be upgraded, but this potentially modifies the complete exchange. Even if an exchange would only like to add a single feature or do a small modification, the complete exchange would be different. This limits code reusability and makes it difficult to maintain a single codebase for all exchanges built on Loopring.
Thus, we want to make the exchange contracts more modular. Exchanges will be able to choose which functionality they want to support and simply add those modules to the exchange. If an exchange wants to create some custom functionality, they can easily do so by creating a much simpler smart contract and adding it to the exchange, while keeping all the other exchange logic untouched. This not only simplifies the protocol design, but it also means more code can be reused which has numerous benefits.
4. Research and Development of Zero-Knowledge Proof Systems
In the last couple of months, there has been an immense amount of new developments in zero-knowledge proofs. ZKPs are of course the core of protocol v3. In protocol v3 we use the Groth16 proof system on the BN254 curve. This system provides the smallest proofs (just 256 bytes!), has cheap proof-verification on-chain, and has quite fast proving times. The proving times largely dictate the off-chain costs of the protocol so they are very important. The biggest drawback of Groth16 is the necessary circuit-specific trusted setup (for which we held a ceremony in November). Each time we modify the logic inside a circuit, no matter how small the change, we need to do a trusted setup for the circuit. This limits us in how fast we can take new exchange functionality into production. For each trusted setup we also need help from participants in the Ethereum community. So a trusted setup not only takes a significant amount of time to organize, but also requires external people willing to spend some of their time to help us.
In contrast to Groth16, new proving systems like PLONK or STARK do not need a trusted setup for each circuit independently. However, they also have different proof sizes, proving times and proof verification costs on-chain. The way the circuits are programmed also often differs from Groth16, which means that operations efficient in Groth16 are not necessarily efficient in a different proving system and vice versa. Strong progress is also being made with recursive ZKPs in proving systems like HALO which is certainly a very interesting and promising direction.
Picking a new proving system (which is production-ready) is no trivial task. New zero-knowledge proofs developments will no doubt keep coming in 2020 and we’ll continue to follow them closely. We hope to switch to a new proof system without circuit-specific trusted setup, faster prover times and increased scalability.
5. Smart Wallet Integration
We’ll develop, open-source, and deploy contracts for our smart wallet. Our priority is the integration with Loopring 3.0 so the app will make DEX user on-boarding much easier, especially for those in China lacking access to Chrome or MetaMask. Integration with other DeFi protocols will follow. The smart wallet integration will also entail making LRC more front of mind, and its staking mechanics much more user-friendly.
Relayer & Backend /2020
Hoss Ma, Loopring’s Chief Scientist, will lead the development of the Lightcone relayer.
1. Lightcone 3.0
The research and development of the Lightcone relayer will be the top priority of 2020. Recall, the relayer (aka ‘operator’) is what handles all the off-chain components, which, in protocol v3, is substantial — including the ZKP handling. It is a tremendous engineering feat to construct a reliable and efficient relayer in the context of v3. Lightcone is our internal relayer we have been building for over one year. While any Loopring DEX is free to build and operate their own relayer, they can also use Lightcone’s relayer-as-a-service.
The current version, Lightcone 2.0, will very soon provide services for WeDEX and Hebao. In 2020, we will start the development of the next major Lightcone version — 3.0. We will redesign the Lightcone API and publish related documents. We hope that third parties (dApps, wallets, aggregators, etc.) will be able to use this documentation to easily integrate with any DEXes being supported by Lightcone.
Lightcone’s services can be consumed in a modular way. For example, when Dolomite, a DEX built on Loopring v2, will upgrade to v3 at some point in 2020, they will use Lightcone for the batching & ZKP proving, but will use their own order-matching engine that they’ve already developed.
We hope to improve the automatic deployment, monitoring, and alarm systems to ensure that the ‘relayer-as-a-service’ concept of Lightcone is better implemented. Ultimately, we plan to support 2 to 4 Loopring v3 exchanges in 2020. The performance and availability of Lightcone relayer will be further improved, reaching the following technical capabilities:
- Single relayer cluster will not perform more than 3 rollbacks & the total loss of assets for operators and exchange owners due to hacking does not exceed $15,000 combined. [Note: this $ figure is for the operator’s assets/gas; the users’ assets are of course 100% secured by the protocol].
- Online recovery completed within 4 hours;
- Relayer’s availability exceeds 98%;
- Single relayer cluster TPS is not less than 200 transactions per second [The highest throughput of Loopring 3.0 is 1,400 trades per second; currently, Lightcone is the performance bottleneck].
Lightcone also plans to support more protocol features, including:
- Offline withdrawal requests. This can increase the withdrawal speed and reduce user withdrawal fees;
- All three LRC staking mechanisms to help LRC holders maximize returns;
- Stop-limit orders, other advanced order types, OTC trading
We also plan to enhance our data operations and analysis capabilities to provide better decision support for DEX operators. We will further improve the elastic computing framework for ZKP generation, and strive to reduce the generation time and cost. All the while, the relayer engineers will also be committed to providing feedback to our protocol engineers to improve the protocol’s design and implementation.
In general, the research and development of the Lightcone relayer is our major focus. The team is fully aware of the importance of a well-implemented relayer for the commercialization of the Loopring protocol, therefore we will ensure sufficient investment of resources. The fact is, building a v3-compatible relayer represents a monumental effort, and thus, we must ensure our solution is ready off-the-shelf for would-be DEX owners to employ. We imagine Loopring+Lightcone powering some of the most successful, high-performance non-custodial orderbook exchanges on Ethereum.
2. DEX Browser
We will open-source and launch a DEX Browser for Loopring 3.0. The DEX Browser will provide data query services (mainly account balances and trading history) for each exchange built on top of Loopring 3.0. It is basically a block explorer, making Loopring DEX activity knowable and searchable through our ZKRollup construction.
Super importantly, this tool will also provide users with “withdrawal services”: in case any exchange is closed (their website is completely offline and their operators are unreachable), all users’ assets can always be reclaimed through such a withdrawal portal. The ability for users to claim their assets even in the worst emergency situations is one of Loopring’s most powerful features, made possible by the protocol’s enforcement of on-chain data availability. Note: even if our DEX browser was down, funds are always accessible by interacting with Ethereum itself. Any third party can also provide a UI to make the claim process (which involves generating a Merkle inclusion proof) simple for users.
3. Hebao’s Backend
We’ll also develop a closed-source backend for the Hebao app for coordinating user activities. The backend, of course, never takes custody of users’ assets.
After more than two years of research and four versions of protocol development, we now have some audacious objectives.
Our goal is that by the end of 2020, the average daily trading volume of Loopring DEXes will be 50 times that of Ethereum’s most voluminous DEX, Uniswap. We imagine this would be driven by 2 to 4 DEXes on Loopring Protocol 3.0 and the Lightcone relayer.
While this may seem overly optimistic, the volume we hope to capture is not an apples to apples comparison with Uniswap: Loopring is for high-performance orderbook trading, Uniswap is more for simple swaps. As described in our recent post, Trading vs Swapping are different niches — with different user intent — and so a more fair analogue would need to include some centralized exchange trading volume figures. Binance, for instance, a leading CEX offering the scalable orderbook experience Loopring v3 approximates, currently does over $1 billion daily volume on average.
LRC Staking for Protocol Fees (Type 1)
Calculated at 50 times the daily trading volume of Uniswap, $1.5 million, our daily average trading volume by the end of 2020 would be $1.5m*50 = $75 million. Monthly, that equates to $75m* 30 = $2.25 billion. With the current protocol fee set to 0.05% (for both parties of a trade), the monthly fees accrued would be $2.25 million.
Of Loopring’s protocol fee:
- 70% will be rewarded to LRC stakers;
- 20% will go to the Loopring DAO (to be built);
- 10% will be burned.
Until our DAO is live, the 20% allotted for it will either go to LRC stakers or be burned. If LRC price remains unchanged ($0.025), a 30% protocol fee burn would burn 27 million LRC per month — ($2.25m * 0.3)/$0.025 = 27m. And 63m LRC , 70%, paid to stakers. Of course, if the price appreciates due to trading volume taking off, the fees for burning would translate into fewer LRC burned. On the other hand, a lower price will mean bigger burn.
Our expectation is to burn 10 to 25 million LRC by the end of 2020.
LRC Staking by DEXes for Economic Security (Type 2)
In addition to staking for the protocol fee distribution, LRC is also used by DEX owners as a bond for economic security. The locked LRC is partially or completely slashed when a DEX violates protocol rules, such as failing to submit a proof for a committed block on time, or having a reversion. Note, this economic security is only ‘in play’ until the zkSNARKs are submitted, which provides final, cryptographic proof. (See here for a list of protocol parameters / slashing costs.)
The amount of staked LRC thus reflects a DEX’s likelihood or disincentive to violating protocol rules. The more LRC locked, the more reliable a DEX should be. Of course, this reliability relates to service performance / execution guarantees — user funds are 100% Ethereum-level secured by the protocol.
Similarly to how this signal informs users, it also informs Loopring in terms of which teams / DEXes are making the most significant commitment to the ecosystem — leading us to be more inclined to support DEXes that lock substantial LRC. This support comes in the sense of potential ecosystem grants (LEAF), and willingness to support their DEX with Lightcone relayer, among other technical assistance. Needless to say, any and all projects are free and welcome to build on Loopring’s open source protocol (smart contracts and ZKP circuits), but LRC at stake can help discern between serious operators, and those less so.
It is hard to estimate at this time how much LRC will be slashed from this mechanism.
LRC Staking for Protocol Fee Reductions (Type 3)
The final type of LRC staking is for lowering the protocol fees on a specific DEX. This LRC would be staked by exchange owners, market makers and high-frequency traders, to realize reduced protocol fees on said venue. While it is exchange owners who need to pay protocol fees for all orders to settle, other stakeholders/users of a DEX would be incentivized to stake there if it meant protocol fee reduction was passed through via lower trading fees.
We envision that via staking types 2 & 3, exchanges and market makers will lock in no less than 50 million LRC by the end of 2020.
Disclaimer: Please note the above goals are just that, goals. They depend on several factors, both endogenous and exogenous to our efforts, including systemic factors such as the general adoption of DEX trading & DeFi, regulation, progress of competitors, and more. Needless to say, we do NOT guarantee any figures and the above does not constitute investment advice for purchasing LRC. We simply wish to illustrate the mechanics of staking with some examples.
Change update frequency from Bi-Weekly to Monthly
As the development of the Loopring protocol has achieved a very important milestone in v3’s deployment, and because the relayer and DEX products are closed source development, beginning January 1, 2020, Loopring Foundation will switch from bi-weekly updates to monthly updates. We believe that in the process of productization, Loopring-based DEXes will release operational information of their own, lessening Loopring’s need for the same frequency of updates.
If you’ve been paying attention to Loopring’s bi-weekly updates throughout 2019, you will find that the focus of those updates and this 2020 roadmap are highly consistent. This shows that after two years of exploration, Loopring has gained significantly more confidence in our R&D direction and achievements.
It also shows that the new decade will bring a new phase for Loopring — out of the lab and into real usage. 2020 will mark the beginning of Loopring’s productization. Initial products focus on the large, nascent China market, but balances a global, borderless view.
We’d like to end by sincerely thanking our community, and the Ethereum, DeFi and ZKP communities for tremendous support and progress. The Loopring Foundation will continue to devote itself to powering the highest-performing non-custodial orderbook exchanges on Ethereum. We truly hope and believe that our hard work and your patience will both pay off.