Synthetix L222 Roadmap
0x42f9
January 25th, 2022

ok I need ROADMAP PUBLISHED. like VERY SOON. I cant take this anymore. every day I am checking for roadmaps and it is staying the same. every day, check roadmap, same roadmap. I cant take this anymore, I have over invested, by a lot. it is what it is. but I need the roadmap to BE POSTED ALREADY. can devs DO SOMETHING??

2021
So yeah 2021… Enough said there I supposed. Once again we find ourselves at the start of another year with no roadmap. Apparently, there is a rumour that the CC’s have decided to publish a roadmap this year, soonthetix, honestly, I will believe it when I see it. In the interim here is your bootleg former semi-benevolent dictator version.

L222
As a starting point I think we can all agree we are finally seeing some momentum on L222 (pronounced Layer Two, Twenty Two, or el-to-to-to), it took much longer than everyone expected or hoped, but we have multiple viable scaling solutions within the Ethereum ecosystem now. Also, L221 is just an objectively much worse meme 👇

L221
L221

So maybe that whole painful scaling year was worth it after all. My favourite of these scaling solutions is still Optimism (Full disclosure I am an investor and advisor to Optimism) I think it will gather even more momentum over the next few months, but it is awesome to see a competitive market for Ethereum scaling solutions, regardless of which one you prefer, this is an amazing outcome for Ethereum.

Competition
However, the Ethereum scaling delays of 2021 were not without cost. We now have multiple nascent but viable independent communities building around Eth Killers. This is no EOS in 2018, except for the downtime and broken fee markets, that is the same. In spite of these issues, there is real activity in these ecosystems. Many of the users of these chains have NEVER used Ethereum L1. So how do we win them back? The same way we built our lead in the first place, by being the best place for the best engineers to deploy contracts. While it is true that users don’t care about decentralisation, I firmly believe the best engineers will gravitate towards the best place to build.

There is no question building on Optimism is going to be the ideal choice for most engineers this year. But it is going to take some time to reel this situation back in. We need to be mindful of the momentum we have lost and we need to be welcoming of users from other ecosystems. They can keep their VC’s though.

The momentum these networks have built now means ignoring them and hoping they go away is no longer viable, if it ever was. One of the key things we need to do is make it easy to move between networks. If we try to wall ourselves off from them we will never regain some of these users. So we have to trust that Optimistic Rollups will win out over the medium term against Alt L1’s and sidechains, and build viable bridges to syphon these users back. Over the longer term, it is extremely likely zero-knowledge scaling solutions will be dominant but it is still unclear what the timeframe for this looks like. To this end, I recently proposed a cross-chain bridge solution using Synthetix. Tweet thread here, essentially what I proposed is leveraging the cool property of Synths to do what we have already done with Curve cross-asset swaps and build fast and secure bridges between Optimism and the various other networks that have traction today. Starting with those firmly in the Ethereum ecosystem but expanding out to the other EVM chains as quickly as possible. If we are high conviction that the Ethereum ecosystem is the best place to build then making it easy to get from any network to the ETH L2’s is critical for raising awareness.

So with that preamble out of the way, let’s dig into what I think is planned for Synthetix this year!

V1 → V2 → V2x → V3
In the middle of 2021, we split the V2x and V3 teams. For those unfamiliar with the Synthetix version naming conventions:

  • V1 was the Havven stablecoin system (03/2018 - 12/2018)
  • V2 marked the transition to Synthetix (12/18-06/2021)
  • V2x (07/2021 - 03/2022) the remaining scope before V3 migration
  • V3 a completely new protocol architecture (06/2022 - TBC)

This split has proven extremely beneficial as it allowed us to make significant progress on a number of components of Synthetix V3, including the V3 Governance Module “V3GM” (more on this later!), while maintaining and extending the existing network.

V2x Phases
V2x eventually split into three phases, the first being the initial Optimism migration work and fixing a bunch of tech debt. The second phase is focused on the debt synthesis which merges L1 and Optimism. We have also split Synthetic futures into its own workstream. Phase one is complete, Synthetic futures is a few weeks away from being code complete but about half the contracts have already been audited. I am loath to provide dates for futures as it has been frustrating for everyone involved to experience continual delays as issues were identified in the spec. However, I can say that the finalised code is far more robust than anything else we have delivered since the launch of the original Synthetix contracts in late 2018.

Remaining Scope
Phase two contains six SIPs listed below:

  1. SIP-185 Debt Shares
  2. SIP-148 Upgraded Liquidations
  3. SIP-165 Debt Pool Synthesis
  4. SIP-198 Atomic FX transactions
  5. SIP-TBC Synth Teleporters
  6. SIP-TBC Debt Migration

SIP-185 replaces the legacy debt register with a more modern tokenised debt system which will allow for both the debt synthesis and upgraded liquidations. It also brings the mechanism closer to that which will be deployed in the V3 contracts.

SIP-148 was originally descoped last year but given the price action over the last few months, it was brought back into scope. It is a significant improvement over the current liquidation mechanism and makes liquidations far less dangerous and ideally even beneficial to the network as it significantly rewards stakers who maintain a healthy ratio. It does this by transferring the liquidated collateral to all active stakers proportionately and having them assume the debt of the liquidated staker. This prevents liquidation cascades and ensures that the value of liquidations remains in the network rather than being lost to liquidation bots.

SIP-165 Introduces a new Chainlink oracle that tracks the system debt across an unlimited number of networks and tracks the debt shares to enable staking to occur on multiple networks. While it is currently expected that all staking will be migrated to Optimism in the near future this SIP allows for flexibility in when and how this is implemented and enables the community to decide to enable staking on other networks in the future.

SIP-198 Removes fee reclamation from some currency Synths like sEUR on L1. This paves the way for more cross-asset swaps, something that Andre has been pushing us to enable for years now.

Synth Teleporters will initially enable synths to move easily between L1 and Optimism, but as mentioned in the tweet thread earlier this is only the beginning. These teleporters pave the way for Synthetix to act as a cross-network bridge for almost any asset that is supported as a Synth. The combination of Synthetix and Chainlink cross-chain messaging is extremely powerful.

Debt migration will allow debt positions to be migrated from L1 to Optimism without burning debt which will significantly improve the process for the whales who have remained on L1.

The final aspect of V2x is the rollout of new Synths to Optimism. Chainlink recently finalised their deployment of OCR feeds to Optimism, this will allow for a rollout of a number of new feeds to Optimism in the coming months. We have been surviving with only three assets on Optimism for the last year, but the community is obviously extremely excited to expand this, especially as these assets can then immediately be implemented into other protocols like Lyra.

There is one last SIP not yet prioritised, SIP-202. I authored this SIP as I believe it is a critical change to the monetary policy of the protocol at a time when we are finally about to deliver on a number of features that have been in development for a long time. The purpose of SIP-202 is to establish a target staking ratio. The staking ratio is the percentage of SNX tokens used as collateral within the network, at the peak in 2019 this ratio was over 90%. For clarity, this is different to the network collateral ratio which is the ratio of debt vs collateral in the network. SIP-202 proposes a target ratio of 85%, if the staking ratio is below this percentage the inflation rate increases the following week. We have witnessed numerous experiments in designing more responsive systems over the last few years, and it is my view that a static emission schedule is a missed opportunity to achieve the stated outcome of inflation which is to encourage a high staking rate and reward those who participate in the network at the expense of passive holders of the token. We witnessed the impact of the introduction of inflation in march 2019, the rate of staking increased significantly and the influx in new stakers propelled us into DeFi summer a year later. I believe that tapering inflation was premature given that building the network has taken significantly longer than expected, realigning inflation now will drive similar levels of adoption and a renewed awareness of the protocol.

Synthetix V3
All of the above takes us to the end of Q1. What are we planning for the rest of the year?

The first thing is the introduction of the V3GM to govern the 2nd Epoch of the year. It is not an exaggeration to say that the V3GM is the culmination of four years of governance R&D within Synthetix as well as observations from the many governance experiments carried out across DeFi in that same period. We think that V3Gm is so important that not only are we planning to use it for all future Synthetix governance but we are planning to spin it out with its own dedicated team as a standalone protocol. It will enable the next generation of projects to launch with fully on-chain governance that is flexible and decentralised. No more multi-sigs all the way down… It puts the ultimate power of governance in the hands of token holders while allowing them to delegate key governance functions to specialist groups. It does this by granting rights to make changes to specific roles, but always having those rights revocable and vetoable by token holders. StakingRewards.sol is probably the most used contract in DeFi, V3GM says hold my beer.

The V3 rollout is going to be the first migration process for Synthetix since late 2018, typically our releases are modifications to the existing contracts. The release of a completely new architecture necessitates a migration. The plan is to make this process as seamless as possible for stakers, but there will be user intervention required. The trade-off is that migration allows us to discard all of the accumulated tech debt over the last few years and start with a clean architecture that incorporates all of the learning from implementing V2x as well as the state of the art for DeFi engineering in 2022.

V3.0 Scope
As we get closer to the release the first SIPs for V3.0 will go into much more detail about the scope and specifications for each change. The following is intended to be a high-level overview of the current thinking. Many of the changes have not undergone feasibility studies so may change as the community discusses and reviews them ahead of deployment.

Proxy Architecture
At the base layer a completely new proxy architecture has been developed. This is going to significantly improve the developer experience for both implementing new SIPs as well as integrating with Synthetix contracts for external projects.

Tokenised Debt Pool
Extending the debt shares contract in V2x to support a fully tokenised debt pool will make staking much easier and allow for a far better user experience. One of the biggest points of confusion within Synthetix is the automated staking of SNX in your wallet. This new mechanism will require sending SNX to a staking contract rather than staking in your wallet a UX pattern that is now very familiar to most users.

Differentiated Debt Pool
The current debt pool is completely undifferentiated, this makes hedging significantly harder for stakers. The new debt pool design will allow stakers to opt into whichever pools they want and they will then only receive fees from these pools. It will also mean that inflation will need to be directed to specific pools based on demand. Which leads to the next SIP.

Inflation Weighting
While the fees generated by each pool will remain in that pool, it is less clear how to determine where the inflation should be directed. We have seen over the last few years that the veCRV style gauge weights while not perfect create a strong incentive to stake and therefore control the distribution of inflation.

Inflation locking
We have also observed the power of vote locking schemes like that used by curve. Again this SIP proposes a slightly different model to that used by Curve where stakers can elect to lock their inflationary rewards for different durations and get a multiplier bonus. The exact parameters are still to be determined during feasibility but it is possible that some stakers may be willing to lock their inflationary rewards for even longer than the current locking mechanism of one year. While other stakers may prefer to get lower inflation but have it instantly unlocked.

LP token staking
This is a fairly speculative SIP, but it would introduce the ability to stake LP tokens from an SNX:ETH AMM pool in order to capture both inflation as well as liquidity incentives and fees from the AMM pool. While this introduces the potential for IL it is likely that a meaningful number of stakers will choose to participate increase the on-chain liquidity of SNX significantly.

Synth Decay Function
This is a potentially controversial SIP that is designed to encourage idle synths to be moved into the market by stakers. While it is possible that this could create some integration issues, there are now many rebasing tokens that are accommodated by AMMs and remain composable. The impact of this on stakers is to create a constant debt reduction mechanism increasing the EV of staking over the long run significantly. The exact decay rate is to be debated, or even whether to implement one for all Synths or on a per synth basis. But this change is potentially a significant win for stakers and for Synth liquidity.

Continuous Staking
The introduction of continuous staking will remove a number of issues in the protocol, the primary one being the perverse incentive to not fix a low c-ratio before a snapshot. This mechanism reduces the fees earned by stakers continuously the further they are away from the target c-ratio in either direction.

There are a few other minor changes expected for V3.0 but they will be introduced as we get closer to launch. We do, however, have some major features planned for V3.1.

Single Asset Debt Pools
One of the major criticisms of Synthetix is that it can only support highly liquid assets with deep off-chain liquidity. This is a fair criticism of the current mechanism, but the introduction of Single Asset Debt Pools not only addresses this criticism but actually takes the best parts of both Synthetix, UMA and Mirror and combines them to enable permissionless asset creation that is cauterised to a single pool but that inherits some of the liquidity of sUSD. This combined with Inflation weighting would allow for more exotic assets to be supported permissionlessly provided a Chainlink Oracle existed, and in future iterations, it might even be possible to have assets that have no oracle at all but are completely derived from on-chain data.

Exotic Collateral
Speaking of exotic assets, Spreek recently proposed a new approach to pricing architecture for exotic assets that would split the price oracles required for maintaining these collateral positions into three components. While Synthetix currently supports ETH, renBTC and LUSD as alternative collateral types, there is no reason sUSD could not be borrowed against almost any asset provided the liquidation and pricing process was robust.

Bilateral Trading
One of the challenges of being dependent on Oracles is that when tradfi markets are closed it is not possible to trade Synths that track these assets. Introducing bilateral trading would allow for the protocol to switch from Oracle pricing to bilateral trading when tradfi markets were closed. Meaning that trading in Synthetic Equity Futures could happen 24/7.

Dynamic Wrappers
An upgrade to the existing wrappers that factors in the directional skew of the debt pool to incentivise arbitrage that makes the debt pool more neutral.

Cross Asset Swaps 
Cross asset swaps have existed in a number of different forms for around year, but recently they were significantly improved with the deployment of SIP-120, though inly in the last week has the cap on transaction volume per block been lifted. In that time volume on L1 has increased significantly for the first time in several months. SIP-198 looks to extend this further by removing fee reclamation on L1 for all FX transactions, this will allow for routes between Euro and ETH/BTC in a much more efficient manner. As mentioned earlier the introduction of Synth teleporters combined with cross-asset swaps will enable volume to be generated across a number of networks with the Ethereum ecosystem and potentially beyond.

There are a few other wild ideas being discussed within the community but I will leave it there for now as many of these ideas may not make it past the initial ideation phase.

End State
So let’s wrap this up by skipping ahead and imagining what Synthetix will look like on 31/12/22.

  • The biggest change will likely be that the Synths will be exchangeable on many different networks. And if all goes well Synth Teleporters could be the primary means for moving assets between these networks, especially rollups with challenge periods.
  • We will have transitioned to V3 and significantly improved the underlying architecture of the protocol and reduced some of the glaring issues in the current implementation.
  • Synthetic futures will be live and dozens of assets will be supported.
  • The APYs from both inflation and fees should be significantly higher than they are today with the majority of SNX staked as collateral, allowing for an expansion of the Synth supply and reducing our dependence on wrapped assets.
  • All aspects of the protocol will be composable allowing for even more integrations like Lyra and Thales.
  • We will have a purely on-chain delegated democracy supported by the V3GM and a growing ecosystem of projects running on that governance framework.
  • We will have permissionless asset creation and far easier staking for new users who only want to hedge certain assets.
  • We will almost definitely discover a number of new use-cases and features to implement as the deployment of V3.0 and V3.1 progresses.
  • The #L222 narrative will dominate as the honeymoon period of the latest batch of ETH killers wanes and the reality of just how difficult it is to scale a smart contract platform correctly becomes increasingly obvious. Most projects will migrate away from L1 and new projects will not even bother to deploy there at all. This will have a compounding effect where progressively more volume, liquidity and transactions occur on L2’s and the investment the Ethereum community has made in scaling is validated.

We can’t control the macro market, but as we demonstrated in 2018 and 2019, the Synthetix community is extremely resilient, and that is even more true today than it was then. Regardless of which direction the market moves this year I am confident Synthetix will deliver as it has every year in spite of the many challenges that must be overcome to scale a complex smart contract suite. 2021 was frustrating for many reasons, but this year we will reap the rewards from the investment we made in optimism and continuously decentralising governance. wagmi.

Arweave TX
JFJMSZzfuGBaK76ehYWtDwmeTxiEkdhhZv_B09JRyKA
Ethereum Address
0x42f9134E9d3Bf7eEE1f8A5Ac2a4328B059E7468c
Content Digest
uM1hi6YIcjOfYAY4FwWqvW9u2uDywv3sv38cLX73jEE