Decentralised synthetic assets

State of Synthetix H2 2020

State of Synthetix H2 2020

This post is partnered with the Roadmap Update for H2.

One of the things I look forward to most is writing the state of Synthetix blog post every six months or so. It forces me to reflect on everything we have deployed recently, which is not something we typically spend much time on internally. We might spend a few minutes discussing a recent milestone to see how we could have optimised the delivery of it post facto, but typically any achievement gets washed away by the wave of new things we must deliver almost immediately. The other reason I enjoy this post so much is that it allows me to pull together all the disparate threads across the project and, hopefully, weave them into one coherent narrative. It is often hard for a new SNX convert to absorb all of the context and understand the implications and reasoning behind past decisions and future plans, but this post twice a year probably gets closest.

The community is pretty aware that we got off to a somewhat slow start in 2020. 2019 was extremely hectic and everyone on the team was suffering from various degrees of burnout, ranging from extreme to debilitating. We took a few weeks off in January to recover and it took until late that month for us to really regain our momentum. I am happy to say that we managed to not only do that but to accelerate. We started recruiting in earnest over the last few months and have been very lucky to find a number of high quality core contributors. So before diving into the state of the project, I’ll review the current core contributor breakdown. Ev and David are two new frontend engineers. Andrew and Alix are contributing to operations and marketing, with Alix focussed on expanding our Korean presence. Anton rejoined the project which was a huge win as he led the delivery of binary options and is a key member of the synthetic futures project. Recently Alejandro joined the project in a solidity engineer role, and we have one more solidity engineer coming on board soon. The breakdown of core contributors is now; five core engineers, four frontend engineers, four operations and marketing and then myself with no discernible skills or role. In addition to the core contributors funded by the synthetixDAO we have over fifty external contributors across every dimension of the project, from our Guardians in Discord to the grantsDAO members funding public goods. I’m happy to say Synthetix is in the best position to execute on our vision since the project was conceived in late 2016.

I’m going to review the 2020 roadmap categories for the year one by one to see what has shifted but first a quick aside about a potential white swan, i.e. a positive tail event, in this case a feature that could significantly alter the growth curve of protocol volume. The success of Curve over the last six months has been pretty astounding even to the most bullish defi advocates, but it is still somewhat hindered by the fact that each curve pool is isolated to one asset class. So e.g. you can trade $5m of USDC into sUSD in one trade on chain, or $2m of sBTC into WBTC via 1inch, but you can’t trade $10m USDC into $10m renBTC, because there isn’t a shared asset in each pool that can be used to bridge them. Except… sBTC and sUSD are essentially the same asset due to their ability to be converted with zero slippage at the current oracle price. They can, therefore potentially be used as a bridge between the two pools. This combined with an sETH:WETH pool, would provide something that is currently missing in DeFi. The ability to do very large trades between ETH and Stablecoins. Right now Curve does a lot of volume, but it is pretty obvious if such a feature were implemented this would exponentially increase trades. And every one would go through Synthetix, and critically, only Synths support this functionality and only Synthetix has enough liquidity in Synths to be able to deliver the required liquidity into the Curve pools. There is an issue of course, which is the unintended consequence of fee reclamation on composability. However, working with Micheal from Curve and Anton from 1inch, Justin and I think we have a solution that uses virtual tokens to enable composability despite the limitation of fee reclamation.

We are hoping to roll this out in the next 6-8 weeks. This is why DeFi is so exciting: the solution space is so unexplored there can easily be hugely impactful mechanism designs sitting right in front of you that can go unnoticed for a long time… Exposure to white swans like this is why DeFi experiments are so critical.

With that out of the way, on to the categories from the roadmap:

  • Growth / adoption
  • Mechanism / incentive design
  • User experience
  • Protocol security
  • Decentralisation
  • System Optimisation
  • Composability

Growth / adoption

While organic volume has been steadily increasing, there have been some headwinds lately. Gas prices on Ethereum have been a hindrance for trading over the last few months. I will address our plans for this in the UX section later in the post, but while slower than we would like we are still seeing a recovery of the volumes seen in 2019 but with a much reduced threat of frontrunning. Fee reclamation has significantly reduced, but not eliminated, frontrunning as there is a trade-off between UX and the fee reclamation window. We have opted to keep the window small to reduce the UX impact, and we have several other changes planned including next price fee reclamation to reduce this further. In addition to growth in Synth trading the launch of Binary Options was particularly successful, significant improvements to the UX have been made recently and we expect that volume will continue to grow. We still need a dedicated venue to trade options once the bidding phase is completed, but hopefully a project will be launched soon to tackle this challenge. We are seeing strong demand for indices like sDEFI which is exciting as they represent novel assets with strong differentiation. We have parked leveraged synths in favour of focussing on synthetic futures, the spec of which is almost complete and some early implementations are already underway. We expect to launch a testnet trading competition in the near future.

Mechanism / incentive design

We have widened our incentives, launching new pools in Curve and Balancer as well as Uniswap V2. This has created deeper liquidity across the entire Synth ecosystem and also stabilized the peg consistently for months with only a few minor deviations. In addition the launch of the sBTC and sUSD curve pools have created incredible liquidity for Synths which has had a flow on effect where more platforms are becoming comfortable integrating Synths. Our upcoming partner volume program will double down on this with rebates paid to integrators from the sDAO. This coupled with direct trading incentives which will launch after synthetic futures should continue to drive trading volume. We are about to relaunch the Ether collateral trial, which was interrupted due to a bug in the liquidation mechanism that caused us to close the trial early. This second trial does not incorporate sUSD borrowing yet which will need to wait until later in the year, during Q4 most likely.

User experience

We launched an extremely successful L2 scaling trial of the OVM several months ago, and are now planning for a Staged rollout onto a limited mainnet. This will allow for a siloed debt pool to be created on L2 and reduce the cost of staking for smaller SNX holders. This coupled with the launch of xSNX, a grantsDAO funded project, will provide more options for SNX holders impacted by high gas prices.

Limit orders will be launching later this week, which is a huge UX improvement for users. Once this has been implemented more advanced order types will become possible.

By far the biggest UX improvement though will come with the launch of synthetic futures, which will add the ability to trade with leverage for the first time. This mechanism will have some significant advantages over existing centralised futures platforms, and should contribute to the further growth in Synth exchange volume.

Protocol security

We announced a number of improvements to our opsec and monitoring suites during the Opsec and Dev sprint in March as well as over the last few releases. This combined with regular wargames has ensured that the protocolDAO member readiness is extremely high. Response times and resolution are down to less than ten minutes across multiple trials. In collaboration with Chainlink we have extended this further and now have much closer communication across both teams during regular maintenance and downtime.

Decentralisation

We have accelerated our transition to decentralised governance and the recently announced decommissioning of the foundation has further cemented the shift to a DAO governance framework. This coupled with the launch of the three governance DAO’s, the protocolDAO, grantsDAO and synthetixDAO, ensures that the protocol is ready for the next phase of governance directly by token holders.

In addition we have made a number of changes to the dApps to remove areas of centralisation, deploying them to IPFS and switching off all tracking services. We will continue to push in this direction, including funding alternate interfaces to ensure users always have multiple options to access the contracts. This includes the deployment of a number of subgraphs which remove the need for centralised API’s.

This process will be extended further with the launch of the delegated migrator and token holder veto power in one of the upcoming releases.

The final aspect of this is completing the transition to Chainlink Oracles, which is scheduled for our next release in Pollux. This process has taken longer than we initially planned but we are really excited to finalise the transition.

System Optimisation

We have made some progress with system optimisation, however, this has not been a core area of focus due to the urgency of other areas of focus. We have made some gas optimisations in places and deployed the contract address resolver.

Composability

We have migrated to new proxies which are erc-20 compatible improving the integration experience for other projects. We have also made significant effort to implement improved documentation, and are continuing to extend this.

Looking Ahead

So now we need to tie all of the above functionality and upcoming work into a coherent plan for the network. This exercise is slightly different for a protocol versus, say, a marketplace or a SaS business. Because while traditional startups try to be convex to market volatility, the reality is this is very hard. Most startups are fairly aggressive directional bets on a specific set of assumptions. Pivots are possible but they are somewhat constrained. The convexity of startups comes from zooming out and looking at the aggregate of all startups to enable a more societal level convexity.

With a protocol, these trade-offs are less stark. The reason is fairly simple: a protocol is an enabler of new systems, not the end product itself. Crucially, when we look at something like the iPhone we see that it did not truly go parabolic until it shifted from a product itself to a semi-open platform. It is simply impossible for closed-directional bets to compete with open-ended convex bets like open DeFi protocols. This thinking has directed much of our effort over the last six months, and it will continue to do so for the rest of the year. We of course want to build dApps that showcase the power of the platform, but we have reached an inflection point in terms of liquidity and functionality where exposing the protocol to the collective imagination and creativity of the Ethereum community is by far the optimal path. This is how you get solutions like dHedge, xSNX, snx.link, snx.tools and many others. By creating a permissionless platform for any protocol or dApp on Ethereum to integrate you get the same benefits as you do from transitioning from centralised governance to governance by an open community. This is something we have put significant effort into.

So we have two parallel streams: one which is focussed on building out the best application layer to showcase the power of the protocol, and another which is focussed on accelerating adoption at the integration level. Let’s focus on the former first.

Application Layer

The Synthetix brand is almost two years old now, and was originally conceived and delivered in about a week or so. It is long overdue for a refresh, and we are going to take this opportunity to refresh the dApps. In addition we will separate the exchange brand from the parent brand, this is a conscious decision to remove the ambiguity and overlap between staking and trading in the protocol. While they are two sides of the same coin, there are differing and incompatible incentives between stakers and traders. We will thus split out the exchange and rebrand it. The reasoning behind this move is to remove Synthetix.Exchange from a privileged position in the ecosystem. It will eventually become one among many interfaces to access Synths, from mobile wallets to dedicated interfaces for specific functions like binary options. I am confident that with the new look and feel, Synthetix OG’s will come to love the new exchange as they get used to the improved UX it brings.

In addition to this we will be refreshing all of the existing sub brands within Synthetix. This includes:

  1. Discord -> community.synthetix.io
  2. dashboard -> stats.synthetix.io
  3. mintr -> staking.synthetix.io
  4. sips.synthetix.io
  5. delegatr -> delegate.synthetix.io
  6. docs.synthetix.io
  7. research.synthetix.io
  8. NEW -> status.synthetix.io
  9. synthetix.io

This will include a new more unified feel across all of the protocol interfaces. It will also include better UX for staking and with the upcoming transition to the OVM a far lower cost to access the network.

We have scaled up the frontend team to take on this major refresh of all of our existing properties, but we plan to add some additional core contributors to ensure we can execute all of this over the coming months.

Integration Layer

The protocol has now reached a level of maturity where it is able to support both direct integrations as well as implementation overlays. While we made significant progress through 2019, in 2020 we have accelerated liquidity for Synths’ on- and off-ramps into the rest of the Ethereum ecosystem via our integration and liquidity incentives with Curve. This has resulted in tens of millions worth of liquidity in Curve pools and the ability to support on-chain trades larger than $1m with minimal slippage into and out of the protocol. It has also ensured the peg has been far more stable over the last six months, and while not perfect with some minor deviations over the last few weeks, longer term volatility is at an all time low. The combination of liquidity and stability have enabled us to plan the launch of our first volume incentives for native integrations. This incentive will launch in the next few weeks and will have a number of well known DeFi launch partners who have already natively integrated Synth exchanges into their dApps, mobile wallets and protocols. This volume incentive program enables applications to generate revenue from the Synth volume they capture, vastly expanding the reach of Synths and the Synthetix protocol.

This program also enables us to begin looking for grant recipients to build dedicated interfaces for Synth functionality. This new grants program will be handled by the synthetixDAO and will help fund seed stage projects to launch new interfaces and protocols built on Synthetix, that have an inherent value capture model through the volume incentive program via fee rebates. We are looking for several teams to take on various interfaces, specifically a dedicated futures interface for the upcoming launch of Synthetic Futures, as well as a dedicated interface to enable trading of binary options through the entire life cycle of options phases including trading the ERC-20 options generated by the protocol. We are open to other proposals and will announce the details of how you can apply to this accelerator program shortly. The first cohort will likely consist of three teams, based on the initial success of the program it will expand from there.

Conclusion

While we have undoubtedly achieved a lot as a community in the last six months, we still have much to do and there remains significant uncertainty. However, the problems we are now addressing are all fairly well understood, and it is a question of execution rather than an existential one. We must resolve the scaling and gas cost issues, we must finalise the functional upgrades to the protocol and we must deliver exceptional application layer UX. Provided we can achieve this we can expect the growth rate of adoption and volume to continue to accelerate. The final component is to smoothly transition to a fully decentralised governance framework while maintaining focus on all the remaining challenges. This is of course a non-trivial exercise, but if it were trivial it wouldn’t be fun.


comments powered by Disqus