Osmosis Updates from the Lab Recap, After UST, May 11, 2022
Osmosis Updates from the Lab occurs every Wednesday at 1 PM EST (5 PM UTC) on the Osmosis Zone Twitter Space. Replays are available on the Osmosis YouTube channel or the podcast.
UST is dead. Many in the Osmosis community lost substantial sums of money, either in the Terra ecosystem itself or in UST and LUNA pools that experienced catastrophic impermanent loss. We likely all lost money this past week, so let’s be kind to one another.🫂
In general, as many on Twitter have mentioned, the Cosmos tech stack held up extremely well to the volatility and extreme use: Osmosis had an all-time daily volume high of $500m during the incident.
Information about the collapse is widely available, so we won’t recap it here. As always, caveat emptor: every commentary is colored by its author’s economic and/or ideological biases.
What has the Osmosis community learned, and what is the plan going forward?
We will keep building and iterating as planned, but with a watchful eye on minimizing risk. To this end, this Updates we talked about things like a volatility-aware AMM, being more strategic with our incentives (as planned), and having governance be more nimble in the face of potential black swans. And we will continue to build out the Osmosis suite with external teams through the Osmosis Grants Program and through developing tooling for teams like Bitsong, our guest today, to build out front-ends that use Osmosis. Outsourcing Osmosis ecosystem development will allow the core devs to focus on making the chain and the AMM as safe and advanced as possible.
Why, despite its risky mechanism being widely known, did UST had become the primary stablecoin on Osmosis? The reasons below seem directionally correct, even if they may not be equally true for all members of the community.
- Lack of IBC-compatible dollar stablecoin alternatives: Ironically, the Cosmos focus on safe inter-operability led the Osmosis DAO to be careful in choosing a canonical bridge, which delayed the arrival of non-Cosmos stables.
- Osmosis permissionless ethos and laissez-faire incentives: Any project can list on Osmosis and provide external incentives, and the Osmosis DAO has admitted most popular Cosmos projects to its algorithmic incentives process. This incident further proves (as we have been emphasizing for the last months) that we must be more strategic with OSMO incentives, directing liquidity where we expect it to improve the quality of the DEX suite, rather than voting in incentives for every coin.
- Cosmos native tech: Terra was the largest Cosmos chain, the biggest example of a project deciding to use the Cosmos stack to build. The Terra community did amazing work popularizing CosmWasm, building out tooling and educational materials for it, and providing the templates on which future Cosmos smart contracts will be built.
- Intertwining communities: As cousins in the Cosmos ecosystem, the communities had always overlapped to some extent, though the narrative favored Terra as its own ecosystem. The communities became closer after Terra activated IBC last fall, and the recent integration of CosmWasm on Osmosis was beginning to bring smart contract developers closer together.
- Hype: Cosmos projects are notorious for focusing on tech at the expense of promotion. Terra gave the rest of the Cosmos a taste of what popular adoption might look like. Further, while Terra’s leadership and backing was less decentralized than other crypto-native stables and the mechanism proved unsound, its goal of a worldwide, uncensorable currency had broad crypto appeal.
- Time and custom dulled the sense of risk & alternatives were nearly ready: The continued smooth operation and expansion of UST through its peg test in Spring ’21 and beyond, the building up of its collateral, the deep pockets of its backers, and its general acceptance or heavily qualified criticism among wide swathes of crypto thought leaders (with the notable exception of the Maker team) contributed to the impression that Terra would continue to defy the odds, at least for as long as we needed to bridge and incentivize collateralized stablecoins. After all, we were only a week away from completing the process when UST collapsed.
Emergency Proposals
Proposals to remove 50% of incentives from the biggest UST and LUNA pools have now passed and been enacted under the emergency hard-fork rules. OSMO/UST →OSMO/USDC, OSMO/LUNA →OSMO/ETH, and LUNA/UST →OSMO/DAI. LP providers now have the option to immediately de-pool in UST pools (but NOT immediately unbond), locking in their impermanent loss before the potential de-peg to zero.
In their proposal, Figment expressed some concern that allowing instant de-pooling would set a precedent of moral hazard. It is true that allowing emergency de-pooling could encourage extra risk-taking in the expectation of future bailouts, but the community (including Figment) voted YES overwhelmingly despite these concerns.
Presumably, people feel confident that emergency measures will be limited to the unique circumstances, namely that the one available dollar-denominated IBC stablecoin in Cosmos death spiraled. Going forward, non-stable holders should not expect bail-outs for rug-pulls, nor should should stables, since users can now price the risk between the multiple dollar-stables now on Osmosis.
Finally, it should be noted — since there were many comments on Commonwealth and social media that misunderstood this point — that instant de-pooling is not the same as instant un-bonding. All de-pooled assets must still be unbonded for users to access them.
Keep Buidling
One thing the UST incident hammers home is that we need to keep building as fast as we can. Moving forward, Osmosis will continue to build out the interchain DEX suite. The stableswap etc. upgrade, summarized last week, will likely be pushed a week or so, since the developers have had their heads down on the emergency hard-fork coding. The Osmosis Grants Program released a new round of grantees, of which Phase is perhaps the most exciting — but we will wait until next week for a full discussion.
vAMM: Given the volatility of the last weeks, it is a good week to think about a volatility-aware AMM — something that Sunny has mentioned before in passing. A vAMM would algorithmically add liquidity in stable conditions and remove it in volatile ones. This would allow LP providers to act more like traditional market-makers, who protect themselves in times of volatility by thinning the order books.
I am curious whether there are MEV games to play in triggering the algorithm into thinning or thickening liquidity, but we will likely here more about the potential mechanism another time. If it cannot be gamed, a volatility aware AMM would benefit passive liquidity providers at the expense of traders —to some extent the opposite of the Uni v3 approach to concentrated liquidity. This aspect would need to be fine-tuned because too much of a bias toward LP providers might drive traders to other exchanges. But if it prevents OSMO from being exit liquidity in times of turmoil, this might be a feature, not a bug.
Superfluid success: While the conservative 50% safety factor for superfluid staking can be unsatisfying, it proved its worth in the toughest of tests. The safety margin is designed to ensure that no attacker can trick the chain into thinking that it has over one-third of stake (i.e. cause a chain halt) from a rugged or attacked superfluid pool. In other words, even with LUNA cratering to almost zero, no attacker could buy enough LUNA to trick the LUNA/OSMO pool into thinking it had one-third of all staked OSMO.
Moreover, if an attacker had taken over Terra and infinite-minted LUNA and UST, it might have been able to attack Osmosis through a superfluid pool if it only had to overcome the regular-staked OSMO on-chain. However, the additional security provided by superfluid staking added to the regular stake, would have been enough to withstand any attack..
-Random question: could we lower the superfluid safety factor if the underlying OSMO were recalculated more often?
A note on ION: In other risk management news, OSD, the Osmosis dollar, was a big topic of last week’s Updates. It may make sense to push this over to ION DAO as it gets its governance set up. The proposed synthetic asset protocol based on Terra-style minting and burning is clearly too risky at this point. Algo-stables can be used to experiment with new units of account and new monetary policies based on algorithms instead of central banks, but the extreme reflexivity of every design so far has been too dangerous. A stablecoin was one of the main early ideas of the ION working group, so we shall see what ideas come forward.
CosmWasm developers welcome! We sympathize with you whether you choose to stay with Terra or find a new platform, but if you are looking for a new home, please consider Osmosis, particularly if you think your app fits in with the DEX suite vision: lending apps, leverage, arbitrage, options, derivatives, etc.
Osmosis can serve as an incubator for teams that want to find product-market fit before launching their own app-chain in the Cosmos. Whether you think you fit or not, please reach out to someone (if you don’t know anyone, you can ask around on socials, and our admins will point you in the right direction), and we can talk about how you might get onboarded, or we can point you on your way to Juno, Secret, or your own Cosmos chain.
The outreach has begun in channels like Sunny’s Gateway talk and Chjango’s joint Twitter Space with the Terra community. Check out the Osmosis Grants Program — we have already funded a number of projects and will continue to fund more, and governance is likely willing to direct more funds to program for high quality teams.
Bitsong
Finally, it was a relief to hear some normal good news about teams building on Osmosis during this Updates. Iulian from Bitsong came on to tell us about Sinfonia, their new front-end based on Osmosis.
Initially the Bitsong team thought about building their own DEX, but since Osmosis already has the infrastructure, they decided to make a dedicated front-end for trading fan-tokens that live permissionlessly on Osmosis. Sinfonia launched a testnet last week: 15,000 users registered already, and Iulian invites more people to join them.
Sinfonia fan tokens can be issued by anyone — artists, managers, fan clubs . They are tradeable, and can be used for crowd-funding and token-gating, allowing the artist-fan relationship to expand in new ways. The Sinfonia front-end may also grow to include features like playing an artist’s song, etc., while still having the trade functions leverage Osmosis.
This sort of re-skinning of Osmosis is exactly the sort of thing we want. We have our own front-end, but we want other protocols to use Osmosis. Sunny reports that there is another team building a sports-betting app with Osmosis that may embed ESPN into their app, so that users can watch games/matches while they bet. One can imagine that other apps will also use novel Osmosis front-ends in their apps: for instance, once mature, Regen Network’s carbon markets could benefit from such an arrangement.
That’s about it for this eventful week! I hope everyone has been able to stay safe and healthy and find ways to mitigate their losses. We’ll see you again next week on Updates to discuss our continued plans for building out the Osmosis ecosystem.
Enter the laboratory at Osmosis.zone, the first decentralized exchange powered by the Cosmos SDK and IBC. See our published lab reports at the Osmosis blog, our bench notes at GitHub and help plan future experiments in our Commonwealth
Connect with other DeFi Scientists by following us on Telegram, Twitter, Discord, Reddit, and the new Facebook and Instagram pages.
Reach out to the Ministry of Marketing by Email or Twitter and the Support Lab by Email or Twitter
