The Umbrella Advantage for DeFi Projects
“…feeds are only set up when there is popular demand and there is a sponsor in place to cover the operating costs.”
The Block, An Analysis of Chainlink Node Operator Earnings
In DeFi, we should be able to trade everything and anything. That is the beauty of Uniswap and nearly all of the projects that followed. Yet, our data providers don’t seem to agree. They want to curate what data is available and how often that data is refreshed. They want to control what we do and who we are able to do it with.
It is Umbrella’s belief that The Oracle Problem is curtailing innovation and therefore growth in the DeFi market. Yes, we’ve hit $15 billion of TVL. But surely that is just the beginning considering the S&P500 is a $30 trillion market by itself. Limitations on what data is available for application developers and how often that data is refreshed restricts what DeFi application developers can do. Furthermore, the fact that the data used to settle contracts, rebalance collateral, close existing positions, and more is controlled by a VERY centralized party has to give us all pause for concern.
DeFi projects have come up with some interesting workarounds to overcome the limitations of the data available to them. Projects like Harvest Finance and Compound relied on less secure but more up to date data sources like Curve and Coinbase Pro. This ended somewhat fatally as money was drained from both through a combination of flash loan and manipulation of the data source.
One could argue they should’ve used Chainlink. However, it is likely that the staleness of Chainlink data could’ve had a similar impact.
Andre Cronje implemented an interesting workaround. He took the 24 hour average of less liquid pairs for his Keep3r oracle, an oracle that came with an unusual disclaimer: “UniswapV2Oracle is an oracle of last resort and should be used for pairs not normally supported by other oracles.” That probably does not give DeFi projects the warm and fuzzy feeling one should have from their data providers.
But most commonly, projects steered away from taking less liquid assets that are subject to volatile swings in price and harder to establish a secure price for. They steered away from ideas and projects that required large datasets that aren’t available with current leading oracles. And similarly they did not venture into areas requiring high frequency of updates.
Some of the hottest ideas in DeFi are currently: options, indexes, VIX and other derivatives like Fixed for Floating. These all require data that is not frequently provided by oracles today. As an example, Hegic chooses a static option pricing model according to their white paper. I suspect they use this naïve method as an abundance of caution. Unfortunately, that makes their prices expensive and uncompetitive with their centralized counterpart Deribit Exchange.
At Umbrella Network, we use a system of Merkle Trees to anchor hundreds of data points onto a single Ethereum transaction. For details of how this works, take a look at our demo product and user guide: taurus.umb.network
We can batch data and proofs so that it is easer and cheaper for DeFi application developers.
Given Umbrella’s natural advantage, we will make 36 of the most commonly used crypto data pairs available on our layer 2 system for free. Furthermore, we will provide more than 40 of the most common data pairs on chain updated every 5 minutes once again for free to all developers. These will include Fiat Pairs, Crypto Pairs, Implied and Realized Volatility and Indexes like the DeFi Pulse Index.
For those that need less commonly used data pairs, we will charge on a SaaS based tiered pricing. We will encourage early developers by providing a free tier. This is to foster innovation, growth and also to get feedback on what the next generation of DeFi application developers need.
