Data as a primitive building block for dApps — Part 1: DeFi applications
Exploring Decentralized Finance (DeFi) applications that will benefit from decentralized data management standards and Band Oracle integrations.
Today, the vast majority of DeFi applications do not operate in fully trustless and decentralized environments because of their reliance on integrating outside data form the real world. By far the most common need for DeFi applications is to have reliable and secure price feeds, e.g. exchange price feeds needed to determine and facilitate the liquidation of undercollateralized CDPs in the MakerDAO system.
The vast majority of DeFi applications already utilizing oracles in their design cannot function effectively without them, so compromises between decentralization, security and functionality must be made in order for them to operate. This forms a clear weak link in the security and decentralization in the design of all DeFi applications that rely on oracles, in particular for the applications where the oracle controls functions, e.g. CDP liquidations to ensure Dai stability.
To compound this problem the existing oracle solutions available today are either particularly centralized or rely on obfuscated and indirect incentives to keep the oracle data providers honest. Band Protocol data feeds provide the best solution for decentralizing and distributing the risk presented when using oracles with its built in incentive mechanisms, while also providing low latency and up to date data on-chain.
Considering this, let us take a look at some of the most popular DeFi applications and how they could benefit from a Band Protocol integration:
Currently the MakerDAO system relies on fourteen different oracles each providing price data, with the result medianized before used in the system. These oracle data providers are currently run by Maker stakeholders/developers, of which the specifics/details of these providers are either anonymous or obfuscated to protect individual participants.
The main problem with this is that in the Maker system the oracle participants are not directly incentivised, only indirectly through being a participant in the system (e.g. employed by Maker) or stakeholder of MKR, which is subject to many other variables to consider other than just the correct operation of the price feeds.
Aside from little direct economic consequence from incorrect oracle behaviour, there is also little reputation at stake for the data providers due to the obfuscation of which parties/participants are actually operating the data feed.
If MakerDAO integrated Band Protocol price feeds it would help mitigate both of these problems as:
- Band data providers are paid directly for their work from the users accessing the data, incentivising providers to give high quality service as they compete for use.
- The data providers in Band Protocol are completely transparent and open, putting their reputation at stake (which impacts their future ability to earn rewards) adding further incentive for honest behaviour, increasing overall security.
- Community participants of each Band Oracle are able to stake the associated dataset token, earning rewards for giving weight to the best performing data sources/providers, such that the oracle is as secure and accurate as possible and are constantly incentivised to maintain this state.
One option that we believe is possible is that Maker could start by integrating a Band oracle into their existing oracle setup, and take steps to slowly integrate them more closely to assist further decentralization of their oracles, which in turn should help the protocol become more robust overall.
Compound is similar to MakerDAO in that it needs price oracles to determine whether or not the margin for a loan is undercollateralized and a liquidation needs to occur. Currently for Dai prices Compound utilizes the Maker oracle, and for BAT, ZRX and REP, the median of Bittrex, Poloniex and Binance prices are used. Aside from the Maker oracle used, the oracles are completely governed by the Compound team but there are plans to move to a community governed DAO structure.
Until this is modified, this places a significant amount of centralizing control and risk in the Compound team operating the oracles. A Band Protocol oracle integrate would be an option to assist decentralizing the operation of the oracles and provides a more simple alternative to building a more complicated DAO governance structure.
Synthetix is a project that creates synthetic tokens pegged to external prices using the SNX token as collateral. Oracles are used to inform the system of the appropriate price for the synthetic asset which matches the price movements of the real world asset traded on exchanges and global markets.
Currently their oracle solution is quite centralized, being controlled and operated entirely be the team as well as in some cases having used the median of only three data sources — which resulted in errant behaviour.
We have discussed this more extensively in a previous blog post.
Synthetix has expressed to their community that they understand these shortcomings and are actively working on and looking for solutions to this. As discussed previously in our earlier blog post, Band oracle integrations could be a solution that will help alleviate these past shortcomings and assist in their goals for further decentralization of their application.
Veil was a prediction market protocol and application, that let anyone create their own peer-to-peer prediction market. Markets on Veil had a resolver that could reference any oracle — this includes an Augur market, Chainlink feed or any arbitrary smart contract state.
Veil used these 3rd party oracles in order to make dispute resolution of an open market more efficient and quick to resolve. This implementation creates somewhat of a trade-off between sacrificing some decentralization by integrating oracles to gain greater market usability/regulatory friendliness — ultimately becoming one of the leading factors for the project deciding not to continue as they deemed it not being decentralized enough for a cryptocurrency userbase to find valuable.
We believe that had Veil had other options available that would allow them to decentralize the available oracle options as much as possible, this would have helped them alleviate one of their weakest links in the security model and censorship resistance of their design.
An approach to create pragmatic decentralization of oracles through Band integrations should be considered in future for other projects to provide the needed censorship resistance, ‘unstoppability’ and practical data inputs needed for an application to be still be considered valuable without having to make as many trade-offs between usability and decentralization.
For more information on Band Protocol integrations, please review our website and documentation.
Published at Fri, 20 Sep 2019 14:15:24 +0000
Bitcoin Pic Of The Moment
✅ This image from Marco Verch (trendingtopics) is available under Creative Commons 2.0. Please link to the original photo and the license. 📝 License for use outside of the Creative Commons is available by request.
By trendingtopics on 2019-04-09 14:04:45
