Although the term “oracle” may sound foreign to some within the crypto circle, it actually isn’t anything new. That being said, it has seemingly gained significant momentum recently – mirroring the elevated interest within the decentralized finance (DeFi) sector this year. So, what exactly are DeFi oracles? While many would automatically assume that it can predict the future or foresee the currently unseen (because of its name), oracles in the crypto space are much simpler and less-mystical than that.
Join us as we take an in-depth dive towards the true definition and function of DeFi oracles.
What are oracles?
Oracles in cryptocurrencies are the ones responsible for the provision of real-world off-chain information to blockchains, which will then be used by its respective smart contracts. Basically, crypto oracles serve as data sources crucial for the establishment of a smart contract. However, it is worth noting that oracles aren’t data sources by definition. Oracles are but layers that check on-chain data concerning the real-world events.
As it currently stands, DeFi-based organizations are heavily reliant on oracles because of its ability to supply accurate real-time data. Notably, seven out of ten of the most used DeFi applications today are supported by either centralized or semi-centralized oracles for crucial off-chain data. This development can be attributed to the lack of reliable and efficient decentralized oracles today.
Demand for oracles is surging due to the fact that several blockchains today are still missing stored on-chain data within their very ecosystems. In reference, this data is usually found on prominent crypto exchanges such as that of Coinbase and Binance. This is because they have the application programming resources that enable oracles to execute accurate queries.
Now that we have defined what oracles really are; let’s identify its various types and unique applications
The different oracle types
Looking into the current DeFi landscape and identifying the oracles utilized within the sector, we could generally segregate oracles into five distinct categories. DeFi oracles are either centralized, decentralized, distributed, prediction markets, or delegated proof-of-stake (DPoS).
Let’s look at each of them:
- Centralized Oracles
Centralized oracle is the simplest form of oracles. Centralized oracles are where a single party, typically a whitelisted third-party provider or the very protocol that utilizes the oracle itself, provides the needed data. Due to its nature, centralized oracles are generally more responsive and faster compared to its other contemporaries, which are more decentralized.
Although faster typically means better, there are some lingering issues with centralized contracts that most should be aware of. While using centralized oracles, one must commit an extensive amount of trust. This is because doing so would mean that there would now be a decentralized smart contract framework supplied by a key point of failure, which is the oracle itself. Basically, the central party could censor the protocol’s data if they so wish. Centralized oracles are indeed more capable and significantly more responsive, but it comes with a higher risk due to its very poor liveness guarantee.
- Decentralized Oracles
In direct contrast with the previous entry, decentralized oracles are much slower and expensive to use but have a significantly higher level of liveness guarantee. This type of oracle is a system supplied with an intensive open network of data providers that acquires unanimity the same as that of a blockchain utilizing economic incentives and game theories.
Decentralized oracles are dependent on several external sources to elevate its data credibility and, thus, its much higher costs. It is also worth noting that this type of oracle functions on the Schelling points game theory wherein all participants must supply data without getting in touch with one another. After that, the Schelling game would determine whether the agreement data point or the proposed adjustments regarding the software are viable or even allowed right after it eliminates any accuracy issues. Basically, decentralized oracles perform the job better in exchange for a far slower and expensive approach.
- Distributed Oracles
This is the type of oracle where several parties are whitelisted to place the data on-chain, enabling the user to carry on any transformation methods before actually using the said data. It is deemed as a much better design than that of the centralized type but is still considered as likely to be manipulated due to its limited liveness guarantee. Its degree of centralization is still relatively high to be exempted from the glaring risks plaguing centralized oracles.
That being said, there are variations of this oracle type aiming to provide better service – particularly the ones that are utilizing the relaying mechanism. This particular variation is where multiple parties are signing the prices while one pushes the value within the chain. However, the issue with this design effort is that it actually makes it that even more vulnerable to manipulation due to the addition of a particular relayer. Although this variation may prove more useful, its liveness guarantee is much lower than that of its original distributed design.
- Prediction Market Oracles
Prediction market oracle is where users are given the right to vote on a particular outcome through bet placing. This means that whenever you assume 51% honesty within the system and its parties don’t want to lose any assets, then the oracle would work and would be decentralized. This oracle type has a much higher level of liveness while also being less susceptible to any forms of manipulation compared to the other types. That being said, it suffers from poor regulatory, liquidity, speed, and cost issues, which makes it ultimately unsustainable for several DeFi protocols today.
- DPoS Oracles
This is an oracle type where whitelisted staked nodes provide accurate data as they would lose their stake if the data they provided is deemed inaccurate. This exact element is what led many to believe that this type of oracle is better than the previously discussed distributed model. Within DPoS oracles, data providers are incentivized for their proper contributions. However, it is still crucial for the user to pay great attention to the whitelisting process and the whitelister itself. Just as with any other DPoS systems today, it is integral for the user to understand how its nodes communicate with one another and how easily they could conspire.
Most prominent DeFi protocols that utilize oracles
MakerDAO and Compound are among the most popular DeFi protocols today that openly use oracles for their everyday operations. MakerDAO, in hindsight, uses a particular oracles module to identify the real-time price value of several assets. This module is built by plenty of whitelisted addresses of aggregator contract and oracles. The oracles would then deliver periodic price updates to the aggregator responsible for the median price identification. This median price would then be used as the reference price of the given platform.
Much like MakerDAO, Compound utilizes oracles to aggregate price information that would then be promptly forwarded to its respective price feed. Notably, these price feeds are controlled by the so-called “administrators,” which are actually holders of the protocol’s native token, the COMP tokens.
The risks surrounding DeFi oracles
Summarizing the things we have discussed so far, oracles can easily be classified into two clashing categories: slow but secured or fast but risky. Either way you look at it, there would be downsides coinciding with its upsides.
Basically, the problem with oracles starts because of the conflict of trust that centralized outside party systems bring into the blockchain and smart contract systems that are naturally decentralized. And because the data supplied by the oracles are directly integrated into the developed smart contracts, which are notably crucial for it to function, it is only evident for oracles to hold higher power regarding the execution of said smart contracts. Considering that, it is critical for DeFi protocols only to utilize oracles that are trustworthy and highly-efficient.
Let us take two of the most common type of oracles today as an example to highlight its respective risks, centralized and decentralized oracles. Firstly, centralized oracles utilize a single point of failure, which heightens the risk of illicit hack attacks. This lack of security makes them less attractive to DeFi protocols, despite its much faster and more responsive qualities. On the other hand, decentralized oracles are incredibly vulnerable to party collusions and other internal struggles because it relies on several independent sources. On top of that, there is no retaliation framework implemented as of yet within this type of oracle once targeted by a hack. Although decentralized oracles are more in-demand today due to its enhanced reliability and functions, DeFi protocols using this model only have a small margin for error, as even one incorrect value could lead to devastating results.
DeFi protocols today are more than willing to commit tons of cash just to get their hands on the tamper-proof and reliable data that oracles possess. However, it is apparent that present models still need to elevate their security levels and come up with more reliable and innovative frameworks. That being said, there is no denying that oracles are very interesting and a proven profitable market. With more and more DeFi protocols looking into these complex prospects, it’s only fitting to say that the future of the crypto space would heavily be influenced by oracles.