IOSG Ventures: A Discussion on the Oracle Track Map

Author: Sally, IOSG Ventures

After recognizing the importance of the role played by oracles, we believe it is necessary to conduct a deeper research and summary of the oracle track. Therefore, we first systematically sorted the mainstream oracle products in the market (see the figure below).

Oracle Map

Here we mainly classify oracles based on three criteria (specific functions, data providers, data processing methods):

Specific Functions

According to the specific functions of oracles, we can divide them into five types: Credit Oracles, Privacy Oracles, Optimistic Oracles, NFT Oracles, and Decentralized Oracle Networks.

a. Credit Oracle

This type of oracle can calculate users' credit scores or credit ratings by utilizing data from the real world (off-chain) and the blockchain (on-chain), while maintaining security and anonymity.

For example, CreDA provides on-chain credit ratings by minting users' credit profiles into NFTs, allowing users to unlock favorable interest rates and incentives in various use cases, such as reducing borrowing rates on DeFi platforms.

At the same time, it has created a DID tool to assist in the exchange of identity information with other systems' DIDs, such as Elastos DID, Microsoft DID, etc., ensuring that the process is implemented transparently.

CreDID Mechanism

b. Privacy Oracle

Privacy Oracles provide services to securely and privately import off-chain data into the blockchain world.

Privy serves as a good example, as it tends to bridge access between public (on-chain) data and private (off-chain) data. It can encrypt user data directly from the platform's front end and privately associate the data with on-chain addresses.

Privy SDK Interface

c. Optimistic Oracle

Optimistic Oracles are DeFi oracles used solely to resolve disputes regarding liquidity. They can typically provide services for any arbitrary data on-chain.

For example, UMA's dispute resolution system, the DVM (Data Verification Mechanism), will not be activated unless there is a dispute. Once UMA token holders vote on the price of an asset, the DVM typically resolves the dispute within 48 hours. If the price returned by the DVM shows that the disputer is correct, then the disputer will receive a reward, while the proposer or liquidator will lose their collateral.

UMA Workflow

d. NFT Oracle

NFT Oracles provide NFT data from off-chain oracle AI nodes and offer accurate valuation and risk assessment of NFTs.

A typical example is the Banksea oracle, which integrates NFT data analysis, NFT valuation, and comprehensive risk assessment of NFTs, providing users and partners with secure, objective, and real-time NFT evaluations. It also has high scalability, suitable for various use cases such as wallets, trading markets, and loans.

Banksea Main Functions

e. Decentralized Oracle Network (DON)

A Decentralized Oracle Network (DON) can be simply understood as an oracle solution with unlimited application scenarios and maximum functional services, as it provides external data to the blockchain through trustless off-chain computation and can be integrated into any smart contract. Chainlink elaborates on its positioning as a DON in its white paper. To implement this positioning, it has launched a series of products and services such as VRF, Keepers, and CCIP.

Source: https://twitter.com/kyleodesign/status/1182189126419828738

Data Providers

Based on the data sources used by oracles, we can classify them into three types: First-party Oracles, Third-party Oracles, and Multi-party Oracles.

a. First-party Oracle

First-party Oracles provide a platform for data providers to directly push their data on-chain, allowing requesters to query the data immediately. The advantage of this solution is that it can directly use first-hand data, and the service fees charged are relatively lower than those of third-party oracles. The downside is that it requires direct integration with exchanges, handling updates and other issues, which can be cumbersome. Typical examples of this type of oracle include: Pyth, API3, DIA, Umbrella, etc.

b. Third-party Oracle

Third-party Oracles involve third parties to pre-validate first-hand data sources. After the third-party validation is completed, the oracle delivers the provided data to the industry chain. The advantage of third-party oracles is the authenticity and reliability of their data, while the downside is that they are overly centralized. Typical solutions in this category include: Chainlink, HAPI, DOS, etc.

c. Multi-party Oracle

Multi-party Oracles provide both first-party and third-party data sources, allowing users to freely choose their preferred model. A typical representative of this solution is Flux.

Data Processing Methods

Based on the data processing methods of oracles, we can classify them into four types: Game-theory Oracles, Reputation Oracles, Proof-of-Stake Oracles, and Proof-of-Work Oracles.

a. Game-theory Oracle

Game-theory based oracles provide non-hostile economic incentives to validators, with typical products such as: NEST, WINKLink.

b. Reputation Oracle

This type of oracle limits hostile nodes by lowering reputation, typical examples include: Witnet, QUE, Lithium.

c. Proof-of-Stake Oracle

Proof-of-Stake based oracles require participants to hold assets to increase credibility, with typical products such as: Band, Razer.

d. Proof-of-Work Oracle

Proof-of-Work based oracles allow nodes that solve puzzles the fastest to win the computational competition and provide data, with typical products such as: Tellor, Gravity.