DataDEX’s data market making mechanism

Recently, I was designing the economic model for data exchange with the team and discussed in depth the characteristics of the combination of data in DEX and AMM. I have some thoughts and summaries to share with community members.

Data pricing
The issue of data pricing has always been a difficult problem. From the current point of view, there are no successful cases that can explain a general pricing strategy for data. People always use oil as a metaphor for data resources in the new era, not because of its commodity attributes, but because of its strategic value. Therefore, data is not commodity, and we cannot use the idea of ​​commodity pricing to design a price making mechanism for data.

The value of data cannot be directly measured. The value of data is determined by market supply and demand. In addition to the “Quadruple V” attributes of big data, data is also scarcity. Therefore, the value of different data in different scenarios is very different. In particular, personal privacy data is scattered on the edge of each user in terms of ownership. Due to privacy protection issues, it cannot be collected to the central server. A mechanism is needed to aggregate the value of the data to form a more valuable Dataset. In order to provide value for data users. In this way, a new collaboration is formed. DataMaker establishes and promotes Datasets, and reflects the data supply and demand scale. Data users can buy data under an open and transparent pricing policies, and DataOwner can register the owned data to different DataSets to obtain benefits. This collaboration needs to be based on a computing network that can perform federated computation on edge resources and data, reward is recorded in a decentralized ledger, and support smart contracts so that DeFi applications such as DataDEX can run.

AMM in the DeFi era brings new opportunities for data universal pricing. Automated market making mechanisms, such as constant function market making algorithms, can reflect the scale of supply and demand. It is a simple liquidity market making mechanism based on smart contracts, which is efficient and transparent, and is very suitable for the exchange of data accessibility.

Data value measurement
The data has no value in the original storage format. Only after it is computed and the aggregated result is applied in the physical data scenery can it be valuable. Moreover, personal privacy and commercial confidential data have definitely legal ownership, and the exchange of data in plain text is meaningless and has great legal risks. How to measure the value of data is a question worth exploring.

Data is exactly copyable, so the value of data cannot be measured in data units such as storage space, rows or columns. Those methods are suitable for cloud storage services and cannot reflect the value of data, ignoring data scenarios and scarcity. For example, during the ‘Black Firiday’, brand company used consumer data, through insight, cleaning, clustering and inference algorithms, to target those customers who are willing to buy, and the transaction volume increased significantly. Small sellers also use consumer data to simply filter out certain behavioral history customers to push advertisements, with limited effect. Therefore, it can be seen that the same data is used by different algorithms, and the value of the output varies greatly. The use of data is generally related to consume of computation, that is, computational complexity, which can be measured by the amount of data input and the amount of computation tasks.

Multi-party experience friendly
Data users are not just arbitrageurs. Most users have physical and complex business use cases based. Datamaker is not just a simple sales agency, but a professional toB field expert. So multilateral friendship is very important for DataDEX. Naturally, due to the price difference between DEX and traditional data market, some of arbitrageurs will exist and swap DataToken in DEX.

We will develop the slickest UI of DEX to make it easy for DataMaker to create Datasets and publish DataTokens. AMM’s algorithm is expected to be concise and user-friendly, so the CPMM market-making based on single asset trading pairs is one of the ideal choices, similar to UniSwap. But we will not support multi-asset allocation, which is unfriendly to users. For example, Balancer, corporate customers cannot establish a financial department to operate data resources. They prefer to use a single master Token to trade DataToken. The experience of data customers cannot be sacrificed in order to concentrate liquidity.

Data-based financial and derivative applications can be explored in the future ecological development. We will also establish a DAO to ensure the sustainable development of this ecology (discussed later).