Published in


Ontology Tech Viewpoint | A Brief Analysis of Web3 Credit System Construction Ideas, Difficulties, and Solutions

With an increasingly enthusiastic and in-depth discussion around Web3, users have developed a deeper understanding and as such now have clearer and higher requirements for Web3 projects. This also makes project parties and users turn their eyes to some topics and fields still under exploration, such as “reputation” and “social”.

After Vitalik published an article about the concept of SBT, “Web3 reputation system construction” was pushed to the forefront of the topic. This paper will discuss the definition of reputation, development status, existing problems, and solutions through two articles respectively.

The definition of reputation will not change because of Web2 or Web3, as its existence is essentially due to the needs of social order and system maintenance. For individuals, reputation is a certificate to prove that they have no bad behavior and is widely used in daily exchanges and transactions between people and financial institutions. For enterprises, project parties, and organizations, reputation involves competition in the market and the collaboration between them. It is a tool for enterprises and organizations to obtain universal social recognition, so as to obtain relevant resources, opportunities and support, and then complete value creation.

Reputation is a basic means to maintain social order, the basis of distribution and spontaneous social control. Its influence covers a wide range of things, from interpersonal communication to the maintenance of relations between organizations and countries. Therefore, the reputation system is necessary for the normal operation of society, which is particularly important in both Web2 and Web3.

The Web2 reputation system is characterized by having a large scale, can be applied in many scenarios, and has limited display time for poor credit records. Let’s take the world’s largest personal and corporate reputation system as an example:

Image source: Official website of the People’s Bank of China

By the end of December 2020, the People’s Bank of China had collected credit information on more than 1.1 billion people and 60.9 million enterprises and organizations. The average daily number of credit investigation reports for individuals and enterprises had reached 8.66 million and 190,000 respectively. It can be seen that the social demand for credit data has been very large.

With the development of science and technology, and the exploration of credit use scenarios, in addition to the common traditional financial field, credit systems are also widely used in more life-like scenarios such as rental platforms, and travel and accommodation apps.

Meanwhile, bad information can only be displayed on the credit report within 5 years from the date of the termination of the bad behavior or incident, which also means that the negative impact of bad information on individuals and enterprises in the Web2 reputation system is limited. Compared to Web2 reputation systems, which are widely used and have strong supervision, the construction of Web3 reputation systems is still in its infancy, and there is no cross-field, cross-project reputation system with credibility at present. The ideal Web3 or new generation reputation system should have comprehensive credit data, strong technical support and reasonable supervision process:

First, create a comprehensive three-dimensional credit portrait.

The future reputation system should include both off-chain and on-chain data to record individual and corporate credit behavior from all dimensions. For example, on the basis of traditional financial data, related interactive data on the chain should also be included in the credit report. In addition, because on-chain data is difficult to tamper with, the negative impact of bad behavior on individuals and organizations is often permanent. Data integrity and the difficulty of tampering can also ensure that individual and enterprise credit behavior is better regulated, so as to consciously resist bad behavior.

Second, have strong and privacy-preserving data storage and technology.

This is the most lacking point in the Web2 reputation system at present, and it is also the reason why we repeatedly discuss and explore the Web3 reputation system.

The current distributed storage can ensure that the possibility of loss is much lower than that of centralized storage when the data storage volume is very large. More importantly, the use of blockchain, VCs, DIDs and other technologies can ensure the authenticity of credit data, and users can also maximize the control of their own data, in line with the Web3 vision.

For example, users can choose to disclose only the necessary credit data to pass to the relevant credit authentication, or even choose to disclose only the credentials that meet the conditions to pass the authentication.

Third, establish a new generation of regulatory norms and roles.

Regulatory agencies or organizations need to be responsible for the development of a new generation of reputation system norms and regulatory roles, in order to avoid project parties creating reputation systems that only benefit themselves. With a professional regulator, the reputation system can be adopted and tested more quickly and comprehensively by more project parties and fields, so as to create a set of Web3 reputation systems recognized by the public.

In the Web3 world, the use value of a reputation system will be further extended on the basis of Web2.

Individual users can choose to expose their credit score or rating on social media and Dapps, present a good profile in Web3 socials, and use their credit report for potential benefits. For DAO and Web3 project parties, they can better understand their community through user credit reports, so as to make targeted promotions and function developments to improve community member engagement. More importantly, they can use smart contract technology to automate the implementation of norms and rules related to credit ratings.

For example, the DAO project can increase the voting proportion of members with good credit, and improve their autonomy and sense of belonging. As you can see, we have a very clear need for a Web3 reputation system, but there is no consensus on a reputation system for Web3. What is the reason for this? What’s the difficulty?

As the interactive data on the blockchain is open and reputational, most of the Web3 project parties involved in the reputation score are developed around these open data on-chain, and only for a specific field to develop the reputation model.

Image source:Spectral Finance

Spectral Finance, for example, trained a scoring model for credit risk using machine learning and named it Multi Asset Credit Risk Oracle score (MACRO score). In the latest open beta model, the reputation scoring system for user interaction on Ethereum, referring to 4 types of credit risk-related behavior, 2 types of dimensions related to the history of decentralized data clients, and market macro environment factors, a total of about 100 data sources, and support multi-client connection.

However, the reputation system should not only include the above data, but also include interaction data in other areas of Web3, such as contribution interactions within the Web3 community, involvement in intra-community fraud, dApp interactions, and so on.

However, the reason this data has not been used yet is that there are so many existing Web3 communities and many community platforms, including Discord, Telegram, Notion and self-built forums, where there is no common data to capture the user community behavior perfectly and to add observation and analysis. Also, this part of data is not completely open and transparent, which becomes the most important part of the data missing in the existing Web3 reputation system.

In response to this data, some project parties chose to use NFTs to record user behaviors, such as POAP and Noox. The benefit of this format is that different dimensions of user behavior can be recorded, linked, and authenticated, and the project can further classify community members based on whether or not the user holds certain specified NFTs, as well as targeted maintenance and promotion. The disadvantage is that there is no uniform indicator to measure which criteria a user needs to reach to qualify for a claim, so the NFTs are likely to be used only in closed loop on the platform.

In addition, from the perspective of DAOs, the project side collects the corresponding user behavior data according to the characteristics and requirements of different DAOs and sets the score. For example, Karma has customized models for several DAOs, such as Aave, Bankless, ENS, etc. These DAOs can use this customized model to continuously observe the member activity and facilitate the discovery of new members who contribute more to the organization. Each member can also refer to the contribution score to further improve their contribution to the DAO to get more revenue. On-chain data is difficult to obtain, and off-chain data is even more difficult to obtain.

Currently, Web2 data are basically held in the hands of centralized organizations (such as Facebook and Twitter), so it is difficult for Web3 reputation systems to obtain the most core off-chain user reputation behavior data. At present, the reputation system can only access the off-link data through the open API, which also requires user authorization and undoubtedly adds a lot of obstacles to building a new generation of reputation systems.

At the same time, the biggest difficulty in the data aspect of the reputation system is the balance between data volume and user privacy. On-chain data that requires user authorization is even more criticized. In the future, Web3 reputation builders will need to work closely with Web2 applications and think carefully about the reputation model architecture.

Only with a comprehensive and large scale user on-chain and off-chain data can we accurately quantify user reputation according to user behavior, and promote the use of the same reputation standard system in all fields.

As we all know, on-chain behavior is bound to the address of the decentralized data client. The same person can register multiple addresses for interactions. While off-chain behavior is bound to the identity system of each country, each person has only one unique identity code, but each person can have multiple Web2 social application accounts, so the reputation system often cannot capture the complete behavior of users, and the authenticity of user identity cannot be guaranteed.


As a decentralized identity identification protocol based on W3C DID specification, ONT ID can better integrate the identity of Web2 and Web3 by recording the address associated with DID in real time and the interaction records of Web2 and Web3 platforms, so as to build a more comprehensive reputation system. At present, Web3 reputation system infrastructure projects such as Orange Protocol have been equipped with ONT ID as the unique user account.

Image source: BrightID

Another way to solve this problem is developed by BrightID, which acts as a Web3 authentication system that requires users to participate in verification meetings, establish connections with friends, and verify their identity based on social relationships. Its advantage is that it does not need to collect any personal data or third-party certification, which protects users’ privacy to the greatest extent. However, its disadvantage is that the verification process is tedious and tests users’ patience.

Source: Gitcoin Passport

The Gitcoin Passport allows users to authenticate multiple apps on a single platform by aggregating their identity data offline. The platform already integrates Web2 and Web3 applications such as Twitter, Google and BrightID. In addition, users are allowed to select data according to their own needs to generate VC authorization. For example, users can choose to only verify the Facebook account name, the number of friends or the entire account, which also protects the privacy of user data.

These projects can be used as the basic system and identity verification tools of the new generation of Web3 reputation system, and integrate the on-chain identity system to further expand the availability and authenticity of the reputation system.

The third difficulty is that Web3 reputation systems lack regulatory processes and regulations.

This is not just a problem in the field of reputation right now, but a bottleneck in Web3 as a whole. In the absence of supervision, it is difficult for the reputation system to be authoritative and truly promoted, and the user behavior is difficult to be constrained.

At present, the blockchain field is still in the early stage of development, and the projects of each track are also in the process of exploration. Projects with reputation requirements will develop a set of favorable systems according to their own characteristics to improve their growth and retention, which cannot be used across platforms and fields.

However, the institutions responsible for supervising the Web3 reputation system should involve members from different countries, and formulate relevant regulations that are in accordance with different national conditions, so that the reputation model can have a “global consensus”. The intervention of authoritative regulatory agencies can not only ensure the fairness and justice of the reputation system, but also further improve the problems of difficult data acquisition and inconsistent identity systems, which is the basis for promoting a new generation of reputation systems.

The construction of a new generation of reputation systems is necessary. It will be based on the existing Web2 reputation system, with the development of a Web3 field, giving full play to the characteristics of Web3, to create a set of systems with perfect data, privacy protection, and reasonable supervision.



Ontology is a high performance, open source blockchain specializing in digital identity and data. ONTO: http://onto.app/downloadpage/TW Telegram: http://t.me/OntologyNetwork

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store