Hong Kong Polytechnic Professor Cao Jiannong: Web3’s decentralized data infrastructure

On September 19, 2023, the “2023 Shanghai Blockchain International Week·The Ninth Blockchain Global Summit” opened in Shanghai. Cao Jiannong, Professor of Data Science at the Poon Letao Charitable Foundation of the Hong Kong Polytechnic University; Chair Professor of Distributed and Mobile Computing in the Department of Electronic Computing, gave an online lecture titled “Decentralized Data Infrastructure of Web3: Current Progress and Future Directions”.

Golden Finance conducted on-site tracking and reporting of the entire meeting. The following is a summary of the speech content.

Dear guests, good morning! I am Cao Jiannong from the Hong Kong Polytechnic University. First of all, I am very grateful to Wanxiang Blockchain for the invitation. The title I shared is “Web3’s Decentralized Data Infrastructure: Current Progress and Future Directions.”

Everyone knows that Web3.0 has become very important recently. First of all, in terms of technology, we have seen Gartner’s technology maturity curve. Web3.0 reached its development peak last year and this year. Governments in various places have also regarded Web3.0 as a key target, especially Hong Kong. Hong Kong issued a Web3.0 strategy statement this year to build Hong Kong into a center for the conversion and retention of virtual assets. In May this year, Beijing also published a white paper on the innovative development of Web 3.0. Internationally, the British new technology department has proposed a strategy for Web 3.0 and a national development strategy for the Metaverse.

What exactly is Web3.0? Web3.0 can be regarded as the next generation of decentralized Internet. On the first generation of the Internet, you can read the information you want to see. Web1 is equivalent to a structured file system, which structures the required information to facilitate our search. This is a Read only static Web. When it comes to Web2, in addition to reading, you can also create content, so it is called Read write Web. In the third generation, it is Web3.0. In addition to being able to read and write, it also has created content, so it is called Read write own value Web. The content we create becomes very valuable.

Web3.0 has three very important characteristics. The first is de-platforming and decentralization. There is no centralized platform to control and manage user data. The second is self-control. Self-control means that users have complete control over their own data and identity. The third is tokenization. All data and content created by users can demonstrate their value through tokens. These three characteristics determine that Web 3.0 is a decentralized new Internet where users can own content and share value.

In the Web3.0 environment, we can see many different innovations. The first is innovation in upper-layer applications. There are many innovations in decentralized finance, GameFi, and SocialFi, all of which revolve around Defi. Defi uses blockchain, smart contracts, and digital identities to remove unnecessary aspects of traditional finance. intermediate channel. GameFi and SocialFi apply Defi concepts and technologies to gaming and social aspects respectively. For Defi, there are NFT and GameFi. In addition to innovations in the application layer, innovations in Web 3.0 can also be seen in corporate organizations, society, economy, etc. The most representative one is DAOs. DAOs allow an organization to self-organize and run through software without centralized management.

The economic and social architecture of Web3.0 has also received support from everyone. The next innovation of Web3.0 is based on blockchain platform, smart contracts and digital identity. But in addition to blockchain, we also need other architectures, such as digital infrastructure and computing power infrastructure. Especially the recently common generative AI, which goes from data to models, as well as campaigns created by users using models, are very good innovative applications of Web3.0.

It can be seen that in the Web3.0 world, in addition to blockchain smart contracts and digital identities, we also have an important component. This component is data, and data is value in Web3.0. We can see that users are using various data. The data can be Pos (location) and Values (value). Under Web3.0, data can be tokenized, and tokenization becomes the user’s digital asset. At the same time, the user’s identity can also be digitized, so there is a decentralized digital identity. The first step is to establish a correlation between the two, which is that the user owns the digital assets it creates. After that, other users will also have their own digital assets, and a Web3.0 network can be formed, and users can share and use the data they own. The value generated by the data each owns will be recorded, traded, and fairly disseminated. Therefore, Web3.0 is a world that revolves around numbers.

The first is the tokenization of digital, then the correlation and authentication between digital value and user identity, then the use of digital assets by each user can be traced, and finally the value generated through use and sharing can be fairly Distributed automatically.

Under such circumstances, why does the Web3.0 field use blockchain and smart contracts? Because blockchain and smart contracts have been used to support a range of behaviors based on numbers and identities. Blockchain, smart contracts and digital identities are the main parts that everyone focuses on and applies. Digital architecture includes digital storage, digital sharing, and digital security. So far, it is the main part that Web3.0 lacks.

In the new environment of Web3.0, a decentralized digital architecture is needed. Based on this, I proposed a three-layer Web3.0 digital architecture.

The first is the digital storage layer. On the storage layer are the usage layer and the sharing layer. Above that is the value layer. At the storage layer, data is stored in a distributed manner, which involves a lot of content, such as how to efficiently and reliably store user data in a distributed manner. The sharing layer can find sharable data and how to protect user privacy and data privacy during the search process. The technologies supported by this three-layer architecture include blockchain, smart contracts and digital identity. In addition, support from artificial intelligence, the Internet of Things, and collaborative edge computing systems is also required.

Under such a Web3.0 digital architecture, what is the current development situation of the three-tier architecture? Current development is limited to the data storage layer. Mainly the development of decentralized storage systems, there have been distributed data storage systems many years ago. Such distributed storage allows users to disperse data to different nodes. How to maintain correspondence between digital identities? In addition to the distributed storage in the past, blockchain has been added. Now with blockchain and distributed storage, users first have to pay tokens to store data in the distributed storage system, and at the same time they also know how to obtain or find the data. The nodes of the blockchain are mainly used to store the data and information that users want to store, and tokens can be obtained on the blockchain. This is the main mode of operation now. But this is not enough. What it lacks is: first of all, the current system is mainly a data storage system, which does not take into account the sharing of data and the distribution of data value. Moreover, it mainly considers general data storage and does not consider Web3. The storage of special data in 0 environment, such as semantic data, has not been fully studied.

What is the main lack of Web3.0 now? There are two sides. The left side is the characteristic of blockchain. Data can go to every node. The data is highly structured and stored on the blockchain. The data stored in these stores is also open source. But in the Web3.0 environment, what characteristics should data have? First of all, a large amount of data is required, and it is not traditional data. There are pictures, videos, etc. as user data assets and campaigns (activities). The data I just talked about has very rich semantics, and the semantics are not just simple 0101 data. Next is the data generated by Web3.0 users, which has security and privacy requirements. This is what is needed to explore what is needed between the two sides regarding the digital architecture of Web 3.0.

What we lack is how to store shared and automatically assigned value in a safe, efficient, scalable, and reliable way, which requires a large amount of semantically rich, security- and privacy-friendly data.

Although it is a simple sentence, it contains many technical and theoretical research results. How to store large amounts of data using a decentralized platform? At the same time, reliability, access efficiency, and data semantics must be ensured. In the context of rich semantics, whether new search, packaging and consensus methods are needed, as well as the privacy and security of Web3.0 data, etc., pose new challenges in the context of data sharing.

I just mentioned the decentralized architecture of Web3.0. First, how to have a decentralized distributed data storage. Second, how can we enable them to be found, shared, and used.

This is just a digital architecture. In a blockchain environment, we also need a Text posls (text position), including redesigning the network, because this is a blockchain analysis, storage and search structure. Can it be used on the network? Give support. For example, does the network protocol know that the storage semantics of Web3.0 can be incorporated into the network protocol? At the top, we need to provide some service layers to support applications under Web3.0. This is a complete Web3.0 digital architecture.

In the future, in addition to digital architecture, Web 3.0 will also include scientific discovery and scientific calculations, which will require the support of computing power. More and more Web3.0 applications, including Metaverse and AIGC, require AI services. Does the Web3.0 environment need to support data processing and data servers? Web3.0 is an ecosystem that integrates data, computing power and AI services. So we see this picture. What we just talked about is the data architecture. When data is used, computing power is required. The result of the computing power is processed data, which can be stored in the data architecture. What are the needs of AI services? When the data model is used, the data needs to be accessed and the generated content is stored. AI requires the support of computing power. Therefore, data, computing power and AI services are organically combined. In the future Web3.0 world, these three important parts must be considered uniformly.

Our university recently applied for a research project in Hong Kong called collaborative edge computing. This concept is decentralized and can connect distributed computing power to form a network so that resources can be shared. Users can enjoy the data, computing power and AI services in this network after connecting to any node of the network. This vision is similar to the computing power network proposed by the mainland, but our construction is built from the bottom up following Web3, that is, independent computing power and data can be connected and shared. Under such circumstances, we can provide a technical or system sample for the future Web3 computing platform and data platform. This is also a project we jointly developed with the University of Hong Kong. We hope this project will contribute to the development of Web3.0.

in conclusion. The first thing I will talk about today is that Web 3.0 is emerging as a technology and application and has become very important to the government. Web3.0 is supported by blockchain and smart contracts, which are essential. But blockchain and smart contracts alone are not enough. The underlying infrastructure is also needed, including data architecture, computing power architecture and artificial intelligence architecture, to form a complete ecosystem and support Web3.0.

The current Web3.0 data architecture supported by blockchain still faces many challenging issues, such as the Volume and Full replication of large amounts of data, and how to handle it effectively and scalably. The balance between rich and heterogeneous and highly structured data in Web3.0, and how to handle it reliably and securely. Finally, data and all assets can be contributed in the Web3.0 environment. How can we ensure security, user identity, and privacy of user data? These are all challenging research topics.

Finally, there are hopes for the prospects of Web3.0. Integrating data, computing power and artificial intelligence services under a collaborative edge intelligence platform - we hope to make it possible in the near future.

I will share it here today, thank you all!