Accelerate the construction of data infrastructure to support the high-quality development of the digital economy

Jiang Qiping

Accelerating the construction of data infrastructure is a major technical need for the construction of digital China, and it is also a strategic starting point for achieving high-level scientific and technological self-reliance and self-reliance.

The Decision of the Central Committee of the Communist Party of China on Further Deepening Reform and Promoting Chinese-style Modernization, which was deliberated and adopted at the Third Plenary Session of the 20th Central Committee of the Communist Party of China, proposes to improve the system for promoting the deep integration of the real economy and the digital economy. It emphasizes the construction and operation of national data infrastructure and the promotion of data sharing. Accelerate the establishment of systems for the identification of data property rights, market transactions, distribution of rights and interests, and protection of interests, improve the capacity for data security governance and supervision, and establish efficient, convenient, and secure mechanisms for cross-border data flows.

The difference between data infrastructure and digital infrastructure

The framework design of the technical layout of the continental data space should be analyzed from the "techno-economic" paradigm, not just from the engineering perspective. For the data space, the technology that constitutes the space for economic and social activities is the first priority. The "National Data Space Development Strategy Study" divides the infrastructure in the data space into two categories: digital infrastructure and data infrastructure. The former is infrastructure from a purely technical point of view, and the latter is infrastructure from a functional point of view.

Data infrastructure is relative to digital infrastructure. Digital infrastructure is primarily a technical facility, while data infrastructure is a techno-economic facility. The former is mainly composed of hardware, software and other technical systems, while the latter emphasizes the realization of technical and economic functions.

Digital infrastructure includes technologies such as the Internet, big data, cloud computing, artificial intelligence, and blockchain, while data infrastructure is an infrastructure defined from the perspective of technical and economic functions, and is the core component of data space. Data infrastructure is a new type of infrastructure that provides integrated data aggregation, processing, circulation, application, operation, and security services for the society from the perspective of the value release of data elements, with the support of network, computing power and other facilities, and is an organic whole including hardware, software, open source protocols, standards and specifications, and mechanism design.

From the perspective of technology in the technical economy, data infrastructure can be further refined and defined as a technical architecture, which is based on the construction of data resources, and the core part is the network facilities, computing facilities and circulation facilities represented by the future network, computing power network and data element field, as well as the two pillars of endogenous security and standards that run through it, to jointly support the value release of data elements. Among them, future network, computing power network and endogenous security are the products of the evolution of traditional information infrastructure technology, while the data element field is a new type of digital infrastructure.

From the economic perspective of the technological economy, data infrastructure is an organic whole including software and hardware, policies, protocols, standards, mechanisms, data resources, and participants. With the support of data infrastructure, the economic and social practice system in the digital era can be modeled as a dual-track system of institutional system and technical system, and all aspects of data production, distribution, exchange and consumption jointly drive the efficient release of the value of data elements, becoming a "steam engine" in the digital era.

From the perspective of planning and layout, the data infrastructure should be expanded into a technical system in the sense of "technology-economy" on the basis of the technical system of engineering significance (including future network, computing power network, data element field, and endogenous security). In terms of functions, the "data interaction" function of the data field is mainly highlighted, especially the functions covering data exchange, openness and sharing.

Improving the data infrastructure mechanism is the key

Facilitating data sharing is the goal of building and operating a national data infrastructure. Data infrastructure facilitates data sharing, including the free sharing of data as a public good on the one hand, and the paid sharing of data on the other hand, that is, the sharing of paid and charged data according to the effect of use.

To understand data sharing, we must first understand the concept of a data element field. The data element field refers to a new type of infrastructure that provides integrated data aggregation, processing, circulation, application, operation, and security assurance services for the society. It is the theoretical basis of data infrastructure.

The core feature of the data element field is also the core feature of the field, and the two are interacting, that is, interacting. It can be considered that a series of behaviors from aggregation, processing, circulation, application, operation to security assurance services all have the essential characteristics of interaction. The field is the space provided for interaction, and it is not a material space, but an activity space, that is, the space where people interact with each other and between people and things. The data element field mentioned here is the connotation of circulation infrastructure.

What is the value of the field? According to the description of the academician of the Chinese Academy of Engineering, it has the functions of correlation release, fusion release and multiplication release for the value of data and data elements. The "Data Element X" action is to release the value of data elements in the way of association, fusion and multiplication.

With an understanding of the above principles, we can further deepen our understanding of mechanism design. The first is the determination of data property rights, that is, the confirmation of rights. The property rights in the determination of data property rights are bundles of rights composed of the right to hold, use and operate data resources, and the confirmation of rights includes not only the confirmation of the right to hold data resources, but also the confirmation of the right to use data resources and the confirmation of the right to operate, the latter two require that in the process of using the data elements, they should be confirmed according to the application effect (that is, the transformation of the data elements into the sales revenue of final products and services).

The second is market trading. Market transactions in data infrastructure (especially circulation infrastructure and data element farms) are not the same as data transactions. According to the Opinions of the Central Committee of the Communist Party of China and the State Council on Building a Basic Data System to Better Play the Role of Data Elements, the circulation and trading mechanism of data elements includes four forms: openness, sharing, exchange, and trading. Among them, openness refers to the authorized operation of public data, and sharing refers to paid sharing.

The third is the distribution of rights and interests. The distribution of rights and interests can be determined according to the contribution of the right holders of three different rights. For the holder of data resources, the data element can be reused multiple times according to the one-time fixed asset investment of the data element (including the multi-scenario application and multi-subject reuse of the user right holder), and the appropriate fee (such as membership fee and usage fee) can be charged according to the use effect, and the data element can also be directly traded as a product in the process of data transaction. For the right to use data resources and the right to operate them, they are more allocated according to their contribution to value realization (mainly realized into sales revenue), because they take high risks in the process of use and operation, so they should obtain high returns.

Fourth, the interest protection system. The object of the interest protection system should include all right holders. The core interests are the residual right of claim, and the others include the interests involved in the user's privacy rights, data portability rights, and the right to forget, as well as the interests of the enterprise to authorize the operation.

The layout of the technical system should do something and not do something

As far as China is concerned, in the construction of data infrastructure, the layout of the technical system must adhere to the principle of doing something and not doing something, and the overall idea includes the technical layout of follow-up and super-model technology.

In terms of follow-up technology layout, it is mainly undertaken by digital infrastructure, including hardware system and software system. The main strategy is to follow the leader closely, and the main goal is not to be strategically widened by the leader. In the process of following, pay attention to effectiveness and cost. It is necessary to adopt the strategy of "comprehensive follow-up and partial breakthrough": first, in terms of technical layout, not to seek to be comprehensive, the state focuses on strengthening the layout in basic frontier fields and key core technologies; Give better play to the role of the market, so that enterprises can independently choose suitable application technologies. Second, in key areas such as artificial intelligence, while following, while changing lanes. Changing lanes refers to taking a different path of Chinese modernization from the West in the paradigm of artificial intelligence.

In terms of the technical layout of super-models, it is mainly undertaken by data infrastructure. The main strategy is to adopt the strategy of Tianji horse racing in the "technology-economy" competition, and make up for the relative disadvantage of technology with the comparative advantage of economy (super-large-scale, super-large-scale market demand, and the application ability is stronger than the technical ability), and win the overall competition. The main goal is to be far ahead of strategic competitors in the application of technology. Use technology as a means to obtain economic benefits. Specifically, it is necessary to achieve a leading edge in the transformation of technology into emerging industries and future industries.

We believe that it is necessary to strengthen the technological layout that focuses on economic functions. In the following two aspects, we will give full play to the advantages of China's technological layout with the idea of data field.

The first is to put the circulation infrastructure in an important strategic position of the infrastructure. The main idea is to build a layer of application infrastructure under the public infrastructure and above the final application, which is called the data element field in theory and the circulation infrastructure in practice.

The first outstanding feature of this intermediate layer is that it has the dual attributes of public goods and private goods, providing an interface between the two for convection. To this end, we will realize the systematization of API technology and middle platform technology in terms of technology. It is necessary to pay attention to the systematization of intermediate technologies with interface functions as much as we attach importance to chips and operating systems. Give full play to the advantages of data technology to serve the real economy, and closely integrate it with industry applications.

The second outstanding feature is that it has a core function marked by interaction, and this function is systematized. The purpose is to use the cross-network externalities to transform them into superposition, multiplication, and amplification of value. Through the construction of circulation infrastructure, we will strongly support the realization of new business formats, upgrading to service-oriented primary, secondary and tertiary industries, comprehensively empowering industrial digitalization, and promoting the transformation of enterprises to achieve excellence.

The second is to guide the active diffusion of employment-oriented and fair-minded technologies. The idea of the layout is to take advantage of the bias of technology (or biased towards capital, or biased towards technology) to strengthen the social choice of technology. Technological policies are used to regulate the proportional relationship between automation technologies (capital-biased technologies) and diversified technologies (labor-oriented technologies). Using the reuse characteristics of data elements, we should comprehensively promote a technical system that is conducive to supporting the paid sharing of digital means of production (such as digital twin of means of production, API systematization, etc.) in all walks of life. the use of technology to promote the equal use of resources and opportunities, and promote the organic combination of universal services and commercial operations in the socialist market economy; Leverage diverse technologies to create new jobs and create more jobs.

(The author is former director of the Information Research Center of the Chinese Academy of Social Sciences and researcher of the Institute of Quantitative and Technical Economics of the Chinese Academy of Social Sciences)