Why not use the domestic operating system? What is the domestic operating system in our mind?

I have experienced 20 years of domestic operating system research and development.

(This article is taken from the article "UNIX, Linux and Domestic Operating System" written by Sun Yufang, Chairman of Zhongke Hongqi Software Technology Co., Ltd. ... In recent 20 years, Professor Sun has actively participated in the introduction of UNIX and the development of domestic operating system COSA, and is also the main planner of the release of Hongqi Linux. As one of the witnesses of the history of domestic operating system, Professor Sun gave a detailed and incisive exposition of the history, lessons and future development of Linux in China with more than 30,000 words. )

The research and introduction of UNIX in China began with 1979. Since the Sixth Five-Year Plan (198 1), the state, Chinese Academy of Sciences and local governments have carried out a series of research and development around UNIX. Since the "Seventh Five-Year Plan", the state has mobilized manpower and material resources to carry out the key project of UNIX/POSIX standard for domestic system software. These R&D efforts have made some technical achievements, which have a little influence in some fields, but their market share is not large. Although it has a strong dominance and a major technological breakthrough in Chinese localization, it has not played a decisive role in the establishment of national basic software.

In view of the fundamental situation that the national information security is threatened, in order to ensure the security of the network and information, the profits of the software industry and even the whole information industry, to gain the initiative of application software development, and to grasp the dominance of the software market, it has become the consensus of the competent departments, researchers and users of China government to develop an independent operating system.

The question now is whether we can conscientiously and objectively sum up the experience, especially the unsuccessful lessons, and explore the development of autonomous operating systems with new ideas. Although talking about Linux and the "free software" and "open source" movements it represents provide us with a good opportunity to develop independent system software, we will still make the same mistakes if there is something wrong with the dominant thinking. It is not easy for me to think carefully and sum up the experience and lessons, even a little painful, but it is a reality that I must face for future success.

Below, the author expounds the lessons from five aspects.

First, there is no guiding principle of establishing market and user demand orientation.

Although the market has been analyzed to some extent in the project proposal and feasibility report, this analysis is subjective and lacks powerful, feasible and concrete measures and means. It is very passive to find users or even "beg" others to use them during the project. There is no mutually beneficial and stable alliance with domestic mainstream PC manufacturers, resulting in a situation of "you attack you and I do our city", and there is no ventilation between them. In fact, the use of users is the biggest driving force to promote the progress of the system. Look at the popular version on the market now, which one is not constantly discovering problems and deficiencies in use, and then constantly improving and enhancing its functions? If users don't understand the system, they can't dispel their doubts about it, let alone use it. The result of the vicious circle is that there is no market and users at all, let alone market share and share.

Second, people's ideas have not completely changed.

The ideas of personnel (especially the general group responsible for project implementation, including authors) have not been completely changed to adapt to the new scientific research management and market operation mode. Almost all the people who participated in the research came from universities and research institutes. Although it is led by ChinaSoft, a subsidiary of the former Ministry of Electronics, the main developers are scientific and technological personnel. To tell the truth, even though this project is mainly supported by ChinaSoft, the management and R&D mode of the Second Development Department is not much different from that of general scientific research units.

Although the members of the overall group are almost all famous experts in this field in China, they are diligent and dedicated to doing a good job; However, due to the older age, it is deeply influenced by the original scientific research and teaching system, and its thinking is not open enough, and its marketization concept is poor; More emphasis on technical achievements and neglect of market development, the final project can be said to be completed from the technical goal, with a certain technical level, but no market share. In the first two years of the Ninth Five-Year Plan, the market problems were noticed, but few leaders of the overall group really "groped" in the market, had no experience, did not know how to effectively open up the market, and did not know how to effectively convert scientific research results into commodities.

Third, the technical route is biased.

One-sided emphasis on independent copyright from the beginning has bound hands, feet and brain. It seems that everything should start with ABC, instead of drawing lessons from others' achievements and experiences to the maximum extent.

For example, at that time, there were many "free" softwares available on GNU of the Free Software Foundation (FSF) in the world. After 1992, Linux became popular on the Internet. Some experts in the overall group also proposed to study this situation, but the overall group did not pay attention to the overall guiding ideology. Especially these things from abroad, I'm afraid they will lose their independent copyright. Although our whole project concentrated on 17 units and more than 200 scientific and technical personnel, faced with huge tasks and pressing time, almost everything was started from scratch, and it was actually too poor to cope, not to mention more coordination and innovation, and finally a system was integrated. But from the perspective of commercialization, firstly, it is fragile and has many defects; Second, thin, the function can not meet the needs of users; Third, the action is slow, and the version upgrade can't keep up with the rapidly changing market demand.

Fourth, the organization and management are weak.

Although the research group on localization of system software is well organized, there are leading groups and expert committees for departmental coordination and technical guidance. The general project team carries out the specific organizational management and technical coordination of the project; There are various special groups "dividing troops to guard" to tackle key technical problems and complete projects; But on the whole, the old management mode is still used, and the organizational system can't quickly respond to the needs of users and markets and the trends of technological development at home and abroad. How does this scattered manpower and material resources compare with some foreign companies' concentrated advantages and planned and step-by-step commodity development work? Although the whole group meets every three months, coordination and response on major technical issues are still insufficient. The research team is scattered all over the country most of the time, and there is little communication with each other, especially in-depth understanding, and the overall group control ability is not strong, so it is impossible to effectively organize this team. Moreover, the technology, experience and knowledge of this team are uneven, and each team has its own "current management" superior leaders, often acting in their own way. In fact, the group as a whole cannot implement strong control measures.

Fifth, financial support is not strong.

In China, especially in the first two decades, the development of system software can only rely mainly on the input and support of the state. An important reason is that our software company has just been born. In the face of fierce competition in the market, software is cheap, and Chinese people lack a correct understanding of the huge cost of software research and development, which leads to serious piracy. These software companies are busy "looking for food to survive" and have no economic strength to independently develop operating systems and related other system software. The development of this basic software requires huge capital and technical investment, and is accompanied by high risks. Although the relevant departments of the state have made great support in this respect, compared with hardware projects such as semiconductor integrated circuits, the state has invested too little in such a major and basic system software.

In recent years, the relevant state departments have paid more attention to the development of software industry, especially 1999. Around the development of high and new technology, especially software industry, many preferential policies and corresponding positive measures have been introduced. This is undoubtedly a great impetus to the development of independent basic software.

The author suggests that for the development of basic system software, especially operating system, we should concentrate our advantages, strengthen investment, make a major breakthrough in a short time, and realize the leap of China's software industry as soon as possible.

Domestic system software has not reached the expected market and has not been recognized by Chinese people. The author lists the above five basic reasons. In my opinion, among many reasons, the most important one is the first one, that is, market orientation and concept. If we first start from the needs of users, formulate and conscientiously implement product development plans according to market rules, coordinate the efforts of all parties, strengthen progress control, strictly manage and take product quality assurance measures, and plan marketing strategy steps and methods, then the domestic operating system is likely to succeed.