On August 2, 2021, the State Council issued the guiding opinions on improving the evaluation mechanism of scientific and technological achievements (hereinafter referred to as the opinions), which is the latest official policy arrangement on the evaluation of scientific and technological achievements. With regard to the evaluation of scientific and technological achievements, the opinions lists 10 systematic and comprehensive policy guiding suggestions. Moreover, the proposal clearly stipulates that all relevant departments and localities shall complete the revision of relevant rules and regulations of the industry and the region within half a year. Objectively speaking, there is still a long way to go from the guidance and suggestions at the macro level to the specific refinement at the meso level, and then to the implementation and practice at the micro level. After all, changing a scientific research habit formed over the years is by no means an easy task. In view of the urgency of this work, the author would like to give some personal thoughts on the second article of the "suggestions" on "perfecting and perfecting the classified evaluation system of scientific and technological achievements".

According to the international general classification standards, scientific research activities can be divided into three categories according to their attributes: basic research, practical research and experimental development research, and scientific research achievements can be divided into three categories accordingly. On this basis, the proposal clearly points out that the basic research results are mainly peer review; The applied research results are mainly industry users and social evaluation; The achievements of technology development and industrialization are mainly user evaluation, market inspection and third-party evaluation. Objectively speaking, the classification evaluation system has clear levels and clear structure; In theory, the evaluation subject of various achievements is also clear. However, it may not be as easy as expected in the specific implementation process.

Take the peer review on which the basic research results are based. In practice, there are two problems. First, the disciplines are increasingly divided, and the peers are seriously subdivided and diluted. In a highly subdivided field, the peer evaluation is almost transparent. How to ensure the objectivity of the evaluation results? As for encouraging international "small peer" evaluation mentioned in the proposal, the plight of the shortage of domestic small peers can be solved to a certain extent through international capacity expansion, but its operation cost is also extremely high, resulting in problems in its wide application; Secondly, peer review itself is also a historical evaluation tool that needs to be reformed. Peer review can only ensure the compliance and preciseness of the results to the greatest extent. Due to the limitations of any individual cognition, once subversive results are encountered, peer review is often in a state of failure. Many subversive results in the history of science are not recognized by peers at first. Due to the extreme conservatism of the scientific world, peer review has survived to this day.

Another example is the industrial user and social evaluation of applied research results, which is based on the principle of directly linking the production end and the consumption end. It should be said that this evaluation model should be the most objective. After all, users do not want to spend money wrongly. They will certainly give the most severe evaluation on the results and try to lower the value of the results, so as to maximize the benefits of users. As for social evaluation, due to the wide source of the evaluation subject and the uncertainty of time, such evaluation is unable to be controlled by the production end of the results, so it is objective. However, due to the difference of everyone's knowledge background and experience, the evaluation result is not necessarily correct. The first mock exam is the price we must pay to avoid the impact of social factors on the evaluation results.

As for the first mock exam of the achievements of industrial development and industrialization, the three in one mode is proposed in the proposal. The essence is to highlight the role of the market. This is also a standard practice with international standards. As long as the market continues to deepen, the right of evaluation can be handed over to the market. The government only needs to provide rules and regulations for the formulation and implementation. What needs to be added here is the third-party evaluation. This idea has been put forward for many years, and the effect is still not ideal today. The key reason is that China's degree of marketization is seriously insufficient, which makes it difficult to guarantee the independence and core competitiveness of third-party institutions. We can't even think of any famous third-party achievement recognition institutions in China? Therefore, the internal mechanism and principle of the effective role of third-party institutions still need to be further studied. In addition, in terms of the division of responsibilities for the evaluation of scientific and technological achievements, the government should focus on basic research, and the evaluation of the other two types of achievements can be handed over to the market and industry users, so as to untie the government in order to have time to formulate better policies.

The reason why China is facing the dilemma of scientific and technological achievement evaluation today is that all evaluations are concentrated at the end of scientific research activities, all kinds of knowledge in the upstream are constantly mixed and imported, while the downstream evaluation export is single, resulting in the "intestinal blockage" of the whole scientific research activity system. The key to solve this problem is to adopt distributed evaluation and then cooperate with classified evaluation, so as to truly resolve the evaluation pressure at the end. The proposal of distributed scientific research evaluation model is based on the decomposition of the structure of scientific research activities: the front end of scientific research production, the middle process and the end of scientific research output. Combined with the sorting out of the problems existing in the classification and evaluation system of the above three types of scientific and technological achievements, the achievement evaluation should start from the structural diversion, so as to alleviate the great shock and confusion caused by the reform of the evaluation system to the scientific and technological community. Therefore, combined with the author's thinking in previous years, we might as well make some fine adjustments from the following three points, In order to avoid the congenital defects in the three types of evaluation to the greatest extent and make it run faster.

First, the achievement evaluation should be appropriately transferred from the scientific research terminal to the scientific research front end, and the small science and technology loan system should be started. This is a viewpoint put forward by the author in 2008. Needless to say, as the most important supporting means of scientific research activities, the evaluation of fund projects has a great impact on all scientific and technological activities. The essence of its operation is the predetermined mode of the state for future achievements. The advantages of launching micro science and technology loans mainly include the following two aspects: first, the scientific research risk is appropriately shared between the state and researchers; Second, it can reduce the internal volume of project application and support those who really love scientific research. After all, only true lovers will risk this limited responsibility. Its operation is also relatively simple. A basic access threshold is set. Qualified people can apply for small loans, and the lender of last resort can repay the loan with the recognized results. This process can be entrusted to a third-party organization and subject to social supervision. Microfinance is equivalent to diversion from the front end of scientific research. At the same time, it is also a beneficial exploration of the reform of government scientific research funding model.

Second, appropriately expand the redemption mechanism of achievements and leave a channel for recognition and compensation for disruptive innovation achievements. In 2007, the author proposed the "redemption system of scientific and technological achievements" in an article. In recent years, progress has been made in the redemption system of achievements, such as various late subsidies, but the strength is seriously insufficient. Combined with the thinking in recent years, it is found that this model is particularly effective for applied research achievements and industrial development and industrialization achievements. It has long been widely used in the market, such as patents and technology transfer. For example, angel investors and venture capital institutions mentioned in the proposal may evaluate scientific research achievements only for the purpose of investment return, However, after all, we should take real money to do this. Therefore, the evaluation of the quality of scientific and technological achievements is much higher than the current inefficient evaluation. In this sense, as long as we improve the technology market, intellectual property protection and the standardized operation of scientific and technological financial investment institutions, we can greatly improve the evaluation quality of the latter two types of scientific and technological achievements, After all, the market and users need real results. In addition, the redemption system of achievements can also improve the evaluation failure that often occurs in the evaluation of basic research achievements, so as to play a good role in checking loopholes and filling vacancies, which is of great significance to those engaged in basic research, especially those who have achieved subversive results and are not recognized by the world.

Third, use time for quality to improve the quality of three types of evaluation. The so-called exchanging time for quality means that it takes time to show the quality and performance of achievements. If we extend the display time of achievements, those with poor quality will be eliminated by time, while those with high quality will be gradually recognized by people, which can reduce the number of achievements to be evaluated, It also allows some social factors that affect the objectivity of evaluation to be filtered out in the filter of time. This is especially true for the evaluation of basic research results, which is the reason to let bullets fly for a while. After all, quality is the child of time, not the offspring of powerful people. What can be left in time is basically of high quality. The reason for the success of the Nobel Prize is based on this lagging evaluation.