In 1889, the first International Metrology Conference (CGPM-1) provided the earliest definition of meter based on the international meter original instrument. In 1960, Krypton-86 wavelengths were used to define the meter. In 1983, the CGPM-17 adopted a new definition of the basic unit of length (meter): "The meter is the length of the travel of light in vacuum in (299792458)-1 seconds". This is to take the speed of light in vacuum as an accepted convention, that is =299792458m/s. Since the value is specified, length units can be derived from time (frequency) units. The improvement of the definition is a reflection and result of the continuous improvement of measurement accuracy, and it is understandable that the metrology community has a sense of accomplishment.
It has been 38 years since 1983, and the problems of the current definition of meter have gradually emerged. First, experimental studies in the first decade of this century actually falsified the invariable principle of the speed of light, seriously undermining the theoretical basis of the current definition of the meter. Secondly, there are many doubts about the constantcy and stability of the speed of light in vacuum. For example, the definition of "in vacuum" does not specify what the vacuum is, and in 1983 it could only have been an engineering vacuum. Now we know that when we think about the concept of vacuum in quantum physics, c is a fluctuating value, not a constant. It is also confirmed that the Casimir effect plays an important role in the quantum vacuum, which leads to the superluminal phenomenon. If the effect of vacuum polarization is added, it can be concluded that the speed of light in vacuum cannot keep its constant value and stability. Furthermore, it is simply impossible that the speed of light in a vacuum, once specified, will never change. In addition, the unit of length (meter) and the unit of time (second) are both basic units. They are independent and have no influence on each other. However, according to the current definition of meter, it contains the saying of "how many seconds", which makes the definition of meter lose its independence. This cannot be allowed.
This paper also holds that it should not be absolutized and idealized to set up the basic units from the basic physical constants, for there has long been a saying of "inconstant constants" in the physical circle. The improvement of the metre definition could be linked to the proposed "improvement of the second definition". In recent decades, optical frequency measurement technology has developed rapidly. Atomic clocks have developed from hydrogen clocks, cesium clocks and fem to second optical combs to strontium lattice clocks. The uncertainty can reach 10-16(or even lower), and the problem of redefining "second" has been put on the agenda. The second definition can be modified, as can the meter definition.