Function Points are sufficient to consolidate a software measurement program?

Function points are too empirical a unit to be the only correct way to measure a program. It is also possible to consider the thematic of the application, and the expected knowledge for its implementation (frameworks, languages, mathematical modeling, knowledge of thematic as legislation, circuits and biology) versus the knowledge of the allocated developers, modeled as a calculation of Vector distance. Limitations of the implementation, such as RAM limit and continuous operating time, can also be considered. Remember that function points are abstract and personal measures.