Reducing the cost of designing, implementing and modifying intelligent systems by the synergistic use of advanced abstractions

This paper investigates and describes how advanced programming language capabilities can be used to significantly reduce the cost of implementing computer programs for large complex computer programming requirements of the Department of Defense. Representative programming tasks that are discussed include the Navy AEGIS Weapons system and an Air Force planning benchmark used to asses various DARPA supported automatic planning systems. These examples show that programming languages, such as Logistica, which allow a wide variety of synergistic features can provide the most effective way for reducing the complexity of the programming task. It is also suggested that the complexity of the programming process is worse than linear with respect to the size of the resulting program code, the size of the code is inversely proportional to the used design space of the programming language, and the design space provided by a programming language is exponentially related to the number of synergistic features it has