Constructive Dynamic Program Slicing Research

Program slicing is a promising technique for providing automated support for various important software engineering activities. There exists hundreds of scientific studies on the theory of slicing. However, slicing is still not widely applied in software industry. Therefore, there is a need to survey and characterize the conducted research and to analyze what problems hinder progress. This paper answers to that need by providing a survey of constructive dynamic program slicing research focusing on publication series with high visibility. The survey identified 12 dynamic slicers related to which 12 aspects were studied. The analysis revealed many issues which can be seen both as concerns and possibilities for improvements. These included the following: 1) Scarcity of high-visibility journal articles. 2) General lack of focus of the conducted research. 3) Limitations regarding the supported programming languages. 4) Mainly unutilized potential synergy with established research regarding various aspects; including visualization and navigation features. 5) Limitations of the conducted empirical studies. Paying attention to these issues is important for being able to make faster scientific progress in the area and to have bigger practical impact on software industry in the future.