Conclusions and Future Work

In this chapter, we presented an outline of a methodology and a case study of incremental change. We observed that domain concepts play an important role in incremental changes.

An interesting question is the optimal sequence of incremental changes that minimizes the changes' impact and any rework needed. Certain concepts depend on others, such as the concept "tax" depending on "item." In that case, the concept "item" must be introduced first, because without it, the concept "tax" is meaningless.

Another interesting question is the relationship between refactoring and change propagation. One of the purposes of refactoring is to minimize the propagation by localizing concepts in one or a few classes. However, it is not clear whether it is always possible to shorten change propagation by refactoring. There may be certain changes that are always delocalized and are "refactoring resistant." Additional case studies are needed to answer this question.

A hope is that the methodology of incremental change will improve the current situation, in which the incremental change is largely a self-taught art. This will enable the accumulation of knowledge in this important field.