implementation. The implications of this are that discrepancies
between pre-implementation design and post-implementation
design are inevitable. In some cases the implementation is an
improvement over the design, but a more alarming situation
occurs when the implementation is worse than the design.
The analysis presented in this case study leads to the proposal of a
practice to be performed by the team members. The goal of the
practice is to trace back the origin of the information needed to
develop a software component and monitor the relationships
between the implemented component and the design, when
available. This practice, which could be part of a routine review
process, may help improve the quality of the product.
It is difficult to measure the real benefit of a practice by repeating
the same project with and without the new practice, particularly in
software engineering. To validate the benefit of this practice, we
consider two avenues for future work with professional software
developers. One approach is to record the anomalies reported
during this practice and evaluate the number of components that
benefits from implementation improvements. Another approach is
to analyze the components undergoing maintenance activities and
determine if the maintenance issues could be reduced or avoided
if the practice was in place when the component was
implemented.