**11. Conclusion**

We can think of the meta-structure attached to the system itself as a measurement program. There is a growing awareness that such systems pay off but not without some investment of both time and resources. As the benefits of applying meta-models become more evident, the establishment of such structure will become more essential for evolvable systems to retain their value to a satisfactory level and to remain competitive in the market. With the growing experiences and maturing systems, multi-dimensional data will become available for further research. This, in turn, will make it possible to develop better models. Although it is typically too costly to run carefully controlled experiments on capturing meta-data covering the entire development cycles, it is definitely worth the investment for any long-running systems. Certain meta-level system development practices must be established so that we can fully understand the system, manage it in a more cost-effective way, and readily adapt to a changing environment. Although our proposed model does not cover every aspect, it opens up a wide range of options for the system architects or stakeholders to deduce their own sets of metrics and to utilize these measures to fine-tune systems to extend their lifecycles. Eventually, the systems being instrumented to support software evolution will become more complex with higher capabilities to configure themselves with more feature-rich services. Thus, a growing proportion of resources will continue to be needed for maintenance of these systems.
