The GDMS packages offers a suite of time series analysis tools intended for the processing of data from various geodetic sources. The package was designed and implemented to be as flexible and platform independent as possible given the time constraints under which the group were working. We aimed to provide a package which would complement and enhance currently available analysis tools. This involved providing both time and frequency domain analysis, windowing and interpolation functionality. The loose object oriented design provides the GDMS with a great deal of flexibility and extensibility. The documentation provides both users and developers with an understanding of how the system was constructed facilitating any changes that may be desired. The success of this project can be measured by a number of factors. Perhaps the first and foremost concern in a software application is whether or not it achieves its stated objectives. In the case of the GDMS, we started with a broad requirements description, which were refined as much as possible in the early stages of the project, however, many were undiscovered until much later in the development process, leading to the delay in completion of many of the mathematical modules due to their complexity. In order for the application to be considered useful, it was essential to implement as much functionality as possible yet as the deadline drew near, we were required to assess and implemented the most important functionality for the application. This was largely a role that Professor Peter Morgan played, being the key consultant in project. A number of valuable lessons have been learnt from the GDMS development process, not all of which relate directly to application development itself. That which we consider to be the most critical lesson here, is planning. As previously mentioned, the original requirements were broad and did not cover many of the finer details required. It was during our initial requirements definition process, that we should have sought to discover as much as possible about exactly what was involved with various features in the broad specification. Our mathematical body of knowledge before the project began, was considered at least enough to proceed, yet perhaps not quite enough to foresee these finer specification details. A more comprehensive requirements definition phase in this project would quite likely have reduced the number of problems that arose later, with missing or incorrect functionality. Programming in a Linux environment, for the duration of the project, has also provided us with an opportunity to expand our knowledge base. We have also furthered our experience in C++ object orient design. Furthermore, our knowledge of geodetic data analysis and processing has benefited, not only in terms of the theory and processes involved, but also the background, applications and its significance to geodetic community. From the very beginning, the GDMS was always intended to be an evolving system and it was not our goal to provide a definitive time series analysis package. What we have provided, however, is both a fully functional system that complements and expands upon existing data analysis application, as well as one which can be further built upon. The project has also served to enhance our knowledge of system software design and implementation. In addition, we have gained valuable experience in the area of time series analysis. As we have achieved our stated objectives, we feel that the project has been a success.