JAKS defines software reverse engineering as the process in which the existing legacy system is re-implemented to make it more understandable and maintainable. Reverse engineering a subject system involves three distinct phases i.e. the identification of current components and their dependencies, the extraction of design information and system abstractions, and redevelopment of system. As part of this re-engineering process, the system may be re-documented and /or re-structured. The system may be translated to more modern programming language, implemented on distributed platform rather than a mainframe or the system data may be migrated to a different database management system. Inputs to the reengineering process may include source code files, database files, screen-generation files, and similar system-related files. When the process is complete, it generates all system documentation, including specification and design, and new source code. Since fully automatic reengineering is unlikely in the near future, the process must involve a combination of transformation and human interaction.

The first step of JAKS approach assures an understanding of the customer environment. This includes determining the system attributes such as the computer languages involved, the number of lines of code, the presence and accuracy of documentation, the experience level of personnel, and other system sizing information such as number of user interfaces, data files, data elements, and programs. We have found it is also important to know what existing tools are being used by the organization, the current software maintenance processes, and future plans for the system, any development process, time should be set aside to revise the methodology based on actual experience in executing it the first time or two.