The Simplification of Data Processing Systems
K. A. Robinson · International Journal of Mathematical Education in Science and Technology · 1974
Currently, data processing systems are very complex things. Much of this complexity arises because the systems are seen as computer based systems which do jobs for the users rather than user systems, parts of which are implemented on computers. Various techniques are proposed for reducing the complexity of systems such as ‘what’ oriented specification rather than ‘how’ oriented specification, structuring of data and procedure definitions to hide irrelevant parts of the system, and the use of an automatic system/ data dictionary processing the data and procedure definitions to cross reference and invert the structures so that different parts of the system may be brought into prominence or hidden under different views of the system. Conclusions are made that the entire systems analysis process should be oriented towards resolving the system complexity by structuring the system, that the project control of systems analysis should be based on this orientation, and that the system/data dictionary produced by the analysis provides a model of the system suitable for educating people about the system and for design of the computer system.