Warnie orr methodology and its disadvantages Essay Example for Free

Warnie orr methodology and its disadvantages Essay Is a bottom up methodology used to model data and procedures where you identify the output and the processes used to generate them and their inter-relations. Its emphasis is on the output. It is analyzed thoroughly. The logic of the program is built around the structure of its data. That structure is obtained by analysis the output and going backwards to find out what the input could have been. The analysis of the output includes: †¢ Output Oriented Comment The output is the core of this system development methodology. We spend all our energy to capture the true picture of the output requirement. This is done even before we think of the physical design. We use rigorous methods to define the data output through a well designed form known as LOS. We capture details of: †¢ Output definition forms †¢ Output frequencies †¢ Data Elements It is assumed the expert can define these requirements with the help of system users who help him fill the mandatory details of these forms. The system users know what they want, but they have little knowledge of computers and computer terms. The software expert knows little about the area to be computerized and can easily misunderstand some concepts. For example computational requirements, the frequency of data use, frequency of data change. Besides when we collect the data from an existing system which is inefficient, the data is inconsistent and unacceptable. The system owners are unhappy with the output generated and that is why they want to change it in the first place †¢ We must obtain the logical output structure first before we think of its physical outlook. The physical problems of the data remain the headache of and hidden by the implementer. Ours is to give the general guidelines on the data and process requirements. This is easy said than done. We cannot claim to be general when we spend all our time worried about our output and its properties. At that level it is difficult to talk about generalities as well. We have to generate dictionaries with specific data structures. Software compatibility with these structures is mandatory. †¢ The approach is data centered, that is the processes are looked at last Take a critical look at the output before any programming activities are undertaken. We break the program into three parts †¢ The middle mapping. also known as the logical output mapping. It deals with how we produce the output and give details of how calculations are done. It assumes the physical mapping will give it the perfect data and in the correct order it is wanted. †¢ Physical output mapping: Formats the data and sends it to the output device. It is the duty of the physical output to know the details of the data format and how to obtain it from the output device. These details should be hidden from the logical mapping. The physical aspect depends on the operating system platform being used. Which tasks are taken by the operating system and which ones are not and that must be a headache of the programmer †¢ Physical input mapping which knows where and how the data is actually stored. If the data gets reorganized or changed in any way the physical input mapping is updated. When changes occur to any part, it is independent of the other parts and carries no ripple effect. Reusable code modules and design elements can be employed. Generic code can be supported to be used by many programs. Having given a summary of the method let us analyze possible problems with this type of method: Obtaining a complete System design before the physical Design This has always remained elusive in most system development methodologies. System design process continues even after implementation because new facts are discovered during actual implementation. Even when prototypes are used, still more facts and faults come into surface. Separating the logical, physical input, physical output mapping Again we need to combine data and processes together, the reason we call them objects, in objects in object oriented systems. How do obtain inheritance, aggregation and software reuse if we separate them. References 1) Ian Sommerville, Software Engineering 2) Aron, J. D. The Program Development Process 3) Booch G. , Rumbaugh J. : The Unified Modeling