If software engineering wants to become a real industry rather than an art form, then software engineers have the responsibility to help users use a complete, detailed and effective method to define user requirements. A professional software engineer should insist on using effective requirements collection methods, such as Joint Application Design (JAD), Quality Function Deployment (QFD) and requirements review, and consider them as his own responsibilities. Software engineers are also responsible for reminding users of potentially harmful requirements.
A very common phenomenon is that software requirements literature is usually negative and makes an incorrect assumption that users can identify requirements 100% effectively. This is a very dangerous assumption. The requirements identified by users are never complete and are often wrong. If Website production If a software project wants to succeed, the requirements must be collected and analyzed in a very professional way, and software engineering is a profession that must know how to do these things well. Software engineers have the responsibility to insist on using appropriate requirements analysis methods. These methods include data mining of legacy applications, joint application design (JAD), quality and function deployment (QFD), necessity analysis and requirements review. It is helpful for requirements analysis, and it is also recommended to let users participate in the development team (for agile development). Use case is also a recommended method.
Users of software products are not software engineers, so we cannot expect them to know how to express and analyze requirements in the best way. Therefore, we need to ensure that the software requirements collection and analysis results done with professional standards are handed over to the software development team. In 2009, almost half of the software applications were transformed from legacy applications, some of which have been used for more than 25 years. Unfortunately, few software specifications or requirements documents are available for these legacy applications.
Due to the lack of data to describe the characteristics and functions of legacy applications, a new requirement analysis method has emerged. This new method extracts business rules and logic from data mining of legacy applications. Because of this method, data mining can also be used to evaluate the size of software function points and code statements.
