Software engineering [Technology 1998 analysis and forecast]

The paper argues that 1998 is the year to reject concepts previously chosen as the basis for implementing high-visibility projects and to postpone projects so that resources can be reallocated to address Year 2000 problems. Systems to store and manipulate data will be based on complex, multi-tiered client-server solutions instead of the simpler two-tiered architecture. Some work originally planned for C++ will be performed in Java. Short development cycles will require developers to make extensive use of third-party components. Information systems departments will stress outside training and certification for both employee and contractor software engineers. Year 2000 paranoia will grow as corporate management recognizes the magnitude of the problem