SPP 1593: Design for Future - Managed Software Evolution

该优先计划旨在开发软件工程的全新方法，重点关注长期软件系统。现有的软件工程研究并不能解决遗留软件的严重问题，软件对新平台的适应，以及软件系统相对于不断变化的需求、新兴技术以及新软件、硬件和系统组件的集成的不断发展。因此，需要新的基础、方法和工具，以便能够开发“永远年轻的软件”，保持其最初的功能和质量，甚至在整个生命周期内不断改进。我们打算为软件和软件/硬件系统的持续发展开发一种方法，从而使系统能够适应不断变化的需求和环境。在软件工程的学科中，有几个领域特别适合于为这一目标做出贡献。在不要求完整性的情况下，现有的需求工程、体系结构设计和管理、基于模型和模型驱动的软件工程、质量管理、人工制品管理和演化以及再工程的方法已经对管理软件演化产生了强烈的影响。然而，这些方法需要进一步发展和整合，特别注重长期的软件系统。将建立一种新的模式，使软件及其平台的开发、调整和演变与操作、监测和维护不再分离，而是一体化。一个目标是定义元模型，用于保存和访问在系统开发过程中提供和获得的知识。此外，方法和过程模型，以及合适的基础设施，必须提供全面支持软件开发和演化的集成。在本优先方案中，我们不打算探讨与应用无关的方法。我们在这里集中在两个应用领域：信息系统和自动化工程中的生产系统。

This Priority Programme has been established to develop fundamentally new approaches in software engineering with a determined focus on long-living software systems. Existing research in software engineering does not solve the severe current problems with legacy software, the adaption of software to new platforms and the continuous evolution of software systems with respect to continuously changing requirements, new emerging technologies, and integration of new software, hardware and system components. Hence, new foundations, methods and tools are needed to be able to develop "forever young software", which maintains its initial functionality and quality and is even continuously improved during the whole lifetime. We intend to develop a methodology for the continuous evolution of software and software/hardware systems, thus making systems adaptable to changing requirements and environments. In the discipline of software engineering, several fields are especially suited to contribute to this goal. Without claiming completeness, existing approaches for requirements engineering, architecture design and management, model-based and model-driven software engineering, quality management, artefact management and evolution, and reengineering already have a strong impact on managing software evolution. However, these approaches need to be further developed and integrated with a special focus on long-living software systems. A new paradigm will be established where development, adaptation and evolution of software and their platforms, on the one hand, as well as operation, monitoring and maintenance, on the other hand, are no longer separated but integrated. One aim is to define meta-models for preserving and accessing the knowledge provided and gained during the system development process. Furthermore, methods and process models, as well as suitable infrastructures, have to be provided to comprehensively support the integration of software development and evolution. In this Priority Programme we do not intend to explore approaches isolated from application. We focus here on two application domains: information systems and production systems in automation engineering.