Beschreibung
In his famous paper ”No Silver Bullet“ from 1986, Frederick P. Brooks uses the legend of werewolves to point to the complexity of software projects. Interdependencies between requirements engineering and organisational conditions contribute to this complexity especially in distributed projects. This complexity is the very nature of software development projects — thus, it is futile to run for a silver bullet to lay the werewolves to rest. Instead, the question is how we can deal with the werewolves. The answer in the world of Harry Potter is to accept werewolves and to know the human-werewolf-cycle; similarily, in the world of software development it is necessary to accept and understand the given complexity of software projects – so action can be taken at critical points of time. The model of distributed requirements engineering developed here describes how distributedness of software development projects and requirements engineering practice influence each other. It consists of carefully conceptualised categories that are grounded in empirical data. The categories can be used to gain orientation in the field of distributed requirements engineering, to analyse a given project and to understand how and why particular phenomena occur. The model also provides a new perspective on distribution: The focus moves from physical distribution, which is typically at the center of distributed software development discussions, to a general notion of distributedness.