Abstract
Schema evolution has recently gained much interest in both research and practice. However, most of the existing works deal with separate aspects of the problem such as evolution of XML schemas or evolution of conceptual schemas.
In addition, all of them view the problem only from the perspective of a single application. In this paper we show that schema evolution has several different levels at which it can be performed and that are highly related. Secondly, we show that schema evolution is not the problem of a single application, but multiple applications having the same problem domain can influence each other as well.
In particular we deal with five levels. We describe the particular levels, how they can be modified and the respective propagation of the modifications to other levels and applications.
We also show which of the situations have already been discussed and solved in the existing works as well as which of them still remain open.