ABSTRACT
In the past years, a number of use cases from the area of distributed data mining were collected and their requirements were analyzed. Along with the classical requirements related to scalability and speed-up, the following additional requirements were defined:
1. the domain experts should be able to build distributed applications by using a generic editor i.e. a front-end, which should hide the complexity of the underlying distributed computing infrastructures;
2. sharing, discovery, reuse and modification of existing distributed applications should be made possible;
3. distributed applications should include mechanisms for identity management, trust, and provenance tracking, if they are to be used securely, trusted and if they are to be interpreted unambiguously and reused accurately.
