Towards an AI-Specification of Intelligent Distributed Learning Environments

This paper reflects on the possibilities to specify IDLEs (Intelligent Distributed Learning Environments). These are distributed multi-party eLearning environments with AI-components (generative expert systems, learner models, etc). IDLEs offer intelligent, learner adapted, anytime services to learners (eg. e-diagnosis of solution proposals, e-consulting in the case of errors). Specifications seem to become necessary to assure consistency among content providers and developers with respect to content, methods, knowledge and response spaces of learners. In this paper we want to demonstrate how to specify AI-components of IDLEs on a logical meta-level and the DLE-part of the system with UML/XML. It will be argued that the use of the new emerging eLearning specification language like the Educational Markup Language (EML) is not advisable when specifying AI-components. A more promising alternative could be software patterns or specifications in logic (eg. the situational calculus).