Özet:

With the emergence of social networks, MOOC, informal learning through networks and connectivist approaches to learning, Digital Learning Environment (DLE) analysis is becoming more and more complex. The models which were previously used to account for the activity instrumented with cognitive purposes, are now showing their limits (Vygotsky, Leontiev, Kuutti, Engestrom, Rabardel...).The massive aspect of a MOOC is difficult to represent in such models. Its « open » character is no less difficult to model. The evolution of these environments is chaotic and their effects appear unpredictable. But chaotic does not mean hazardous. A complex system is naturally chaotic and it is not possible to predict the outcome of the process directly, by calculation. But it is possible, from systemic modelling to develop plausible scenarios based on the analysis of available data (training trace, trends, emerging properties, etc.). As Paul Valéry once said, “We are only reasoning on models », whether or not the models correspond to reality. The models are constructed from perceived realities, but enable one to assess emerging properties in a projective fashion; their counter-intuitive effects, for example. We hypothesize in this contribution, that the paradigm of systemic modelling of complexity (Edgar Morin, Le Moigne) appears more than ever as a framework which is suitable for the representation and analysis of a DLE of the last generation. In fact, by applying the theory of complex systems to the modelling of a MOOC taken as a case study and considered as a last generation DLE, we will report, in a projective manner, some of the emerging properties.

Anahtar Kelimeler: