Structuring teaching and learning corpora for a better research sharing

When analyzing situated interactions in online learning, one should link the data to some contextual information of the learning situation. This is an essential condition if we wish to construct a research object that would become usable by various research teams and disciplines and that would enable replication of experiments and cumulative or contrastive analysis. Interaction data are generally not contextualised, not exhaustive or simply not freely accessible. We define a “Learning & Teaching Corpus” as the description of its five required components: research protocol, learning design, structured interaction data (interactions, productions and log files), licence and analysis. The resulting structure makes this corpus usable by the research community and permits the capitalisation of analyses. This proposition is illustrated by the construction of a data corpus based on an authentic learning situation named “Simuligne”. This work is then discussed according to different viewpoints: ethical and rights issues, standardisation efforts in the distance learning community and the new trends on track analysis tools interoperability