We have developed the algorithm for recommendation sections to be added to enwiki and we're working on the frwiki part now. The quality of sections recommended is high based on the human evaluations, therefore, we want to go ahead and do an experiment where (new or less experienced editors) can receive these recommendations in an experimental setting, to test for the impact of these recommendations on this population.

For this, we need to think carefully about the design of such a system and we may need to work with the community of editathon organizers to make sure we capture their needs with the designs (if the direction of the experiment goes towards in-editathon experiment).