Discussion with Gilles Dowek : behind the platform FUN with the MOOC Lab

The platform FUN, with its 25 MOOCs, will come to life soon. The MOOCs will begin in January 2014. It is not a small matter to deploy a national infrastructure for these MOOCs and to offer the elements to ensure their progress. The Ministry for Higher education and Research, on the initiative of FUN, decided to work with the Mooc Lab from Inria to set up such a platform. We were glad to meet Gilles Dowek, director of the Mooc Lab. He told us about the rich past of this still very young team, created only last May. Their research projects around MOOCs leave us to discover exciting themes at the crossroads of data processing and pedagogy.


October 2nd, 2013: platform FUN is open. Lauched by  France Digital University, its purpose is to host, coordinate and give visibility to the courses made by professors of universities and French schools. More than 25 MOOCs will be launched at first; they will start at the beginning of January 2014. To deploy such an infrastructure for MOOCs at a national scale, as well as allowing their progress for students and for teachers, is far from being a small matter. Therefore, the Ministry for Higher education and Research called upon institutional actors to help with this approach. Thus, on June 18th, 2013, the  project of the Mooc Lab to install an internal platform became a national scale initiative.

Gilles Dowek is the director of this very young dynamic team of Inria. The Centre Vichow-Villermé works in collaboration with  the Mooc Lab since its beginnings to build its first 4 MOOCs on public health. We were eager to interview Gilles, to know more about the Mooc Lab and its young but rich history but also its enthralling research projects on this new topic of MOOCs…


Could  you present yourself as well as the team of the Mooc Lab?

 I am a researcher at Inria, and I share my time between the Deducteam team and the Mooc Lab. In general, I always shared my time between a research activity, (today in Deducteam), and an activity which benefits the community, such as teaching, science communication, management, etc., and today it is the Mooc Lab.

Created last May, the Mooc Lab is made up of about ten people. Half of the team has a developer’s profile, the other half is centered on the question of pedagogy, but of course there are no clear boundaries. Moreover, its in the interfaces where the most interesting things occur: the force of the team is its diversity. We want to learn together while gradually correcting our errors, because no member of the Mooc Lab is an expert in MOOCs, even if most of us taught and thought about teaching and pedagogy.


Could you explain to us the genesis and the missions of this young team in particular the various actions carried out with FUN (France Digital University)?

The Mooc Lab was created within Inria on May 1st, 2013. Our roadmap consisted at this time of four axes. The first related to the realization of MOOCs. The second was about the installation of a platform and its development. The last axis related to the research activities to be developed around the set of themes of MOOCs. We had chosen at the beginning to work with the Centre Virchow-Villermé for the realization of contents for the MOOCs. Indeed, it appeared interesting to us to learn how to make a MOOC in contact with teachers of disciplines different from our own, data processing and mathematics, and to include varying teaching approaches. For the second axis, we had undertaken the installation of a platform based on OpenEdX, called InMooc at the time.

This first roadmap evolved six weeks later following a decisive appointment with the Ministry for Higher education and Research, where we decided to install FUN platform for the Ministry. The internal project InMooc was thus transformed into a national initiative.


How did the deployment of such a platform happen considering the short time available (June-October 2013)

The site FUN was opened on October 2nd, 2013. The deployment was thus carried out in only a few months. The fist steps required much work. Indeed, such a platform requires powerful servers and a 24/7 availability. It was not possible for each university to locally host its own servers and it was thus necessary to cooperate with a Cloud provider. OpenEdX was set to run on Amazon, but that was not appropriate, because it was imperative to us that students’ data are hosted by a french and public operator. We thus, collectively, decided to call on the CINES (National Data-processing Center of Higher education.) We had to adapt the platform so that it would be able to function on another Cloud provider than Amazon. For the videos, a similar problem arose. OpenEdx was conceived to function with Youtube, but we wanted to host our videos with Dailymotion. Thus, we had to adapt the platform again. Finally, RENATER, which provides Internet access to Universities and research centres, was also involved. A MOOC platform indeed generates a lot of traffic, and it had to be made sure that the network was dimensioned for that.


Why have you chosen to use the open source code of OpenEdX?

We installed and tested several platforms and our choice fell on OpenEdx, originally developed by EdX, i.e. by Harvard and MIT. On the one hand, OpenEdX met our needs, on the other hand, an open source licence was essential, so that we could contribute to he development of the platform after its installation. Lastly, choosing OpenEdx offers interesting possibilities for a cooperation with the important OpenEdx community.

In an open source platform, information is also better controlled, which gives an additional guarantee to keep students’ data secure. These data are, indeed, much more sensitive than data coming from social networks, for example, and reveal a large amount of information about the capacities of a student. These data must be controlled by the students themselves, and nobody else should be able to decide to communicate them to a company or to a headhunter. These questions regarding security and the protection of privacy raise new research topics which MOOCs helped emerge.

Now that the platform FUN is delivered, we will focus on our initial roadmap including the making of contents, the development of the platform openEdX, as well as the research part.


What will this research on MOOCs entail?

The research has many elements which are very different from each other. The first one relates to these questions of security, protection of privacy and authentification during an exam. Another one is the man-machine interaction: how to use the outstanding opportunities which data processing offer to leave frontal pedagogy where a teacher is filmed making a course. That is essential in certain fields, for example in geometry, where it is necessary that the students themselves draw the figures. The third element is the self auto correction and the personalized training where the potentialities of machine learning are immense.

We are organizing a first seminar of research on this last subject on January 13th, 2014. This seminar is the occasion for us to launch cooperations with external researchers. Such cooperations seem essential to us in order to understand how to take advantage of the multiple possibilities that data processing offers and to invent the teaching of the 21st century.


I would like to thank Anneliese Depoux, Gilles Dowek and Ingrid Pfeifer for the attentive proof readings .
Licence Creative Commons
Ce(tte) œuvre est mise à disposition selon les termes de la Licence Creative Commons Attribution – Partage dans les Mêmes Conditions 3.0 non transposé.

Post comment

Your email address will not be published. Required fields are marked *

© Virchow Villermé 2014 -2016 Legals Sitemap