Ignoring the MOOCs for a second, isn't this essentially how a seminar has traditionally operated? Everyone reads the assigned papers/books/whatever on their own

Ignoring the MOOCs for a second, isn't this essentially how a seminar has traditionally operated? Everyone reads the assigned papers/books/whatever on their own time, and then shows up in a room for a discussion. If so, these were my favorite classes. The class operated as follows. For each class meeting, the syllabus listed 3-5 "core" papers that were "officially required" reading, along with a smattering of background material. One or two students would give a brief overview of the material for ten or fifteen minutes before they (and the profs) led the discussion. - There needs to be a hook for the discussions. In many cases, the profs baited that hook by choosing papers that contradicted or complemented each other. For example, it's fairly easy to get a self-sustaining discussion going on whether Theory A seems plausible. Papers 1-3 show data that they claim supports it, but the experiments in 4-5 seem to refute it. Hashing this out--and watching others do so--is probably even more useful than whatever the ostensible topic of the class may be. This works really well when there is a legitimate debate. It can sorta kinda work if there was some debate but the question has been settled pretty conclusively. Some students can be asked to play the losing side (e.g., the pro-luminiferous aether folks in a physics class). It does not work well at all for things like methods papers or situations where the debate is so remote or unfathomable (e.g., I would not teach calculus or programming like this) You can sometimes get away with "what did you like or dislike about this paper?", but this often veers into a discussion of methodology (which may or may not be the goal). However, I think a good selection of papers typically led to good discussions, while even the best-prepared list of discussion questions couldn't do much to provoke interesting debates about a random walk through the literature. - Everyone needs to have a pretty solid background. We tried a similar format with 1st and 2nd year undergrads and it was not a smashing success. The conversation kept bottoming out in places where students did not know enough to ask/answer questions or connect it with other things that weren't explicitly in the reading (This is probably more of an issue with science or social science; I imagine it happens less in literature classes). The undergrads were also very reluctant to question published papers. One class attempted to circumvent this with a mixed format. The Thursday classes were lectures. A guest professor would lecture for ~90 minutes on their field of research. These served as background material for 3-5 papers that were the assigned reading. On the following Tuesday, students (+ the professors coordinating the course) discussed the papers. The guest lecturer would return for the discussion and act as a sort of 'expert witness' who could answer questions. - This format works well when the atmosphere is fairly casual and the students are comfortable with each other. This probably imposes an upper limit of 15-20 students/class or so.