Wouldn’t it be great if cMOOcs could be made more ‘productive’ – instead of advancing many people’s knowledge a little by re-creating the same (or similar) new knowledge again and again, can MOOCs be structured to stimulate the creation of new knowledge in a more coordinated way. Can you bring the learners together to produce something entirely novel as they learn? This is in the true spirit of connectivism.
I think this question could be made more precise by introducing Margaret's Boden's distinction between psychological and historical novelty (I am not sure about the exact wording and haven't got the book at hand). The former refers to something I discover which may not be new in a generic sense but is novel to me. Historical novelty refers to something that is novel to everybody (as far as it is humanly possible to establish that, but that's another matter). What Colin seems to refer to here is psychological novelty at the group level. So if someone in a company discovers something that is psychologically novel to her, she might share it with her co-workers whom she knows to be unfamiliar with this. That way, there is knowledge production at the company level. If everybody does their share, you get a multiplier effect, which is the effiency Colin refers to. This is an interesting thought, something which would certainly be in the interest of companies (or other organised groups of people).
First of all, this thought has no bearing at all on xMOOCs, which really aren't about knowledge production but about knowledge consumption. And even if new insights happen to be produced that go beyond that what the teachers intends to transfer, the individual character of such MOOCs guarantees it is the individual and only the individual whose knowledge gets increased. So we definitely are talking cMOOCs here. Then, I guess it depends on the question of how the cMOOC is organised whether Colin's suggestion makes sense.
Suppose, a cMOOC-like setup would have been followed within the confines of some company which sets up a professional development exercise this way. Even if the exercise is open in the sense that non-company participants may join in, it still should be possible to arrange for a mechanism that ensures that novel knowledge gets shared. Content curation is a way to do so, blogs are another. Even in situations of self-organised groupings, such mechanisms could be put in place, as long as someone takes the trouble to do so. I think Stephen Weber's in his The Succes of Open Source has revealed such mechanisms, certainly for the Linux developers community. A big question is whether people will _want_ to share their flashes of novel insights this way. After all, the novel knowledge gives them an advantage over their colleagues, which may come in handy in many ways. But that is a problem for all instances of sharing knowledge among co-workers, not just for cMOOCs.
So, I see no reason why psychological novelty may not also occur at the level of closely interacting individuals. I am not sure, though, that you need interdisciplinarity in particular. It is well known that creativity thrives in the presence of heterogeneity, all kinds of heterogeneity, not just of the disciplinary kind although it does help. So, when contemplating measure to foster psychological group novelty, we should focus on a more general kind of heterogeneity of such groups. (@pbsloep).