Stephen Downes

This (10 page PDF) showed up in the Learning Engineering forum today and has the unusual distinction of beinga  post-mortem of a project that is being retired. There's a lot to like about the article and the project. It was called the MOoc Replication Framework (MORF) and was basically a way of providing MOOC data for analysis in a way that respected student privacy through the use of data enclaves: "secure, controlled environments where sensitive or confidential data can be accessed and analyzed by researchers." It wasn't really a success; the only studies to make use of it were those at the universities where the project was housed. It was technically difficult to operate and involved large amounts of data, so researchers found it easier to use more accessible MOOC data. This may have had an impact on the sort of studies they ran; "the cost of learning to work with MORF may simply have been hard to justify for external researchers who could simply investigate different research questions." I appreciate the honesty of this paper and wish we saw more like it.