MOOCs, crowdsourcing systems, online games, and the maker movement are just a subset of the many genres for mass collaboration that have emerged over the past few years. All of these environments succeed in generating large participation and production. However, they vary considerably in the extent to which they are well-designed learning environments. I enumerate a set of findings from the learning sciences about the characteristics of good learning environments and then apply them to critical analyze current mass collaboration environments. After identifying the apparent strengths and weaknesses of these socio-technical systems, I describe how future mass collaboration environments might better support both mass production and mass learning.