I'm slightly torn on models like this. On the one hand, the cost savings and potential to improve outcomes by scaling the impact of the best classes is obviously wildly important. At the same time, it does seem like there is hard to quantify value in the social aspects of a traditional school. As I write this, however, I realize that there are plenty of private and parochial schools where the class sizes are so small that you could probably replicate the social value pretty easily, since you wouldn't need too many students in the same place to do things like "class trips".