Yes. As mentioned above, this is definitely the direction we're hoping to take the Popup test. The tricky thing is developing the backend grammar map so that it accurately captures what people need to know instead of becoming a list of very arbitrary pieces of random knowledge about Chinese.

Usability-wise, I'm not terribly pleased with the current distinction between the Popup Test and the popup review thing. I suspect that as we move forward (and as the test improves) we'll gradually erode the distinction until there is a test mode that simply works and a more seamless series of study activities which are recommended based on what people study and how they work. Figuring out a good default way or recommending next lessons steps (lesson clustering? vocab similarities, etc.) is probably a more realistic first step.

We definitely believe that the best way to structure things is to impose structure on the study process from outside the lessons rather than within them in the form of an explicit lesson series. So I think we're on the same page here.