Do you want me to focus exclusively on the data? Okay.
On the data area, there are areas for which the department had data available that it did not share back with communities when it did the analysis of the data. That should have been done and needs to be done, and it will be done.
There are areas in the data collection, though, as I've explained, where providing longitudinal data, such as how the individual performs over a period of five years.... As I explained, there are challenges, I understand, on how one captures that data, given a highly mobile population. The example I gave was that if you have an individual in an individual community on a reserve who takes a training course in that community, we can capture the data of how they did in the year and how they performed because it's within the year that they took the course. If they subsequently move to a larger community, the organization that's delivering the training doesn't have access to that individual anymore and can't measure it.
Those data things are challenging to collect longitudinally, and the way we are going to go about that, given that we can't directly access the Revenue Canada database to be able to tell you how that individual did after two years, three years and four years—