Again, because of the nature of this program being experimental, we take the same models and we apply them in all sorts of different conditions. We will run the same program in aboriginal communities, rural communities, northern communities, and in urban environments, the whole point of which is to test those models and understand exactly the kinds of parameters you're looking at.
We are very much interested in taking a SNAP or a YAP or a YID project and running them in multiple sites to test the very kinds of things that we need to understand. What does it mean if you don't have that density of population? That tells us things about cost. It tells us things about the kinds of resources we need to see available in communities. It tells us who the best deliverers of those kinds of resources in a community are. Again, in a rural community, you don't necessarily have that concentration of the different kinds of services.
That is very much a part of what we're doing right now in trying to build this database and in building the evidence that tells us what works in Canada. We are running these sites and doing them in different contexts so that we really do understand those variables.
As I said, we've only been building up this database since 2008. We're only starting to have an understanding of what works. I think it is a little bit of what allows for the conversation you're having today; that is, where you go from here in terms of understanding what works in the Canadian context for crime prevention, what we can measure, and what kinds of investments we want to make in Canada.