Sure, just very briefly. I would say, just in a slightly broader context, that the world of intelligence collection and national security protections that we live in now is one that absolutely requires intelligence agencies around the world to engage in the collection of datasets. I think it's better to have this presented to the public in legislation as an acknowledged fact of life in the intelligence world. The challenge is how to control the use of datasets, which mostly consist of metadata.
From my perspective, the legislation does not do a bad job in that regard. The biggest concern I have about it is that the legislation would like to draw distinctions between foreign datasets and Canadian datasets, and surround each of them with different legal protections. I understand that in theory. In practice—and I've raised this with CSIS and CSE officials—I'm not sure how that's going to be done because we're talking about a much more blended pool of information.
I would encourage the committee to hear some more precise testimony from officials involved in thinking through the dataset problem, as to whether you can really distinguish between these two things. If you can't, the legal surround that we're trying to provide for it doesn't make a whole lot of sense.