Public Safety committee  But it's in the context of a very specific and narrow understanding of that interplay of the risk. It's not just all information in that context, but also that the sharing of that information will be injurious to national—

Public Safety committee  Certainly the intent is to be very economical with any redaction that might be required. You can look at examples in recent U.K. reports where it's a few words and in some places it's more, because of the nature of some of the reports they've done. In other cases, no element of a report is redacted even though it contains references to what I am sure my U.K. colleagues would view as very sensitive, but there's still a way to provide as much information publicly as possible.

Public Safety committee  I'll let Ian respond on the drafting question, but I'll say two things very quickly. One is that I think on the nature of the information, the threshold would be very high. The expectation is that it's not just that it's inconvenient. The other is a specific suggestion you had in terms of classified: it's probably all classified.

Public Safety committee  I believe it was the Department of Public Safety, but I can confirm that for you.

Public Safety committee  That would be correct. I'll confirm that. If it's incorrect, we will correct the record.

Public Safety committee  If you'll permit me, I would also add that I think the committee in the proposed legislation that's before the committee is in fact in many ways substantially more broad than the committee in the U.K. in terms of scope and the degree to which the mandate of the committee...ability in terms of looking at, as the minister's been very clear, ongoing operations.

Public Safety committee  The presumption is in fact that the requests, because each minister is responsible for each of the agencies, technically would flow that way, but practically, our expectation is that each of the heads of agencies and their staff, as appropriate, responding to whatever demands for information that a committee may have, would appear before the committee on a regular basis.

Public Safety committee  Malcolm Brown, deputy minister of Public Safety Canada.

Public Safety committee  I think there's lots of evidence that demonstrates grassroots, community-based efforts on counter-radicalization are the most effective. One of the challenges is not to design a “we're from Ottawa and we're here to help” program across the country in terms of counter-radicalization, but to build on best practices to share that the vast majority of it is grassroots, led by family members who have gone through the experiences you've just described.

Public Safety committee  Well, Michel may want to add a bit. The minister and Michel have both said that it's never been used. I think the question is difficult for officials to respond to. There are scenarios that could be described, and the green paper—

Public Safety committee  There is a convention that advice to a previous government stays with the previous government, and most governments like it, particularly when they are no longer the government, which happens from time to time. I would encourage my colleague to—

Public Safety committee  —restrain himself.

Public Safety committee  Let me take a stab at this because we have to be careful about “would we have any concerns”. There is the old maxim that silence deems consent. We have to be careful about that. We have to remember that SCISA covers more than the service. It covers 17 departments and agencies, many of which don't have any definition at all.

Public Safety committee  Frankly, on the collection of metadata and the extent to which it's undertaken, I think you really need to turn to our colleagues at CSE. I don't think a huge amount of metadata is collected. I'll let my colleagues correct me, but generally speaking, it is essentially data that is the equivalent of non-personalized information about a device—the length of time, where that device is, and that kind of thing.

Public Safety committee  I'm not trying to pass the buck here, but I think if there are those kinds of questions about the theory around metadata, it would be more helpful if you had the experts from CSE to try to answer your question.