Thanks very much.
Professor Scassa, I normally agree with everything you say. In this case, I agree with almost everything you say. I have one challenge. I'm probably speaking into a bit of a void here, because I don't think we're going to get to a place where these applications are as effective as they maybe could be, but one challenge getting in the way of that is the idea of requiring opt-in systems.
You are absolutely right that there are very different tracing applications. We can look to a decentralized system like the DP-3T standard, say, or look to using Bluetooth, which is more accurate than using GPS data. If we have a data governance framework in place that respects every principle on purpose limitation, that ensures that information will be deleted at the end of this pandemic and that has strong oversight from privacy commissioners or privacy advocates like you, and if it is true—and this may not be true, though there is some research out of Oxford that it is—that an adoption rate of 60% or higher is required for this to really have an impact and be successful, and we could save lives as a result, why is an opt-out system so important? Is it not a balance? Are we drawing the lines even before we get to the important question of what is effective overall and where the balance should be struck? Are we just saying, right from the get-go, that we can't even have this conversation because it should be opt-in?