Let me say briefly that part of it is in the selection of the project. You want to make sure that you're selecting projects whose use is fairly certain. This doesn't mean there is not flexibility.
If I use the example of the CSEC facility, it's unlikely that over a 25- to 30-year period the intended purpose of that facility is going to change from its being the headquarters of that government institution. Do you need flexibility? Absolutely you do—they could change approaches. But the model and the procurement contract defines a performance availability; it's defined in that way so that there is flexibility available for the user—in this case, CSEC—to call for changes because of their approach. It doesn't mean they're breaking the contract or renegotiating to do it.
Another example would be a hospital. From my personal experience, I haven't seen many hospitals constructed that, within a 25-year or 30-year period, aren't still a hospital. Are there changes in operating room approaches or technology? The model has to be able to allow you to evolve with these.
Another example might be that if we built a new home for the CBC, with broadcast studios and all kinds of things, the technology might change, but it's such a special-purpose asset that you would only do it if you thought there was going to be a longer life.
So I think it's the selection of the project. The risk would come if you're just doing it for a short-term issue or something that really doesn't take any special type of facility.