There are two fundamental approaches in the way universities manage intellectual property, and then there is variance within the two. The two approaches are inventor-owned and institution-owned; that basically says who owns the intellectual property. Then, within them there are various gain-sharing rights that have been negotiated at each university.
Typically, in the U.S., by comparison, intellectual property is institutional. That's as a result of a federal act called the Bayh–Dole Act in the U.S. Canada is somewhat unique in having a larger percentage of inventor-owned IP policies. Within the inventor-owned category, there are different degrees of control that the institution can exert.
In some cases—for example, at Waterloo—they say, “Inventor, you do whatever you want. You don't even have to tell us; you just do whatever you want with the intellectual property. We'll help you, if you tell us and you ask for our help, but you don't have to.”
At the University of Alberta, there's a little more control. The IP is owned by the inventors. They can do what they want, but they have to tell the institution that they have it. The institution makes sure that the ownership rights are clear and so on—that they don't, for example, have a graduate student also licensing technology at the same time as his professor is, or that sort of thing.
However, I'll come back to the point I made earlier. If you look at the numbers of the created spinoffs that are sustainable, it doesn't really matter what the intellectual property policy of the university is. I know from my experience at the University of Alberta that the leadership of the university is very supportive of commercialization and speaks about it, highlights it, celebrates it, supports it when it's happening. That makes more of a difference than what the actual words in the IP policy say.