I can start by saying that there clearly has been, over the last number of years, an increased emphasis in this area. Certainly, I feel it on my own campus at the University of Ottawa, and we see it through incubators that exist in a number of communities across the country or on individual campuses. I also think that we have recognized over time the need for real investment on the research side, and at times there has been increased focus on where some of those dollars are going, as well as an emphasis on the commercialization piece.
To bring back the cross-border issue that you referenced, I saw that during the last hearing there was some discussion about people often looking.... Bayh-Dole seems to be a model, and we just heard examples in the United States. I hadn't heard of the idea of eight and four, although the idea of not getting paid for four months sends a bit of a shiver down my spine. I also would suggest that it runs counter to the model that many have, at least on university campuses, where much of those eight months is dedicated to teaching and interaction with students, and much of the research takes place over that four-month period. The idea that we should shift away from that model where we have people having the time to do that is a bit concerning.
What we've seen in the United States, especially under Bayh-Dole, is that there are a couple of winners where at times universities have hit the golden ticket: Stanford and Google are always used as a classic example. The reality is that many universities haven't generated significant revenues out of this. We have hits and misses in Canada as well. All of this comes back to the point that I thought was fairly consistent from a number of panellists, that if we are looking solely at commercialization coming out of universities and colleges, we're looking in the wrong place. It's businesses that do this well. This can happen, as you've heard, by transferring this in students; by specific projects; through open access, OA; and through ensuring that we remove some of the IP impediments that I tried to highlight.