People make decisions that may have one intent and yet are somehow perverted into something else. And sometimes it's because of design. Sometimes it's because of happenstance. But very often, it's mysterious to them.

I mean, if you are directing actors to do one thing and then directing them to do something else entirely because the one thing you wanted them to do may not work, then you are just shattering their confidence in the project.

I think in the '80s, when I started making films, we were all suspicious of these technologies. We were all convinced they would filter out any emotion and sense of intimacy, and the films I made during that period reflected that. In fact, what has happened is the opposite. I think we're saturated with a degree of intimacy we would never have expected, and we're trying to sort through this idea of complete access to each other's lives on an ongoing basis. Our emotions aren't filtered out at all. They're actually accelerated.

I think if you look at the themes that are presented in the film, some are inherently social, and I think that any film which deals with the family is dealing with the smallest social unit in our society - and in a sense it is a question of scope.

Every actor has a different temperament. Part of my job is to know what those boundaries are. The actor has to know you'll be there at the other end, that you're trying to represent them in the best light, who they are as they're harnessing these roles. The methods vary from actor to actor.

The film camera's ability to physically move through space, not zoom through space - every time we have a video camera the movement is through zoom; every time we have a film camera it is a physical movement.

I think the situation in Toronto is such that there are funding organizations which make it easy for a film to raise more money than it needs and very often that works against a film.