I guess there is a situation that might come about where you want programmatic control of the camera, but at times you want to be able to interact directly for debugging or whatever. In QC you just put the whole renderer chain inside a Trackball Macro. I'm not sure how you would do it in Vuo, would you need two cameras inputing to renderer, one being driven by nodes and the other by direct interactions?
I suppose one solution would be to make a node that takes mouse and key events directly and converts that into transformation messages that can be feed into the draggle camera node, that way you can switch between the two to override one or the other. could get messy though… leave it to the brains trust to figure this out.