It does not surprise me that the discussion on interactivity following Rod's paper drifted towards reopening the C/I debate, to the exasperation of some. But, I believe this debate is at the core of the application of educational technology and that it is important to have a proper perspective. As a halfway convert towards constructivism, the extent to which belief in its powers has reached almost religious proportions, scares me. There can be no doubt that constructivistic learning methods are powerful, but that is also where the danger lies, sometimes to the point of disqualification. As a physics lecturer I know with certainty that if the class I used to teach would be left to discover the laws of nature on their own, the vast majority would discover nonsense. The reason is that natural phenomena are very often counter-intuitive. There just aren't so many Newtons and Einsteins around who could arrive at the "truth" through counter-intuitive evidence.
Take circular motion as example. I'm prepared to guess that millions of people believe (wrongly) that a body moving on a circular path experiences a force acting radially outwards, the centrifugal force. They have constructed this misconception through driving their motorcars around bends or observing other rotational phenomena. Ask any physicist how difficult it is to convince people that the force acting is directed inwards to the center of the circular path and that the centrifugal force does not exist.
I believe that any constructivistic learning process should be carefully guided to catch any embryonic misconception at its origin and to prevent it from being internalized. This is where the technology at our disposal is so helpful. Clearly this is also where Rod's interaction proves to be indispensable.