I'm to the point now where I can see the movie in my head (and hear it as well), but obviously a lot of this won't come together until post-production. I'm 99% positive, for instance, that our script will change as we work to integrate interview material or discover that some footage just doesn't work, etc.

I really liked CKRtech's suggestions about balancing narrated and hosted material. I'm not sure if Bill or I had thought much about actually being in this ourselves, but I've kept the possibility open for situations exactly like that D-Pad segment you brought up. It'd certainly be..er, handy, to have that demonstrated rather than just a still shot of a controller. Indeed, I think most situations involving a "still shot" might be better with some type of demonstration footage, provided it's done well and not cheesy or amateurish looking.

I also agree about not having the whole thing be 100% eye candy. If we're trying to get across a difficult topic, cut back on the visuals and maybe just show a talking head, still shot, anything that won't be too distracting from the message. On the other hand, I'd convinced that it's usually better to show rather than tell, so if we can somehow get across the "message" in visuals, that's better than just putting up text/narrating something. Better to keep it simple on screen; if people want to dive deeper into complexity....buy the book! :)

This whole project is fascinating to me because my skills are in writing, using visuals only to supplement the text. Now I'm being asked to do the exact opposite. Challenging, but also exhilarating!