On the one hand, I agree with all of Donald Clark's criticisms of the recent crop of educational videos, especially those in MOOCs. He's quite right when he says there's too much talking head, too much cognitive dissonance, a dull presentation style and poor editing. He has this research and that to show that elements of video design impact retention. And yet... on the other hand, I have to ask whether improving video quality would produce enough of an improvement to justify the time and expense. Yeah, sure, if I'm consuming video like it was TV, then maybe. But if I'm in the middle of a project and I just need a clip to show me, say, how to fogment my doodad, then all that matters is that I can see what's happening. It's the act of fogmenting the doodad, not watching the video, that leads to retention.