[noahl]

As for what goal you're trying to accomplish - I think it would be worthwhile just to see what sort of software people start writing when the expected level of programming knowledge starts increasing. With text editors, for instance, we already know there is a spectrum of possible solutions, some of which are easy to learn but not very productive, and some of which take a long time to learn but are very efficient (Emacs). What happens when millions more people have enough knowledge to use Emacs effectively? Or what would happen to spreadsheets if most people knew enough to make their own one-off calculating programs? I don't think they'd go away, but they might look different. I will be very interested to see where this goes.

But yes, you're certainly right that problem solving is the most important thing to teach. I think your idea of a comprehensive problem-solving approach is interesting - it sounds like you want to make that a theme across several classes. One of them could be programming, but perhaps other sorts of engineering could use it too.