I must agree with your main point, which is that the typical user just won't have the expertise nor the energy to dive in and modify the code, making the GPL mu

I must agree with your main point, which is that the typical user just won't have the expertise nor the energy to dive in and modify the code, making the GPL much less useful for them. There are two obvious ways to solve the problem: find a way to make developers make more software that is useful to end-users (basically the Apple route). Or, find a solution so that end users themselves can modify their systems. That second route is currently close because of the ridiculous size of current systems. (The volume of a typical GNU/Linux distro is the size of a whole library, and Windows and MacOS are fare marginally better.) The obvious solution there is to reduce the size of our software. http://vpri.org/ http://vpri.org/ is currently attempting to do an OS in 20,000 lines of code (a middle-sized book), and they are doing quite well. Note that doing the "Let's Remove the Fat" route, if successful, will massively reduce the amount of necessary programming work. Plus, more of that work will be done by the end users themselves. That could have dire consequences if we don't prepare for it. (A similar example would be automatic Google cars driving truckers out of business –no pun intended.)