Myth of Interactivity

In rereading the end of chapter one, I noted what I think is an important point. Manovich reminds us that all art is "interactive," and that heralding electronic media's interactivity accomplishes little. He also makes a rather complex point about why the "interactivity" of hypertext achieves less than the interactivity of, say, a novel.

He suggests that, as part of the modern move to standardize processes, hypertext takes associational memory and makes it physical. "Before we would look at an image and mentally follow our own private associations to other images. Now interactive computer media asks us instead to click on an image in order to go to another image. . . . We are asked to mistake the structure of somebody else's mind for our own"(61).

Does this argument make sense? I think it's a cogent explanation of why the interactivity of hypermedia doesn't really break into new boundaries. I also think it begs the question of what real interactivity would be. Would it be something like GTA? I've argued that one way to do interactivity would be to get a set of basic rules and then "let 'em go!" Thoughts?