I just came across an interesting article by Daniel Kennett about the problems with Reeder for Mac's UI. He uses Fitts's Law as the basis for describing why the app's interface doesn't work. I agree with his assessment in theory. But in practice, I've been using Reeder for Mac since it came out, and I haven't had any problems or frustrations at all (as it pertains to hitting targets with a mouse).
I haven't studied Fitts's Law extensively, but it seems to me that it may not be nearly as relevant as it once was. Certainly, I'm not the average user. I'm a developer and use the computer all day. But even so, most people in my generation probably use a computer just as much on a daily basis. Sure, they may not know all the ins and outs, but they're more than a little familiar with using the mouse and I don't think they'd have any difficulty using the app either. Fitts's Law was formulated in 1954 before anyone had ever used a mouse based interface, before they'd ever seen a computer. Undoubtedly, it was important for the first users, and still is important for people new to computers, like children and the Amish. That's why those interfaces typically have large, easy to hit buttons (for children I mean, I've never seen an Amish interface). But I think long-term experience has to be taken into account. Here's an interesting line from the wikipedia article:
It describes untrained movements, not movements that are executed after months or years of practice (though some argue that Fitts's law models behaviour that is so low level that extensive training doesn't make much difference).
Seems to me that is the key point: is Fitts's Law just as relevant when the user has extensive experience using a computer, possibly 10-20 years? I'm not really sure, but my gut says no. If I analyzed every app I used for adherence to Fitts's Law, I'm sure they would all fail, but I don't have problems using them. Any web browser would most likely be deemed unusable due to the size of the close buttons on the tabs, but most users seem to do just fine.
Of course, following Fitts's Law can't hurt, and I think it's good to keep in mind as you design an interface. If you follow it religiously with large targets with extra padding, your app would no doubt be easier to use. But I feel that it alone shouldn't be used as justification for why an interface doesn't work. In Reeder's case, I feel the buttons are a good size and well spaced. The app's not perfect (or finished), but none of my problems with it have to do with using the mouse.