To re-phrase what I was saying:

- Is that type of consistency wanted, needed, desired? UIs often use inconsistency purposely. A red button among ten black ones may be a very desirable thing. A larger Enter key is inconsistent and yet a very good thing.
All I mean to say is: you can't generalize this way. Inconsistency per se can be good and deliberately designed into a UI of any sort.

- A paradigm may be: you start apps at the bottom, you operate inside them at the top of your screen. I don't mind it at all. You may very well dislike it; both our positions don't need to be generally right or wrong. A paradigm is there in both cases. The interesting question becomes: is there, and if so which one is it, a paradigm which is more effective? To establish that, not always but very often you can only do field tests. Often, you'll be deeply surprised. They way human brain and perception works is often surprising an unexpected. Huge, centrally placed buttons go unnoticed, minor details are perceived as dominantly important, wording is misinterpreted in ways you would not imagine. Not always, but surprisingly often.

- What do you mean exactly by "just does not fit"? Maybe users take more time to start/close apps, or navigate among them, or maybe they get confused or click the wrong UI element more often with one solution rather than with the other? Or, any other measurable advantage or disadvantage? I personally can't think of any one in particular, but there may very well be.
All I am saying is: either you find measurable parameters according to which the various solutions can be rated and evaluated, or a solution will fit just as any other, and yours remains a very respectable subjective opinion, absolutely relevant for your personal UX, but not sufficient to call a UI solution bs.