I see your point. Considering the article Werner recommended, I understand the idea. And I have nothing against changing shortcuts for something more logical (as long as it is reasonably unfrequent. A major release, such as Prime, is a good time to do so). Not changing things because "that's the way we do / it's been like this forever" is the worst reason ever, and I fight people who think that way in my company on a regular basis (and complain about this attitude on an almost daily basis). But maybe I say that because I never used Mathcad-not-Prime 
.
Anyway, French keyboards have "-" and "^" keys, that don't even need an Alt/AltGr/Ctrl to access, but those shortcuts still don't work. Another thing that makes the software look a bit "amateurish", but even it that's not fixed as such, customieable key bindings would be a great way of solving the problem as well as providing an improvement.
(IMO, the interface issue are the ones that harm the software the most. I my mind, a symbolic math engine is something hard, and a bug there is unpleasant, but more understandable. On the other hand, everybody does graphical interfaces, so I [naively?] think it can't be that hard. And messing that up doesn't give confidence in the rest...)
Back to the "." issue, what you say makes perfect sense, and I think nobody using MS Office would think of putting a shortcut on ".", that's just dumb. There is no harm in restricting customizable shortcuts to Ctrl (or Alt, or AltGr, or Ctrl+Alt) + "something". And for people who miss the illogical shortcuts of M15 that do not fall under this category ... too bad.
