As for de facto standards, I don't think we should be paying much attention to them. I think it's very important to draw the distinction that software development currently has no standards (you might be able to argue for TCP/IP, where alternatives are only ever used because they fulfill some unique use case not covered by TCP/IP), because for as much as we want to call this industry "software engineering", we sure as hell don't treat it like any other field of engineering, for many of the reasons already mentioned.
I think the first step is to sit down and agree on some terminology. You can't even get two programmers to agree on what "Object Oriented Programming" means. No wonder we aren't treated like professional engineers. We don't act like them.
Still though, we use some languages for teaching (like Pascal or Modula-2) which don't have the kind of industrial usage levels of C or C++, so we might see Python being taken up sooner rather than later. I think there are one or two courses already using it over here on a trial basis.
Just because C is hardcore and Lisp is amazing and Haskell is mind-exploding doesn't mean that Python isn't a great language.