I you feel like OOP and template metaprogramming are being publicly derided, then you're welcome to use plain C. It's still going strong after 40 years and will

I you feel like OOP and template metaprogramming are being publicly derided, then you're welcome to use plain C. It's still going strong after 40 years and will probably still be around in another 40. I'm joking a little, maybe, but the point is that languages don't need to change in a hurry. People still start new projects in Fortran and Lisp; there are natural islands in the North Sea newer than these languages. I think it's easy to feel like C++ is due for replacement when it isn't really. Yes, it's easy to be jealous about all the new things going on in other languages but that's in languages that sacrifice performance. If you're using C++ correctly, you're in a performance critical environment (either low memory, fast speed or steady speed). Or you have a lot of integration with C/C++. From this position, the only place you can currently go is from C++ down to C. Honestly, there just isn't an apples-for-apples replacement for C++ at the moment. Rust is the only serious effort out there at minimum/zero overhead programming but it's 6 months away from 1.0, longer still before people get used to large scale design patterns that work well in the language and type system and much longer before it approaches the level of third-party library support that C++ has. Yes, Boost needs a lot more work. C++ has traditionally had a woefully inadequate platform independent standard library, relying too heavily on Win32 or Qt or Carbon for things that should have been platform independent. But frankly, C++ is doing okay. It has never been the easiest of languages to use but it hasn't gotten any worse. It's actually gotten better with C++11 and C++14 and with clang on the scene and libc++ providing a libstdc++ alternative and Microsoft finally trying to reach standards compliance.