I agree with the statement that C++ can be used to simplify. However, it can also be used to make it more complex. As an example, in the scientific community p

I agree with the statement that C++ can be used to simplify. However, it can also be used to make it more complex. As an example, in the scientific community people rewrite Fortran and C codes to C++. This is good in the first place because one can get maintainable, readable and efficient code with high level templated libraries. However, once you write something wrong, the compiler error output kills you (yes, one should blame the compiler, not the language). As an effect, the next generation can no more maintain the code and either goes back to good-old-Fortran (or replace with "good-old-C" in the context of the Linux kernel). Using fancy C++ idioms (sorry for when I call traits "fancy", but for physicists it is) makes it sometimes (!) worse :-(