Coming from a C & Java (hacking, not really professional unfortunately) background and using C++11 for a performance-oriented application (data processing & com

Coming from a C & Java (hacking, not really professional unfortunately) background and using C++11 for a performance-oriented application (data processing & compression), the experience is a bit harrowing. It feels like, compared to C, C++ has very much moved in the direction of "you don't have to worry about pointers (almost)." And the "almost" kills. In C, it was very clear what was data sitting on your stack and what was a pointer to somewhere else. The syntax and semantics were very consistent and straight-forward. C++ seems to try to hide most of this, and it succeeds, almost, except that occasionally you'll mess it up and really blow your foot off. Also, any C++ programmer making fun of Java should go and compare the syntax for HashMap<T> versus unordered_map(). Hash maps are probably one of the most widely used data structures and the C++11 implementation is much clunkier than Java's.