Why should I have written ZeroMQ in C, not C++ (part II)

Re:Why should I have written ZeroMQ in C, not C++ (part II) by Xcelerate on Thursday August 30, 2012 @05:11PM (#4456842)

C++ is a ridiculously complicated language. And I mean that in a bad way. I don't normally use C++ as my primary language, but there are certain cases where I need the speed or low-level access. And I feel like every darn time I do, I need to relearn all of the language's weirdness and idioms. From order of operator precedence, to best practices with classes, to what a "char" is on which platforms, to the various 101 meanings of "static" and "const".

It's an unnecessary time drain and I can't but help wonder how many man-hours a year are wasted by people learning all the intricacies of C++. (Heck, just check out this guy's FAQ on all the things you probably never knew about C++ http://www.parashift.com/c++-faq-lite/). I could almost memorize a list of x86 opcodes faster.

Complexity != powerful language. Clojure makes this point very well.

[ Parent | on HN ]

Re:Why should I have written ZeroMQ in C, not C++ (part II) by ansible on Thursday August 30, 2012 @07:33PM (#4457424)

C++ is a ridiculously complicated language. And I mean that in a bad way.

What really gets my goat is the minimal differences between 'struct' and 'class'. Why do we need both again? If you're going to allow structs to have member functions and such, then you should never have had a 'class' keyword to begin with. Ugh.

[ Parent | on HN ]