Quote:
You've missed the point, and to be fair, I phrased it badly. The top guys don't need to be uber-productive. They write the infrastructure that allows the rest of us to productive. This article explains it better.
You indeed phrased it badly; the article in no way says that productivity is not important. Only, writing good code is
more important. If C++ is the only language that can possibly do the job, then of course it is the better choice. You seem to imply, however, that using C++ has merit of its own, forgetting that it's merely a tool like many other. If the same results could be achieved using a more productive language/method of development, then yes, switching to that language/method is critical, no matter how 1337 you think C++ is.
Tim Sweeney at least thinks that productivity is a good thing, and keeps an open mind about the tools he's using.
Quote:
One could argue that the competitors to C++ are not languages which "stay out of the way". They have, for example, no multiple inheritance or operator overloading not through a technical limitation but exactly because they want to limit the power of the user.
That's not really "staying out of the way".
C++ doesn't have higher order functions, garbage collection, a decent standard library, type safety, transactional memory, built-in concurrency primitives, proper tail recursion, working encapsulation or any number of features those other languages have.
Does that mean that C++ is trying to limit the power of the user? Of course not. It just means that you don't write code exactly the same way in C++ as in those other languages. Saying that language X is less powerful because it doesn't have operator overloading is just stupid. Is C++ less powerful than Haskell because I can't use it to work on infinite lists, or less powerful than Lisp because I can't use it to manipulate its own parse tree?