Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
C++0x Final Committee Draft Approved (herbsutter.wordpress.com)
27 points by EricBurnett on March 13, 2010 | hide | past | favorite | 10 comments



One thing I didn't realize until now is that the "0x" is not some reference to 0x1234568, but rather, it was supposed to be the last two digits of some year between 2000 and 2009. Oops.

Anyway, I still don't see the point of C++ anymore. Is C++0x just for making legacy C++ codebases slightly easier to maintain by adding a few features over C++ + Boost?


I think the most important change to C++ is the addition of a memory model and standard primitives for threading. In its current state, writing a truly thread-safe, portable program in C++ is nearly impossible.

Edit: For some detail on the subject, see Boehm's paper, "Threads cannot be implemented as a library." http://www.hpl.hp.com/techreports/2004/HPL-2004-209.pdf [PDF]


Lots of us use and love C++ and lots of new things is still being written in C++ — it is not a legacy language.

C++98 introduced us to generic programming and C++0x continues along that path.


Performance. Until someone rewrites the entire Haskell standard library to use unboxed primitives and that stream fusion voodoo then C++ sits alone at the intersection of performance, abstraction and usability.


Shouldn't it be C++xx?


C++0B


This is not backward-compatible. With C++98 that is.


C++0xB?


0xB is hex for 10


I know... his was missing an x is all.




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: