Dominique Devienne wrote:
Glen Fernandes wrote
I would not be in favor of a Boost distribution that requires anything higher than C++11 for existing libraries. i.e. Not C++17 or even C++14.
But that's basically Boost as it stands, i.e. Boost 1.x. I don't think anyone's proposing to drop that.
That's Boost 1.x with more C++11 and less C++03 - with authors who choose to drop C++03 support, requiring C++11 at a minimum as per: https://pdimov.github.io/articles/phasing_out_cxx03.html This is something which Boost authors are actually acting on. e.g. Boost.Math and Boost.Multiprecision as of a future Boost 1.77 release: https://www.boost.org/users/history/version_1_73_0.html <quote> Math * IMPORTANT: C++03 support is now deprecated and will be removed from March 2021. </quote> <quote> Multiprecision * IMPORTANT: C++03 support is now deprecated and will be removed from March 2021. </quote> i.e. This is promising because people support this particular proposal enough to do something, not just for the (potentially unlimited) joy of discussing it on a list.
But isn't this just a "teething" issue? I see Ville already replied in fact.
I'm following this discussion over this in C++ committee EWG reflector. I'm optimistic that people will do the right thing by users. But I prefer to make decisions based on what I know now, and what has bitten users today, with -std=c++20 and GCC 10.1 that has already shipped, not speculation about what will happen. As I said before, this is just me. C++11 (not higher, not C++17 or C++20) is just what I support and what I'd be willing to put in work towards in existing Boost libraries (and even volunteer to do for those that I do not currently maintain, but would welcome the contributions). Other authors remain free to drop support for everything besides C++17 or C++20 if they want. Glen