• I really doubt that such a large change can come before C++ lost all its momentum. (And you could even argue that that already has happened)

    I’m currently stuck at work with C++14 as we need certain compiler qualifications (for safety in automotive). Even if that proposal would be in C++26 I could maybe use it in 2035 ($current_year - 2014 + 2026). In 2035 I doubt anyone in that domain will still write C++ as Rust is already making headway.

    • I definitely gave up on C++ when trying to refresh on it you’re me too go learn C++8 before jumping in to 14.

      If learning something requires learning an old version of it before moving on to the current version, something went very wrong.

      • It is perfectly ok to start with C++14, in fact probably preferable to starting with anything before C++11. The idioms changed a lot in C++11. I think changes since then have been minor and incremental by comparison.

          • When C++11 came out, there was an immediate feeling that the language had received a major overhaul and the best ways to do most things had completely changed. Everything from before that was legacy code, though a lot of it was around. I expect it is still mostly like that.

            Or do you mean about Rust? Yes that is new. I still don’t understand the attraction of Rust over Ada that well.

              • I think the C++11 edition (whichever it was) of Stroustrup’s book TC++PL suggested using C++11 immediately. That is what I would have suggested. I used C++ by necessity in a few projects before that, but I didn’t start actually somewhat liking it until C++11. Everything before that was ugly legacy code.