Correct, so now imagine Python 3 as the adoption of Core Guidelines, which is basically what modern safe C++ is all about.
Lasted surveys, which are anyway only answered by people that are active enough to care about stuff outside job, places use of tooling to validate modern practices at about 20%.
Now do some statistical extrapolation and it is relatively easy to find out how much "modern C++" is actually used in practice.
Bjarne hasn't dedicated two CppCon to the subject to improve adoption of better C++ practices just by accident.
I agree. I think perhaps the one thing that does more uniquely affect C++ is the compatibility with C (at least not newer versions of C) and the strict backwards compatibility with C++ before it. Which is a little strange to say because compilers _do_ deprecate features, so it's not strict-strict, just strict enough for the standards committee to use it as an excuse.
Lasted surveys, which are anyway only answered by people that are active enough to care about stuff outside job, places use of tooling to validate modern practices at about 20%.
Now do some statistical extrapolation and it is relatively easy to find out how much "modern C++" is actually used in practice.
Bjarne hasn't dedicated two CppCon to the subject to improve adoption of better C++ practices just by accident.