the proposal isn't to not implement them at all, it's to not immediately incorporate them in a way that they're permanently frozen. they'll implement them in s

the proposal isn't to not implement them at all, it's to not immediately incorporate them in a way that they're permanently frozen. they'll implement them in some versioned package that everyone can adopt and play with, and then incorporate them. fracturing could happen. but that could happen if they immediately included it in the standard library with warts, too. I can easily imagine people creating multiple packages to overcome shortcomings of a half baked stdlib package. ..so I don't want go to follow some fearful path because of some unquantifiable risk that may or may not come to pass, and could come to pass regardless of which path they take.