> I'd guess the opposite. Might depend on how trendy the new feature is. Because I still see newish Java programmers who apparently don't know that java.date.u

> I'd guess the opposite. Might depend on how trendy the new feature is. Because I still see newish Java programmers who apparently don't know that java.date.util exists, and who also don't know that rolling your own date math is only done by date manipulation library authors and connoisseurs of slow moving train wrecks. A second, related but harder to pin down, issue exists here. I see relatively inexperienced developers who, for instance, get a little confidence with Javascript chaining and callbacks, and work themselves into corners where a given filter method (like the 'Where()' example) doesn't quite do what they need, and suddenly they have no idea what to do. So they hack in something awful in the middle, forget that it is a weird homegrown something instead of a library function, and see profoundly strange problems later. Both of those are things programmers grow out of, hopefully. But they seem to be instances of having APIs with large surface areas causing what amounts to research failures.