> I can't speak for Go's genesis within Google, but outside of Google, this underanalysed political stance dividing programmers into "trustworthy" and "not" underlies many arguments about the language.

This summarises my opinion of Go that has been forming recently, that it's a replacement for Java.

Java was the "safe" choice. Many engineers knew it, bad engineers couldn't be _that_ bad in it and would manage to produce something. They could be handed an architecture spec and implement it word for word without doing much translation to the Java language. But I don't think good engineers could "shine" with it.

It normalised all engineers around some median of productivity, and made projects predictable.

I believe Go is doing the same thing in some ways. It's not supposed to allow good engineers to find great ways of expressing complex architectures (however necessary that may be), it's designed to let engineers of all skill levels hit all problems with a Go shaped hammer and get something that looks inoffensive, boring and predictable.

This is very unsurprising given that it came out of a large engineering organisation. Google have many engineers of a wide range of skill levels, and normalising engineering is more important than doing it better.

I realise Google probably has a higher than average engineering ability, maybe far higher than average, but they still have a wide range, and I also include in this engineers who might be fantastic frontend engineers, who have to get stuck into some system software, for example.