I agree and have made that point before. Java tried to be a language for "enterprise programming", which tried to keep the language so simple that having lots of people wouldn't muck up your code. IMHO, it failed epically at this, and today we would basically laugh at the idea that Java is good at that, but it was in fact a design criterion back in the day. One could make the case that it sort of succeeded at the isolation bit, but at the cost of making everything so verbose that it's absurd and a constant stream of technologies to deal with this has been coming out for Java ever since.
Despite the taint Java left on it, it's actually a good idea! It's a common use case. After years of programming in Perl in a fairly large programmer environment I can see the virtue of the idea of using a simpler language helping keep a lot of programmers in sync in their code base, instead of living in a world where I can look at the source and virtually instantly tell both when and who it came from due to massive differences in dialect. But Java seems to have failed so badly that it soured everyone on the entire idea, and nobody has tried since.
I think Go actually has a better chance at this, not for any large reason but for a series of smaller ones. Defaulting to object composition, once you understand it, is a powerful-yet-simpler approach to putting objects together than inheritance, and makes it much easier to both use the objects of others and to create objects to use for others, because you don't need to worry about what will be F'ed up in the "child classes", as there is no such thing. Implicitly satisfied interfaces don't sound like they'd be a big change over Java, in practice it's night-and-day to be able to declare interfaces that are automatically satisfied by objects you don't control. Contrast how many ways there are to write the same Go function as there is to write it in, say, Scala. And writing some decent multithreading primitives into the core is probably a good idea for this use case, too.
It won't happen tomorrow or next year, but, arguably, rather than Rust or even Python, the language that Go is the biggest threat to in the long term is Java. And it's a threat for precisely the reasons that people on HN tend to complain about. Indeed, I personally have the same objections that most HN'ers do for personal usage, but right now bar none Go is the language I'd most like to work in, precisely because it was written for the Google use case which also happens to be the one I have.