I think the article is quite fair, and captures the situation with Go reasonably well.

But if you step back a moment, the answer offered is, "Go is good because you're not that smart or ambitious (or don't have much time to learn anything new), and you're gonna work as a replaceable cog in a programming machine made from a whole bunch of other people anyway, so, yeah, may as well get to like the lowest-common-denominator-accessible toolchain for that."

It's really not a ringing endorsement once you think about it.

There are all sorts of things it doesn't claim because it can't: Go doesn't provide more powerful abstractions, it isn't more maintainable, it doesn't have a better package system, you can't write correct code in it faster, it doesn't have a better way to handle errors, etc. etc..

Go's superpower is really just that it does a passingly competent job in enough different areas that you've got a good chance at escaping disaster (python: it takes forever disaster; C/C++: footgun disaster; Haskell: I-am-spending-all-my-time-learning-category-theory-now-and-don't-care-about-the-original-problem-anymore-disaster; etc.) and doesn't ask much of you to get to that point.

It's a very weird, yet surprisingly useful, superpower.

It's not one I need personally, though. The time I would have wasted struggling through Go's non-disastrous limitations I already invested in learning Rust and I've not regretted that one bit.