The point -- as is often the case -- is precisely stated at the end of the article: "What this boils down to is that when we approach any of these server dynamic languages to optimize, we can't employ the same tried-and-true techniques we have used for decades (literally) in C, C++ and Java. We need creativity mixed with discipline to make a difference here."

Its not making a point of bad or good. Its discussing the features of the languages that make optimizing them challenging.

In fact, JS doesn't seem to have anything that's specifically suited for async IO at all, apart from clumsy closures.

Also, string handling can often be a bottleneck. Anything dealing with HTTP the protocol, for instance, will have all sorts of crazy hacks to parse that messed-up text format as fast as possible. (And this is a major reason why HTTP/2 isn't text-based.)