> What I'm suggesting is that wasn't demonstrated.
An in-depth analysis of the respective languages APIs for a particular targeted problem isn't enough?
I won't disagree that readers might make a leap to intuit the author thinks Rust is overall better. But that extra leap doesn't mean he failed to show Rust was better at a particular problem. In fact, that is WHY people would make that un-warranted leap.
> But the point of the article wasn't to compare languages, it was to show the kinds of problems exposed in Go when it is used in massive real-world systems.
I mean, not really. Cross platform file manipulation is, maybe not common, but not obscure. And making a web request reliably is also not something you'd expect to be only needed in massive systems.
> An in-depth analysis of the respective languages APIs for a particular targeted problem isn't enough?
It isn't the same. There is another cheeky quote I can paraphrase: Everyone has a plan until they get punched in the face. He is comparing a Go implementation that has been punched in the face in a real-world use case against a Rust implementation that was sitting on the sidelines.
If the point of the article (and the title) was "Go file system API vs Rust API, an in-depth analysis" I would not have made my comment. The thesis of the article appeared to be "pains I felt in Go when I used it on hard real-world problems". All of his points seem to stand completely fine when you remove the comparisons to Rust. For that reason I would have preferred to remove them.
You mean the Rust language that has dozens of cross platform implementations of coreutils binaries? Go might be larger, but I hardly think Rust qualifies as sitting on the sidelines.
That's a fair opinion. I think the article is richer for having shown what a better API can look like for contrast.
An in-depth analysis of the respective languages APIs for a particular targeted problem isn't enough?
I won't disagree that readers might make a leap to intuit the author thinks Rust is overall better. But that extra leap doesn't mean he failed to show Rust was better at a particular problem. In fact, that is WHY people would make that un-warranted leap.
> But the point of the article wasn't to compare languages, it was to show the kinds of problems exposed in Go when it is used in massive real-world systems.
I mean, not really. Cross platform file manipulation is, maybe not common, but not obscure. And making a web request reliably is also not something you'd expect to be only needed in massive systems.