"This is how open source works"

Exactly. It's a client for an open standard. That's exactly why they're open, so that this can be done.

Now, someone explain that to the people who whine and cite xkcd.com/927 every time someone writes a new client.

The naming is very unfortunate since there is already an OSS software around with the same name: [0] and they clash now. [0][ https://hadoop.apache.org/docs/r2.7.2/hadoop-yarn/hadoop-yar...]

See also: https://news.ycombinator.com/item?id=12684980

This is npm's blog, not the initial announcement.

EDIT: responded to the wrong comment, sorry.

"See also" is just a reference to a related discussion, not the same as the brain-dead calling of "dupe" on every link on a subject.

Sorry, I meant to respond to the other comment saying it was a dupe.

Slight name collision with Hadoop's YARN.

Unless everyone commits to using nonsense names for software, collision happens.

Sure my point was just that Hadoops' YARN(yet another resource negotiater) is fairly well known and well-established at this point. I don't actually care what people name their project but I was expecting to see an article on scheduler's is all.

Well said.

here we go again..

"Dupe"; discussion here: https://news.ycombinator.com/item?id=12684980

This is NPM's post about Yarn, not what I would call a dupe.

Its a dupe in that discussion about yarn has already been established and is taking place on another thread.

There's no point in fragmenting the discussion of yarn.

I would've expected the discussion here to focus on npm, their reaction to competitors, and the main content of this article: client-registry cross-compatibility.

Why not add this (npm's response) to the discussion on the other thread?

Edit: which I just did.

Depending on how the discussion grows such additional information could be buried amongst other conversation, and it might be important enough to be highlighted (by having its own thread here?).