I don't get why people say this. You're acting like the npm team is desperate for contributions and fixes and the Yarn people are like "No! We'll make our OWN tool!".
It's the literal opposite. npm hasn't been open to contributions or fixes for years.
I've contributed multiple issues and PRs which have been completely ignored. It seems from the outside that their focus on monetizing causes their API to be purposefully opaque and subject to change. It's been disappointing and certainly turns me off from ever purchasing private repo hosting from npm itself.
I have a deep understanding of how it works now, and therefore more to contribute, but what's the point? I just keep running my own npm(s) instead and contribute to packages that appreciate it.
A strong statement, to be sure, for which one would hope to see equally strong evidence. But two out of those three issues have unfulfilled requests for followup. I'm not sure quite what you mean them to show.
A lot of people say this and I'll repeat the answer: go look at npm's source code and tell me you'd like to "patch in" a large change such as the one Yarn takes on.
No seriously, go look at their source code. In particular go look at the caching code, one of the big things that Yarn fixes. You think their time would have been better spent hacking on that?
Maybe? It'll be interesting to see how many bugs turn up in Yarn that've already been fixed in npm. Until we can perform that kind of comparison, I'm not sure it is possible to know whether building a new package manager from scratch was more worthwhile than improving the one that everyone already uses.
Exactly my point. I am not totally against reinventing the wheel. Sometimes its fun and sometimes interesting results do come out of it. I am especially looking at the servo project.
But considering the amount of collaboration effort it took to build this (fb said they collaborated with multiple companies in different timezones), wouldn't it be better to spend a little more time understanding npm's codebase and get it patched ? Maybe I'm wrong, but yea just a thought.
> But considering the amount of collaboration effort it took to build this (fb said they collaborated with multiple companies in different timezones), wouldn't it be better to spend a little more time understanding npm's codebase and get it patched ?
Better for you? You know that open source work is free labor, right? Why should they prefer to spend weeks trying to understand undocumented, opaque and "clever" code (much of which has never been refactored) when they can start from scratch with a codebase of their own design?
What I'm asking is, what is the value to them? Or to you (assuming you ever do open source)? Unless you are getting paid for your OS work, you're under no obligation to do things any way other than what works for you.
My favorite thing about the post* that introduced Yarn, itself a rewrite of a basic component, is that the author(s) unironically began with "In the JavaScript community, engineers share hundreds of thousands of pieces of code so we can avoid rewriting basic components, libraries, or frameworks of our own."
It's the literal opposite. npm hasn't been open to contributions or fixes for years.