Is nx an overkill for small organizations? At what point would you abandoned traditional way to do stuff(multiple repos and packages to share code) and implement nx?
I personally would never go back to multiple repos and packages…too much effort to maintain it. Even if we talk just about a handful libs.
So my answer would be no -> nx will fit to any size. The biggest challenge is to coordinate the teams to work properly with the toolings
Totally agree with , almost all of my projects are in Nx, unless there is not an integration yet (Oclif CLI or React Native for instance).
Same here our main project is setup in NX (actually migrated to when it came out), at present we are not using it to it’s full potential, but now have that option if we need it, also find the structure of the project that much nicer.
Got it. Just what I needed to hear.
Just to pile on here I’d argue that as soon as you have at least 1-2 apps depending on 1-2 libs, you should already have implemented some kind of tooling to make your life easier.
I’d say the overhead is so low that it’s definitely worth it :slightly_smiling_face:
I can’t agree more. It keeps your repo clean, focusing on your features, no specific webpack config for the front, no dedicated pm2 config for the server… Would like to see such tools years ago.
Yep, the reality for me is setting up an nx repo is insanely easy, and certainly way easier than dealing with setting up a lerna repo or similar to manage libraries, with a corresponding npm registry (jfrog, npm, github packages) to publish them to. Then deal with the headache of multiple versions of libraries and what apps use what versions. etc.
I think the only thing to look at when deciding to use nx or not is are the frameworks you want to use supported with schematics and builders either from Nrwl or the community. If not theres a decision to be made