Maël Nison

Totally! Are you interested into making a PR? I think a good start would be to have a `frontend` workspace that would consume a `my-fibonacci` library

This rfc is pretty old, and predates our workspace feature which is a much safer alternative to `yarn link`. I don't think we'll implement `yarn knit`.

> yarn upgrade A --fresh/transitive - would update A and re-resolve all subdependencies ignoring all existing entries in yarn.lock. That might be tricky because the resolution logic is parallel and...

Hey! Thanks for the suggestion 😃 I do have one big concern, however: this proposal makes it impossible to statically know what scripts are available. We would always have to...

>If we're willing to consider making a breaking change, then we should remove the registry from the `resolved` field. Removing it would simplify and shorten the lockfile, eliminate the potential...

> I was hoping we could let the registry in .yarnrc override the one in the lockfile sooner than that, as a non-breaking change. We already have kinda-breaking changes in...

Hmm I'm not entirely sure I follow - we currently have these informations: registry configuration + resolved field (registry + url). Assuming the registries from the resolved fields should always...

I think the devil lies in the "Unresolved section" paragraph. I'm all for nested workspaces, but we need to find a good way to solve this: > Actually, there are...

Yeah, static analysis isn't doable in Javascript (unless we talk about a super strict subset of JS), and would be kinda out of scope for Yarn. @ljharb Btw, has the...

>There’s an open PR exploring it now, i believe. Seems to be https://github.com/nodejs/node/pull/22112. Nice! I wonder if it could be integrated with Yarn so that we could print a "are-you-sure"...