Proposal: move uvwasi to the nodejs org

[cjihrig]

I'm opening this issue to see if there is any interest in moving uvwasi into the nodejs org. uvwasi is the bulk of Node's current WASI implementation.

Disclaimer: I am planning to limit my involvement on that project moving forward regardless of where it is hosted.

[mcollina]

Are you planning to work on it somehow, or are others planning to do so?

(I'm generically +1)

[cjihrig]

I'm not planning to continue working on it myself, but others are more than welcome to do so.

[guybedford]

+1, I think this would be worthwhile to maintain for Node.js going forward.

[cjihrig]

cc'ing @nodejs/tsc and @nodejs/community-committee for proper visibility of the issue.

[mmarchini]

+1

Assuming there are no objections we can transfer it in 72h (72h is the wait time per our policy, but since tsc and commcomm were only cc'd today I suggest starting to count now so folks have time to review the request).

In the meantime, if you want to speed up the process, you can check if the following files exist in the repository (and create those that don't exist):

• CODE_OF_CONDUCT.md: it can be a reference to the Node.js Code of Conduct.
• CONTRIBUTING.md: if there isn't one already, the contributing guide of Node.js core could be a good example. Consider including the Developer's Certificate of Origin section in the document to avoid potential copyright conflicts.
• LICENSE, or other kind of documents that describe the license of the project.
• README.md

[mcollina]

+1

[mhdawson]

+1

[jasnell]

While I'm generally +1, I'm concerned about the lack of maintainers. We either need to find some people willing to move WASI support in Node.js forward or we need to remove it.

@cjihrig ... would you see any reason why the WASI stuff couldn't be implemented as a native addon?

[mhdawson]

@jasnell I agree with the concern over maintainers, but figure that if what we have is a copy having it in the repo would make sense until we figure out what next steps are. Although maybe I need to better understand if uvwasi is the bulk of Node's current WASI implementation means and how transferring the repo helps support what we already have in core (if it does)

[cjihrig]

We either need to find some people willing to move WASI support in Node.js forward or we need to remove it.

I'm also fine with sending a PR to remove it if that is what is preferred.

would you see any reason why the WASI stuff couldn't be implemented as a native addon?

From a technical perspective I think it's possible.

[jasnell]

I'm also fine with sending a PR to remove it if that is what is preferred.

Don't get me wrong, I'm not saying that removing it is preferred. It just might make it easier to find contributors/maintainers if it is.

[gireeshpunathil]

+1 to moving to core
+1 to implement natively, but not a pre-req for moving

[MylesBorins]

Thank you for all the hard work on this up until now @cjihrig

I think there is a lot of value in us shipping this built into core, WASI + WASM is a potential platform agnostic backend for natively compiled modules. That said we absolutely need to have maintainers. Perhaps we could reach out to Bytecode Alliance as well as W3C folks working on WASI to see if we can get any people interested.

While I can't personally devote the time to work on implementation I would be willing to help find folks to pitch in.

[danbev]

+1 I'm interested in helping maintaining/contributing.

[cjihrig]

If no formal objections are raised, I plan to transfer the repo this weekend.