A home for reference-implementation polyfills

[bkardell]

During last years' TPAC we had a breakout session where a few ideas came about. Now that the standards group exists and things are more setup, it seems like a good place/time to talk about one of them.

Effectively, when we're working out new standards proposals (often in WICG), groups working on that produce a polyfill reference implementation in order to a) help them think through the challenges b) provide better economics for developers to participate more directly by having a thing to try to use, evaluate and comment on. However, this is quite an inconsistent patchwork of "how do you find one" and "how do we know it isn't bit rotting" or "how do we know that repo will continue to exist" or "is someone maintaining it with the spec if the spec advances?" and "how do we know it has good licensing" and lots of other questions.

There were several participants interested in making a smoother path here by making it easy for these things to enter as some kind of openjsf project (maybe) and for the W3C to work on integrating some some kind of optional, but consistently findable 'reference implementation polyfill' link in drafts... or something.

It seemed that everyone agreed there was "something interesting and worth further discussion" here. This issue is in pursuit of said further discussion.

[eemeli]

So is the request here to be receptive towards the acceptance into the foundation of a polyfill project, or to take some more concrete actions towards helping to make such a project happen?

Fundamentally, as with many things, I suspect that this primarily needs someone or someones with sufficient bandwidth to start maintaining it.

[michaelchampion]

I like this idea but it needs some analysis:

1. Who exactly would benefit if OpenJSF were to provide some sort of stewardship of reference implementation polyfills? Is it something JS developers would use, something various JS Foundation projects would use, or something mainly the WICG/ W3C / WHATWG spec community would use....?

2. What would a successful outcome look like? (Narrative is fine, I'm not asking for key performance metrics or anything at this stage). What's a story line where some problem real people suffer from gets mitigated because OpenJS got people together to solve it?

3. Are there examples where this kind of thing has been tried in similar domains? How would this repository differentiate itself from existing resources one finds from queries such as https://www.google.com/search?q=polyfill+repository ?

4. And as A home for reference-implementation polyfills #106 (comment) suggests, who is willing to put energy into creating a home for reference implementation polyfills? I'm an OpenJSF newbie, but I get the impression that projects usually have a community and an active GitHub repo before they are formally incubated.