Consider not adding this as a feature.

[wolfbeast]

A note of caution

This proposal contains two advanced features, WeakRefs and FinalizationGroups. Their correct use takes careful thought, and they are best avoided if possible.

If the features are best avoided, why are they being proposed to begin with? I'd say consider not adding these features to any standard and keep ECMAScript from growing even further out of bounds than it already has. If one particular JS engine decides they want it it's up to them, but in all fairness these kinds of considerations that are very implementation-dependent should not be generalized to a standards proposal, at all.

[ljharb]

If TC39 doesn't add them, WASM will (because they're required there), and they'll be effectively part of JS anyways.

[wolfbeast]

You shouldn't enforce the requirements of a binary machine format on JS. If its only purpose is to cater to requirements for machine translated code sourced in other languages, then it should be contained to that format. WASM interpreters can treat it as a custom thing, since its use would be very restricted -- not so in JS where you can't even guarantee that it will function as you expect based on which implementation you're in.

JS already has too many paradigms and implementation quirks to make it "like other languages", no need to feed more into that with this kind of requirement, IMHO.

[ljharb]

That's not the only purpose - there's plenty of JS use cases for WeakRef.

[wolfbeast]

There's always plenty of use cases for anything you can come up with that isn't part of a language. Doesn't mean it's the right thing to do to just add it to a language that doesn't have it.

[ljharb]

At any rate, stage 2 means the committee has committed to adding a solution to the language, so “not adding it” has already been considered and rejected.

[mathiasbynens]

stage 2 means the committee has committed to adding a solution to the language

Point of order: "committed to adding a solution to the language" is not what stage 2 means. Per https://tc39.es/process-document/:

[Stage 2 acceptance signifies that t]he committee expects the feature to be developed and eventually included in the standard

[mathiasbynens]

@wolfbeast Thank you for bringing this up. I think it would be valuable to link to past TC39 meeting notes where WeakRefs have been discussed at length, as part of this proposal's README. I believe some of the background Jordan is hinting at here is captured in those notes, and we should try to make them more easily accessible to help address your feedback.

[ljharb]

Thanks for clarifying; to me that’s what that process document statement means, but indeed it doesn’t say that verbatim.

[allenwb]

The statement in process document was intended to mean exactly what it says. It was not intended to imply an irrevocable or automatic commitment to including the feature. the only real commitment is advancement to stage 4.