Some thoughts about this proposal

[tc39]

Some thoughts about this proposal

[ljharb]

Let's save that discussion for the yet-to-be-presented proposal :-)

[ljharb]

Thanks for posting this!

The third point is just not a constraint we have; things proceed as fast or as slow as makes sense for the proposal. We've had one proposal move from nothing to stage 3 in one meeting.

Regarding the first point: I do agree that if #x was ever a reified value, and #x in o worked, o[#x] would have to work. However, if any other form (like private #x) is used, then I think that expectation no longer holds - which means that any reification proposal that isn't just "bare #x" would not conflict with #x in o.

[hax]

We've had one proposal move from nothing to stage 3 in one meeting.

The meeting notes record that at least one delegate raise the issue of "point of order" for moving too fast :-)

And in the notes:

WH: Stage 3 is appropriate for feedback etc.
AK: Stage 3 is appropriate for that.
DD: Stage 3 is a good time for feedback.

It's very unclear what "feedback" means. As previous arguments, it seems stage 3 only consider "feedback" from implemetors. I doubt the delegates have the same expectation of "feedback" in that meeting.

Actually if we have unused parameter proposal which @devsnek proposed now, we may not need optional catch binding (just use catch(?) {}) :-P

[ljharb]

a) that someone asked the question, and the proposal advanced anyways, illustrates that it is objectively not a constraint that holds; b) that proposal isn't even presented yet, and we will always need optional catch binding, since an unused parameter must by necessity (according to my mental model) precede a used one, and thus would never be a replacement for optional catch.

[hax]

and the proposal advanced anyways, illustrates that it is objectively not a constraint that holds;

This is interesting, which shows how unclear process affect the thing in random way.

since an unused parameter must by necessity (according to my mental model) precede a used one

Off-topic:

I don't think we necessarily disallow function f(a, b, ?) {}, especially it will have different function length according to devsnek/proposal-unused-function-parameters#6

It also could affect destructuring if we extended the syntax everywhere. [a, ?] = iter will take two items.

[ljharb]

The process document is quite clear about what feedback means: stage 3 says

Indicate that further refinement will require feedback from implementations and users
and
The solution is complete and no further work is possible without implementation experience, significant usage and external feedback.

and while feedback is always appreciated and welcomed, and often sought, the process document does not mention any explicitly prior to a proposal entering stage 3.

[hax]

It's clear literally but not practically, especially how user feedback would be considered after stage 3.

Let's save that discussion for the yet-to-be-presented proposal :-)

Yeah. What I try to explain is if optional catch was still stage 2 or 3 now, i believe it should be considered with unused param proposal together. But it's too late now. Of coz, we will always possibly find some features could had a better form if we can restart again, but i hope the process could help us to minimize such things.

[ljharb]

This proposal reached stage 3 at today's TC39.