Discuss
  • On ECMA International

On ECMA International

# Andrea Giammarchi (12 years ago)

I've tried to find something like ECMA constitution or "how things work here" page in what I think is the official ECMA website: www.ecma-international.org

I could not find what I am looking for, which is an answer to this question:

Does TC39 make decisions based on majority or unanimity?

I believe it's essential to understand this so that any time there is some discussion in this ML, if some member of TC39 is not aware/involved, we can eventually hold or point the discussion if the answer is: unanimity

Thanks in advance for the answer.

Best

# Brendan Eich (12 years ago)

Andrea Giammarchi wrote:

I've tried to find something like ECMA constitution or "how things work here" page in what I think is the official ECMA website: www.ecma-international.org

I could not find what I am looking for, which is an answer to this question:

Does TC39 make decisions based on majority or unanimity?

We work by consensus (ISO and Ecma both do, in general). Consensus means general agreement. Unanimity at every instant is not possible or desirable, but by the end of a spec cycle we need everyone to agree.

In human terms, this means some members of the committee may be happier than others, but no one is fatally unhappy about anything, or about the whole thing.

I believe it's essential to understand this so that any time there is some discussion in this ML, if some member of TC39 is not aware/involved, we can eventually hold or point the discussion if the answer is: unanimity

You are mistaken. es-discuss is free to discuss things (without rehashing, trolling, beating dead horses, escalating to tl;dr -- oops, that happened recently :-/) without any concern about how TC39 works, unless TC39 has heard appeals and rendered some kind of "final verdict".

We are not inclined to do a "final verdict" often. Never say never. What you may have herad, which is not the same thing, is "we're not going to revisit for ES6", if not "please stop rehashing."

What is not rehashing, and has been crucially helpful at times: re-opening an issue that was settled by ended up with problems someone found later, or where there was weak consensus and someone thought later of a way likely to get stronger consensus, or variations on those themes.

Thanks in advance for the answer.

I think you are overthinking things. If es-discuss has to worry about Ecma or ISO process too much, everyone loses.

# Andrea Giammarchi (12 years ago)

Thanks for the answer Brendan.

Without names, since apparently nothing final has been decided yet, it happened recently that everyone agreed here but one single TC39 member didn't manage to participate that thread so, while everyone else thought it was a "thumbs up" agreement, this person that missed the discussion, seems to strongly disagree about the decision.

AFAICT this person, with surely reasonable arguments behind the disagreement, is able to move this already overall agreed decision that made everyone slightly happy back.

If TC39 works like this I personally think that whenever something that's aiming to be or not to be into ES specifications is discussed here, every member should be present or agree here or the discussion, and any outcome, becomes kinda pointless since everything will be re-discussed from the scratch in the next TC39 meeting regardless.

This is not necessarily a bad thing but makes the usage of this ML a bit different in terms of potential influence to the future of the language and it's probably fair, at least it is for me, to properly understand how things work there.

Thanks again and best

# Allen Wirfs-Brock (12 years ago)

When it comes time for TC39 to approve the final spec version of ES6 it is likely that each participant will be able to point at something in the spec. that they would prefer was different. But they will also agree that approving that version is the best overall course of action.

# Andrea Giammarchi (12 years ago)

This seems more than reasonable so, for what I understand, is majority as long as there's no concrete harm/problem/worst damage doing "that way" that could make someone really disappointed.

Aka: consensus

# Brendan Eich (12 years ago)

Andrea Giammarchi wrote:

This seems more than reasonable so, for what I understand, is majority

No.

"Majority" means 51% wants it, 49% doesn't (simple majority).

"Consensus" means "general agreement", and if someone demurs on certain things, they are holding fire and not forming a dissenting minority that opposes the whole spec moving forward.

# Andrea Giammarchi (12 years ago)

OK, in your words is much better than in mine, got it, and thanks again.