OCaml Language Committee: votes and consensus

Committee.md

@@ -43,7 +43,14 @@ After an issue has been tagged for committee attention, the chair chooses a comm
 
 The shepherd's first job is to summarize the question for the committee and make a recommendation of response. Responses need not be "yes" or "no"; sometimes the committee will be asked to choose among several different ideas for syntax, say. The shepherd corresponds with the author until the shepherd understands the issue well enough, and then formulates a summary of this communication and a recommendation. This summary and recommendation are posted to the rest of the committee via the committee's mailing list. The shepherd then sets the label *Under consideration*. This ideally happens within two weeks of the shepherd assignment.
 
-The committee then debates, either via the mailing list or on the GitHub ticket. (The mailing list, though technically public, is a good place for conversations primarily intended to reach other committee members; the GitHub ticket will attract more responses from the wider community.) Hopefully the committee reaches consensus; the shepherd then posts the committee's response back to the author. If the committee is unable to reach consensus, it votes; if there are many options to consider, it may use a ranked voting algorithm, at the discretion of the shepherd. (That is, the shepherd is broadly empowered to choose the most effective approach toward making a decision for the particular issue at hand.) If there is any vote, the shepherd posts the result of the vote, including information about the strength of the win. Because the decision of the committee is non-binding, reflecting the level of committee support in the final answer may be of interest to the broader community in deciding how to proceed. Once the shepherd posts the committee's answer, the issue is no longer under consideration by the committee, and the *Under consideration* label is removed. Ideally this step lasts no more than four weeks.
+The committee then debates, either via the mailing list or on the GitHub ticket. (The mailing list, though technically public, is a good place for conversations primarily intended to reach other committee members; the GitHub ticket will attract more responses from the wider community.) Hopefully the committee reaches consensus; the shepherd then posts the committee's response back to the author.
+
+If the committee is unfortunately unable to reach consensus, it might choose to vote. In particular, if the remaining conflictual points are a matter of personal tastes, the vote is expected to help map the preferences of each committee member. If there are many options to consider, it may use a ranked voting algorithm, at the discretion of the shepherd. (That is, the shepherd is broadly empowered to choose the most effective approach toward making a decision for the particular issue at hand.) If there is any vote, the shepherd posts the result of the vote, including information about the size of the majority or plurality opinion.
+
+If there is no agreement on the technical merit of a proposition, the committee
+might also decide that it cannot take a decision on the proposition.
+
+Because the decision of the committee is non-binding, reflecting the level of committee support in the final answer may be of interest to the broader community in deciding how to proceed. Once the shepherd posts the committee's answer, the issue is no longer under consideration by the committee, and the *Under consideration* label is removed. Ideally this step lasts no more than four weeks.
 
 ## Who is the committee?
 
@@ -53,17 +60,17 @@ You can reach the committee by email at [`[email protected]`](ma
 The current members, including their GitHub handle, when they joined first, when their term last renewed, when their term expires and their role, are:
 
 
-|          | Name     | Handle   | Join Date | Renewal Date | Term End |
-| -------- | -------- | -------- | --------- | ------------ | -------- |
-| <img src="https://github.com/Octachron.png?size=80" /> | Florian Angeletti (**chair**)| [@Octachron](https://github.com/Octachron) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/nojb.png?size=80" /> | Nicol&aacute;s Ojeda B&auml;r | [@nojb](https://github.com/nojb) | 2025/01 | 2025/01 | 2028/01 |
+|          | Name     | Handle   | Join Date | Renewal Date | Term End | Affiliation |
+| -------- | -------- | -------- | --------- | ------------ | -------- |-------------|
+| <img src="https://github.com/Octachron.png?size=80" /> | Florian Angeletti (**chair**)| [@Octachron](https://github.com/Octachron) | 2025/01 | 2025/01 | 2028/01 | Inria Paris
+| <img src="https://github.com/nojb.png?size=80" /> | Nicol&aacute;s Ojeda B&auml;r | [@nojb](https://github.com/nojb) | 2025/01 | 2025/01 | 2028/01 | LexiFi
 | <img src="https://github.com/let-def.png?size=80" /> | Fr&eacute;d&eacute;ric Bour | [@let-def](https://github.com/let-def) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/goldfirere.png?size=80" /> | Richard Eisenberg | [@goldfirere](https://github.com/goldfirere) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/andrewjkennedy.png?size=80" /> | Andrew Kennedy | [@andrewjkennedy](https://github.com/andrewjkennedy) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/fpottier.png?size=80" /> | Fran&ccedil;ois Pottier | [@fpottier](https://github.com/fpottier) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/gasche.png?size=80" /> | Gabriel Scherer | [@gasche](https://github.com/gasche) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/lpw25.png?size=80" /> | Leo White | [@lpw25](https://github.com/lpw25) | 2025/01 | 2025/01 | 2028/01 |
-| <img src="https://github.com/yallop.png?size=80" /> | Jeremy Yallop | [@yallop](https://github.com/yallop) | 2025/01 | 2025/01 | 2028/01 |
+| <img src="https://github.com/goldfirere.png?size=80" /> | Richard Eisenberg | [@goldfirere](https://github.com/goldfirere) | 2025/01 | 2025/01 | 2028/01 | Jane Street
+| <img src="https://github.com/andrewjkennedy.png?size=80" /> | Andrew Kennedy | [@andrewjkennedy](https://github.com/andrewjkennedy) | 2025/01 | 2025/01 | 2028/01 | Meta
+| <img src="https://github.com/fpottier.png?size=80" /> | Fran&ccedil;ois Pottier | [@fpottier](https://github.com/fpottier) | 2025/01 | 2025/01 | 2028/01 | Inria Paris
+| <img src="https://github.com/gasche.png?size=80" /> | Gabriel Scherer | [@gasche](https://github.com/gasche) | 2025/01 | 2025/01 | 2028/01 | Inria Paris
+| <img src="https://github.com/lpw25.png?size=80" /> | Leo White | [@lpw25](https://github.com/lpw25) | 2025/01 | 2025/01 | 2028/01 | Jane Street
+| <img src="https://github.com/yallop.png?size=80" /> | Jeremy Yallop | [@yallop](https://github.com/yallop) | 2025/01 | 2025/01 | 2028/01 | University of Cambridge
 
 <!--
 We would also like to thank our former members:
@@ -137,3 +144,27 @@ Any member of the committee is free to step down at any time; such a member may
 
 There is no process for members of the public at large to directly add or remove committee members. (That is, there is no public vote.) Representative voting across the internet is fraught, and the drawbacks to such a system seem to outweigh any benefits. It is expected that a misbehaving committee (say, one that selects only its friends and ignores other nominations) loses legitimacy and is publicly called into question in an attempt to make changes for the better in its operation.
 
+## Conflicts of interest
+
+The committee hopes to reflect the various interests of the OCaml community from
+academia to the various industrial users. Conflicts of interests might introduce
+(often unconscious) biases that might cloud the technical discussion or reduce
+the diversity of point of views.
+
+Thus we expect committee member to disclose such conflicts. Currently, the
+committee classifies at least the following situations as being unconditional
+sources of conflicts of interest:
+
+1. belonging to the same institution as the proposal submitter
+2. ongoing or recent direct financial ties to the proposal submitter
+
+Any committee member hesitating about a less clear cut case can ask the
+committee secretary for a second opinion.
+
+Committee members in conflict can still participate in the committee discussion,
+but they cannot participate in the decision process. In particular, they
+cannot be a shepherd for the proposal.
+
+Moreover, in order to inform the external audience, the chair should mention the
+known conflicts of interest at the start of the committee deliberation.
+