Wiktionary:Votes/pl-2022-09/Meaning of consensus for discussions other than formal votes created at Wiktionary:Votes

Meaning of consensus for discussions other than formal votes created at Votes
Voting on: Clarifying the meaning of consensus for discussions other than formal votes created at Votes.

Proposal
In the previous vote "Votes/2019-03/Defining a supermajority for passing votes" it was agreed as follows: "A vote passes if the ratio of supports to the sum of supports and opposes reaches 2/3 or more. A vote where that ratio does not reach 50% should be closed as "failed"; a vote that has at least 50% but less than 2/3 should be closed as "no consensus". Abstentions, votes by ineligible users, and votes cast after closure do not count toward these ratios. This concerns votes proper and does not apply to straw polls, RFD and RFV discussions, and anything not on a vote page." (Emphasis added.)

As uncertainty has arisen over what constitutes consensus for discussions which are not formal votes created at "Votes", it is proposed that this should be clarified by amending "Voting policy" by adding one of the three options specified below after the "Voting eligibility" section.

Option 1 defines consensus as the support of more than half of the editors taking a supporting or opposing stance in the discussion; abstentions do not count in the tally. Option 2 defines it as the support of at least two-thirds of such editors. Options 3 is like Option 2 but more flexible and indeterminate.

Rules for voting

 * This vote has no effect on the results of any discussion closed in the past.
 * However, this vote will apply to any discussion that has not yet closed by the time this vote is concluded, and of course to future discussions.

Schedule:
 * Vote starts: 23:59, 28 September 2022 (UTC)
 * Vote ends: 23:59, 27 October 2022 (UTC)
 * Vote created: — Sgconlaw (talk) 16:16, 21 September 2022 (UTC)

Discussion:
 * Wikt rei-artur3.svg Votes/2019-03/Defining a supermajority for passing votes
 * Wikt rei-artur3.svg Beer parlour/2022/September
 * Wikt rei-artur3.svg Wiktionary talk:Votes/pl-2022-09/Meaning of consensus for discussions other than formal votes created at Wiktionary:Votes

Option 1
The following text is to be inserted into "Voting policy" after the section "Voting eligibility":


 * Meaning of consensus for discussions which are not formal votes

Where the consensus of editors is required for discussions other than formal votes at Votes (for example, in discussion rooms such as Beer parlour and on discussion pages such as Requests for deletion and Requests for verification), the support of more than half of the editors taking a supporting or opposing stance in a discussion on an issue shall be regarded as consensus on that issue. Abstentions shall be disregarded when determining whether this threshold has been met.

It is also agreed as follows:


 * This vote has no effect on the results of any discussion closed in the past.
 * However, this vote will apply to any discussion that has not yet closed by the time this vote is concluded, and to future discussions.

Support

 * 1) . I feel that a clear distinction between the threshold for voting at "Votes" and voting in other contexts is desirable. — Sgconlaw (talk) 12:03, 28 September 2022 (UTC)
 * In the spirit that a proper consensus building exercise is a combination of voting and examination of arguments: this vote has no rationale and points to no place where the rationale can be found. The above provides almost no rationale and leaves all questions open: Why should a threshold for a Beer parlour discussion differ from that of a formal vote? If it should differ, why should it be 50% and not, say, 55% or 60% to achieve a certain stability? Since there is no eligibility rule, why should IP non-contributors or even registered editors with almost no content contribution have their vote counted in a Beer parlour discussion or in a RFD discussion? (The proposal is about mere tallying so their arguments do not matter except as convincing other participants.) And, to some trivial and unimportant, why should the project abuse the terms "consensus" and "consensus process" by redefining them to mean if not their opposite, then at least something sharply different, namely 1) bare tallying, and 2) plain majority? --Dan Polansky (talk) 08:00, 30 September 2022 (UTC)
 * 1)  PseudoSkull (talk) 12:53, 29 September 2022 (UTC)

Oppose

 * 1)  This is an absurd overturn of the 2/3 threshold we have for formal votes. If plain majority passes in Beer parlour, why do anything via votes at all? Just decide in Beer parlour to deprecate all policies and vote in new ones via Beer parour. Why create a formal vote for a proposal when I can get my way via Beer parlour much easier? I know of no environment that lets plain majority decide and call it "consensus". The typical consensus-based decision making in business does not work like that at all. Not even Wikipedia has this absurd definition of "consensus". I wrote more on the talk page of this vote and I have a writeup at  doing an analysis and reporting on what sources say. --Dan Polansky (talk) 13:29, 28 September 2022 (UTC)
 * 2) . Too rigid, and would fundamentally change how RFD works currently, which can lead valid entries to be deleted if they have one more vote than keeps. Also, what about votes that don't involve a large number of editors? There's just too much in the air. AG202 (talk) 15:23, 29 September 2022 (UTC)
 * 3)  My main objection would be that this might lead to BP being used to replace WT:V so that policy changes pass on a lower voter turnout. At least this would bring clarity to RFV and RFD though and I wouldn’t strongly object to this applying for these fora (though I prefer the 2/3rds threshold).--Overlordnat1 (talk) 07:51, 30 September 2022 (UTC)
 * 4)  I don't feel a simple majority equates to a consensus in any meaningful way. -  TheDaveRoss  19:18, 6 October 2022 (UTC)
 * 5)  --Skiulinamo (talk) 09:02, 12 October 2022 (UTC)
 * 6)  Vininn126 (talk) 13:26, 18 October 2022 (UTC)
 * 7)  &mdash;The Editor's Apprentice (talk) 18:06, 23 October 2022 (UTC)

Option 2
The following text is to be inserted into "Voting policy" after the section "Voting eligibility":


 * Meaning of consensus for discussions which are not formal votes

Where the consensus of editors is required for discussions other than formal votes at Votes (for example, in discussion rooms such as Beer parlour and on discussion pages such as Requests for deletion and Requests for verification), the support of at least two-thirds of the editors taking a supporting or opposing stance in a discussion on an issue shall be regarded as consensus on that issue. Abstentions shall be disregarded when determining whether this threshold has been met.

It is also agreed as follows:


 * This vote has no effect on the results of any discussion closed in the past.
 * However, this vote will apply to any discussion that has not yet closed by the time this vote is concluded, and to future discussions.

Support

 * 1) I support the 2/3rds consensus, which seems to be the usual interpretation of the rules. I also prefer the clarity here, as ‘set in stone’ for option 3 is a bit too vague. Overlordnat1 (talk) 07:38, 30 September 2022 (UTC)
 * The problem with that is the proposal defines no quorum, and no eligibility requirements, so IPs and registered users with no or very little mainspace contribution would be able to vote and have their votes counted in the tally, and the closer could do nothing about it. An inclusionist IP could vote bare "keep" in our RFDs and there would be no remedy. I do see benefits in having inflexible vote counting, but then it would need to have proper eligibility requirements like our formal votes have: a registered account with at least 50 content contributions, or even 500 since small useful dictionary edits are so easy to make. On the other hand, one could argue that if option 2 passes, eligibility requirements can be added via a subsequent vote.
 * A deeper problem is that option 2 is unlikely to pass given the opposition in Votes/pl-2022-01/RFD voting policies. Given the vote, option 3 looks most promising. --Dan Polansky (talk) 08:10, 30 September 2022 (UTC)
 * There’s nothing in any of these proposals about who eligible voters are and the wording in the main section of our policy is ambiguous wrt votes on RFV and RFD, though it does impose requirements for WT:V. If I need to change my abstain vote to a support one for tactical reasons for section 3 then I will happily do so but that point hasn’t been reached yet. --Overlordnat1 (talk) 11:22, 30 September 2022 (UTC)
 * Fair enough. One may interpret options 1 and 2 as being against indeterminate eligibility criteria rather than empty eligibility criteria. And indicating one's preference by supporting something and switching later to 2nd best option makes perfect sense, or else you support both options equally, which is not the case.
 * What do you think of strength of arguments in RFD? What about bare keeps? --Dan Polansky (talk) 11:28, 30 September 2022 (UTC)

Oppose

 * 1) . As I support option 1, I do not support options 2 and 3 which are incompatible with it. — Sgconlaw (talk) 12:03, 28 September 2022 (UTC)
 * Oppose . This is quite tolerable definition of "consensus", but I believe certain flexibility of closure is preferable. I think matters that are mere matters of preference and not accuracy or fact could be decided against 60% threshold. And the need of flexibility was the opinion of the opposers in Votes/pl-2022-01/RFD voting policies; have people changed their minds? --Dan Polansky (talk) 13:02, 28 September 2022 (UTC)

Abstain

 * 1)  This is 2nd best option. My interpretation is that there are unspecified eligibility criteria rather than empty eligibility criteria: no sane closer would accept any votes from any accounts no matter their contribution to our content. I still find flexibility preferable, at least to allow 60% threshold for matters of subjective preference, but better pass option 2 than nothing. --Dan Polansky (talk) 11:50, 30 September 2022 (UTC)

Option 3
The following text is to be inserted into "Voting policy" after the section "Voting eligibility":


 * Meaning of consensus for discussions which are not formal votes

Where the consensus of editors is required for discussions other than formal votes at Votes (for example, in discussion rooms such as Beer parlour and on discussion pages such as Requests for deletion and Requests for verification), the support of at least two-thirds of the editors taking a supporting or opposing stance in a discussion on an issue is a hint for the threshold for consensus, but it is not set in stone. As a result, the consensus determination is somewhat indeterminate and can take into account considerations other than pure tallying. Tallying does play a role.

Support

 * 1) . 2/3 is a fine overridable default threshold. It is the threshold officially approved for formal votes, and the default threshold for Beer parlour discussions cannot vastly differ from that one, or people will avoid formal votes, and then, why have them in the first place. Opposing voters in Votes/pl-2022-01/RFD voting policies required flexibility, and here it is delivered. Flexibility has downsides, but there was no consensus for inflexible threshold in the RFD vote. Options 1 and 2 do not even introduce any eligibility criteria and any quorum for a Beer parlour discussion, so an attacking swarm of anon IPs could land into Beer parlour and turn Wiktionary upside down policy-wise like a breeze; of course, that would not be recognized, but that this the logical consequence of proposals 1 and 2 on the table as written. Those who want the RFD closers to take strength of argument into account (which I think is very difficult to do fairly, but let us see) will appreciate the flexibility allowed in this option. --Dan Polansky (talk) 13:02, 28 September 2022 (UTC)
 * 2)  per Dan Polansky. Whoop whoop pull up Bitching Betty ⚧️ Averted crashes 16:42, 28 September 2022 (UTC)
 * 3)  per Dan Polansky. AG202 (talk) 15:21, 29 September 2022 (UTC)
 * 4)  I think this reflects how we currently operate, and I have no qualms about codifying that. Generally 2/3s is a good benchmark, unless there are compelling reasons to decide some other way. -  TheDaveRoss  19:17, 6 October 2022 (UTC)
 * 5)  --Skiulinamo (talk) 09:02, 12 October 2022 (UTC)
 * 6)  Vininn126 (talk) 13:26, 18 October 2022 (UTC)
 * 7)  - two thirds feels about right for consensus. John Cross (talk) 07:49, 23 October 2022 (UTC)
 * 8)  &mdash;The Editor's Apprentice (talk) 18:10, 23 October 2022 (UTC)

Oppose

 * 1) . As I support option 1, I do not support options 2 and 3 which are incompatible with it. I also feel that option 3 may lead to too much uncertainty. — Sgconlaw (talk) 12:03, 28 September 2022 (UTC)

Abstain

 * 1) This is an improvement on the current situation, as it brings some clarity, but the addendum to option 2 stating that these requirements are ‘not set in stone’ is a bit vague and serves to immediately reduce that clarity. --Overlordnat1 (talk) 07:41, 30 September 2022 (UTC)

Decision
Option 1: failed 2-7-0 Option 2: no consensus 1-1-1 Option 3: passed 8-1-1 --Overlordnat1 (talk) 12:28, 28 October 2022 (UTC)


 * thanks. I have updated "Voting policy". — Sgconlaw (talk) 15:10, 28 October 2022 (UTC)