User:Born2cycle/Yogurt Principle

The eight year saga of discussions and repeated proposals on the titling of Yoghurt, ultimately resolved simply by restoring the original title, Yogurt, inspired the writing of this essay to remind WP:RM discussion closers of the following:

1. The closer's duty is to determine community consensus regarding the title of the article in question, not just the WP:LOCALCONSENSUS of the participants.
2. Though there may be no WP:LOCALCONSENSUS among those that happen to be participating in the discussion, that doesn't mean there is no community consensus regarding the title.
3. Community consensus about the title in question is not determined by counting !votes, but by an analysis of the arguments presented in the discussion, especially by evaluating the arguments based on how well they are based in policy, guidelines and conventions (because those reflect community consensus).

This is especially important to consider in situations where there has been a history of "no consensus" closes, as was the case with Yogurt. There, the title, Yogurt vs. Yoghurt, was disputed for seven years involving no less than eight RM discussions, most closed as no consensus. Arguably, in all of these discussions, the support (move to Yogurt) side was supported by stronger policy-based arguments, particularly arguments based on what is now known as WP:RETAIN, and WP:COMMONNAME, but closing admins seemed reluctant to find consensus because of large numbers of weak opposing !votes. With the notable exceptions of RM #2 (which was reversed by RM #3 by dubious reasoning), and RM #8 (which finally ended the saga), they closed the discussions as "no consensus". See #Yogurt title history.

In all eight of the Yogurt vs. Yoghurt RM discussions, the main support arguments were, essentially:

1. the original established title of the article was Yogurt, and it had been moved, without sufficiently broad input, contrary to WP:ENGVAR and WP:RETAIN, to Yoghurt, and
2. the spelling without the h complied better with WP:COMMONNAME because it was more commonly used overall, and because it was commonly used even in those English speaking regions where the h variant was more common.

Further, it was argued, repeatedly , that if the article was moved to Yogurt, there would be no equally strong policy-based argument to move it back to Yoghurt. Indeed, since the move was finally made at the end of 2011, the eight-year-long dispute about the title of that article was finally resolved.

Once the article was moved from Yoghurt to Yogurt all discussion about the title simply ended, because, as had been predicted for years, once moved there was nothing to discuss: no serious grounds on which to propose a move. Had the closing admins found in favor of the move because of the substantially stronger policy-based arguments, and lack of strong counter arguments that would support a reverse move, that dispute would likely have been resolved much sooner. Years sooner.

Contentious title conflicts could and should be resolved sooner if RM discussion closers would take their duty to determine community consensus seriously, and to not be dissuaded from doing so just because there is no consensus among discussion participants.

Participant consensus vs. community consensus

Critical to understanding the Yogurt Principle is understanding and appreciating the difference between a WP:LOCALCONSENSUS of those participating in a discussion, and what community consensus is regarding the issue being discussed.

Any given discussion on Wikipedia, even the relatively large RfCs involving dozens, is comprised of a small self-selected (not random) sample of the thousands of WP editors that make up the community, and is by no means guaranteed to be a representative sample of the much larger community. This is why determining consensus among the participants of a discussion is not an effective way to determine community consensus about the question at issue. For example, if half of, say, twenty participants favor a proposal and half oppose, that result tells us there is no consensus among the participants, but nothing about whether there is community consensus about the proposal, and, if so, what it is.

This is why discussions closers on Wikipedia have the duty to determine consensus by evaluating the arguments in terms of how well they are based in policy (which reflects community consensus), rather than by determining what, if anything, the consensus of those who happen to be participating is, as reflected at Wikipedia:Closing_discussions#Consensus, which says:

The closer is there to judge the consensus of the community [not of the discussion participants], after discarding irrelevant arguments: those that flatly contradict established policy, those based on personal opinion only, those that are logically fallacious, those that show no understanding of the matter of issue.

Also Wikipedia:Deletion_guidelines_for_administrators#Rough_consensus:

Consensus is not determined by counting heads, but by looking at strength of argument, and underlying policy (if any). Arguments that contradict policy, are based on opinion rather than fact, or are logically fallacious, are frequently discounted.

This idea is reflected at Wikipedia:RMCI#Determining_consensus:

Consensus is determined not just by considering the preferences of the participants in a given discussion, but also by evaluating their arguments, assigning due weight accordingly, and giving due consideration to the relevant consensus of the Wikipedia community in general as reflected in applicable policy, guidelines and naming conventions.

Of course, WP:CONSENSUS also says this, at WP:CONSENSUS#Determining consensus:

Consensus is determined by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy.

The significance of a history of "no consensus" results

RM closers are sometimes understandably reluctant to find that community consensus favors a move in a discussion, based on stronger policy-based arguments, when a consensus of the participants does not support it. After all, those opposing may not take the time and make the effort to fully explain the policy-based arguments opposing the move if it appears there is no consensus anyway. This is likely to be especially true the first time a given RM title change is proposed. But by the time there have been several such proposals made, it is more reasonable to assume that if there are relevant strong policy-based arguments, they have been presented.

This is why when there is a history of "no consensus" results, it's especially reasonable to evaluate community consensus based on the arguments presented in terms of policy basis.

If both sides are truly supported by strong policy-based arguments about equally, then discussion should be closed as no consensus, perhaps with a suggestion that it might be time to look at revising the applicable policies and guidelines, as they are providing conflicting guidance in at least the given situation.

Otherwise, one side is better supported by policy-based arguments, indicating that community consensus supports that side, and the discussion should be closed accordingly, even if there is no consensus among the participants.

Yogurt title history

Main page: Talk:Yogurt/yogurtspellinghistory

The following chronology notes key events in the history of the title of the Yogurt article.

• December 10, 2002 - Article created as Yogurt
• November 17, 2003 - Suggestion on talk page to change spelling to Yoghurt because of pronunciation.
• December 25, 2003 - "Yogurt" spelling changed to Yoghurt
• June 1, 2004: Spelling questioned
• August 26, 2004 - Yogurt spelling restored
• September 18, 2004 -Spelling changed back to Yoghurt
• May 12-17, 2005 - RM #1 result: No consensus (stays at Yoghurt)
• October 10-25 2006 -RM #2 result: Move to yogurt
• October 26-November 1, 2006 - RM #3 result: No consensus (moved back to yoghurt)
• May 14-21, 2007 - RM #4 result: No consensus (stays at Yoghurt)
• June 21, 2009 - July 5, 2009 - RM #5 result: No consensus (stays at Yoghurt)
• July 8, 2009 - RM #6 result: No consensus (stays at Yoghurt)
• Sep 15 - Oct 27, 2011 - RM #7 result: No consensus (stays at Yoghurt)
• Dec 2-10, 2011: RM #8 result: Move to Yogurt

Of the eight RM discussion, closers decided the outcome was "no consensus" six times. There was a finding of consensus for moving to Yogurt in RM #2, but this was reversed by a no consensus finding in RM #3.

Analyses of RM #1, RM #2, RM #3 and RM #4 show that in each of these cases, had the closer determined community consensus based on how well arguments were based in policy, particularly by ignoring JDLI and arguments that gave no reasons, rather than by counting !votes, the finding would have been that consensus supported moving to Yogurt.

This was verified once the title was so changed, and there has been no objection whatsoever from the community about that title ever since.

Conditions required for Yogurt Principle to apply

In order for the Yogurt Principle to apply to a given situation, the following conditions should be present in a proposed move:

1. Most if not all relevant strong policy-based arguments are likely to have been presented in discussion, such as when the proposal to re-title the article from the current title in question to the proposed title has been challenged in repeated RM discussions.
2. Support for the proposal is based on strong policy-based arguments.
3. If the move were made, there would be no conceivable significant policy-based arguments to initiate a reverse move discussion comparable to those currently favoring the move.
4. Assuming the move is made, strong policy-based arguments currently favoring the move now would still apply to defend the new title.

For example, in the case of Yogurt, the strong policy-based arguments favoring Yogurt were presented in every one of the eight RMs, including the first:

1. It was the original title of the article, community consensus to move this article from Yogurt was never demonstrated, and it has been stable and non-controversial since it was restored in 2011 (WP:RETAIN, WP:ENGVAR)
2. It meets WP:COMMONNAME and WP:CRITERIA better than "yoghurt" (it is natural and recognizable even in most English speaking regions where the h spelling is more commonly used)

There are no strong policy-based arguments to move this article from Yogurt, and there never were any to keep it at Yoghurt when that was its title.

In other cases, other strong arguments are likely to apply.

In the case of Ivory Coast...

Examples of titles with histories of "no consensus" closes

The following are examples of RM proposals that were repeatedly closed as "no consensus" even though moving was supported better by policy. Most were eventually moved as proposed, which settled the matter in each case.

• Major cities in the United States:
  • Chicago, Illinois → Chicago
  • San Francisco, California → San Francisco
  • Seattle, Washington → Seattle
  • Las Vegas, Nevada → Las Vegas
  • Boston, Massachusetts → Boston
• Napoleon I → Napoleon
• H0 scale → HO scale
  • Borderline case since there was only one previous challenge
• Yoghurt → Yogurt
• Public House → Pub [1] [2] [3]
• Independence Day (film) → Independence Day (1996 film) [4] [5] [6]
• Hillary Rodham Clinton → Hillary Clinton [7] [8] [9] [10] [11] [12]
  • While consensus was found to be in favor of the move in the most recent move proposal, a move review reversed that finding. That decision was questioned.
  • Finally moved June 11, 2015. [13]
• Clock Tower, Palace of Westminster → Big Ben [14] [15]
  • Borderline case since there was only one previous challenge
• Côte d'Ivoire → Ivory Coast
  • RM, Côte d'Ivoire → Ivory Coast, No consensus, 18 November 2005, discussion
  • RM, Côte d'Ivoire → Ivory Coast, No consensus, 17 January 2007, discussion
  • RM, Côte d'Ivoire → Ivory Coast, No consensus, 5 July 2010, discussion
  • RFC, Côte d'Ivoire → Ivory Coast, No consensus, 20 Jan 2011, discussion
  • RM, Côte d'Ivoire → Ivory Coast, No consensus, 19 July 2011, discussion
  • RM, Côte d'Ivoire → Ivory Coast, Moved, 8 July 2012, discussion
  • MVR, Côte d'Ivoire → Ivory Coast, Upheld, 18 July 2012, discussion
    • This is a quintessential case of policy-based argument (WP:COMMONNAME) versus non-policy-argument ("official name" which is specifically addressed in WP:COMMONNAME).

• Beyoncé Knowles → Beyoncé
  • A history of 9 RM discussions before it was finally moved. This is not the best example, since she usually used her last name early in her career, and gradually used it less and less as time went on.

• New York → New York (state)
  • Discussions took place in Oct–Nov 2004, Feb 2005, Feb–Sep 2005, Mar–Aug 2008, Oct–Nov 2010, Aug 2011, Jan 2013 and Jul 2013. In June-July 2016, the article was moved to "New York (state)", but this decision was overturned at move review. Then a more detailed July 2016 move request was closed as no consensus. A subsequent August 2016 RFC reached consensus that the State of New York is not the primary topic for the term "New York", advising further discussions. In October 2016, the status quo outcome of the July discussion was endorsed at move review. Finally, after a drafting debate and a renewed move request, the state article was moved to "New York (state)" on July 19, 2017, and "New York" now holds the disambiguation page.
  • SEE: Talk:New_York_(state)/FAQ
  • During the discussions about New York state JFG presented, at Talk:New_York_(state)/July_2016_move_request#Opposite_move_would_have_no_rationale, an argument that essentially reflects a core aspect of the Yogurt Principle:

    Imagine for a minute that the long-standing status quo dating from 2001 were different, i.e. that New York were the dab page and New York (state) or New York State were the state page. Would there be a clamor to change this and make New York the title of the state page? Proponents of such a move would have to argue that the state is either the overwhelmingly dominant meaning of New York (even in today's debate, nobody seriously claims that) or that the state has a longer-term significance than the city (quite hard a position to defend, as the state historically inherited its name from the city). Would people also complain that such a title arrangement "irreparably harms the encyclopedia"? No, the encyclopedia would serve readers well, especially considering that all internal links would be pointing to the right place. This little gedankenexperiment effectively counters the "if it ain't broke don't fix it" arguments: first we have many policy-based arguments that the current situation is indeed "broke", and second if the situation were reversed it would most certainly be defensible as "not broke". — JFG ^talk 11:12, 19 July 2016 (UTC)

Examples of determining "no consensus" of participants rather than community consensus

In the following RM proposals, the closer finds the support side to be better supported by policy, but never-the-less determines there is no consensus. These are examples of the closer determining the consensus of the participants rather than what community consensus is based on the arguments.

• Wikipedia:BOLD, revert, discuss cycle → Wikipedia:Bold, revert, discuss cycle "This [oppose] argument also has no basis in either policy or normal practice on other pages. The support camp at least has WP:BOLD on their side so has a better basis in policy. But given the level of opposition, I don't think that this can be closed in favour of moving." [16]