Wikipedia:Pending changes/Request for Comment 2012

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

Background: The idea of WP:Pending Changes (also referred to in some implementations as “Flagged Revisions”) has a long history on Wikimedia projects. Originally developed at the urging of the community, the idea of having edits be reviewed prior to going “live” has been around the English Wikipedia since at least 2008. Discussions since the concept of the tool was introduced have often revolved around what, if any, implementation would be suited to the English Wikipedia community. The topic has been hotly contentious for nearly as long as it has been discussed. A trial of what was termed “Pending Changes”, a customized version of the original “Flagged Revisions” concept, was run in 2010.

At the close of the trial, community opinions varied on whether the trial had been successful and, if it was felt to have been successful, what form “Pending Changes” should take going forward. As a result of this community indecision, and at the urging of Jimbo Wales in September of 2010, Pending Changes entered a kind of limbo, in which it remained on some pages but had no clear guidelines for implementation or usage. A 2011 Request for Comment addressed the limbo state in which Pending Changes had been left and brought about the removal of Pending Changes from the English Wikipedia.

On the basis of Newyorkbrad’s closing statement in that RfC that "this community decision [to end the trial of Pending Changes] is without prejudice to any future discussion relating to adoption of pending changes or flagged revisions or some similar system based on a new discussion. There seems to be a consensus that a bit of a break from this discussion would be a good thing", a new RfC was begun roughly one year later, on 23 March 2012, to address the essential question of “should Pending Changes be activated on the English Wikipedia?”

Basis of the Current RfC: Formatted by Beeblebrox as a choice between three mutually-exclusive options, the 2012 RfC was intended to produce "an actionable result" regarding whether Pending Changes should "be permanently deployed here". Commenters were encouraged to support one of three possible options:

1. Position 1: "The negative aspects of pending changes outweigh the positive. Therefore the tool should not be used at all on the English Wikipedia."
2. Position 2: "Despite the flaws of the trial period pending changes has proven to be a useful tool for combating vandalism and other types of problematic edits. The tool should be used in accordance with the following draft policy. This policy is intended to reflect the community input in discussions. It is not set in stone and after use of the tool is resumed there may be unanticipated problems which can be corrected through normal consensus gathering processes."
3. Position 3: "Pending changes should be kept in the long term, but the draft policy is insufficient and/or out of step with what the community wants from the tool. Pending changes should not be rejected entirely but should remain unused until such time as there is a more complete policy in place that has been explicitly approved by the community."

The RfC remained open to comment for 60 days, and was accompanied by a concurrent "Discussion" section which remained open until the posting of this close statement. Users were encouraged to focus on supporting one of the three options and to direct commentary to the RfC’s "Discussion" section. Users were not permitted to add additional positions to the RfC beyond the original three.

Raw results of this RfC: The numerical results of this RfC are as follows, after IP votes, duplicate votes, struck votes, and commentary are removed from the positions’ "support" lists:

1. Position 1: 178 supports
2. Position 2: 308 supports
3. Position 3: 17 supports

This gives a total participation in this RfC of 503 distinct user "support"s, of which roughly 35% supported Position 1, 61% supported Position 2, and 3% supported Position 3.

Analysis of arguments: The closers of this RfC (DeltaQuad, Fluffernutter, The Blade of the Northern Lights and Thehelpfulone) considered all of the available positions in light of both raw numbers and strength of argument. As such, the closers acknowledge the following arguments as among the most notable:

• Position 1 supporters stated the belief that the idea of Pending Changes – that is, of edits not necessarily going "live" when made, depending on user status – is counter to the ethos of Wikipedia, the "free encyclopedia that anyone can edit". This was countered by supporters of Positions 2 and 3, who felt variously: that accuracy of information, especially in relation to BLPs, trumps openness to edits by anyone ("Wikipedia is first and foremost an encyclopedia (pillar #1) and ought to be reader-centered, not editor centered" [1]); that the limitations set by Pending Changes were no philosophically different than the limitations set by semi and full-protection, both of which have been deployed on Wikipedia for years; and that Pending Changes actually left articles more open to editing by users who would otherwise be shut out by our current page protection mechanisms.

• Some Position 1 supporters felt that the implementation of Pending Changes would put in place a hierarchy of editors where one hadn't existed before. Some of these arguments were based on the assumption that those given "reviewer" status would use it to force through their preferred content or that those able to turn on the "reviewer" right for other users would hold it hostage. Supporters of Position 2 countered this by pointing out that "reviewer" would be no different a right than any other that could be turned on by administrators. The closers of this RfC reject, as a violation of WP:AGF, the blanket argument that those who support Pending Changes should be assumed to be willing to abuse the system or the giving of user rights.

• Position 1 supporters stressed that the sheer volume of articles on Wikipedia is staggering, and that if even a small proportion of them are put under Pending Changes, the community will not be able to handle the volume of edits in need of reviewing. Many of these comments cited experiences on other WMF wikis which have already implemented some variety of Pending Changes. Supporters of Position 2 argued that Pending Changes, when used sparingly and in accordance with the draft policy presented on the RfC, may actually streamline the acceptance of edits in comparison to those on semi-protected pages. A number of Position 2 supporters cited their experiences on other PC-active wikis as evidence of how Pending Changes implementations can be functional and useful. It must also be kept in mind that each wiki which uses Pending Changes is able to customize its implementation, and the implementations used on other wikis may not be useful evaluators of the proposed implementation on English Wikipedia. Supporters of Position 2 also suggested that the issue of backlogs be addressed by an "end" to the queue of Pending Changes, such that all changes not approved after a certain amount of time should be automatically approved.

• A point of essential disagreement between supporters of Positions 1 and 2 is whether Pending Changes is preferable – that is, adds value – to any case in comparison to semi or full-protection. Arguments from Position 1 supporters highlighted that few users had requested Pending Changes, as opposed to protection, during the PC trial, and that if the Pending Changes queue backlogs, application of PC may be less useful than page protection. It was also suggested that the lack of being able to see one’s edits "live" may be demoralizing to new users. Position 2 supporters felt that Pending Changes did have value beyond that of page protection, especially in the areas of discouraging vandalism (arguments suggested that if a vandal knows their edits will never be displayed to the world at large, they are less likely to make them) and encouraging edits by new and inexperienced editors (who would otherwise need to request the confirmed user right or use a template to request that another user make their requested edit).

• A number of Position 1 supporters stated that though the notion of Pending Changes is useful, and the general concept of lightweight content controls even more so, they were unable to support Pending Changes as it was proposed and/or felt that the system needed so much work to become useful as to be essentially unsupportable currently.

• Many supporters of Positions 2 and 3 felt that the potential damage done by vandalism, especially damage to BLP articles, outweighs the potential inconvenience to editors who are unable to have an edit go immediately live under Pending changes. This was countered by Position 1 sentiment that openness is the basis of Wikipedia, and to lose some amount of openness would fundamentally alter the functionality of the project.

• A number of issues with the trial implementation and/or the suggested Draft Policy of Pending Changes were raised by supporters of all three positions, including:

1. The possibility of ongoing review backlogs and the lack of guidance provided in the Draft Policy for addressing them.
2. Lack of detail as to the responsibilities of reviewers and the qualifications a user must possess to be given the “reviewer” right.
3. Lack of detail as to when Pending Changes should be preferred to semi-or full protection, and how administrators and those who request the implementation of some form of protection/pending changes should make the decision about which to apply. In particular, it was pointed out that though Pending Changes may be particularly useful in two types of cases, the draft policy gives little consideration to the types of cases in which it may be most useful, described by one editor as: "One, where the vast majority of edits are good faith, but bad faith ones are causing exceptional damage. Two, where the vast majority of edits are bad faith, but the article nonetheless has a history of productive anon edits." ([2])

Consensus of this RfC: The notion of "consensus" on Wikipedia holds unanimity as an ideal, but acknowledges that unanimity is often impossible to achieve in the real world. Consensus in our real world, then, is "determined by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy". In contentious cases such as this RfC, then, the closers must take into account not only the raw number of editors supporting each option, but also the strength of arguments presented by supporters of each position and the content of those arguments, to reach the most appropriate decision – the decision supported by the strongest arguments, and with which the largest number of editors can abide. The closers of this RfC commend the RfC’s participants for working constructively with each other for a sustained period, often striving to reach common ground and agree as to what specific outcomes of each issue might be acceptable on both sides (or all three sides) of the aisle. Your efforts at explaining your positions, both to us and to other users, and your careful reasonings for holding the positions that you do, were immensely helpful to our ability to close this RfC.

Our findings as to the consensus of this RfC are as follows:

• The consensus of the community is that Pending Changes should be implemented on the English Wikipedia, based on both the majority of votes and the strength of arguments presented

• However, it is clear that though the community supports the adoption of Pending Changes, the community is also well aware that both the Pending Changes Trial of 2010 and the draft policy as presented in this RfC suffer(ed) from weaknesses.

• Therefore community consensus on this matter is determined to be that the community should dedicate itself to determining the implementation of Pending Changes that it wishes to be turned on. Many opinions have been presented in the various areas of this RfC as to what works and doesn’t work in Pending Changes as implemented/proposed; the community must now focus its energy on optimizing the implementation of Pending Changes that it wishes to see go live. Because the community must have time to discuss this, and in order to avoid the holiday season which would interfere in both devs’ and editors’ schedules, Pending Changes will become live on 1 December, 2012. To allow developers enough preparation time, we recommend that community discussion about changes to the draft Pending Changes policy be concluded no later than 1 November, 2012. If the community has not, at that time, reached a consensus about how to change the draft policy, Pending Changes will be implemented according to the terms of the Draft Policy until the community can find a consensus.

The closers stress to the community that consensus does support the implementation of Pending Changes, and we urge all the participants of the RfC– as well as its developers- to participate in the upcoming discussion about how to implement the software version of this consensus. We strongly recommend that this RfC be considered the endpoint of up/down community discussions about whether Pending Changes should be used on English Wikipedia for a period of time of no less than two years to allow community focus to move to the use, adaptation, and evaluation of the Pending Changes tool.

We thank the community for offering its input into forming the consensus and hope to see the community come together for the finalization of the policy.

• The Blade of the Northern Lights (話して下さい) 03:01, 23 June 2012 (UTC)[reply]
• A fluffernutter is a sandwich! (talk) 03:03, 23 June 2012 (UTC)[reply]
• -- DQ (ʞlɐʇ) 03:05, 23 June 2012 (UTC)[reply]
• The Helpful One 11:30, 23 June 2012 (UTC)[reply]