FAMEPedia:Reviewing pending changes

This page explains and provides the guidelines for the review process associated with the pending changes protection feature (enabled on 1 December 2019). Articles with pending changes applied can be reviewed by administrators or users called pending changes reviewers ( user group) who hold the pending changes reviewer permission. The process of reviewing is intended as a quick check to ensure edits don't contain vandalism, violations of the policy on living people, copyright violations, or other obviously inappropriate content.

Currently, there are Special:ListUsers/sysop administrators and Special:ListUsers/reviewer pending changes reviewers on FAMEPedia. (In total, 0 user accounts have this right.)

Overview
The purpose of reviewing is to catch and filter out obvious inappropriate edits and vandalism on articles under pending changes protection, a special kind of protection that permits anonymous and newly registered editors to submit edits to articles that would otherwise be semi or fully protected under one or more of the criteria listed in the protection policy.

Reviewers do not take responsibility for the correctness of edits they accept. A reviewer only ensures that the changes introduced to the article are broadly acceptable for viewing by a casual reader. The reviewer checks the pending change(s) for an article and can then decide to either accept it, revert it or modify it then later accept it. Reviewers are not expected to be subject experts and their review is not a guarantee in any way of an error-free article. They are expected to have a reasonable editing history, distinguish what is and what is not vandalism, and be familiar with basic content policies. Reviewer rights are granted by administrators; and in cases of misuse of the right or to protect FAMEPedia from possible misuse, the rights can be removed by an administrator. The permission can also be removed at the request of the user, or the Arbitration Committee.

Reviewing process
Articles with pending changes are marked as such in watchlists, histories and recent changes. In addition, there is a special page, Special:PendingChanges, which lists all articles with pending changes. Clicking on [review] at Special:PendingChanges or [pending revisions] in watchlists, histories and recent changes will return the diff between the latest accepted revision and the last revision to the page. Most of the time, you should be able to complete the process from the diff alone, while in more complex cases you may have to check the recent history or edit the article.

General criteria
As a general rule, you should not accept the new revision if in analyzing the diff you find any of the following:
 * It conflicts with the biographies of living persons policy.
 * It contains vandalism or patent nonsense.
 * It contains obvious copyright violations.
 * It contains legal threats, personal attacks or libel.
 * It conflicts with the reason given for the protection.

Reviewers should take special consideration of the reason given for protection and attempt to uphold it. The protection policy reserves pending changes protection to clear cut cases, so interpretation issues should be minimal. For example, if the article is protected because of repeated inappropriate edits by a sockpuppeteer, and if the same type of edits are made by a newly registered or anonymous user which you suspect is the same person, you should not accept those edits.

When reviewing, it is crucial to first check the number of users having edited the article; this information is provided in the middle of the page just before the diff content starts: "(X intermediate revision(s) by Y user(s) not shown)".

Also note that you remain subject to editing policies like edit warring and ownership of content.

For a list of tips and optional criteria you can apply when reviewing, see tips for pending changes reviewers.

Reviewing edits by a single user

 * Technical note on the reviewing interface: If you revert ("Revert changes" button), the comment you provide is automatically appended at the end of the standard revert edit summary, and you are asked to confirm your action. If you accept ("Accept revision" button), the comment is entered in the review log.

If all the edits were made by one editor, then check if they are clear vandalism or not. If the most recent edit is clear vandalism, it is reasonable to assume they are all vandalism, and you may revert the changes without comment. If it isn't clear vandalism, then you should check if there are any of the obstacles to accepting described above (sneaky vandalism, BLP violations, etc.). If there are obstacles to accepting, then you should revert with an explanatory comment or edit the page to ensure compliance, such as by reverting citing BLP violations, or modifying the text to remove copyright problems. After a revert the new revision is automatically accepted, while if you edited the article and dealt with all obstacles to accepting, you may subsequently accept.

In ambiguous cases, reverting is not the default option; you should properly investigate the case or leave it for a second opinion. For example, if information is modified without a new source, which may be sneaky vandalism, you should not presume vandalism but check whether the article has an existing source for it, which may have changed as well (e.g. number of YouTube views, box office results, etc.). If no source is provided, you may search for one and if none is straightforward but there are no reasonable reasons to believe the new edit is vandalism, it is acceptable, but if on balance you estimate that the edit is more likely to be vandalism, you should not accept and may revert. Removal of content ('blanking') should be properly examined. If this is vandalism the new revision should not be accepted, but blanking can be perfectly legitimate, such as removal of BLP violations. Test edits falling under patent nonsense criteria are not acceptable, but you should make sure that this is not a mistake, as sometimes unintentionally inserted wiki commands are part of an acceptable edit, in that case you just need to remove or replace the undesirable syntax.

Acceptable edits
If there are no obstacles to accepting, then it is presumed that the new revision is acceptable. You should treat the edits as you would habitually, following the appropriate policies and guidelines. It is not necessary for you to ensure compliance with the content policies on neutral point of view, verifiability and original research before accepting, but of course you are free to uphold them as you would normally with any edit you happen to notice. For example, in case of additions for which you can find no reference in the article but estimate unlikely to be vandalism, treat them as you would treat any such edit: do nothing, tag as needing citation, provide an appropriate citation, or revert – depending on the situation at hand. In general, there are three options:
 * If you intend to do nothing regarding the changes, then accept the new revision. Accepting doesn't prevent you from later editing the article to address any concerns you may still have, or bring up an issue with the user, or at the article talk page. Consider thanking or welcoming constructive new users.
 * If you intend to ultimately revert the changes, then you may do so from the reviewing interface with an appropriate explanation, but as for all reverts they must be supported by policy. It isn't necessary for you to accept the revision before reverting even if you determine that there are no obstacles to accepting, as reverts are quick, accepting or not would yield the same end result and no policy prevents users (therefore reviewers) from editing pages with unreviewed revisions.
 * If you intend to edit the article regarding the changes (such as adding a citation, citation needed, fixing typos, removing some of the additions, etc.), then you may accept immediately then edit the page. While you may also edit then accept, note that during this time the edit remains unreviewed, so it should be avoided if you anticipate taking some time.

Accepting, immediately or after some modifications, is the default position, and even if an edit may appear suboptimal, this is in itself not a reason to revert, as for all edits, since they may yet be improved.

Reviewing edits by multiple users
If the pending edits were made by multiple editors, bear in mind there may have been a good edit that has been removed by subsequent vandalism. Do not rely solely on what you see in the "pending review" diff page, instead:
 * 1) Check the page history regardless of whether the version you see contains vandalism.
 * 2) Review each series of edits by individual users from the page history (diff from the latest accepted revision to the last revision by the first user, and so on). Undo any edit that is vandalism, a BLP violation, or unacceptable according to reviewing criteria. Each undo will create a new edit under your username, but will not be automatically accepted. Leave acceptable edits in place, unreviewed.
 * 3) Once you are satisfied that all inappropriate edits have been undone, you will be left with acceptable edits. Review the most recent pending edit as you would in case of a single user and you're done.

For any specific issues related to reviewing, please use FAMEPedia talk:Pending changes. For discussing the guideline itself, please use FAMEPedia talk:Reviewing pending changes.

Editing pages with pending edits
If you edit a page with pending edits, there will be a note mentioning this between the page title and edit window; you can click to show the diff between the latest accepted revision and the last revision, and review pending edits. There is an option to accept the new revision you will save below the edit summary at the right of "watch this page". Be sure to have reviewed pending changes before clicking it. If you don't click it, after saving the software will ask if you want to accept the new revision.

Unaccepting (reversing an action to accept)
Unaccepting a revision is reversing an action to accept a revision, whether manual or automatic (hence you cannot unaccept a revision which has not been previously accepted). It can only be done from the reviewing interface and is unrelated to the action of reverting an edit. You should generally not unaccept revisions, except to undo yourself if you realize you have made a mistake, because it only pushes the page back to Special:PendingChanges on latest revisions and has no effect on old revisions. If you have concerns with an accepted revision, then edit the article to address the concerns. If you think a revision should not have been accepted, you may discuss the issue with the reviewer if you feel this is needed. Automatically accepted revisions should generally not be unaccepted, even if they were vandalism, because there is no benefit in doing so (it only removes the [automatically accepted] tag appended to it in the history). If you think that an editor has incorrectly reviewed an edit, you can make them aware by putting Uw-wrongreview on their talk page.

Becoming a reviewer
During the trial period for Pending Changes and the Reviewer permission, criteria for requesting the reviewer permission were specified as follows:


 * 1) You have an account, and routinely edit.
 * 2) You have a reasonable editing history – as a guide, enough edits that a track record can be established.
 * 3) You have read our policy on vandalism and understand what is vandalism and what is not.
 * 4) You are familiar with the basic content policies: Biographies of living persons, Neutral point of view, No original research, Verifiability and What FAMEPedia is not.
 * 5) You are familiar with the basic legal policy: FAMEPedia:Copyrights.
 * 6) You have read the guideline on reviewing.

No consensus to change these criteria came out of the RfC following the trial period, so they are still in effect. Therefore:

If you meet the above criteria, then ask! Add your name to the list of requests at FAMEPedia:Requests for permissions/Pending changes reviewer.

If you have rollback or autopatrolled rights, you are a good candidate for reviewer rights as well – the level of trust is similar; though it is not necessarily the case that you will be granted reviewer rights if you have rollback or autopatrolled rights. Administrators automatically have reviewer rights.