4

Wikipedia's premise is that anyone can edit an article. While it's true, whoever tries to edit articles (especially of living people) knows it's not that easy to make a claim stick. There is a ton of community made editing rules, and the willingness to enforce those rules varies a lot. For some pages, it sparks really long discussions, and you can see the politics unfolding in the articles' talk pages where a consensus has to be reached. In the end, it feels like a game of whoever gathers more "editing time": the "party" with more time has the resources to do the "final" edit, so they "win".

In that system, who has the final say in such editing disputes?

0

3 Answers 3

3

Typically you end up with a giant so-called "request for comment" or similar which is eventually closed by an uninvolved administrator. That decision is then implemented. The "mass killings under communist regimes" controversy was the most recent highly publicized dispute. You can see the dispute was whether or not to delete the article. There was a "AfD" (article for deletion) nomination which was immensely long and eventually closed by four administrators (an unusually large number because of how controversial it was). The result was no consensus, so the status quo prevails and the article is kept.

1

Discussions rarely end in "requests for comment". Are you referring to any discussion in specific?

The English Wikipedia follows a community norm called "BOLD, Revert, Discuss" (BRD) in which all editors are encouraged to be bold when updating the encyclopedia. If another editor objects or if it breaks another community norm, they will revert (undo) the edit. At that point, instead of reverting back and forth, which is known as "edit warring", the article should be left in its status quo state and an editor (usually the one who is proposing the change) should start a talk page discussion about it.

Usually other editors chime in there and the point is to find a consensus agreement on including neutrally phrased information that has been covered by reliable sources. Most discussions end this way. One side peters out either for lack of sourcing or lack of interest or because of a convincing policy-backed argument.

Claims very often do stick on low traffic pages. On high visibility political articles, the content has already been so heavily edited over time that most major discussions have already been had, and any controversial change requires more background justification than simply relying on "BOLD, Revert, Discuss". So to your question, in the rare occasion that there is a long talk page discussion, consensus is not measured as a poll or vote but by weight of policy-backed arguments. For example, Reddit brigades a discussion with "votes" in a certain direction, they will not weigh as heavily when an experienced, neutral community member (usually an admin) formally "closes" a contentious discussion after either a week or 30 days. The more contentious the discussion, the more scrutiny is put on the closer to accurately summarize and represent the strengths of the viewpoints presented. Again, discussions rarely reach this level of discussion.

More often, if you and another editor are going back and forth on a talk page, having reached an impasse, the next step is dispute resolution. Usually this entails finding a neutral third party opinion by posting a neutrally phrased request for a third opinion on a variety of related noticeboards. "Requests for Comments" is another option for creating an open invitation. There are also other escalations for behavioral issues.

The best way to break a the uncommonly long discussion with the same voices on repeat is to declare an impasse and try one of those solutions that bring in outside opinions. Otherwise the "discussion" will continue to drone on, producing text but no resolution. If the subject isn't pressing, sometimes another reader/editor will find the stale discussion months later and resolve it then.

0

Here I would probably divide it into disputes and discussions. In disputes, it usually happens that both sides see it through to the end, that is, both sides put in the time and it comes to some kind of end.

If you have a dispute with someone, you resolve it with them on their discussion page. If it doesn't end, you can ask for help from the administrator, you can use the third opinion procedure, or a request for comment (these procedures are available on the Czech Wikipedia, but I don't know how it is on others, for example, the English one). Usually, a dispute can be closed by requesting a comment where other community editors have their say. Sometimes disputes end up in so-called arbitration, in which the arbitration committee usually issues a decision, that the parties to the dispute must comply with. If they do not follow it, the arbitration committee acts again, or the administrators enforce it.

However, there are also a number of other discussions on Wikipedia, for example, discussions about the form of articles or the rules themselves. Here it usually ends with some kind of agreement or consensus, but it often happens that not everyone has the time and energy to carry such a discussion to the end. It often happens that when it seems that no one is protesting and it is possible to establish a new consensus, people speak against it and the discussion has to be extended. Adopting new rules or modifying existing ones is a very time-consuming process and should be started by someone who has time and patience.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.