Wikipedia:RFC/How to present a case

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Please note that this guide is descriptive, not normative. That is to say, it explicitly does not deal with the ongoing controversy over how the community and those involved in RFCs should behave. It is based on empirical observation of what arguments have worked and which ones have not, as well as upon discussions with editors about what they pay attention to.

There are three very important things to realize about RFCs:

  1. They are a request for comments, to the entire world, not a court hearing;
  2. Product is much more important than process.
  3. If they do not resolve the situation, they may lead to ArbCom cases (see ArbCom's guide to arbitration).

Almost everything you need to know about making or responding to an RFC is an extension of one of those three facts.

The absolute best way to fare well in an RFC is to not do things wrong. If you are trolling, actively POV pushing, or generally being an asshole, there is no guide in the world that can help you. Don't do that.

The perception of your case[edit]

At any time, there are twenty or more active cases in Arbitration, almost 200 closed ones, a stack in mediation and a handful of RFCs under way. Individuals cannot keep close track of all of these. If you mention anything to a user not intimately involved in the current process, they are not likely to remember particular details of the argument or which POV that user was advocating.

Write your evidence and your proposals so that they help jog everyone's memories. Assume that every time a user pulls up the RFC page, or the discussion page, they won't remember what was concluded last time or elsewhere. They are not clueless – but they need information.

What users will and won't look at[edit]

RFC pages, and their discussions, can become very long. Nobody has time to read through 100 KB evidence pages, and they especially don't have time to reread them after they've forgotten what was what.

Don't let your page get to 100 KB. Be concise, be direct, be clear. You do not need to cross every t, dot every i, and show every single instance of a given user being a problem — doing so can be counterproductive. Pick the clearest examples you can, and present them with as little commentary as is necessary.


Users do not read up on disputes that might reach RFC eventually. It is very unlikely that they know the history of the dispute going in — that someone is a known advocate of a point of view, that someone has a history of defending problem users, or that everybody who has ever dealt with a user recognizes them to be a complete lunatic. Point these things out to them. If you point to an edit that comes after a month of heated discussion, it may not make sense to someone who was not a part of that discussion.

Take care with evidence that requires context. If there is better evidence for the same point, use that instead. Otherwise, be ready to explain the context. Note that the more explanation a piece of evidence requires, the less likely anyone is to have time to pay attention to it.

Expertise of the users[edit]

Most users are not subject experts, but some are. This is why RFCs, unlike ArbCom cases, may come to conclusions on the basis of article content. In practice, users are likely to be cautious about basing a ruling on the ground that one side is right in a content dispute. There are exceptions to this—in general, we have looked unfavorably upon people who are using Wikipedia as a platform for advocacy, as well as people who accuse Wikipedia of offering safe haven for a conspiracy to suppress their point of view.

Wikipedia is not collectively hostile toward the documenting of minority views—only toward those who break fundamental Wikipedia principles (such as neutrality and personal attack policies) in their edits relating to such views.

Content issues are complicated and take time to figure out. Other approaches may be indicated. Instead of making the argument that somebody is advancing a nutty conspiracy theory that has no credibility, find statements on talk pages where they express a desire to advocate a cause, instances of them removing well-sourced information, instances of them accusing those who disagree with them of conspiracy, and other more concrete and self-explanatory things. Almost none of the cases which fail resolution at RFC and become Arbitration cases have actually required careful attention to content issues in order to get the necessary result.

Effective arguments[edit]

The community generally believes that the Wikipedia method works; that Wikipedia is, on the whole, a successful project, and that admins are generally trustworthy. They explicitly choose any outcome that will result in Wikipedia working better.

Pursuing arguments that oppose Wikipedia's basic principles, suggesting the existence of a massive cabal of rogue admins, or treating the dispute resolution process as if it were an end in itself will not work.

It is mistaken to argue on the assumption that an RFC functions like a court of law. See Wikipedia:Wikilawyering.

Arguing about flaws in the Mediation and RFC process is usually a waste of time and will make editors, admins, and eventually Arbitrators look dimly upon you. Take the time that you could spend arguing about the details of the process and apply it to trying to gather useful evidence. The first to try to rules-lawyer the arbitration process invariably loses — because they wouldn't be rules-lawyering if they had a case, and the same may be taken to be true of RFCs, with the addition that rules-lawyering an RFC tends to predict the progression of the case to ArbCom and might reasonably also be used as a cue to take it there, rather than waiting for tardy responses to be completed.


Clear and persuasive presentation of evidence will almost always be more effective than whatever is said on the talk pages. Almost nothing useful ever comes out of any argument between and among parties on the talk pages. The chance of any argument being made on other pages getting noticed or being cared about drops dramatically the longer or the more numerous those arguments get. If you must engage in discussion, short and simple and on the RFC talk page is probably the most effective approach.


Questions about the conduct of RFCs – the process in general – are best taken to a sympathetic forum elsewhere, as if they are placed on the talk or RFC page they may have an unfortunate appearance of wiki-lawyering and prevarication – see above.

Desired outcomes in user RFCs[edit]

Many editors that start RFC/User pages are understandably very frustrated with behavioral issues. At some level, they really want the outcome to be a vindication for their work and punishment for the other editors.

However, RFC/U discussions are not designed to impose solutions on the unwilling. "The community" may facilitate the discussion, or it may provide views by outside editors, but it will not force a user to change his/her behavior through the RFC/U. The goal is dispute resolution through working together, not outside punishment.

Consequently, the desired outcome needs to be both something that could be agreed to by all of the involved parties, and something that could implemented by the involved parties themselves.

Some examples of possible outcomes
  • The editor will voluntarily limit his work on This article to WP:1RR for the next month.
  • The editor will voluntarily limit his assistance at This article to making suggestions on the talk page.
  • The editor will agree to provide high-quality sources whenever s/he adds potentially controversial information to articles on This subject.
  • The editor will agree to make an effort to comment on the content, not the contributor.
  • The editor will agree to stop using profanity on Wikipedia.
  • The editor will agree to remove offensive text from his/her talk page.
  • All editors will voluntarily follow the Bold, revert, discuss system in This article to reduce edit warring.
Some examples of impossible outcomes
  • We get to revert whatever this editor does if we don't like it.
  • The editor is forced to apologize against his/her will.
  • The editor is forced to re-write This article to our specifications.
  • The editor will be blocked.
  • The editor will be permanently topic-banned.