Wikipedia:Non-sysop closures

On Wikipedia, editors who have the user right sysop are often referred to as administrators, and the sysop user right grants those editors additional technical tools. Because sysops on the English Wikipedia are vetted by the community, they tend to be elite editors with a lot of trust and social capital. This trust and status within the community has the effect of turning the sysop user group into a social position, and editors will often ask sysops to perform actions which require their social capital rather than their technical tools. The conflation of the technical group sysop and the social group "administrator" can be confusing since the social role of administrators can be performed by anyone the community trusts. One area where this difference has been contentious is non-sysop closures.

Discussions on Wikipedia are closed when an editor discourages further comments and summarizes the discussion, often using a template like {{archive top}}. The summary will often include a course of action that participants came to a consensus on, and closures by uninvolved editors are often considered binding precedent when making future decisions. This is helpful because it prevents duplicate discussions and helps future editors understand decisions without reading hundreds of pages worth of discussion. But while these decisions are more formal than other discussions, a closed discussion is no more binding than an unclosed discussion (though see WP:CONLEVEL). The difference is that closed discussions make it easier for uninvolved editors to intervene in future conflicts. Sysops attempting to stop disruption will be able to do so faster if they can be pointed to a summary of consensus (closure) rather than having to read the entire discussion themselves. While not every discussion requires closure, it can help in some cases where a definitive decision is needed or a summary of the discussion would be helpful in the future.

The weakness of closures is that they rely on the ability of the closing editor because the community is trusting that editor to write an adequate close. A problem arises when the community, or a subset, doesn't trust an editor who wrote an otherwise fine close. Because of the lack of trust, one side will likely challenge the close leading to extensive discussion and ultimately wasted time. Editors have tried a number of methods to resolve this problem, but a routine request is for closure by an administrator. While no tool bundled with sysop helps with this task, the social role of administrator does help.

Requests for closure by a particular kind of editor ("admin", "experienced editor", etc) are made because social capital is needed to mobilize the editorial community, not because it requires special tools or tenure. Ideally, closures should be uncontentious because everyone already came to a consensus on what to do, but in cases where there is a rough consensus or the community does not agree, one side of the argument may not be willing to abide by the discussion outcome. While nothing stops a newly auto-confirmed account from closing a contentious discussion, editors are more likely to challenge the close regardless of the merits. Challenging closures by more experienced editors becomes more fraught because of how social capital is leveraged. Firstly, experienced editors have more community trust, and so their closures are often given the presumption of validity unlike closures by newbies. This makes it harder to overturn the close since a no-consensus outcome retains their close. Secondly, any discussion involving them is likely to attract editors who know them to the discussion leading to wider and often more sympathetic discussion than a closure by a newbie. Thirdly, experienced editors likely know the proper venues and individuals to contact to seek enforcement of the closure. These all apply to closures by sysops, but often sysops have more extensive community ties and the technical ability to enforce the decision themself. These social mechanisms result in a wikt:cline of stability where editors with more social capital produce more stable closures because others are more willing to trust their judgement.

The essay on non-admin closures presents guidance for navigating these problems without naming them, and the continued conflation of technical and social roles leads to convoluted guidance. In particular, it cautions non-sysops to avoid closing discussions which are contentious, without guidance on how to evaluate that. Non-sysops often close contentious discussions with minimal fuss, and previous discussion has found "non-admin closure" to be an insufficient reason to overturn a close. The problem is that sometimes participants want a stable close, and an easy way to achieve that is through leveraging the social capital of sysops. Because non-sysops might not have the necessary social capital to enforce a stable close, they are cautioned against closing any contentious RfC. This has the perverse effect of increasing the social capital and workload of sysops at the expense of other experienced editors. While using sysop---a technical permission---as a proxy for social cachet is convenient, it harms the community by reinforcing the idea that sysops, because of their account permissions, have greater social power than other editors.

Improving the system of closures and close challenges requires understanding why sysop closures became the gold standard for a process that doesn't require any of the tools bundled with +sysop. How has the non-admin closures essay impacted the social division between The Cabal and non-administrators? How can we classify discussions so that sysops are not the only people qualified to close "contentious" ones? How can we counteract the "cline of stability" where greater deference is given to closes by elite editors? Would doing so help or hurt the project?

See also

edit