User:Theleekycauldron/Just pull it

One of the things I was most surprised about when I was being onboarded onto the Arbitration Committee was the fact that, as a matter of standard practice, oversighters don't deliberate and waffle before hitting the suppress button. Unlike pretty much every other advanced tool on the project, from CheckUser to rollback, oversighters hit the button first, and if it was a borderline call, they'll bring it to the other oversighters for discussion afterwards. But the more I weighed the potential harms, the more that made sense. If you suppress and you're wrong, the worst that happens is that people can't find some information for a little while. If you don't suppress and you're wrong, the worst that happens is harmful information about someone being irreversibly leaked all over the internet.

One area that badly needs that philosophy is the high-visibility Main Page. Millions of visitors stop by every day to read blurbs that are checked by a handful of editors on a punishing 24-hour cycle rather than relying on years-long eventualistic improvement. It should be incredibly important that we get things right in the face of high stakes and high difficulty. But when people come to WP:ERRORS to complain about bolded articles with unreliable sources, BLP violations, or live factual errors, admins waffle. The calculation should be the same: if you pull and you're wrong, the worst that happens is a blurb's run being delayed or cancelled. If you don't pull and you're wrong, the worst that happens is an uncorrected error being shown to hundreds of thousands more people, maybe even an error that harms a living person. But instead, we ping the nominator and reviewers to hear their input, we debate about the spirit of some policy, we let things slide because tracking down and verifying complaints takes time we're already pressed for.

This is a mistake. When someone brings a credible complaint to ERRORS about a serious issue in a live blurb that can't be fixed quickly, like a BLP violation or a grave factual error, don't waste time nitpicking or pinging other people – just pull it. If consensus is that it can be restored, maybe with some modifications, that's fine. That doesn't mean you were wrong. It just means you were being proactive instead of reactive – or worse, indecisive.

When to waffle

Obviously, the community still trusts and wants administrators to use their heads. I don't want people to walk away from this essay thinking that I'm telling admins not to consider all of their options, and some due diligence is required to make sure a complaint isn't bogus. There are lots of situations where a pull might be inappropriate as an action of first resort:

• If no urgent action is required. Some complaints don't actually need to be handled right away, and some don't require any action at all. If a bunch of people are complaining that a blurb mentions a swear word, they're just wrong as a matter of policy. If someone makes a complaint about whether "[x] job title should be capitalized" and you're not sure if they're right, that doesn't mean you need to pull the blurb while that's being discussed. In general, if an issue doesn't rise to the level where you feel the need to do something about it, don't.
• If you can fix it quickly. If someone points out a mistake, and you're comfortable loading up the relevant template and correcting the blurb, go for it! If it's a simple source interpretation error, you might just be able to reword the blurb to make it accurate. Or, if a bolded article has an unreliable source, you might be able to just get away with deleting the sentence(s) relying on it and letting other people work that out. If it's too much for a reasonable admin to handle on the fly, though, you might still want to consider a pull.
• If the blurb isn't live. Errors in upcoming DYK or OTD sets are serious, but they're not as serious as errors that are live on the Main Page. You can take a bit more time to think and discuss, and in the case of DYK, you can push the hook back a couple sets to buy yourself some more time instead of pulling.
• If a pull does harm. Obviously, TFA/L/P blurbs can't just be pulled, although they rarely have the same kind of huge issues ITN, OTD, and DYK have with accuracy. But even pulling blurbs on ITN can make that section look woefully out of date – if a city gets nuked, an error in the casualty count doesn't justify a preemptive pull. Also, pulling blurbs on ITN, OTD, and DYK can upset the balance of the Main Page, which is a secondary concern but not a non-concern. If a pull is the only option, OTD generally has alternative blurbs available in the template, and DYK might have alternative blurbs available on the nomination page or in the queues. In the rare event a TFA/L/P blurb needs to be pulled, there are always months' worth of blurbs ready in their queues.

Exceptions notwithstanding, what you should doing with a serious complaint on a live blurb is handling it, however you choose to do that. What you should not be doing is letting it sit while you try to get the broadest possible range of input.