Skip to main content
Meta Stack Exchange

You are not logged in. Your edit will be placed in a queue until it is peer reviewed.

We welcome edits that make the post easier to understand and more valuable for readers. Because community members review edits, please try to make the post substantially better than how you found it, for example, by fixing grammar or adding additional resources and hyperlinks.

Required fields*

5
  • 4
    the goal of voting rings is to inflate the reputation of 1 or more accounts, often for the purpose of getting out of a question ban or making spam look more credible or inflating the imaginary internet credentials of a user. Making it cumbersome to start one is certainly one effective way to prevent it, we do have an example to look back on with Teams where the ability to easily get a voting ring going using the association bonus from teams (which was silly to exist in the first place) caused problems. Commented Sep 20, 2023 at 19:45
  • 2
    When I say that using reputation to prevent abuse is a simplistic solution, that's what I'm getting at. Yes, sock puppets and voting rings are real concerns because, people do not necessarily use votes as we permit - sometimes they honestly assume it's "acceptable" while others do it after being told to stop repeatedly. This test is a sort of cost/benefit analysis of removing this solution. We don't know anything about how potential voters would vote. We can guess. We can poke at unregistered and low rep user votes... but we can't actually know. Commented Sep 21, 2023 at 6:21
  • There are a lot of assumptions out there about the amount of tooling that would be needed to address the new issues caused. We could invest a lot of time in creating those tools and automations and making sure they work and then remove the reputation to vote to see what happens. Best case, we've built all the right things and no additional tools are needed and all invalid votes of all types are handled and all is well with the site. This would be amazing but I think it overestimates our ability to predict human behavior. It doesn't allow us to test the impact on engagement. Commented Sep 21, 2023 at 6:34
  • The other end of things is we just make the change with no improvements in tools and figure out what we need based on the test, leaving the sites to deal with the issues. This would save us from building tools that we thought would be helpful but ended up missing most of what they were intended to catch. It also ignores that we have existing tooling needs to identify vote fraud that mods have been really clear about needing. This is also higher risk as it could leave the sites unable to deal with issues until tools were built or would make rolling back the test more likely. Commented Sep 21, 2023 at 6:40
  • 1
    Our proposed test plan is in the middle of these two extremes. We're building improvements to tools we already know we need to address current levels and types of vote investigations and let mods use and suggest improvements to the tools to ensure they're useful. When the test starts, we hope the bulk of cases will be identified by the existing tools and, when they don't, we seek out the new cases and build tools specifically geared to them. While there is some risk, it's not as high as the prior option but we get data without first spending 6-8 months guessing what tools we need. Commented Sep 21, 2023 at 6:54