A good solution might be an "exponential back-off" temporary ban. In this, an open-source project adopts a policy that the mental health and wellbeing of its d

A good solution might be an "exponential back-off" temporary ban. In this, an open-source project adopts a policy that the mental health and wellbeing of its developers is a priority, and in particular abusive language from end users is not accepted. Specifically, if a user communicates with such negativity (e.g. commenting this way on a github issue), two things happen: 1) That comment is deleted 2) That user is prohibited from posting on this project's issues for 30 days. That's for the first offense. If, after 30 days, the user does something like this again, they are temporarily banned for 60 days. And then 120. You get the idea. If you don't like 30 days as a base, then it can be 24 hours, or 90 days, or whatever the project decides is appropriate. Regardless of these or other parameters, this strikes a balance between accepting needed feedback from the community, and "canceling" someone from contributing useful feedback in the future. In fact, I bet you this will have a role in conditioning certain people to communicate in more nurturing and kind ways. A permaban would likely NOT do that, but a short temporary ban that increases on repeat offenses probably will. I am not sure if Github supports this already, or provides some mechanism that could be used by the project maintainers to manually implement it... But if you are developing on any platform where end users sometimes communicate toxically with volunteer developers just trying to make the world a better place, maybe this idea is worth considering.