Indeed. Nobody has decried the fact that moderation happens; it happens, it is necessary, and it keeps the discussions productive. The largest problem is the silence and concealment.
It is interesting that this thread got squelched for EVERYONE in response to one single person posting inflammatory words. In effect, everyone who wished to discuss this topic was punished for the actions of one person, instead of just that one person being punished. If this was a fully automated action, then why is Discourse configured to act this way, and how can it be reversed?
Discourse is configured this way to reduce the need for immediate human moderation. The idea is that even if no moderator is currently online, a heated discussion can be forcefully cooled down.
I would not say that this is a “punishment”. Not being able to write messages in a specific topic for 4 hours does not cause you any harm, nor will it kill a discussion if the discussion is actually important.
It looks like these algorithms can be configured somewhat, but not disabled. Given that there are very few moderators in this space and they can’t be here all the time, I think some automation is a necessary evil for a space like this.
In principle, if a user’s flags are routinely overturned by moderators[1] they’ll be discounted in the future. I don’t know how big an impact that has here.
I’ll also say it’s helpful to the moderators to get a breather. If we get flooded with flags then a throttling mechanism makes sure we don’t get overwhelmed.
It makes sense. One thing I’m curious about though is the possibility of flagging itself being abused. Occasionally I’ve seen a post, then seen it was flagged, and wondered why that was, as it didn’t seem flag-worthy to me.
The reason I mention it is that I’m a moderator on another community (unrelated to Python) and we’ve had some discussion there among the mods about the phenomenon of people using flags as a “super downvote”. It seems to sometimes happen, for instance, when people make political statements that aren’t prima facie disrespectful or out of line but that some others vehemently disagree with. That’s not something that comes up here per se but the underlying problem can still arise.
I assume that you are aware of this possibility, but just in case. . .
Note that discourse also has specific features like weighting the votes of different users differently, as a way to reduce the chances for abuse - users whose flags tend to be rejected get their voting power reduced.
For those curious[1], the user who flags a post is visible to the moderators, so they would likely notice if someone was abusing the system.
Of course reasonable people can disagree about what should be flagged and someone getting a little flag-happy doesn’t necessarily merit action–the weighting system should help that.
