Honestly, I think the large-scale nature of communities is a factor. Moderators now are also more willing to ban accounts that are non-spam. Certain views are

Honestly, I think the large-scale nature of communities is a factor. Moderators now are also more willing to ban accounts that are non-spam. Certain views are "harmful" and apparently people cannot be asked to not react to seeing things, or even ignore people, whether via some tool or just by saying "Oh, it's her again, I'll just skip over that." This places pressure upward on the moderators to make certain people go away. Folks love gaming the algorithms to decide what will and will not be read and algorithms come into play because of the aforementioned large scales. Face it, the Internet runs on ad money and that can be tight, so everyone gets the bright idea to either let the algorithms do the moderation or to let people who have free time on their hands and the motivation to do it. Usually the motivation is expressed in some seemingly-altruistic manner but it always boils down to pushing their own views. Eventually some views become acceptable, others anathema, and you get that ghastly distillation where more moderate voices are driven off until finally you get a kind of ghost-town. This is a pretty tough nut to crack from this vantage point. One of the ways I see out is the admittedly computationally-intensive tack of making the algorithms only affect a user's personal view rather than making it site-wide. You can still run into the echo chamber issue, only in this case it is a bunch of people in their own private speech bubbles in one large chamber. I've spent a lot of time thinking about this because I have watched this sort of thing kill communities for the past thirty years now.