Very interesting read. However, I wonder if there is even a use for shadowbans in Discourse.
Yes, there is a shadowban plugin but it’s hardly used. It has one (1) Github star and the statistics we have within Communiteq reveal that the number of Discourse forums that are actively using this is less than 0.5%.
So why is that adoption of shadowbanning within the Discourse platform so low? It’s not because shadowbans are bad (which they are), it has another reason. From a technical point of view, shadowbans are very easy to detect on Discourse. Simply visit the forum without logging in or using a different account and spot the differences. And that is possible because Discourse is not a social media platform that uses an algorithm which decides what you will see and in what order it is presented. In Discourse, the content you are shown does not depend on what you have done / seen / read or written before. And if there is no algorithm, there is no place to downplay certain content. There is only the option to hide it completely instead of showing it.
I don’t support all forms of transparent moderation either, because transparent moderation can still be subjective.
Besides my work in the Discourse ecosystem I’m also founder of a company that combats online disinformation and hate speech. During talks, I am often asked for examples of foreign interference on the democratic processes in this country (the Netherlands), where the people asking the question always expect me to tell some story about Russian trolls. But in fact, the main examples of interference in the democratic processes by foreign actors we have seen in this country, are social media platforms like Twitter or YouTube banning or shadowbanning Dutch politicians, without any kind of review by a court or other independent body.
And that’s bad.