2 ms·
> The notification emails contain context that explains that if you reply to the email, your message will be published on https://radiopaper.com https://radiopa
by kazinator 5y ago
> The notification emails contain context that explains that if you reply to the email, your message will be published on https://radiopaper.com https://radiopaper.com
This (or this part of it) is not so radically different from the concept of posting to a mailing list, and having that show up on the web archive.
Also, if you write to a moderated mailing list, your conversation is initially between two people: you and the moderator.
> more resistant to trolling and abuse, is that messages are not published until the counterparty replies or accepts your comment.
While that's certainly resistant to trolling, it's not resistant to the abuses. Certain users can favor certain other users and promote their comments while suppressing other comments.
If we think about HN for a moment, that's completely opposite; here you cannot even be downvoted by the counterparty to your comment (the user to whom you are responding). There is a good reason why it is that way.
Being able to veto people replying to you sounds like a good recipe for the nucleation of echo chambers.
I don't believe there is any magic formula that eliminates trolling, yet respects everyone's freedom. The closest thing is the model of Usenet, with personal kill files. Everyone has a right not to read what they don't want: what they find disruptive, annoying, offensive or whatever. Everyone also has a right not to have someone else prevent them from reading what they want. The only way to achieve that is for everyone to have their own personal filter.
You get shadow banned when you annoy everyone, ending up in everyone's kill file. That is fair; you cannot point fingers at an unfair moderator. You were talking at a group of people, each of whom preferred not to listen. The weak spot in this idea remains, of course, uthentication; users changing their identity to evade being blocked by others.
An interesting concept might be bayesian filtering. Every forum participant has their own filter. Instead of blocking/killfiling users, you indicate to the system that some content is unwanted, and that trains your filter against that sort of content. A troll who uses multiple identities cannot evade that filter as easily, because to do that they have to change their content and style. And anyway, if thereby they cease trolling, that may well be fine.
Finally, even the perfect scheme still has the problem that people may be harming themselves if they shun whatever they disagree with. What is a "troll"? Sometimes it's actually just a bearer of an uncomfortable truth, perhaps without the best of manners. But are the manners the real reason you want to block the troll, or the uncomfortable truth?