Let’s hope human cloning will arrive within dang‘s lifetime.
Re:Trust-Based Moderation Systems
by capableweb on Sunday December 10, 2023 @01:45PM
(#38593764)
Meh, fine-tune a LLM on dang's comments and call it a day. Ship early and ship often, adjustments can be made as we discover it doing the wrong thing.
Re:Trust-Based Moderation Systems
by pavel_lishin on Sunday December 10, 2023 @02:52PM
(#38594383)
Funny, but taking it seriously: it should be trained on dang's moderation actions, not comments.
Re:Trust-Based Moderation Systems
by capableweb on Sunday December 10, 2023 @02:56PM
(#38594424)
Most of dang's comments (that I've came across at least) are moderation actions, like telling people how they're not following the guidelines and so on. But yeah, also the actual backend moderation actions should obviously be included in the training set.
A sort of Dang Acevedo? :)
Re:Trust-Based Moderation Systems
by MichaelZuo on Sunday December 10, 2023 @07:51AM
(#38591191)
Metafilter also seems to work fine. Just charge money for each account creation, eventually the repeat trolls will get tired of paying over and over again.
Excellent at allowing the same 10 guys (and their alts) to dominate almost every discussion. For truly serious topics, this place isn't all that different. Manipulation is a powerful weapon.
I've never seen this to be the case. Could you elaborate on how dang does this?
Re:Trust-Based Moderation Systems
by mistermann on Sunday December 10, 2023 @11:17AM
(#38592556)
Excellent at producing a specific outcome (optimizing for certain variables) - whether that specific goal of an outcome is anywhere near optimal in a comprehensive sense is another matter entirely.
Of course, dang based moderation also works well but you need a dang for that.