Luca RossiApr 6, 2024
https://www.nature.com/articles/s41586-024-07229-y
It looks like that if you remove all that makes them different... All platforms look the same. I'm probably wrong but isn't this expected when you don't count or normalize everything that makes them different?
Persistent interaction patterns across social media platforms and over time - Nature

Long conversations online consistently exhibit higher toxicity, yet toxic language does not invariably discourage people from participating in a conversation, and toxicity does not necessarily escalate as discussions evolve.

Nature
Show threadbastosApr 6, 2024
@LR I’d still expect to see more variance in the toxicity indicator.
Show threadLuca RossiApr 6, 2024
@bastos usenet is also a selection of groups - alla fairly toxic - and, besides few good places usenet was a family toxic place with very strict culture full of Rtfm and similar stuff.
Show threadbastos
@LR I think what they’re getting to is that the diversity of opinions is a driver in toxicity because it’s a requirement for polarization. It’s not all that surprising. It just reads like a surprise because most social media research is ahistorical. They nailed that by covering a few decades so this becomes abundantly clear.
Apr 6, 2024 at 8:49amWeb
Show threadLuca RossiApr 6, 2024
@bastos that's an interesting observation but far from unexpected. Framing this as "platform doesn't matter" is stretching it a little too much.