Hi @dinosaur this is an excellent and important proposal. Thank you for spelling it out so clearly and opening the topic for discussion.

actionable, specific, and kind communication

I love your emphasis on kind communication.

Two questions -

• Do we also want this to be automatically actionable (eg I set my client to never show messages flagged as "abuse") or does it always have to require human intervention?
• How do we stop potential abuse of this mechanism?

Also have you seen this related discussion? It's about using tags and/or flags to voluntarily tag/flag posts as NSFW and hiding (completely or requiring a click to reveal) this material. There is also the suggestion that the same mechanism could be used to hide posts flagged/tagged with a Content Warning.

It feels like both proposals might possibly benefit from being merged if only to make the user experience less confusing.