What makes you think this is a terrible feature. This is a federated service and the company is actively attempting to get the service to a state where it can exist independently of them, preferably without the service fracturing into a bunch of islands, mostly separate from each other (cough mastodon cough).
What solution would you propose that's better than this one and how is this solution a net negative (personally I think labelling services are a great approach to the problem)?
Community notes have been very effective in retaining public trust while combating misinformation on Twitter/X.
This is the polar opposite of that. Delete all opinions you disagree with, no matter the reasons.
Echo chambers are not productive. Full stop. It will lead to worse outcomes.
I would have a different opinion if they weren't advertising this as a general-purpose way to shut people up. This could work if it were just a method to remove "rule-breaking" posts where some authority would still evaluate the reason for removal.
Community Notes, and their abuse, is one of the reasons I left Xitter. I was a CN editor and saw that it just became another battleground for ideology
Bluesky, with ATProto, is putting more control into the author and user hands. You may not like this design change, but many people do. The nice thing about ATProto is that you can write a different UI that doesn't hide or allow for this, and still belong to the social network fabric at large. You don't have to rebuild a new, separate social network from scratch
They already stated they are planning on adding community notes along with further extending their labelling services.
Community Notes doesn't address dog piling though. And unlike twitter, Bluesky & ATProto have no concept of private accounts so when you get dog piled you can't just go priv like you can on twitter until things cool off.
This is specifically addressing that problem. i.e. Addressing toxicity. It's not about trying to correct misinformation.
The post mentions misinformation in this section only in the context that QRTs are a common way of addressing misinformation and that allowing you to detach could reduce QRTs ability to do that.
So while Community Notes should be a preferred method of addressing misinfo, it doesn't do anything to allow users to disengage or avoid toxicity (which is the point of this blog post).
What solution would you propose that's better than this one and how is this solution a net negative (personally I think labelling services are a great approach to the problem)?