Post

This is definitely a great feature, but there's something about it that makes me uncomfortable. Couldn't this be used for harassment? Imagine an app that "labels" people in derrogatory ways.

1 repost12 likes

It's always up to the user to decide what labels they're going to subscribe to. I don't think there would be any turned on by default. Obviously you have the problem that the designer will subsequently decide to change it in unwanted ways, or will be convinced to hand it off to someone malicious.

Yeah I'm not too worried about that since seeing these labels is an opt-in thing - it's not much different from someone having a website somewhere else that says nasty things and links to Bluesky profiles

Is that materially different from a hate group using a private WhatsApp group or publishing a Google Doc with a list of accounts to harass? Maybe in turns of making it slightly more convenient for them to remember who to target

Seems like this can maybe counteracted by using a blocklist of people who have subscribed to the labelling account. I wonder if this is technically feasible (generating a automatic list of people subscribed to an account).

otoh, i don't think WHICH accounts have been labeled by a labeler is public, and if a bad labeler were to publish a list of accounts, you're just back to the "mean site" problem. (i don't necessarily disagree with you; i'm just thinking out loud)