Private School Labeler on Bluesky | Hacker News Reader

Private School Labeler on Bluesky

simonwillison.net

40 pointstosh13 comments

This coule easily be abused, however, it strikes me as a good balance between speech & reach [1]

If you are nobody, you don't really need to worry about a label. But if you are a politician with some agenda, you can't take those labels off you. If you have a broad reach, you must bare those labels everyone put upon you, this imitates IRL social pressure to an online virtual world.

I wonder if labels could be reversed, see who labeled me, and dispute wrong labels? The "reach" is an unexpolored setup in social networks.

[1]: https://atproto.com/guides/overview#speech-reach-and-moderat...

> On the one hand, there are clearly opportunities for abuse here. But given the opt-in nature of the labelers, this doesn't feel hugely different to someone creating a separate webpage full of information about Bluesky profiles.

It requires very little imagination to see how this could be abused. Does the bluesky client also display labellers a user subscribes to?

Then at least it would work in both directions and it would be obvious who is using a labeller to label people for distasteful reasons.

I'm afraid you're going to need to spell it out to me. What are the opportunities for abuse here that aren't already present in the form of private WhatsApp groups or shared Google Docs with lists of accounts and abusive notes about them?

I'm not saying there aren't any, but my imagination hasn't been able to come up with anything more than "it makes it slightly more convenient for co-ordinated harrassers to remember who they were planning on harrassing".

I think there's more than a quantitative difference between "slightly more convenient" and "with no effort". A centrally maintained list of labelled "enemies" that you get forever with a click is materially different than having to research people who gave you a reason to do so.

Using your logic, a shared blocklist could also be a way to "harass" people.

Is your thesis that since we live in a world of infinite possibilities, any device that makes harassment more efficient is not worth measured consideration for guard rails because there’s always a different and worse way to it?

The labels feature was originally designed to help PROTECT against harassment - people being harassed can use labels to share notes with each other about hostile accounts.

Does that positive use of the feature outweight any potential negative uses?

It doesn’t enable anything a browser extension couldn’t do as well.

> It requires very little imagination to see how this could be abused.

Pray tell, how do you feel how labelers can be misused as opposed to an app like TweetDeck etc that simply leverages the features of the platform?

This got flag killed when I linked directly to the labeler: https://news.ycombinator.com/item?id=42196313

HN is confusing sometimes.

Could this be worked around to add private notes to people’s profiles? My fav feature on RES was the ability to “tag” accounts on Reddit

On the surface, it sounds useful, but this sounds like a case where the potential harms greatly outweighs any other use, inc. doxxing, harassment, bullying, libel ... Unfortunately, there's no way to stop it cold.

I still don't get it. Right now you could create a Google Doc full of malicious information about one or more Bluesky profiles. Why is this different?

To clarify: these labels are maintained in systems outside of Bluesky, and are only visible to users who deliberately "subscribe" to a specific labeler.

Reply on news.ycombinator.com