←back to thread

Bluesky is ushering in a pick-your-own algorithm era of social media

(www.newscientist.com)
190 points amichail | 2 comments | 20 Nov 24 12:59 UTC | HN request time: 0s | source
Show context
hntempacct99 ◴[20 Nov 24 15:16 UTC] No.42194606[source]▶
From what I have seen using Bluesky this isn't true at all. It's brutally censored, even more than Twitter was in 2021. Or are there other relays and appviews I can use that aren't? Is there a comprehensive list of Bluesky infrastructure that isn't run by Bluesky themselves (excluding a PDS)? Or is it totally centralized for now?
replies(10): >>42194628 #>>42194642 #>>42194671 #>>42194682 #>>42194690 #>>42194767 #>>42194774 #>>42195031 #>>42195409 #>>42195473 #
happytoexplain ◴[20 Nov 24 15:20 UTC] No.42194642[source]▶
Wow, "brutally censored"? This is the first I've heard. What are you referring to? (I don't use Bluesky).
replies(1): >>42194808 #
1. hntempacct99 ◴[20 Nov 24 15:36 UTC] No.42194808[source]▶
Check what the default moderation service is flagging. I understand that the standard Bluesky site moderates however they want, and that's fine, but this is a decentralized network right? So a productive discussion is to discuss what other relays and appviews are currently running where the users can pick and choose that algorithm, as content exclusion is perhaps the single most important part of any content algorithm, and the defaults on standard Bluesky are pretty locked down.
replies(1): >>42195439 #
2. happytoexplain ◴[20 Nov 24 16:21 UTC] No.42195439[source]▶
This? https://bsky.app/profile/moderation.bsky.app

These looks like reasonable defaults - frankly I'm a bit delighted they are configurable. A lot of these are things most social platforms would outright ban without an opt-out. I think it makes sense to start medium-narrow and let users broaden it (not to mention it's kind to new users - though I understand that kindness is a bit dead in our culture currently, since it's been falsely accused of being mutually incompatible with having hard, real conversations). And I do get the pros and cons - I get the argument about starting broad and making the user narrow it down. But, specifically, I think "brutally censored" is pretty dramatic.