Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Visualizing 13M Bluesky users

joelgustafson.com

Post reply on HN
Print view View original

71–80 of 252 posts

Re: Visualizing 13M Bluesky users

#71

Post by rsynnott »

andreygrehov wrote: post #31
jamwil wrote: post #23

Earlier quoted context omitted.

Hate speech is derogatory or harmful speech targeted at a marginalized community. Hating a company or product is not the same thing.

I just clicked "Discover" to see popular posts. https://bsky.app/profile/lepapillonblue.bsky.social/post/3la... https://bsky.app/profile/sethabramson.bsky.social/post/3larj... https://bsky.app/profile/browneyedsusan.bsky.social/post/3la...

… So, wait, two political cartoons (the first one’s been on the go for years; it’s pretty much a classic at this point) and a cat? Like, unless you’re going to claim that ol’ minihands is a marginalised group of one, very hard to read any of these as hate speech. Actually, even if one _was_ to accept that obviously silly premise, still probably not. Maybe the cat, at a serious push.

Re: Visualizing 13M Bluesky users

#72

Post by zft »

typical182 wrote: post #40

Bluesky and atproto seem to be built to be hackable. Someone in the community recently built a searchable directory of Bluesky "Starter Packs" (which are a way for a user to publish a set of interesting people & feeds to follow, primarily to help newcomers bootstrap their experience): https://blueskydirectory.com/starter-packs/all Dan Abramov posted about it earlier today, saying he liked it and: "the fact that it ca…

Starterpacks are great, what about seeing what is current top engaging accounts ? https://www.graphtracks.com

Re: Visualizing 13M Bluesky users

#74

Post by andreygrehov »

CuriousCosmic wrote: post #54
andreygrehov wrote: post #18

The visualization looks absolutely fantastic. Great work. Would love to dive deeper into the tech behind it. I'll just share some irony. They say X/Twitter is full of people spreading hate speech. I just logged in into my old BlueSky account. My entire feed is full of people saying how much they _hate_ X/Twitter.

That's a current wave going through. There's a bit of discourse going on where some users were rage baiting by constantly posting twitter screenshots over to bluesky and people who left bluesky are pushing back about how they don't want to see that content. That discourse comes in waves each time a major migration happens from twitter to bluesky but it settles down fairly soon after each time. Give it a few days and…

That's a reasonable explanation. Thanks.

> Also it's worth noting that the "Discover" feed is trained specifically per person so while the defaults aren't great, if you use the "show more like this" and "show less like this" options on posts (under the triple dot) then pretty quickly it tunes in towards content you care about vs content you don't.

This is the first time I clicked "Discover". I haven't logged into BlueSky for almost two years.

Re: Visualizing 13M Bluesky users

#75

Post by jsheard »

nuz wrote: post #55
typical182 wrote: post #40

Bluesky and atproto seem to be built to be hackable. Someone in the community recently built a searchable directory of Bluesky "Starter Packs" (which are a way for a user to publish a set of interesting people & feeds to follow, primarily to help newcomers bootstrap their experience): https://blueskydirectory.com/starter-packs/all Dan Abramov posted about it earlier today, saying he liked it and: "the fact that it ca…

Another argument is that it could lead to a bot problem 10x worse than twitters ever was.

The Twitter bot situation only seems to have got worse since they shut down free API access. LLM engagement farming bots everywhere in replies, hordes of scam bots replying if you use certain keywords, porn bots following and DMing everyone non-stop...

Evidently the people running the bots don't really care whether or not you give them an API to work with.

Re: Visualizing 13M Bluesky users

#76

Post by Karrot_Kream »

TinkersW wrote: post #67

[flagged]

Huge social media sites are not going to be legible without some sort of preference function from the individual. Whether that's follows, hashtags/keywords, or engagement-oriented algorithms, that's the way. If you're not the kind of person who enjoys socializing In The Large that's fine. That's what topical sites (like Reddit, HN) or group chats are for.

Re: Visualizing 13M Bluesky users

#77

Post by openrisk »

> Building and querying the quadtree is intrinsically heirarchical

Glad to see I am not the only one having problems with hierarchy.

Interesting work at many levels (no, no pun): starting with the bluesky data availability, the processing and the visualization algorithms.

But its not quite clear where to place these visualizations in the data science spectrum. Conventional numerical graphics have (over time) developed a sophisticated grammar that allows fairly precise reasoning and inference. So they are heavily used in scientific publications, in the financial sector etc. for real information transmission (People might even reverse engineer a plot to recover data!).

With networks and graphs, besides a general feel for the topology / connectivity or clustering its kinda hard to pin down what is the transmitted information. Not clear if useful grammars covering such large graphs are yet to be invented or if this is the nature of the beast.

Re: Visualizing 13M Bluesky users

#79

Post by throw7 »

jamwil wrote: post #23
andreygrehov wrote: post #18

The visualization looks absolutely fantastic. Great work. Would love to dive deeper into the tech behind it. I'll just share some irony. They say X/Twitter is full of people spreading hate speech. I just logged in into my old BlueSky account. My entire feed is full of people saying how much they _hate_ X/Twitter.

Hate speech is derogatory or harmful speech targeted at a marginalized community. Hating a company or product is not the same thing.

So derogatory or harmful speech toward white people is perfectly fine.

Re: Visualizing 13M Bluesky users

#80

Post by sneak »

nuz wrote: post #55
typical182 wrote: post #40

Bluesky and atproto seem to be built to be hackable. Someone in the community recently built a searchable directory of Bluesky "Starter Packs" (which are a way for a user to publish a set of interesting people & feeds to follow, primarily to help newcomers bootstrap their experience): https://blueskydirectory.com/starter-packs/all Dan Abramov posted about it earlier today, saying he liked it and: "the fact that it ca…

Another argument is that it could lead to a bot problem 10x worse than twitters ever was.

Computers that are operated by humans that are using APIs designed for computers is not a bug, but a feature.
Post reply on HN