The Bluesky Bubble: This is a relapse, not a fix
theatlantic.com
theatlantic.com
The irony of publishing an article intended for a mass audience about how people shouldn't talk to each other. Obviously journalists are the only ones who can be trusted with such power...
Both of them are people putting/ publishing text out to a mass audience, in a public way.
Some interesting feeds I've seen are only posts by mutuals, lots of curated feeds that just show a subset of posts by a small number of accounts (whether or not you follow them), a following feed that filters out reposts/replies, a custom open source Discover algorithm based on your likes.
This feels way more interesting to me than anything Twitter/X has ever tried to do, even if you really like their single algorithm and the way their following feed works.
I have fully switched to Bluesky at this point anyway, but the list approach worked fine for me on Twitter long after most users were complaining about the feed algo.
Back when I was younger and addicted to Reddit I remember unfollowing all my subreddits and hacking together a little script that hid the follow button. That friction made Reddit a lot less desirable.
I’ll no doubt eventually try out Bluesky, but I don’t think it could ever capture Twitter’s magic pre-Elon.
Counterexamples include LinkedIn, HN and various subreddits (not all) - they're operating at large scale, very lively but low levels of nasty (as a percentage).
IME, LinkedIn suffers from a different problem: on there, "everything is awesome", and the toxic positivity embedded in its DNA always snuffs out any authentic rancor.
So yes, it's free of bad behavior -- but it's also free of "normal" conversations. It's just full of "thought leaders", "visionaries", and "visionary thought leaders".
haven't all of us of a certain age participated in a forum that eventually devolved into drama, and then everyone picks sides, and someone makes a new forum, and half the group leaves?
That's all that's happening on X. Now we have a bunch of twitters: X, Bluesky, Truth, Fediverse.
They're all twitters and the X exodus is just a twitter split. X will go on, but smaller, and people will share posts between the twitters the way they always have shared posted between social media sites: with screenshots.
Nature is healing.
He has the resources to keep it going for years as a forum for his groupies, but if anyone else were in charge it would be a die-off for sure.
I don't think Bluesky will succeed unless it offers something different to the basic Twitter experience. It's hard to picture enough people giving up Twitter (or using both) for it to reach the critical mass required to sustain it.
If you changed your status on the old messengers e.g. "making a cup of tea" then it had some utility at any rate, a status was a kind of temporary notice, to be replaced by another status at another time. In an IM application you would see your online and offline contacts and see their status underneath their name. People had a few different unconnected IM applications. You could use the API to set your status in twitter and it would be set in AIM (or was it both ways?)
You could SMS in a status/tweet from a phone, many people loved that.
After that it became a way to record or broadcast other temporary notices, and then to a kind of realtime status. The concepts of a conversation happened then, talking about a topic but without direct refering other people and then with mentions @handle threads on twitter happened afterwards.
The twitter tech folks were considering making the metadata around a tweet as more important than the inside content. (e.g. RDF, foaf, geo, links, wikipedia uuids, subject topics, etc etc)
And then the masses joined in a few years after with Oprah and it was all about the message.
After that the push to VC eternal growth and the algorithm powered feed was what encouraged bad behaviour and a range of un-satisfactory responses to that behaviour which continues to this day.
I wasn't persuaded, because I remember ello.co [0] (which shut down in July 2023). It too was run by a public benefit corporation, but it couldn't sustain itself because t-shirt sales and affliate links weren't bringing in enough revenue, and they never got their freemium model off the ground.
No great loss, which is what I'll say when Bluesky starts to rot because its operators need to provide ROI to the VCs backing them [1].
[0]: https://en.wikipedia.org/wiki/Ello_(social_network)
[1]: https://decrypt.co/288157/bluesky-raises-15-million-no-crypt...
"About the Author
Ian Bogost is a contributing writer at The Atlantic."
As an oldlennial, I am not sure I care for this term.
Meanwhile, my YouTube feed regularly gets attacked with AI generated crypto ads using the Tesla or SpaceX brands.
All social networking sites seem inevitable till they fail, remember digg?
Bluesky and threads are building interesting sets of communities, and with the number of active users they have now, it's likely that they'll stay there.
There are 15k+ starter packs on the directory now https://blueskydirectory.com/starter-packs and just a quick scroll through them shows the diversity of groups using the platform.
Bluesky appears (for now at least) more open to API driven interaction and doesn't default to an algorithmic view, avoiding some of the more annoying things on twitter.
Now as they grow there will be inevitable problems with scammers/spammers and the rest but all popular sites that allow user generated content have to deal with that eventually.
Is there such thing as accurately "reflecting society's pulse" within individual posts? Sure, maybe you can pull big data to understand public sentiment on certain topics or events. But the platform's goal is to present users with content they are most likely to engage with, so that they spend as much time on the platform as possible (and therefore view as many ads as possible). That content does not reflect society—it's often targeted ragebait. When users are goaded en masse by ragebait and engagement farming, some percentage are going to develop a distorted view of public sentiment, and their own engagement will further the issue.
The fact is that content discovery algorithms and moderation largely shape the tone and personality of a platform, both for a viewer and for a contributor (who is psychologically inclined to want to fit in).
Twitter isn't going to collapse and die overnight, but this could be the start of a 5-year trend where the "utility" of Twitter over other platforms gradually shrinks as reputable brands and public figures diversify their presence.
Assume it’s true for a min, why does a society need a website reflect its pulse? It is useful for company and govt to harvest the data, but I failed to see any meanful way common individuals will benefit much from it.