I think Meta’s AI strategy is to advance AI/ML that is beneficial for them, no more charity.
They need a good-enough LLM (llama) to cut content moderation costs, they need a good segmentation model (segment anything) for photo filters, AR/VR and photo/video content moderation.
For LLM frontier, they can wait it out to see AGI become a commodity they can buy after it is ready.
Just to say the obvious: the new CEO is a VC partner and former CEO of Automattic. That seems very bad, no matter how "committed" they are to the vision of Bluesky.
Ultimately the goal "build a nice community where people can enjoy social interactions" is fully incompatible with "build the next Everything For Everyone Social Website like twitter/facebook/instagram/youtube/tiktok/etc so that we can get 5 billion users and start pushing ads at people". Unfortunately once you take VC funding, you no longer have the option of doing the former.
From an actual content perspective Bluesky is fine, but there's no investor who would take a look at the site's user statistics[0] and say "oh yeah things are going great." There needs to be drastic changes if investors hope to have any return on investment.
Another problem is that Twitter's demise left people who liked the format disenchanted and suspicious (and rightly so), and because of that, trying to recreate Twitter is bound to fail, at least until some more time passes.
I just checked https://apps.apple.com/us/iphone/charts/6009 and X (formerly twitter) is the #1 news app followed by substack, CrimeRadar Dispatch Audio, and coming in at 4th place is reddit.
So if twitter's dead, what does that make reddit, 3 spots behind it? Well, not dead, obviously. Pretending that twitter is gone or dead is just not rational behavior.
That's a two year old article and X is the #1 news app today. How can you possibly construe that as "shrinking fast" if two years later it's in the top spot literally today? It seems like wishful thinking on your part rather than being reasonable based on first principles and the data at hand, from where I'm sitting.
The ranking compiled on Wikipedia is from a couple months ago. X is now behind Pinterest, Reddit, and LinkedIn (and of course, the major social players). Also, why would it be "wishful thinking"?
BTW, on US iOS App Store, Claude today is way ahead of Chrome, Google Maps, YouTube, Gmail, TikTok, Instagram, and WhatsApp. Surely you don't think it means it's used by more people.
Again, the ranking from Apple is from today. Not "a couple months ago". And again, it's #1 while reddit is #4. So which is incorrect, Apple, or Wikipedia? It has to be one of them.
>Also, why would it be "wishful thinking"?
Because you've Motte & Balley'd twice now, each time in the direction of downplaying X's success. Because X is objectively doing great. #1st place is objectively great.
>BTW, on US iOS App Store, Claude today is way ahead of Chrome, Google Maps, YouTube, Gmail, TikTok, Instagram, and WhatsApp. Surely you don't think it means it's used by more people.
Are any of those news apps? This is the third Mott & Bailey. Again in the direction of denigrating X with bad data. So first principles and neutral data sourcing cannot be the reason for the inaccuracies - I dare say lies. It's flailing at this point.
The discussion is about trends, and the most recent datapoint shows it at a point of primacy when the argument was it was in decline two years ago. Clearly that claim was incorrect.
One of Twitter's strengths was that it was a constructive community where near services & informational/radiative bots chilled. It was a connective fabric, it made information available.
That's all been gone. The algorithm fav'ing paid blue check users massively made things worse from there.
Bluesky attempts to be better on all fronts here. Interesting apps/services are welcome, permissionless. There is no top down pro-facsism pro-racism pro-MAGA finger-on-the-dial algo-shaping.
Sure there's some who will just be burned out & not interested. But there's so many interesting structural safeguards & such a openness to play & creativity & tuning... I really encourage folks to give it a time. I would definitely hope that "bound to fail" is perhaps not a cast die, that, we tried something great once, it's gone, never again, is not how this works.
I remember being in 20 years old, at the start of my career, and complete broke. I thought Twitter was just a toy website, until one day I radically changed my mind.
I was a customer of a bank that treated me with nothing but contempt. Whenever I called the bank because of a problem, I would stay on the line forever to eventually talk to an unbothered representative. One day, instead of calling, I complained on twitter and tagged the bank. Half an hour later the bank apologised and fixed my problem.
Elon Musk is all of those things personally (see: that heil-handing incident), and it's public knowledge that he made the Twitter engineering team amplify his own posts.[1]
For whatever faults the old Twitter had pre-Musk, it did establish a certain critical mass for a certain type of short form threaded discussion which seems to be largely dead at this point.
Jack seemed interested in the protocol side of the house and making a good product that was in spirit of the internet, it also had a mix of people you might know IRL (with reasonable privacy defaults) and official sources/public figures. I don't think he was much interested in the censorship, it feels they got run over by a bunch of activist types during covid (who decided it was de rigeur to censor real doctors for perceived 'misinformation'). Jack started work on Bluesky and now is involved in Nostr
Speaking personally, supposedly Twitter now (X) still has a bunch of censorship and I don't especially like Musk (but what he did was valuable, showing Jay Bhattacharya he'd be put on a trending blacklist) and the site is... well, I should be able to follow threads without having an account but they crippled it so much. It reminds me of Instagram, "log in to see any PUBLIC page"
Court orders are different from psyops. The Bhattacharya thing, and the entire narrative around that stuff was essentially a psyop. Geo-restricting a Turkish politician because the authoritarian govt arrested him and gave a court order to restrict his account is - first of all - what all social media platforms must adhere to legally, and second of all is currently being contested by X in court.
I don't know if it was a psyop. Twitter banned tons of people during covid for completely ridiculous reasons, unless it was somehow all forced by court order or something
It was the censorship. Which isn't gone, but is way less restrictive than it was. And I've actually started seeing bangers (auto) translated from Japanese, French, Spanish, and Portuguese lately, which is fantastic. I don't really want an English hivemind. I want to see what the whole world thinks. It's kinda fantastic tbh.
Who needs censorship when you have an algorithm to feed people what you want them to see and you've self-selected for only people who aren't morally opposed to the new site?
Don't be fooled into thinking you're getting a dose of unfiltered reality on X.
>Who needs censorship when you have an algorithm to feed people what you want them to see and you've self-selected for only people who aren't morally opposed to the new site?
It feeds you what you engage with, and it changes surprisingly quickly. It caught onto my ARC raiders interest almost instantly. I engaged with a Portuguese post once, and now I get wonderful translated posts in Spanish, French, and Arabic too.
>Don't be fooled into thinking you're getting a dose of unfiltered reality on X.
What evidence could you possibly have that I'm not? There's lots of "politically incorrect" things which is a symptom of low filtration. Besides, you can't have seen my feed. Completely baseless allegation. So what's the real reason for taking the anti-X stance?
> There's lots of "politically incorrect" things which is a symptom of low filtration.
Politically incorrect things might be a symptom of low filtration on almost any other site, but not one run by Elon Musk. He has a clear agenda and is not shy about putting his finger on the scale at X. It's so blatant and well documented that it's almost hard to imagine you could be commenting in good faith.
>Politically incorrect things might be a symptom of low filtration on almost any other site,
Why would that change anything? I've always found political incorrectness to be a symptom of free speech.
>but not one run by Elon Musk.
Why would that be any different? Same symptom. Same free speech as far as I can tell.
>He has a clear agenda
What's the agenda?
>is not shy about putting his finger on the scale at X.
What instances of him putting his finger on the scale do you have? He gets community noted hilariously often.
>It's so blatant
What makes it blatant?
>well documented
By people who clearly hate the man and have lost their ability to reason over it. Like the ones who lost the narrative control of twitter.
>it's almost hard to imagine you could be commenting in good faith.
Having different opinions than you isn't bad faith. I brought up that the censorship is better than before (but still not great), and mentioned some cool new developments I've seen. You've attempted to steer the conversation to be about Elon Musk or myself. These are both ad hominem attacks, which is textbook bad faith.
That user number is not reproducible last time I tried (~8 months ago). I was looking at the code the other day and saw what I believe is one of the reasons, but I still couldn't find several million accounts (>10%), which is pretty hard to lose. (8+ bsky run pds equivalents)
This also does not account for (1) people with multiple accounts (labellers, feeds, bots, intent) or actual activity (significant % are likely churned, didn't delete)
Even if the stats are off by a factor of 10 it wouldn't matter. You could remove the numbers from the user axis entirely and it would paint the same story: there were massive user influxes after the 2024 US election and the inauguration in January, but user retention has been on a steep decline for the year+ since.
Again, this is not a reflection of anything bad about Bluesky as a user. IMO a smaller and more focused is a good thing for the actual community, hence why I read/post on HN and not Reddit or Twitter. However as an investor there's basically no way to interpret those statistics as anything but bad.
Strong agreement, though I would say it looks to have reached a stable level for now. I've found several subcommunities that I can get good info from. I'm curious how the '26 election cycle will affect things, already seeing increase political discourse.
I don't think it's quite stable. The external events that caused waves of new users to arrive from X are getting rare and bringing in fewer. When those aren't happening it's been a slow, gradual decline.
>From an actual content perspective Bluesky is fine
From a content perspective nothing important is permitted to be discussed there. It's just another hivemind with the exact same opinions as reddit and HN. Completely pointless and nothing more than the output of a temper tantrum over not getting to be the censors in charge and the whole world knows it.
To be honest, I was never entirely on board with Jay's almost exclusively cryptocurrency background. I think she's done an acceptable job as CEO, but I have also felt that leadership at Bluesky was never good enough to see legitimate success.
Today, Bluesky remains largely undermoderated and they have managed to bake in more toxic features Twitter ever did in such a short timespan. Its success is largely driven by having a UI closer to Twitter's original UI than any other alternative, and taking a stronger stance against far-right rhetoric than Twitter.
The only technical saving grace is the broad control you can take over the algorithm to avoid the content you don't want to see, but Bluesky is generally covered with more calls for violence than their nascent content team could ever actually deal with.
And I have yet to actually see a real use of ATproto that isn't just immediately blown out of the water by ActivityPub.
But I digress, the new CEO pretty much hammers that final nail in the coffin for me. I have zero belief in Bluesky to be anything but another awful corporate corner of the web that I should avoid.
> Its success is largely driven by having a UI closer to Twitter's original UI than any other alternative, and taking a stronger stance against far-right rhetoric than Twitter.
These things are very valuable, and if Bluesky can't succeed doing them, I hope someone else can.
> And I have yet to actually see a real use of ATproto that isn't just immediately blown out of the water by ActivityPub.
ActivityPub doesn’t remotely even try to solve problems solved by atproto. What are you talking about?
In short, atproto makes apps interoperable by default by decoupling data hosting from applications. This means that apps become projections of everyone’s data, and can embed and interpret typed data from each other. ActivityPub doesn’t offer anything close, which is why you don’t have projects like http://leaflet.pub, https://standard.site, https://tangled.org, https://semble.so in the AP ecosystem.
If you genuinely want to learn about atproto, I have two longreads for you:
Fair enough, I've never been involved in a CEO recruitment, I can't imagine the candidate pool tends to include people like the previous CEO of Bluesky
There was some WP drama between Automattic and the WP community a while back.
Also the whole point of Bluesky is that they aren't supposed to be a big evil silicon valley tech company. But now you have a silicon-valley VC running the thing.
> As Bluesky matures, the company needs a seasoned operator focused on scaling and execution
Translation: enshittification
That’s the other shoe where they will iterate on ways to monetise the party. Ads, paid “verification”, making users pay to use atproto apps (or making developers pay to use the managed storage)… the sky is the limit.
In a way I’m happy Bluesky never took root and outside a few enthusiasts in my bubble it’s practically unknown.
Yes, but its gonna be weird. I've done very tangentially related work with game controllers, and my suspicion is you're gonna have to try a few (dozen?) before you find one that even sends reasonable commands. I can imagine a FireTV stick sending game controller inputs. I have no idea, but that's the type of thing that might go wrong.
A MX3 air mouse might be the exact thing you want though.
Obviously not the main point, but I've been reading watch media online for over a decade now, I've read or heard this "Quartz Crisis" story hundreds of times and never ONCE read about the coincidence with the Bretton Woods agreement. Makes sense though, its basically oral history.
What’s shocking about it? Seems like the usual culprit— a bad config rollout. Took a long time to identify, so maybe that’s shocking. But I can attest that sometimes, you get into fight or flight mode and miss the obvious when trying to diagnose a disruption like this.
That said, nowadays, the first thing I do is spawn an agent to look through the most recent commits and try to identify something that could be the cause of a service outage.
This one seems like something Claude Code or Codex would have quickly flagged.
Agreed, we've all been there, but 4 hours! For a network config change. No one raised their hand and said "hey I just toggled this thing maybe we should look, I did it exactly when our entire region went had down"
