I don't understand how Bluesky is going to continue to exist past 2026, based on the sharply declining usage, their headcount, and the amount of funding they've taken.
It has apparent value propositions past the social network, but none of those use cases are visibly taking off and none of them appear to be monetizable. The social network itself is what will be evaluated when they go out for more funding. And I don't see how you can raise at all for a social network in 2026 with flat numbers, let alone the declining numbers Bluesky actually has.
I've been dual-wielding Twitter and Bluesky for about a year (after a year off Twitter where I was mostly Mastodon), and, anecdatally, we've hit a point where the engagement and volume of stuff I see in Bluesky is lower than what I was getting even on Mastodon. Earlier on, there was some truth to the idea that Twitter had a much larger audience, but you'd get better engagement on Bluesky. I now get better engagement on Twitter. I can see people I had followed into Bluesky moving back to Twitter.
I have no idea what's going to happen, but I'm curious to hear a coherent story about how Bluesky isn't cooked.
Usage has absolutely declined from peak switching periods where inevitibly some users won't stick around, but that's to be expected. Most stats seem to be leveling off (which isn't exactly stable growth either so the rest of your points stand).
Yeah, I think what you needed to do here is zoom out. That's a sharply declining chart.
I understand that as a Bluesky user the peak and dropoff doesn't hurt the experience. But investors are going to put money in with the expectation of a return and what they're going to look at are the derivatives of the adoption curve: how quickly is it gaining users, and is adoption accelerating?
What I've found is that most of the people I really want to follow are exclusively on bsky. BUT most of the people I'd prefer to casually follow either exclusively or mostly only posts on Twitter. And the 2nd group is a much larger pool than the first group.
I really dislike Mastodon so gave that up a while back. I know there are a few people I'd like to follow who only post there, but such is life.
I've found Bluesky does a really bad job at not showing me stuff I don't want to see. Furry p*rn on the "cute internet cats" feed? Yup. Heaps and heaps of political rage bait on the main discover feed? Yup.
I wanted to like BlueSky, but it's such a bizarre echo chamber of people who left Twitter for ideological reasons that it basically filters for people that I actively don't want to engage with.
Those types of people are still there on Twitter (mostly on the other side these days), but I don't see them because the algorithm filters them out.
> I don't understand how Bluesky is going going to continue to exist past 2026, based on the sharply declining usage, their headcount, and the amount of funding they've taken.
You can just Google these things. Nate Silver posted (grim looking) usage numbers --- followers and posters --- just a couple weeks ago. They very publicly raised a $15MM (priced) A round a year ago, after raising what I understood to be a comparable amount of seed funding. There was talk early this year of them raising again at a $700MM valuation, but I don't see a subsequent announcement that that happened.
You can generally take a headcount number and assign a fully loaded cost to it (say, $200k, conservatively) and just math it out. And of course that analysis assumes their infra expenditures round to zero.
So no, I'm not just making stuff up. I could be wrong! I feel like I was open about that.
Also, linkedin says 53 people, though crazies like to say they're employed places. It does say 29 in the US so that's more likely real. Assuming that skews eng a bit, that's probably (225 fully loaded eng, 150 fully loaded the rest) $6m in payroll alone. Not to mention server expenses. That's tough.
The case for more funding: 1 - yolo; 2 - non-economic investments; 3 - Musk gonna Musk, so expect Twitter to further shed users. There may be a business to be built there if you can run significantly more efficiently than Twitter which was a shockingly poorly-operated business (Zuck was dead on re: the clown car. Except maybe more like a silver mine, not a gold mine.). Oh, and 4 - the EU is pretty hostile to Twitter and Threads, so maybe there's some there there. Dunno.
Yeah, this is basically where I'm coming from. There are ideologues and stans arguing both ways whether Bluesky is a moribund platform. I don't think it's intrinsically moribund. But I think that's the wrong question. The more important question is: is it succeeding to the point where it will attract another round of investment? That case is I think pretty hard to make based on the available public evidence --- but the Bsky people obviously know a lot more about this than I do!
> I've been dual-wielding Twitter and Bluesky for about a year (after a year off Twitter where I was mostly Mastodon), and, anecdatally, we've hit a point where the engagement and volume of stuff I see in Bluesky is lower than what I was getting even on Mastodon.
What do you count as "engagement"? Views/likes? Those can be produced. Interesting conversations that aren't obviously LLM can't. That's the metric I use for anecdotally seeing Bluesky (and Mastodon) as immensely more engaging in a signal/noise ratio.
> I don't understand how Bluesky is going going to continue to exist past 2026, based on the sharply declining usage, their headcount, and the amount of funding they've taken.
Do you have any data to back this up?
Also famously twitter is losing value and advertisers and users still it’s not stopped twitter from existing
Regarding engagement doesn’t twitter have lot more low quality engagement vs Bkuesky
I have the opposite problem. My Bluesky (and Mastodon) feeds have gotten so active that I have a hard time catching up these days. And Twitter seems to be mostly full of bloviating narcissists who are addicted to feeling like “influencers” of the unwashed masses, not co-equal participants of public discussion.
More broadly, Twitter’s problem is that it carpet bombed the bridge with a substantial portion of its intelligent population. I’m pretty sure half the tech, game dev, and research people I follow on Bluesky/Mastodon will not return under any circumstance, myself included.
No. Government should be using a self-hosted instance of something standard and federated (ie, an ActivityPub implementation such as Mastodon). That can be federated out to the walled gardens (Twitter), vulture-capital places (Bluesky), etc as needed.
I think platforms like that will survive on essentially rage and bitterness. People will always flock to platforms like X, Bluesky, iFunny, and others to vent and rage. There's so much anger in nearly every post that I don't see any of it as particularly useful, or a good use of one's time.
Social media is bad for you. Though I'm not sure that social media enjoyed a "social media is good for you" phase. There are certainly studies I'm sure that are out there, but the general consensus seems to be negative.
Bluesky is a public benefit corp which at least nominally means they don't have to maximize shareholder value at the expense of everything else.
Decentralization is not a priority for most people. If anything, they actively want centralization, because it's easier. To get those people to decentralize, the solution will have to be dead easy and invisible. The AT Protocol being developed by the Bluesky people looks promising.
I don't work for Bluesky, I'm not on Bluesky, and I don't particularly care, but I found your comment unfair after reading about ATProto on HN literally yesterday.
Opening an account on Mastodon was a pain even for me, a long-time Internet user and a nerd. I'm confronted with a choice of servers, but if I love cheese and golf, do I open my account on the cheese server or the golf server? What will be the disadvantages of choosing one over the other? Instant blocker right in the sign up process that made me actually give up several times. If it was like that for me, imagine how it could be for an average user.
The Fediverse will only be popular if someone releases a client that makes it as easy to use as X and Bluesky. Not sure if it's technically feasible (I don't know much about the innards of the protocol) but it doesn't seem to have happened at the moment.
I did sign up for a mastodon account, but the server I used simply disappeared a couple of months later. I haven't worked up the motivation to bother doing it again.
> Opening an account on Mastodon was a pain even for me, a long-time Internet user and a nerd.
I was in the same boat as you, and my experience was completely different. Mastodon's federation model reminded me of IRC, except nowhere near as balkanized.
So how did I wrangle the supposed complexity? I started out on one of the main instances and just started people-watching. Over time, I took note of which server contained users whose content I enjoyed over time, then I just joined the server.
Joining the server got me a slower federated feed that was both more pertinent to my interests that also functioned as a de-facto community space. I also found the moderation to be more to my own preferences. But the bigger server wasn't _bad_, I just preferred the smaller server because it was more personable.
I don't use Mastodon much anymore, but that's more because a good chunk of my social circle left for BlueSky than any gripes I had with the platform. I don't know where they will go if BlueSky goes belly up, but I can tell you that it won't be back to Twitter.
This is the reason why it was a pain. You or I might think "this is a decentralized platform, I should look at all the servers and find the best one for me" and immediately get choice paralysis looking at who has what rules, who federates with who, who runs which one, etc. An average user will probably stick to the main instance and not even think about it.
As for another data point, opening a Mastodon account for me was as trivial as choosing one random link from a webpage, choosing a username, password, and entering my email. If you get frozen on the choice of instance, that has more to do with your mental process than with an effective difficulty for the average user.
If you really cannot go beyond your inclination, and since you are a a long-time Internet user and a nerd, why not host your own instance?
There's already way more stuff on Mastodon for the hashtags I follow than I can possibly consume. If this is "unpopular", I'm happy with the way things are.
Maybe it _should_ be a little tougher to sign up for than the other mainstream options.
They aren't exactly equal evils... and if a federated replacement for Xitter were going to win, I think it already would have. Ease of use (and comprehensibility) are both pretty important in this business, and nobody seems to get that.
Because Threads is Meta's attempt at bullshitting Mastodon users in welcoming a wolf among the herd. Search for "Fedipact": Meta is de facto cut off from many Mastodon instances.
The Twitter buyout should have made this obvious. When one company runs the whole thing, then some rich twat can wreck it on a whim, and there’s nothing you can do about it.
> Trading one corporate overlord for another is not the smartest play.
Only if you expect to be there for ever even as they inevitably enshittify. Be under no illusion that although BlueSky is having its "first they are good to their users" phase now, it is temporary.
So make the most of it now while it's good, but be prepared to move on when that changes. Embracing impermanence is a smarter play. This is nothing new, thus passes all social media.
X still appears above the fold in a special section on Google with custom previews.
Eg. searching 'Anthony Albanese Bluesky' for Australia's leader has a link to X, with custom integrated previews, above the Bluesky link of the PM despite the search explicitly stating Bluesky and despite the account posting to Bluesky.
It's hard for anyone to move over since the lack of engagement is rigged like this.
Btw if you set your browser default search to anything other than Google searching for people works much better. No customized bumping of X based links over other options when searching for people.
Duckduckgo and Bing put the bluesky link as #1 for the above. Seems straightforward to make the switch to me. If you haven't changed your browsers default search engine in the past 5 years now's a good time to do so. Much better results await.
kagi.com is a superior search engine compared with all three, to the point where I pay for a license for everyone in my company to use it.
The kagi assistant is also nice, only responding with AI when you add a question mark to your query, with the option of opening the query in a separate web search RAG w/the LLM of your choice
Interesting, maybe it also shows quality of search. I know that these are barely relevant compared to google, but Bing, Ecosia, etc. are showing Bluesky link.
For years I've tried switching full time to DDG and always found myself typing !g so much that I gave up and went back to Google. For the past year and a half or so it's finally stuck and I cringe it I'm on a different device and accidentally get a Google result. It's become everything it wasn't at launch.
Most scientists don't have a professional profile on Twitter or Bluesky. It's the most self-aggrandizing ones who tend to do that.
The rest of us are just writing papers, presenting at conferences, collaborating with other research groups without any interest in putting it all out there on social media.
This whole X versus Bluesky thing is basically irrelevant. Neither of these platforms are good venues for dissemination of scientific research.
"Neither of these platforms are good venues for dissemination of scientific research."
These "platforms", i.e. other peoples' gigantic websites, are fantasy worlds that cater to self-aggrandisement where relevance requires a presence in that world
The outside world, aka the real world, is under represented and thus ignored
Most of my hard science contacts (physics, biology etc) from my days in student government have moved to Bluesky. Newer academics seem to be starting out there and skipping Twitter entirely.
If you have heard of Metcalfe's Law, you'll understand why this is not good for Twitter long term.
It's like internet media is reaching a new level of maturity in that it is bifurcating into the tabloids and the serious magazines, like printed media was before.
Total activity across the entire site doesn't meaningfully affect users as long as the activity they actually care about is still there, "Bluesky's aggregate metrics are down" and "Bluesky is the best place to talk science" can both be true at the same time. The website you're reading right now has abysmal metrics compared to Facebook and TikTok, but that's not really the point, is it?
yep. This feels like a feature to me, not a bug. As a mostly reader/follower on these platforms I'm surprised when others compare their engagement metrics, rather than the quality of the engagement on these platforms. Obviously their goals and incentives are different than mine, but the more time I spend on the internet the more I value the smaller communities due to the correlation with quality and topical focus.
My academic twitter migrated to Mastodon. Most my colleagues from different universities who used to be on Twitter are there now and I do not miss not being on X. There are even servers run by organizations like ACM (Association of Computer Machinery).
Readit News
It has apparent value propositions past the social network, but none of those use cases are visibly taking off and none of them appear to be monetizable. The social network itself is what will be evaluated when they go out for more funding. And I don't see how you can raise at all for a social network in 2026 with flat numbers, let alone the declining numbers Bluesky actually has.
I've been dual-wielding Twitter and Bluesky for about a year (after a year off Twitter where I was mostly Mastodon), and, anecdatally, we've hit a point where the engagement and volume of stuff I see in Bluesky is lower than what I was getting even on Mastodon. Earlier on, there was some truth to the idea that Twitter had a much larger audience, but you'd get better engagement on Bluesky. I now get better engagement on Twitter. I can see people I had followed into Bluesky moving back to Twitter.
I have no idea what's going to happen, but I'm curious to hear a coherent story about how Bluesky isn't cooked.
Usage has absolutely declined from peak switching periods where inevitibly some users won't stick around, but that's to be expected. Most stats seem to be leveling off (which isn't exactly stable growth either so the rest of your points stand).
I understand that as a Bluesky user the peak and dropoff doesn't hurt the experience. But investors are going to put money in with the expectation of a return and what they're going to look at are the derivatives of the adoption curve: how quickly is it gaining users, and is adoption accelerating?
It looks like Bluesky is going to be shedding active for the near future, probably settling around a million users active per day.
Odd. Here's another BlueSky stats page, and almost every metric is in the red.
https://bluefacts.app/bluesky-user-growth?t=3m
I really dislike Mastodon so gave that up a while back. I know there are a few people I'd like to follow who only post there, but such is life.
I wanted to like BlueSky, but it's such a bizarre echo chamber of people who left Twitter for ideological reasons that it basically filters for people that I actively don't want to engage with.
Those types of people are still there on Twitter (mostly on the other side these days), but I don't see them because the algorithm filters them out.
Are you an insider? Where are the numbers?
how_long_can_bluesky_exists = f(usage, headcount, funding)
Three parameters and you supplied zero of them.
It does sound like you just randomly picked a number 2026 and proceeded to rant on how little engagement you received on Bluesky.
You can generally take a headcount number and assign a fully loaded cost to it (say, $200k, conservatively) and just math it out. And of course that analysis assumes their infra expenditures round to zero.
So no, I'm not just making stuff up. I could be wrong! I feel like I was open about that.
Also, linkedin says 53 people, though crazies like to say they're employed places. It does say 29 in the US so that's more likely real. Assuming that skews eng a bit, that's probably (225 fully loaded eng, 150 fully loaded the rest) $6m in payroll alone. Not to mention server expenses. That's tough.
The case for more funding: 1 - yolo; 2 - non-economic investments; 3 - Musk gonna Musk, so expect Twitter to further shed users. There may be a business to be built there if you can run significantly more efficiently than Twitter which was a shockingly poorly-operated business (Zuck was dead on re: the clown car. Except maybe more like a silver mine, not a gold mine.). Oh, and 4 - the EU is pretty hostile to Twitter and Threads, so maybe there's some there there. Dunno.
What do you count as "engagement"? Views/likes? Those can be produced. Interesting conversations that aren't obviously LLM can't. That's the metric I use for anecdotally seeing Bluesky (and Mastodon) as immensely more engaging in a signal/noise ratio.
Deleted Comment
Do you have any data to back this up?
Also famously twitter is losing value and advertisers and users still it’s not stopped twitter from existing
Regarding engagement doesn’t twitter have lot more low quality engagement vs Bkuesky
More broadly, Twitter’s problem is that it carpet bombed the bridge with a substantial portion of its intelligent population. I’m pretty sure half the tech, game dev, and research people I follow on Bluesky/Mastodon will not return under any circumstance, myself included.
[1] - https://tobacco.stanford.edu/cigarettes/doctors-smoking/more...
Dead Comment
Dead Comment
[0] https://arewedecentralizedyet.online/
Decentralization is not a priority for most people. If anything, they actively want centralization, because it's easier. To get those people to decentralize, the solution will have to be dead easy and invisible. The AT Protocol being developed by the Bluesky people looks promising.
I don't work for Bluesky, I'm not on Bluesky, and I don't particularly care, but I found your comment unfair after reading about ATProto on HN literally yesterday.
The Fediverse will only be popular if someone releases a client that makes it as easy to use as X and Bluesky. Not sure if it's technically feasible (I don't know much about the innards of the protocol) but it doesn't seem to have happened at the moment.
I was in the same boat as you, and my experience was completely different. Mastodon's federation model reminded me of IRC, except nowhere near as balkanized.
So how did I wrangle the supposed complexity? I started out on one of the main instances and just started people-watching. Over time, I took note of which server contained users whose content I enjoyed over time, then I just joined the server.
Joining the server got me a slower federated feed that was both more pertinent to my interests that also functioned as a de-facto community space. I also found the moderation to be more to my own preferences. But the bigger server wasn't _bad_, I just preferred the smaller server because it was more personable.
I don't use Mastodon much anymore, but that's more because a good chunk of my social circle left for BlueSky than any gripes I had with the platform. I don't know where they will go if BlueSky goes belly up, but I can tell you that it won't be back to Twitter.
This is the reason why it was a pain. You or I might think "this is a decentralized platform, I should look at all the servers and find the best one for me" and immediately get choice paralysis looking at who has what rules, who federates with who, who runs which one, etc. An average user will probably stick to the main instance and not even think about it.
If you really cannot go beyond your inclination, and since you are a a long-time Internet user and a nerd, why not host your own instance?
There's already way more stuff on Mastodon for the hashtags I follow than I can possibly consume. If this is "unpopular", I'm happy with the way things are.
Maybe it _should_ be a little tougher to sign up for than the other mainstream options.
Only if you expect to be there for ever even as they inevitably enshittify. Be under no illusion that although BlueSky is having its "first they are good to their users" phase now, it is temporary.
So make the most of it now while it's good, but be prepared to move on when that changes. Embracing impermanence is a smarter play. This is nothing new, thus passes all social media.
Eg. searching 'Anthony Albanese Bluesky' for Australia's leader has a link to X, with custom integrated previews, above the Bluesky link of the PM despite the search explicitly stating Bluesky and despite the account posting to Bluesky.
It's hard for anyone to move over since the lack of engagement is rigged like this.
Duckduckgo and Bing put the bluesky link as #1 for the above. Seems straightforward to make the switch to me. If you haven't changed your browsers default search engine in the past 5 years now's a good time to do so. Much better results await.
The kagi assistant is also nice, only responding with AI when you add a question mark to your query, with the option of opening the query in a separate web search RAG w/the LLM of your choice
Deleted Comment
The rest of us are just writing papers, presenting at conferences, collaborating with other research groups without any interest in putting it all out there on social media.
This whole X versus Bluesky thing is basically irrelevant. Neither of these platforms are good venues for dissemination of scientific research.
These "platforms", i.e. other peoples' gigantic websites, are fantasy worlds that cater to self-aggrandisement where relevance requires a presence in that world
The outside world, aka the real world, is under represented and thus ignored
If you have heard of Metcalfe's Law, you'll understand why this is not good for Twitter long term.
[1] https://en.wikipedia.org/wiki/Metcalfe's_law
Scientist tend to form their own closed communities anyway.
Do you have a source for that?
SaaS, maybe. Tech, absolutely not.
Dead Comment
Dead Comment
[1]: https://bsky.jazco.dev/stats