Most people are probably on the 3 biggest instances and the "official" instances don't have it blocked as far as I can see (https://blog.joinmastodon.org/2023/07/what-to-know-about-thr...). Blocking also seems like a silly thing for a network that is about being open and interoperable.
Quite, and, as I said in my post, I think we (the Fediverse) owe it to the rest of the Internet to enable the deconstruction of the current social media landscape as much as possible.
The protocol is open and interoperable. I can't build a Facebook but nothing is stopping Mark Zuckerberg from building a platform on ActivityHub.
But there would be no point to a federated network if every node were required to accept traffic from all other nodes. That's just centralization with extra steps.. and centralized social media is what most people who move to federated networks are trying to avoid, to say nothing of the cost burden on smaller instances to handle all of that traffic. A lot of people block the big instances as well for similar reasons.
No, but "spam" is subjective and while you can decide that for your own instance, doing it for a bigger instance on behave of your users seems a bit overreaching...but luckily I can just move to an instance where I agree with the rules or host my own.
(Google Groups was to Usenet what Threads is to ActivityPub, and it got collectively blocked by all Usenet servers for spam, then decided to shut down.)
> Because I believe that Meta as a whole is moving to an open systems approach, not because Mark Zuckerberg is a nice guy, but because he is smart enough to go open before he is forced to by regulators, (or broken up).
I think it's both more than that and less than that. Unless I missed some news, Meta's newfound love for product openness has only really manifested in Llama and Threads—I haven't seen any evidence of shifts in their core products.
I think what's happened is that in both cases Meta knows that they don't really stand a chance of actually unseating the big players in that segment on their own, but that Meta can make a dent in their profit margins if they can dilute the value of owning that segment. Llama is squarely targeted at preventing anyone from owning AI. I think Threads is doing the same thing with microblogging. If Meta can't own a segment they're going to make darn sure that no one else can own it and use their profits from it to push Meta out of their core.
If it also makes the company look more palatable to regulators, that's a nice side effect, but I'm not sure it would work as the primary goal.
Well, personally I think that Meta is taking the threat of regulator action very seriously. You only have to look at how Google and Apple are in trouble.
Owning WhatsApp, Threads and Instagram makes them a big target for breakup.
And an open system defuses that significantly.
I'm not sure that it has as much of an impact as you think it does. Threads would not be a significant target for regulator action regardless of openness—it doesn't have enough share of the market to be interesting. Meanwhile, the openness of Threads does nothing to alleviate the concerns of their dominance in every non-microblogging social media category.
Meta could link Threads and Instagram in a heartbeat if they chose, and were't scared of the consequences. Threads is small fry at the moment, but a very good test and precursor to making the same play across the other properties.
I respect your opinion BTW, I don't like flame wars. :-)
It's worth remembering that the Fediverse is not just microblogging, there are platforms for pretty much every kind of social media play.
> If Meta can't own a segment they're going to make darn sure that no one else can own it and use their profits from it to push Meta out of their core.
I think it's more about trying to keep your competitors busy defending their moat - so that they don't go on the offense and attack your moat.
Yeah, that also works as an explanation. Same principle—defend yourself by going on the offensive.
There's a reason why Threads was only conceptualized after Musk acquired Twitter and started talking about building an everything app. Pre-Musk Twitter was very content with its market and so wasn't a threat to anyone else. Musk's X was a different story (if he could execute on it, which at the time seemed more likely than it does now).
I may be wrong but I remember it almost exactly the opposite i.e. that Musk was clearly flailing with Twitter and Meta saw that as an opportunity to move into Twitter's market, which they would have struggled to do with competent management at Twitter.
Yea, lol, Musk's twitter was never, ever a clear threat to any competitor and Zuck smelled blood in the water. Elon fanboism has spun it the other way for obvious reasons.
I'll also note that Twitter was pretty open and had lots of goodwill, which would have made competing against it really hard.
Twitter getting rid of 3rd party clients (pre-Musk), and shutting down API access (post-Musk) burned a lot of goodwill, leading to large spikes of growth of the Fediverse, and creating a window of opportunity for a microblogging service from Meta.
“Commoditize your complements” is a related point.
Content is the input to the Meta ad engine. You can generate it on-platform, or you can bring it in from elsewhere if it is commoditized. (And commoditizing your complements often looks like breaking down your competitors’ moats.)
I don’t think it’s as cynical as a “salt the earth” strategy though, surely they (plan to) make profit from Threads and LLaMa regardless of what their competitors do? Threads users will drive ad revenue, Llama (theoretically, eventually) increases the quality bar for content, driving their main flywheel.
> Meta's newfound love for product openness has only really manifested in Llama and Threads
They’re also paying lip service to openness by spinning off their VR OS. Of course that doesn’t mean it’ll become any less proprietary, only available on more hardware.
It’s smart to join up with others when you’re the underdog. I think that’s all they’re doing.
Yes, there's a reason I specified "product openness". React isn't a product and never could have been, so while I'm grateful to them for releasing it open source it doesn't really represent the same kind of play as Threads or Llama.
No, because they’re not exposed to users as features that they interact with. From a user perspective, they don’t ever see if it was made using those tools. They see the AI front and center.
They see AI features front and center. They dont see the backend (which may or may not be running Llama). Same goes for React although, with React, the framework is what the end user directly interacts with. So, by that logic, React is more of a product than Llama.
If you are a user, what feature is react giving you? Pre and post react, Facebook is largely the same to a user.
Pre and post graphql, Facebook is also the same.
To the user, neither of those are expressed as features.
Llama / Meta AI is a feature that changes how a user interacts with the system. When it was added, it’s noticeable. If it were to be taken away it would be noticeable. Nobody was boycotting react.
That is the difference between an implementation detail and a product/feature.
I'm not sure I understand this analogy. If Meta changes what underlying model they use, it'll still be called Meta AI. Having the criteria be "what new features does it offer to the users of the site" means that things like databases, servers, etc. are not products. This is objectively false.
You are not wrong, but I also think there is another part at play here: Facebook started out quite open and then gradually became closed (just like reddit and twitter).
This is a new product, so they start out open and then slowly close down as they acquire marketshare.
I don't see "Threads" on the list of "Core platform services", and I see no evidence of the services that are so designated becoming part of the fediverse.
Meta is a gatekeeper, it applies to their services. This (or an alternative) will have to be rolled out to Instagram. I guess now the smaller Threads is testing ground.
Not sure where you're coming up with that. I'm aware of the "Fedipact" but I'd guess it comprises a fairly small percentage of users, and some users have taken steps to leave instances which block Threads.
I've been intrigued by Threads since it launched last year. Succeeding where there's a competitor as entrenched as Twitter is nigh on impossible. The network effect is massive.
But of course there are chinks in TWitter's armor, specifically the whole blue check debacle and that Twitter itself is rapidly becoming 4chan (or even 8chan) and advertisers are understandably fleeing.
Still, the chance of success isn't great. Launching it under the iG brand was probably a good idea, better than under Facebook at least.
Still, when it comes to open source and integration with third-parties I keep thinking of the quote "open source is for losers" (that might come from this [1]), meaning all but the dominant player embrace interoperability and open source in a desperate attempt to topple the dominant player.
I have no idea what usage looks like but I'm glad they haven't given up. I know Google would've canceled it by now (and probably replaced it with something that looks kinda similar, has a different name/branding but no compatibility with the old thing, if chat apps are anything to go by).
I still say federation is something tech people care about but offers nothing of tangible value to end users. There hasn't been a successful federated technology since email. For a reason.
That predates email and is ... not really a counterpoint to federation sucking for end-users? There's so much spam/phishing and somehow it's no one's legal responsibility.
Again, not really the point.
So long as Meta can say their platforms are open, they have a bulwark against regulators and/or being broken up.
It really doesn't matter to them if it's popular or not.
I would go as far as to say they don't really want it to be popular, but I would love to see enough integration for there to be at least some attempt at competition.
There's lots of Fediverse interaction on Threads. It's a little tricky still because you can follow Threads users from other instances but you can't follow accounts on other instances from Threads. That being said, I see lots of people (myself included) boosting, liking, and even commenting on Threads content from elsewhere in the Fediverse. And now that Threads has just started to publicly show fedi comments on Threads (not just to the OP), we'll get a better sense of engagement on any given post.
What's the true goal here? Given the decentralized aspect of the fediverse, facebook won't be able to sell ad space here.
Is threads hoping to become "the" instance in the fediverse? Then they can impose their will across the rest of the fediverse.
threads: "oh, you have a single user instance and want to federate with us? Pay us $500 per user/year, and use our preferred server which allows us to sell ad space on your server"
Right now fediverse is growing and has very honest algorithms. Once Meta is on the scene, their "decentralized" timeline will look more enticing than others. It will be captivating and you'll prefer to hangout on their server with all nice extensions that just work there. With time, +90% of users won't bother looking on small isolated islands and you'll have facebook 2.0 in place.
Anyways, it is a fair fight. Meta coming to fediverse already means that things on this side are interesting.
Why not? The Threads client will run ads. If users prefer the experience (integration to other Meta products, QoL, etc.) then they will use it and watch the ads. If they prefer other clients they will use those.
Ironically, Facebook was also involved with Google defederating from XMPP, according to an explanation I read many years ago (IIRC, from Larry Page when he was CEO). FB messenger could connect to Google Talk users, but not the other way around - so Google wasn't pleased about freely exporting users' "social graph"[1] to Facebook, and getting nothing in return.
1. The "social graph" was the new hotness then, everybody had to have it, just like AI now. An exec at Google convinced the company that Facebook was an existential threat, leading to some very questionable decisions bourne of a self-induced siege-mentality.
It remains baffling and sad that Google's first response to this was to murder their best social graph in cold blood. Between giving Reader the axe, and the real names fiasco on Google+, the company has destroyed the necessary trust to ever operate a social graph worth having.
Please don't tell me that YouTube is a social graph, by the way. Words should mean something. If YouTube were a social graph, I'd see videos from the accounts I follow.
Are you not logged in? I see plenty of videos from the accounts I follow.
Youtube is obviously a social graph. Who you follow, and who they follow, how often you interact, whether you comment, etc. directly influences the content that gets recommended to you.
Mark Z was clear on Dwarkesh's podcast that when he has things open he does so because he thinks it will benefit Meta and he explicitly says it's not altruistic and that if it stops being useful he'll stop making things open.
Nostr makes so much more sense to me than the fediverse because your account is not bound to a particular server, so if your admin gets bored you can simply use other relays or host your own instead of being kicked off. And with this you get pretty significant fediverse interop: https://gitlab.com/soapbox-pub/mostr
I've been around BBS, web forums and social networks for almost 30 years now.
Indeed NOSTR is completely fresh air and mindblowing. My account just works across any clients, doesn't need even need Internet nor electricity. I can print my note on a piece of paper and people can still verify that this is a verifiable text from me.
Really in love with the tech, didn't had this feeling for quite a while.
I think this is great. I've been using threads for 6 months, and while I was never a prolific twitter user - I thought it sucked even before the elon purchase, now, it's entirely unusable to me (and I've tried) - I do like its interface, a seemingly "light" (by meta standards) approach to moderation, and now a step in this direction I think is a really good thing.
As to the "why" I think it's fairly obvious. If they integrate with everything, they'll eventually become the defacto platform and/or steer users and engagement to their other platforms with less friction.
Yea, this is a platform-wide thing and across most other platforms as well - I meant in terms of like, forcing you to edit posts with "forbidden" keywords like they love to do on IG (but never tell you exactly what you said wrong), tone policing, hiding comments between users arguing with each other, etc.
You could use the fediverse directly instead of Threads if you don't want to be 'over-moderated'. Some server admins in the fediverse also go over board, but you can host your own instance in the fediverse or search for an instance that align with your topics.
If I understand Threads adoption, it is quite large compared to Mastodon. Will Mastodon be able to handle the load if Threads -- a commercial entity -- is not blocked?
I'm holding off on making the cynical conclusion about Threads' motivations.
89 comments
[ 2.6 ms ] story [ 163 ms ] threadBut there would be no point to a federated network if every node were required to accept traffic from all other nodes. That's just centralization with extra steps.. and centralized social media is what most people who move to federated networks are trying to avoid, to say nothing of the cost burden on smaller instances to handle all of that traffic. A lot of people block the big instances as well for similar reasons.
(Google Groups was to Usenet what Threads is to ActivityPub, and it got collectively blocked by all Usenet servers for spam, then decided to shut down.)
I think it's both more than that and less than that. Unless I missed some news, Meta's newfound love for product openness has only really manifested in Llama and Threads—I haven't seen any evidence of shifts in their core products.
I think what's happened is that in both cases Meta knows that they don't really stand a chance of actually unseating the big players in that segment on their own, but that Meta can make a dent in their profit margins if they can dilute the value of owning that segment. Llama is squarely targeted at preventing anyone from owning AI. I think Threads is doing the same thing with microblogging. If Meta can't own a segment they're going to make darn sure that no one else can own it and use their profits from it to push Meta out of their core.
If it also makes the company look more palatable to regulators, that's a nice side effect, but I'm not sure it would work as the primary goal.
Owning WhatsApp, Threads and Instagram makes them a big target for breakup. And an open system defuses that significantly.
I'm not sure that it has as much of an impact as you think it does. Threads would not be a significant target for regulator action regardless of openness—it doesn't have enough share of the market to be interesting. Meanwhile, the openness of Threads does nothing to alleviate the concerns of their dominance in every non-microblogging social media category.
It's worth remembering that the Fediverse is not just microblogging, there are platforms for pretty much every kind of social media play.
Zuckerberg thinks Threads could be Meta's next billion-plus-user product.
I think it's more about trying to keep your competitors busy defending their moat - so that they don't go on the offense and attack your moat.
There's a reason why Threads was only conceptualized after Musk acquired Twitter and started talking about building an everything app. Pre-Musk Twitter was very content with its market and so wasn't a threat to anyone else. Musk's X was a different story (if he could execute on it, which at the time seemed more likely than it does now).
I don't think Meta would have been so keen to launch a competitor if people weren't so desperate to leave x (nee twitter) for another platform.
Twitter getting rid of 3rd party clients (pre-Musk), and shutting down API access (post-Musk) burned a lot of goodwill, leading to large spikes of growth of the Fediverse, and creating a window of opportunity for a microblogging service from Meta.
Content is the input to the Meta ad engine. You can generate it on-platform, or you can bring it in from elsewhere if it is commoditized. (And commoditizing your complements often looks like breaking down your competitors’ moats.)
I don’t think it’s as cynical as a “salt the earth” strategy though, surely they (plan to) make profit from Threads and LLaMa regardless of what their competitors do? Threads users will drive ad revenue, Llama (theoretically, eventually) increases the quality bar for content, driving their main flywheel.
They’re also paying lip service to openness by spinning off their VR OS. Of course that doesn’t mean it’ll become any less proprietary, only available on more hardware.
It’s smart to join up with others when you’re the underdog. I think that’s all they’re doing.
They’re part of the Rayban partnership. Imho definitely a product.
I feel like thats a pretty strong distinction.
If you are a user, what feature is react giving you? Pre and post react, Facebook is largely the same to a user.
Pre and post graphql, Facebook is also the same.
To the user, neither of those are expressed as features.
Llama / Meta AI is a feature that changes how a user interacts with the system. When it was added, it’s noticeable. If it were to be taken away it would be noticeable. Nobody was boycotting react.
That is the difference between an implementation detail and a product/feature.
This is a new product, so they start out open and then slowly close down as they acquire marketshare.
[1] https://digital-markets-act.ec.europa.eu/gatekeepers_en
But of course there are chinks in TWitter's armor, specifically the whole blue check debacle and that Twitter itself is rapidly becoming 4chan (or even 8chan) and advertisers are understandably fleeing.
Still, the chance of success isn't great. Launching it under the iG brand was probably a good idea, better than under Facebook at least.
Still, when it comes to open source and integration with third-parties I keep thinking of the quote "open source is for losers" (that might come from this [1]), meaning all but the dominant player embrace interoperability and open source in a desperate attempt to topple the dominant player.
I have no idea what usage looks like but I'm glad they haven't given up. I know Google would've canceled it by now (and probably replaced it with something that looks kinda similar, has a different name/branding but no compatibility with the old thing, if chat apps are anything to go by).
I still say federation is something tech people care about but offers nothing of tangible value to end users. There hasn't been a successful federated technology since email. For a reason.
[1]: https://siliconangle.com/2014/05/29/only-loser-vendors-are-t...
I know so many people who simply don't respond if the incoming call isn't in their address book.
I know it’s still early and features are being built out, but my guess is the number is close to 0.
Is threads hoping to become "the" instance in the fediverse? Then they can impose their will across the rest of the fediverse.
threads: "oh, you have a single user instance and want to federate with us? Pay us $500 per user/year, and use our preferred server which allows us to sell ad space on your server"
Right now fediverse is growing and has very honest algorithms. Once Meta is on the scene, their "decentralized" timeline will look more enticing than others. It will be captivating and you'll prefer to hangout on their server with all nice extensions that just work there. With time, +90% of users won't bother looking on small isolated islands and you'll have facebook 2.0 in place.
Anyways, it is a fair fight. Meta coming to fediverse already means that things on this side are interesting.
Probably
> Then they can impose their will across the rest of the fediverse.
Nah. The play at that point would be to just discontinue support for federation and lock all their users inside the new walled garden.
Why not? The Threads client will run ads. If users prefer the experience (integration to other Meta products, QoL, etc.) then they will use it and watch the ads. If they prefer other clients they will use those.
1. The "social graph" was the new hotness then, everybody had to have it, just like AI now. An exec at Google convinced the company that Facebook was an existential threat, leading to some very questionable decisions bourne of a self-induced siege-mentality.
Please don't tell me that YouTube is a social graph, by the way. Words should mean something. If YouTube were a social graph, I'd see videos from the accounts I follow.
Youtube is obviously a social graph. Who you follow, and who they follow, how often you interact, whether you comment, etc. directly influences the content that gets recommended to you.
Mark Z was clear on Dwarkesh's podcast that when he has things open he does so because he thinks it will benefit Meta and he explicitly says it's not altruistic and that if it stops being useful he'll stop making things open.
Indeed NOSTR is completely fresh air and mindblowing. My account just works across any clients, doesn't need even need Internet nor electricity. I can print my note on a piece of paper and people can still verify that this is a verifiable text from me.
Really in love with the tech, didn't had this feeling for quite a while.
As to the "why" I think it's fairly obvious. If they integrate with everything, they'll eventually become the defacto platform and/or steer users and engagement to their other platforms with less friction.
⁂
https://news.ycombinator.com/item?id=41327567
⁂
I'm holding off on making the cynical conclusion about Threads' motivations.