I won’t be happy if they integrate federation. Ever heard the phrase “embrace, extend, extinguish”? It’s a tactic used by large companies to squash growing competition.
Google used it, for example, to squash a growing open-source chat messenger protocol called XMPP. (Think of XMPP like ActivityPub.) Google allowed its Google Talk application to integrate with people using XMPP. (They embraced XMPP.)
Then, they added their own proprietary features that wouldn’t work with normal XMPP users. (They extended, or built on top of, XMPP.)
Then, they cut support for XMPP integration, leaving it effectively dead in the water. XMPP users suddenly had a list of Google Talk users in their friend list who would never appear online again, whereas Google Talk users maybe had one or two people in their friend list who looked like they’d moved on from Google Talk. (They extinguished XMPP.)
Now imagine that happening with Threads. You, a Mastodon user, follow a bunch of people who just happen to be on Threads. There are some things Threads users can do that you can’t, but you don’t really mind. It works well enough. Then, one day, Threads stops working with Mastodon. Suddenly, over half of the people you followed are no longer available to you. The only way you can follow them again… is to join Threads.
Lololol yes. And the ploum article people parrot about xmpp too.
There is a lot of groupthink in the fediverse and not enough critical thinking.
Threads integrating will highlight the serious value in activitypub. I personally don’t know why they would integrate unless they want to just completely dismember Twitter and blue sky or get in front of govt regulations.
The danger here is that Threads will suddenly be > 99% of the activity and user base in one fell swoop. You might find that Mastodon doesn’t scale as well as you think. It might also be that Threads deliberately or half assedly doesn’t federate properly, so that being a non-Threads user means posts aren’t visible inside of Threads or suffer from down ranking or other issues.
I’d add that historically federation hasn’t gone well when a big fish enters a small pond. XMPP was cited above as an example of that. At the time you had proprietary services like AOL / AIM, ICQ, Instant Messenger. XMPP was going to liberate us from proprietary and even Google got on board for a bit before dumping it.
The problem with big companies is they want all the cake to themselves and abhor having to yield control or cooperate with others. Meta might make nice noises about ActivityPub while they’re the underdog to Twitter but you could see them rapidly change their tune if Twitter went under.
I’m surprised I don’t hear as much of “I wan’t literally nothing to do with Meta in any way shape or form” sentiment (which happens to be my feelings on the matter).
Federation is the most interesting thing thats happened on the internet since http became the standard in my opinion, and I’ve been kicking around since Telnet, BBCs, IRC, Etc.
Meta doesn’t want to play nice, they want to see if they can own everything, in my opinion. “Oh, people are doing this cool thing that we can’t yet monitize and make our shareholders richer? See if we can somehow assimilate it, at a loss at first, as ususal of course.”
I’m not sure about embrace, extend, extinguish, but it does sound like it’d be the way here. Either way, why the living fuck can we not have anything interconnected that a megacorp can’t decide they’ll take over? Why can’t we keep that from happening, even when small independent individuals are running the server?
The entire point of ActivityPub is that it’s open and EEE-proof. If the users leave it for something proprietary but better, then it isn’t EEE, it’s just a better product.
Simply being open source is not an achievement in itself. The platform has to be user friendly, stable and future-proof. Most FOSS and federated alternatives create a platform and then endlessly harp on federation like that’s the end. No, that’s the beginning. The point is to make a product better than Big Tech WHILE maintaining federation and Foss status. THAT is what makes a platform EEE proof.
If the users leave it for something proprietary but better, then it isn’t EEE, it’s just a better product.
That’s literally the second E, extend.
Nothing is EEE-proof. If Meta puts even just 10 billion dollars into developing and marketing their fediverse EEE project, it’s going to be better for the average user (I.e: billions of people already using Meta’s services) than what a couple of FOSS devs made for free in their spare time.
That’s not what extend means in this context. In this context, extend means to add non-standard features to the protocol which only your implementation understands.
ActivityPub is open source. This means, by definition, Meta are allowed to use it. They could easily do the extinguish part by just never using the protocol. If another platform existing is such a threat to the Fediverse then we’re doomed to failure in the first place.
While Google removing XMPP no doubt helped sink nails into the coffin, WhatsApp played a bigger role in the death of XMPP than Google removing it from Talk. It was increasingly irrelevant for a now growing number of people.
I would compare Mastodon with Threads to Linux with the Wine compatibility layer.
People used to hold off on using Linux, because they did not want to lose their Windows software. Now with Wine, people are a lot less reliant on Windows and many Linux users and Microsoft competitors profit from this.
The Steam Machine was reliant on native Linux games, but almost nobody wanted to develop for a niche desktop OS, which led to less games on the platform, which made Linux even more niche for gamers …
That’s a big reason the Steam Machine failed.
The Steam Deck side-steps all of this with Wine. Now Linux can grow freely, even when developers ignore Linux completely. Wine gives Linux a fighting chance against Windows.
The same way I believe, that Mastodon will only get a fighting chance, if they can side-step the “nobody I know uses Mastodon” problem. Federation with Threads could be one solution to this.
I won’t be happy if they integrate federation. Ever heard the phrase “embrace, extend, extinguish”? It’s a tactic used by large companies to squash growing competition.
Google used it, for example, to squash a growing open-source chat messenger protocol called XMPP. (Think of XMPP like ActivityPub.) Google allowed its Google Talk application to integrate with people using XMPP. (They embraced XMPP.)
Then, they added their own proprietary features that wouldn’t work with normal XMPP users. (They extended, or built on top of, XMPP.)
Then, they cut support for XMPP integration, leaving it effectively dead in the water. XMPP users suddenly had a list of Google Talk users in their friend list who would never appear online again, whereas Google Talk users maybe had one or two people in their friend list who looked like they’d moved on from Google Talk. (They extinguished XMPP.)
Now imagine that happening with Threads. You, a Mastodon user, follow a bunch of people who just happen to be on Threads. There are some things Threads users can do that you can’t, but you don’t really mind. It works well enough. Then, one day, Threads stops working with Mastodon. Suddenly, over half of the people you followed are no longer available to you. The only way you can follow them again… is to join Threads.
“ever heard of embrace extend extinguish???”
Lololol yes. And the ploum article people parrot about xmpp too.
There is a lot of groupthink in the fediverse and not enough critical thinking.
Threads integrating will highlight the serious value in activitypub. I personally don’t know why they would integrate unless they want to just completely dismember Twitter and blue sky or get in front of govt regulations.
The danger here is that Threads will suddenly be > 99% of the activity and user base in one fell swoop. You might find that Mastodon doesn’t scale as well as you think. It might also be that Threads deliberately or half assedly doesn’t federate properly, so that being a non-Threads user means posts aren’t visible inside of Threads or suffer from down ranking or other issues.
I’d add that historically federation hasn’t gone well when a big fish enters a small pond. XMPP was cited above as an example of that. At the time you had proprietary services like AOL / AIM, ICQ, Instant Messenger. XMPP was going to liberate us from proprietary and even Google got on board for a bit before dumping it.
The problem with big companies is they want all the cake to themselves and abhor having to yield control or cooperate with others. Meta might make nice noises about ActivityPub while they’re the underdog to Twitter but you could see them rapidly change their tune if Twitter went under.
I’m surprised I don’t hear as much of “I wan’t literally nothing to do with Meta in any way shape or form” sentiment (which happens to be my feelings on the matter).
Federation is the most interesting thing thats happened on the internet since http became the standard in my opinion, and I’ve been kicking around since Telnet, BBCs, IRC, Etc.
Meta doesn’t want to play nice, they want to see if they can own everything, in my opinion. “Oh, people are doing this cool thing that we can’t yet monitize and make our shareholders richer? See if we can somehow assimilate it, at a loss at first, as ususal of course.”
I’m not sure about embrace, extend, extinguish, but it does sound like it’d be the way here. Either way, why the living fuck can we not have anything interconnected that a megacorp can’t decide they’ll take over? Why can’t we keep that from happening, even when small independent individuals are running the server?
The entire point of ActivityPub is that it’s open and EEE-proof. If the users leave it for something proprietary but better, then it isn’t EEE, it’s just a better product.
Simply being open source is not an achievement in itself. The platform has to be user friendly, stable and future-proof. Most FOSS and federated alternatives create a platform and then endlessly harp on federation like that’s the end. No, that’s the beginning. The point is to make a product better than Big Tech WHILE maintaining federation and Foss status. THAT is what makes a platform EEE proof.
That’s literally the second E, extend.
Nothing is EEE-proof. If Meta puts even just 10 billion dollars into developing and marketing their fediverse EEE project, it’s going to be better for the average user (I.e: billions of people already using Meta’s services) than what a couple of FOSS devs made for free in their spare time.
That’s not what extend means in this context. In this context, extend means to add non-standard features to the protocol which only your implementation understands.
ActivityPub is open source. This means, by definition, Meta are allowed to use it. They could easily do the extinguish part by just never using the protocol. If another platform existing is such a threat to the Fediverse then we’re doomed to failure in the first place.
While Google removing XMPP no doubt helped sink nails into the coffin, WhatsApp played a bigger role in the death of XMPP than Google removing it from Talk. It was increasingly irrelevant for a now growing number of people.
I would compare Mastodon with Threads to Linux with the Wine compatibility layer.
People used to hold off on using Linux, because they did not want to lose their Windows software. Now with Wine, people are a lot less reliant on Windows and many Linux users and Microsoft competitors profit from this.
The Steam Machine was reliant on native Linux games, but almost nobody wanted to develop for a niche desktop OS, which led to less games on the platform, which made Linux even more niche for gamers …
That’s a big reason the Steam Machine failed.
The Steam Deck side-steps all of this with Wine. Now Linux can grow freely, even when developers ignore Linux completely. Wine gives Linux a fighting chance against Windows.
The same way I believe, that Mastodon will only get a fighting chance, if they can side-step the “nobody I know uses Mastodon” problem. Federation with Threads could be one solution to this.