So what if he doesn’t talk to them? The protocols and code are available for anyone, and instances are open for federation. Facebook could, without any sort of consultation, deploy their own instance of Mastodon with their own fork of the code and keep all their changes to themself. If they’re going to do it anyways, it’d be better to work with them on it.
People tend to forget things quickly, especially if they can communicate with their friends and family from Lemmy. Sooner or later, everyone will give in and just federate with Meta.
That will eventually lead to code changes to cater to Meta’s needs, those changes might not be made public (Mastodon is LGPL 3.0, if you don’t release the binaries, you don’t have to release the source), and those changes will eventually lead to telemetry gatering, incompatibility issues, etc., and that will eventually lead to people steering away from Mastodon… Lemmy and KBin might be soon to follow.
That sounds very pessimistic, I hope that won’t be the case, at least it seems like the mastodon instance I’m on will block it from the start, so that’s at least something.
I’m skeptical that Facebook would want to openly federate with externally controlled services because it’s kind of wild out here by design. However, if they did there would also be upsides. Those people who refuse to use anything but Facebook could be reachable without the rest of us having to go to Facebook, and people who only use Facebook because that’s where everyone else is could migrate away. Platforms opening up is a good thing.
I doubt Facebook would run Mastodon if they wanted to federate. They have an existing system with existing data and they have plenty of development resources to bridge that to ActivityPub. If Facebook did want to run Mastodon for some reason, even if they did open source their changes, which they probably would since they have a history of working with open source, the big changes would likely be unusable for most servers because Facebook scale is completely different from the typical Mastodon server. It doesn’t make any sense for a free Mastodon server with less than a million users to deploy the same kind of infrastructure that Facebook runs for 3 their billion monthly active users all over the world.
I was there for that and you’re misremembering. Before Google, hardly anybody used XMPP or knew what it was. Google came, and then you could talk to Google users on XMPP but still regular users didn’t know what XMPP was, and could sometimes be confused why your e-mail was different. Google left, taking their XMPP users with them. XMPP is still XMPP to this day. Every instant messaging service from that era, including Google Talk, has pretty much died out. XMPP might actually be an exception because there were few users before and the relative decrease in users is probably much less than platforms with more memorable names and better advertising.
I used to use XMPP before Google “killed it”, and my story is that before Google had XMPP, I used XMPP gateways to talk to people on other platforms. Google integrated and I started using Google’s XMPP client on my phone because it was much better than any other XMPP client available at the time. Google discontinued XMPP support and I didn’t move back to another server, but it wasn’t because Google had killed XMPP. I don’t know if I ever had any native XMPP contacts, and I didn’t talk to anybody on AIM, MSN, etc anymore, and I still didn’t talk to anybody on native XMPP, so had no reason to use XMPP. I talked to a few people on Google Talk, people who had never used any other XMPP service, and then Google discontinued Google Talk because that era of instant messengers had apparently ended.
This plan to prevent the same thing from happening is really misguided. You can have few users now, few users later, and few users further in the future, or you can have few users now, many users later, and maybe few users again further in the future. People who are on non-Facebook platforms now are very unlikely to decide they like Facebook better and leave later if Facebook federates and then defederates.
The idea of everyone getting together to preemptively defederate Facebook is also very hypocritical. We have a decentralized, open system where anybody can start an instance and we tell people to find an instance with rules and content they like. Then the possibility of Facebook federation starts being talked about and suddenly we don’t want the same rules to apply to Facebook. People want Facebook globally blocked before they get a chance to federate, and primarily out of fear of Facebook the company or prejudice against Facebook users, not because of the technical concerns around scaling. If the rules only apply to small instances with small budgets, what happens if one of the instances starts to get too successful?
It’s not about getting the code. They have the code, have for years, and hate it because it forces an open system.
This is about forcing people in “positions of power and authority” over mastodon/lemmy/kbin servers to conform to facebook’s wishes so that they can destroy a competing platform.
Google XMPP or Microsoft Word Document style.
It’s been done before, the only reason for people to cave now is they’re getting paid.
Mastodon is AGPLv3. That means if you allow someone to communicate with a server, you must offer them the modified source code. Not just when you distribute the modified code like in the GPLv3. So even if they forked Mastodon their code modifications would need to be made available.
However iirc ActivityPub itself is under a more permissive scheme (I think it’s predecessor was using the MIT license?) so Meta could use the protocol itself.
To get around that they’d have to do something drastic, like getting the owner of the code to change the license in next release, and keep him in an NDA while doing so in order to position themselves when the change happens.
getting the owner of the code to change the license in next release
AFAIK, all contributors need to agree in order to change the license of a codebase. If a contributor disagrees, their part of the code has to be rewritten in order to comply.
So what if he doesn’t talk to them? The protocols and code are available for anyone, and instances are open for federation. Facebook could, without any sort of consultation, deploy their own instance of Mastodon with their own fork of the code and keep all their changes to themself. If they’re going to do it anyways, it’d be better to work with them on it.
The know large instances might defederate from them, that’s why the NDAs.
Eventually, Meta will do to the fediverse what Google did to XMPP. I hope I’m proven wrong.
Wouldn’t the NDA’s just make it even more likely taht people will defederate?
People tend to forget things quickly, especially if they can communicate with their friends and family from Lemmy. Sooner or later, everyone will give in and just federate with Meta.
That will eventually lead to code changes to cater to Meta’s needs, those changes might not be made public (Mastodon is LGPL 3.0, if you don’t release the binaries, you don’t have to release the source), and those changes will eventually lead to telemetry gatering, incompatibility issues, etc., and that will eventually lead to people steering away from Mastodon… Lemmy and KBin might be soon to follow.
That sounds very pessimistic, I hope that won’t be the case, at least it seems like the mastodon instance I’m on will block it from the start, so that’s at least something.
I’m skeptical that Facebook would want to openly federate with externally controlled services because it’s kind of wild out here by design. However, if they did there would also be upsides. Those people who refuse to use anything but Facebook could be reachable without the rest of us having to go to Facebook, and people who only use Facebook because that’s where everyone else is could migrate away. Platforms opening up is a good thing.
I doubt Facebook would run Mastodon if they wanted to federate. They have an existing system with existing data and they have plenty of development resources to bridge that to ActivityPub. If Facebook did want to run Mastodon for some reason, even if they did open source their changes, which they probably would since they have a history of working with open source, the big changes would likely be unusable for most servers because Facebook scale is completely different from the typical Mastodon server. It doesn’t make any sense for a free Mastodon server with less than a million users to deploy the same kind of infrastructure that Facebook runs for 3 their billion monthly active users all over the world.
Please look into the history of XMPP and how Google essentially killed it.
I was there for that and you’re misremembering. Before Google, hardly anybody used XMPP or knew what it was. Google came, and then you could talk to Google users on XMPP but still regular users didn’t know what XMPP was, and could sometimes be confused why your e-mail was different. Google left, taking their XMPP users with them. XMPP is still XMPP to this day. Every instant messaging service from that era, including Google Talk, has pretty much died out. XMPP might actually be an exception because there were few users before and the relative decrease in users is probably much less than platforms with more memorable names and better advertising.
I used to use XMPP before Google “killed it”, and my story is that before Google had XMPP, I used XMPP gateways to talk to people on other platforms. Google integrated and I started using Google’s XMPP client on my phone because it was much better than any other XMPP client available at the time. Google discontinued XMPP support and I didn’t move back to another server, but it wasn’t because Google had killed XMPP. I don’t know if I ever had any native XMPP contacts, and I didn’t talk to anybody on AIM, MSN, etc anymore, and I still didn’t talk to anybody on native XMPP, so had no reason to use XMPP. I talked to a few people on Google Talk, people who had never used any other XMPP service, and then Google discontinued Google Talk because that era of instant messengers had apparently ended.
This plan to prevent the same thing from happening is really misguided. You can have few users now, few users later, and few users further in the future, or you can have few users now, many users later, and maybe few users again further in the future. People who are on non-Facebook platforms now are very unlikely to decide they like Facebook better and leave later if Facebook federates and then defederates.
The idea of everyone getting together to preemptively defederate Facebook is also very hypocritical. We have a decentralized, open system where anybody can start an instance and we tell people to find an instance with rules and content they like. Then the possibility of Facebook federation starts being talked about and suddenly we don’t want the same rules to apply to Facebook. People want Facebook globally blocked before they get a chance to federate, and primarily out of fear of Facebook the company or prejudice against Facebook users, not because of the technical concerns around scaling. If the rules only apply to small instances with small budgets, what happens if one of the instances starts to get too successful?
It’s not about getting the code. They have the code, have for years, and hate it because it forces an open system.
This is about forcing people in “positions of power and authority” over mastodon/lemmy/kbin servers to conform to facebook’s wishes so that they can destroy a competing platform.
Google XMPP or Microsoft Word Document style.
It’s been done before, the only reason for people to cave now is they’re getting paid.
Oh man, I can’t wait until Eugen also turns into a corporate cocksucker and back-knifer.
Mastodon is AGPLv3. That means if you allow someone to communicate with a server, you must offer them the modified source code. Not just when you distribute the modified code like in the GPLv3. So even if they forked Mastodon their code modifications would need to be made available.
However iirc ActivityPub itself is under a more permissive scheme (I think it’s predecessor was using the MIT license?) so Meta could use the protocol itself.
Hey, you’re right!
To get around that they’d have to do something drastic, like getting the owner of the code to change the license in next release, and keep him in an NDA while doing so in order to position themselves when the change happens.
Good thing we’re not seeing that
AFAIK, all contributors need to agree in order to change the license of a codebase. If a contributor disagrees, their part of the code has to be rewritten in order to comply.