I still stand by that defederation as the only line of defense is a losing strategy. Keeping users siloed in Facebook's garden shouldn't be seen as a win for us.
Keeping users siloed in Facebook’s garden shouldn’t be seen as a win for us.
Sometimes the only winning move is not to play. If people hadn't federated with google's XMPP back in the day, google wouldn't have had the same level of control it had to kill XMPP as a competitor.
We need to learn from the lessons of the past, and the past has resulted in the deaths of services when federating with corporations.
I never said defeating them or out competing them should be the goal. The goal should be the survival of services. And corporations will kill these services.
They piggy backed on rapidly growing XMPP and then became lazy with keeping compatible with the rest of the xmpp federation and at some point the s2s connection stopped being feasible as they never implemented TLS for it, and did't really care as most xmpp users were on their server anyways and thus did't use the s2s connection.
Its not a typical nefarious EEE story, but it did a lot of damage to the xmpp federation anyways.
This predates Google Talk and is rather about the XMPP Gmail integration. Back then XMPP was the hot topic in tech circles (Twitter was even prototyped to be XMPP based) and people were switching to it and recommending it to others to replace ICQ/MSN/AIM etc. However, often they recommended others to use the Google XMPP service as back then Google was still naively seen as the "Do no evil" good guy, having just started up recently and giving away free things like previously unheared off 1GB of email storage etc.
So the situation is not quite comparable to AP and Facebook (and XMPP is far from dead), but it is still possible to draw some lessons from it.
|They're welcome to come here whenever they like .
Only if they know it exists and can still connect with the people and communities they care about. This is what the federated approach was supposed to fix, the silos, the community capture.
We know what Meta is, and we know our history, so we know Meta’s goal is to destroy the fediverse. Federating with Meta is not likely to yield your desired outcomes.
Embrace: Join the fediverse with your existing user base that dwarfs the fediverse’s existing user base, and with infinitely more money.
Extend: Use your size, in terms of users and capital, to steer the direction of the ActivityPub fediverse standard to your advantage and your competitors’ disadvantage. You see everyone else as a competitor because you are a corporation seeking to monopolize the user base for profit.
Google is adding code to Chrome that will send tamper-proof information about your operating system and other software, and share it with websites. Google says this will reduce ad fraud. In practice, it reduces your control over your own computer, and is likely to mean that some websites will block access for everyone who's not using an "approved" operating system and browser. It also raises the barrier to entry for new browsers, something Google employees acknowledged in an unofficial explainer for the new feature, Web Environment Integrity (WEI).
I want to share information and to communicate. I don't want every bowel movement tracked and monetizes. I don't want 30 cross site requests when going to a news site. A single story should not require 10MB of JavaScript libraries.
I have no doubt that most of the authors of the original internet are aghast at what their high-minded creation has itself created.
That's a common misconception actually, any and all data available via federation is already public and easily scrapable even without running an instance of one's own. Defederating only hides (in this case) Threads content from users on the instance doing the defederating, but the data is still public. Not to mention copies of it would still be fully available on any extant federated instances.
But they would still be unable to embrace (and, by extension, extend and extinguish) because users from Threads would be unable to interact with users from other instances. Basically, they'd be unable to get rid of a potential competitor using the EEE method.
But how could interoperability lead to extinguishing? That's the part I don't understand. By what means could Threads "extinguish" the network of instances that stay federated?
It seems the idea is that it gets so big that it either can't exist without it or leeches the userbase. I've not really seen any explanation either, but I've come up with an idea around it. For example, in my experience Lemmy.World is filled with the type of people who would use Threads (from responses I've gotten about corporations like Spotify and Apple - heavily praised and no negativity about them). As threads and .world users interact, over time there becomes a dependency between those instances due to the community connections that are made. At a certain point, one or the other does something to encourage usage - that would be Extending.
For how long would something like activitypub be able to hold out? If Meta begins making contributions to it? Or if after that dependency, Meta makes a chance to how their federation works internally and fractures the point of activitypub by making instance runners/users pick one or the other. Or worse, Meta flat out buys Automatic. There goes the Fediverse.
FWIW - I'm not informed or have any idea what I'm talking about in this regard. I'm fully guessing and postulating, I don't even think I'm parroting what I've read somebody else say about it because, like I said, I've yet to see an explanation how the extinguish would function in this example. Historically I have an idea, but the circumstances here are different, ish.
But, this is Meta we're talking about. I don't think we'd be any happier federating with Reddit if the opportunity arose because these companies have historically shown they will pull teeth to get what they want, no matter how many people's teeth they have to pull.
"Well can they?"
I don't know. Maybe not? Do you want to let them try? Why let them? By defederating, it's like having a glass wall where yes, they can see everything looking in, but the interaction is mitigated. Ifnthe example I brought up is accurate, any changes .World decided to make with Meta in mind would not affect the rest of the instances that have defederated, since we don't even see that stuff from them in the first place.
Comparatively, slrpnk.net currently is federated with .World but not Threads, so if .World makes changes, those may be seen from instances that are federated with it?
From my understanding, a specific post on .World that has interaction from Threads and slrpnk.net. Threads and .World would see everything while Slrpnk.Net would only see federated instances and .World comments.
We are about 1.5m here in the Fediverse. Threads is already 100m. That's quite a large number of things to be missing, so it's possible that there's a large number of conversations that defederated users are only seeing half of? That could be another example that pushes Extinguish.
Anyway, sorry for any confusion or nonsense - I wrote this in a hurry on my phone, but I also wanted to lay out my thoughts and understand to see if it's at all in the ballpark. Shit, just use me as Cunningham's Law.
You've basically got it. To use the "Google XMPP" example some others have:
XMPP users existed, and its userbase was growing (similar to Lemmy). Google made Google Talk, a desktop chat application they used to have, compatible with XMPP (which was the "ActivityPub" of chat applications) (embrace).
After a bit, Google started adding their own proprietary stuff to XMPP. (It's similar to how Apple/ Google added proprietary stuff in their respective text message applications, like reacting to a text with an emote.) The XMPP devs, for whatever reason, couldn't or didn't make Google's own proprietary Google Talk features compatible with XMPP, so XMPP users might've started feeling left out (extend).
After a while, Google Talk got rid of its XMPP support, and, as a result, many XMPP users could no longer communicate with many of the friends they had made on the platform. (Since Google Talk users outnumbered XMPP users, there was a very high chance that people you communicated with on there were using Google Talk.) Google Talk users, on the other hand, simply noticed maybe one or two people on their list had gone offline permanently (extinguish).
Yeah this is the one, and it seems easy to see exactly that process taking place. I don't think it's so much the data concerns, alone at least, nor even the potential for content. I think many would agree that, to some extent having a larger user base available could be a good thing. It just so happens that 1) the user base is "more accessible" at best and potentially dangerous at its worst (not all of threads is friendly) and 2) it's Meta. There couldn't possibly be a reason for them to pursue this other than not having their grasp on it. I see no reason to trust it.
Someone you like on Threads and nowhere else. Use it there then. You can view them if it's federated? Will that still be the case in 1, 2, 3 years? At which point you've integrated so much of your instance into Threads that when support for ActivityPub is dropped or whatever change gets made, well, you may as well stick with Threads...
There's just no good outcome. I am an optimist, for the right perspective and reason devil's advocate is always worth a glance... and this? This has no good causes behind it. Man, what is it with all the big corps and apps trying to tie everything into one single spot like WeChat. Can't people just scroll Mastodon then X then Threads then Lemmy then Kbin then Facebook all separately like a normal mass consumer?
I would love to federate with Reddit. I hate having a Reddit account. I hate their website layout and apps. I hate their ads. If I could access some of the niche reddit communities that aren't on Lemmy without using Reddit that seems to have absolutely no downside.
It prevents that specific strategy that would culminate in extinguishing. The idea being to siphon users away from other platforms, then add features that other platforms won't or can't implement, and use that to create an image of their own platform being better, having more features.
If they succeed at having a lot of users oblivious to what's happening, they will use those features, and when they don't work for people on other platforms, they will blame the other platforms instead of their own, further cultivating the image that other platforms are broken/unreliable.
In the end, they leave other platforms unable to compete, forcing users to either have a "broken"/incomplete experience, or migrate to their platforms. (Or leave the fediverse entirely). Or they can simply stop federating at that point, after users have left for their platform, cutting off the rest of the fediverse from content hosted on their platform.
The way defederating prevents a strategy like that is by cutting them off before they can get a foothold - they can't make users feel left out if they don't get to influence their experience in the first place.