- cross-posted to:
- technology@beehaw.org
- cross-posted to:
- technology@beehaw.org
Server indexes of places for newcomers to join can be instrumental for Fediverse adoption. However, sudden rule changes can leave some admins feeling pressure to change policies in order to remain listed.
Maybe I’m naive but I kinda don’t get it. People talk about defederating as if…what, all Meta IP addresses will be magically blocked from scraping your content? Any script kiddie can harvest Lemmy/Mastodon/whatever content.
Has Meta shown itself to be a bad actor? Yes. Should my email provider block all emails from Meta? Well…that’s a bit much I think? If Facebook email still existed, should my email provider block that?
My point is yes, Meta bad, but all Thread users also bad? I thought — and apparently I’m very wrong here — that the Federation paradigm was kinda like email. And the only email I want blocked is a domain where every single user is malicious, not a domain run by a malicious entity which has normal people as users, who aren’t necessarily very tech literate.
I don’t actually care, but I just find it a little confusing tbh.
@qjkxbmwvz I think the main fear is Embrace Extend Extinguish.
It’s not about interacting with Threadworms, it’s about sleepwalking into a situation where Meta is changing the very nature of ActivityPub itself.
I’m actually curious about “Embrace Extend Extinguish”: What can they do? They “extend” the ActivityPub protocol in a proprietary way, ok. Doesn’t mean any other instance has to use that, no? Ok, that would mean if an instance doesn’t follow that extension, it can’t interact optimally with Threads, but how does it matter? To me it seems all that can be lost by that is the content/user base that Threads brings into the Fediverse and then we are at the same point as we would be if we defederated immediately. Maybe I’m missing something here?
I think this article, How to kill a decentralized network, gives one of the best explanations, because it uses a real world example of how it has happened in the past.
I guess it is impossible to say what would have happened if Google never used XMPP. To me it mostly looks like google joined XMPP and made it way bigger than it was before and eventually left it again, making it small again. But is it worse than before Google even joined?
Maybe, but can we say for sure?
Maybe the lesson is not “don’t let the big corporate players in”, but rather “make sure the development of the underlying protocol itself is done in an open way”. If Google/Meta adds proprietary extensions, just don’t add them to the main protocol. If they leave the protocol again or changed their implementation in a way that is largely incompatible with the open version, nothing is lost than what they brought in initially. Doesn’t that make sense?
I agree.
I think a good example is how Slack started off by having good IRC integration, then slowly added features which were incompatible with IRC, and finally terminated IRC integration.
So clearly, Slack killed IRC, right? (…of course they didn’t!)
I see the potential situation with Threads as similar.
the problem occurs when most of the content comes from Meta (they will likely have the vast majority of Fediverse users). especially if major communities exist on their instance. when meta decides to no longer support fedi integration, those in the fedi are forced to decide between staying with their communities by ditching the fedi and moving to threads or having many of their communities ripped away.
meta will do this at some point as a play to draw users to them, but we can decide if we want to be affected when that comes to pass.
Threads exists for the sole purpose of capturing some of the people showing interest in the fediverse as twitter dies and keeping them in the facebook ecosystem. Once it believes it has exhausted this window of opportunity it will defederate just as it de-federated it’s xmmp based messenger service once it thought it had the upperhand.
Every server that defederates from meta preemptively is working to build a resilient community that will survive this inevitable scenario. Every server that federates with meta will become dependent on it then collapse as their users leave to join threads once that becomes their only option to continue interacting with the threads users that their social experience was built on.
Your post only concerns threats to an individual user re scraping or malicious interactions. The threat meta poses to the fediverse is systemic. In the long run the meta-blocking servers are the fediverse. The meta-federating servers might see some short term attention but in the long run will have the same fate as those that hitched their wagons to the metaverse.
I dont care if they scrape my comments I just wouldnt want to see sneaky “promoted” posts aka ads and I enjoy the idea of boycotting facebook.
Ultimately the decision is for the instances owners and admins to make, not ours. I will just migrate to one that doesnt federate with facebook if I have to.
I don’t quite see how that would even work. Those posts would need to be coming from individual users rather than from Facebook itself and you can just block those users. Facebook can display ads in between posts on their own app but those wont be visible to people using other apps.
Here’s one scenario.
Facebook feeds its users content according to an algorithm.
Facebook and lemmy users can interact with the same user content (liking, commenting).
There are vastly more Facebook users than lemmy users.
By dint of Facebook’s greater number of users, lemmy users will see the most popular content that is fed algorithmically to Facebook users.
Conclusion: lemmy users are being fed content by the Facebook algorithm (in this still, thankfully hypothetical, scenario).
Like imagine Facebook promotes some viral post and it gets a thousands of upvotes. Any lemmy user on a federated instance, sorting by upvotes/hot/etc, is going to see that post.
That’s the kind of top-down reach that is so alien to the fedi
Facebook could just create fake users that post ads as content
Hopefully fediverse admins are sensible enough to ban users who are blatantly posting advertisements. I know that a lot are, but I also know that a few of the bigger servers tend to turn a blind eye to that kind of thing.
See Grrgyle’s reply. That would be mine If I could explain things as good as them.
Nobody is forcing you to follow users/communities on Threads.
I like to browse by “all”. And nobody is forcing me to use an instance that federates with facebook either, like I said, I’ll migrate if I have to.
You’ll never get the tech iliterate people to switch to the rest of the Fediverse otherwise. Defederating Threads is about making it as bad as possible for its users - it’s about hurting Meta and stemming its bad influence on the web.
It’s not about looking what’s happening in the garden, it’s about entering in the garden. It’s two very different situations.
I agree, and I predict people will eventually pick instances that are doing what you suggested.
My understanding is that the defederation is to prevent MetaFacebook from getting to a point where they control the entire thing and then destroy it.
I don’t think defederating is the right move for that, but it’s a move
I think most people simply just don’t know how federation works and they imagine that defederating blocks Facebook from accessing your content when in reality it’s the exact opposite; it places one way mirror between us from which only they can see thru. There’s also some great irony in the fact that they’re talking about genocide while advocating for using the nuclear option to block Facebook despite the massive number of innocent casualties it’ll cause.
EDIT: Turns out I was mistaken. Defederation indeed does stop the flow of data both ways.
Brother these things are in no way the same. One is a tech giant knowingly aiding and abbeting governments who are ethnically cleansing their country and another is not being able to see posts from a different instance. The only great irony is you calling them innocent casualties.
It’s a comparison. By definition they’re not the same thing but there are similarities; you’re doing something that affects 100% of the userbase because you have an issue with 2% of them.
You should think of a better comparison cause this one sucks.
Also no. You’re doing something that affects 100% of users because the node these users use is malicious. The problem is the underlying structure not the people using said structure. Maybe this makes them stop using said structure.
Its like being upset that I dont answer unknown numbers. “Well only 10% are scammers so you’re affecting 100% of calls”
This is like comparing requiring students in schools to wash their hands to genocide. The scale is the same but the impact is vastly different, and if you don’t want to wash hands (or be defederated) you can just move. Except for changing activitypub instances is even easier than changing schools and both are easier than leaving Palestine.
What do you mean by this? Even if Meta would collect data from defederated servers (I don’t think they would), it would be massively more complicated than if they were federated.
Federarting means there’s a two-way road between your instance and threads.net and traffic can flow both ways. When you defederate it stops the traffic flow from threads.net to you but the traffic from you to them is unchanged. Even if every single instance defederates them they can still see all the content that’s posted there. Nobody else just wont see any of theirs. Only your instance admins know your email, ip-address and so on but all your posts and messages are publicly available to anyone and you can’t stop them from accessing it.
It’s basically the same thing as blocking an user. You wont no longer see their messages but they will see yours.
EDIT: Turns out I was mistaken. Defederation indeed does stop the flow of data both ways.
No, that is not how defederation works. One server defederates, traffic stops in both directions. It’s not comparable to user blocking.
There’s a big difference between the posts being available publicly on the Web and them being sent to Threads via federation.
I hate to admit when I’ve been wrong but this seems to be one of those cases. I tried to use my lemmyNSFW account to view content on a instance that doesn’t federate with them and I indeed can’t see any. I stand corrected.
Good on you for admitting it - we’re all wrong sometimes :) take it as a learning opportunity
Sir/Madame, not being able to see some online content is nothing at all like having your family members murdered in real life.
Read A Death Sentence For My Father sometime and you will see.
The thing is your article blames meta for not doing something that would be impossible on lemmy by design. Meta didn’t act to silence messages calling for violence but there is no mechanism to do this top down on lemmy only by defederating instances or individual communities/instance admin banning posts. Exactly the same thing could happen here, if the user base ever got large enough.
@VirtualOdour the point of me sharing that article was just to try to put a human dimension on genocide for that callous person above.
Meta have been implicated in at least two genocides now and openly obstructed the International Criminal Court in their investigation of one of them. I think people are only pointing that out to show how evil Meta are.
But if you want to know what specifically they will do to ActivityPub, the other article I shared has more direct relevance: How to kill a decentralized network.
Let’s stick to one topic for now.
If lemmy was as popular as Facebook then exactly the same thing would have happened. Lemmy is designed not to have the top down control which the article says Facebook should have used to hide posts.
You can’t blame Facebook for something if you support an alternative where it wouldn’t even be possible to avoid that thing.
If you’re willing to acknowledge that we can move on and you can try and say in simple terms what you think meta did to obstruct the ICC, try to be accurate and concise.
Your topic’s a false premise. First of all it’s totally valid to criticize someone for something that couldn’t apply in the current situation, because what’s being criticized is the decisions and attitudes that their actions reveal.
Meta’s refusal to moderate a website they control after multiple warnings that it was being used to incite genocide speaks to their institutional values, accountability, and culture.
By contrast, plenty of instance owners have shown responsibility, accountability, and good faith about admins moderating the instances they control.
Lol that’s condescending, and it’s also a bit offputting. I come here to bloviate thank you very much. :)
The thing is though, I’m not part of the wider conversation about facebook above. You glommed onto a very simple, very specific point I made to someone else about the human impact of social media incitements to genocide.
What Meta did to the ICC isn’t even related to my above link (which is about the Tigray genocide, not the Myanmar genocide). But it’s well-documented, and I’m not interested in rehashing it here.
It’s a comparison. By definition they’re not the same thing but there are similarities; you’re doing something that affects 100% of the userbase because you have an issue with 2% of them. Like Israel fighting Hamas and the entire Gaza population having to suffer because of it.
It’s a really inappropriate comparison tho.