For those who don’t want to click through, this is the content of the post:
There is another reason I find the discussion about blocking #Meta’s #ActivityPub project #Threads interesting:
I’ve been saying for a while now that the #Fediverse is a new and different beast, and whoever tries to understand it simply as a direct social media replacement misses the whole picture. We’re also federated communities, just as much.
Today we see a lot of concern about “what will the #Fediverse do” with #Meta. Wanna know what we will do? Everything and nothing. Because the Fediverse is not one entity. This is the essence of its decentralized nature - and that’s cool. If your server intends to block Meta servers completely - cool. If not, cool again.
But if you expect a unified response on something like that, you’re in for a disappointment.
This is not a “schism”, a “problem”, something to “solve”. This is just decentralization in practice. We don’t need to have the same blocklists, and that’s ok. Open protocols are not something you can control, so chill. When the time comes for this subject, choose a server with a policy that you agree with. But if you’re worried that we won’t all have one unified stance… are you sure you actually like #decentralization?
Edit: It looks like the post got copied by Lemmy anyway, but I’ll leave it for now just in case it doesn’t show up on Mlem or Jerboa (or if it gets deleted)
I don’t disagree, but if it takes off there’s going to be a selective pressure on instances to engage with meta’s activitypub stuff and that’s going to let meta scrape data. I think that being able to engage with a wide variety of users, especially on platforms like mastodon or other similar offerings, is going to be a good thing, but the protocol needs implicit ways to protect its users outside of just blocking out other instances, whatever that looks like, so that decentralization can be granular and not just “how much do you want to give your data to meta”
I agree - my main reason for sharing with this post in particular is because the tie-in it has with Beehaw’s recent decision to, at least temporarily, defederate with .world and sh.itjust.works; I just found the framing about decentralization, esp. the fact that the Fediverse is not a monolithic entity mandating a uniformly aligned approach, useful.
On the whole, I do think either ActivityPub’s protocol spec would need some kind of privacy revision, seeing as it’s already been a Problem where microblogging admins have had to block access by servers dedicated to mirroring Mastodon posts which don’t delete their copies after posts are deleted by the user, or the software itself, Lemmy in our case, will have to make adjustments to its implementation of federation like you said. Of course, I’m mostly just conjecturing here and I don’t actually know what either of these might look like 😅
The main part of this which I problematize are the people who are sticking their necks out for Meta and suggesting instances shouldn’t be quick to defederate because this is, supposedly, a good opportunity to bring federated social media into the mainstream. Yet, in my opinion, they’re not making enough of the fact that, even with their open-source contributions, Meta’s software manufactures discord and bigotry on a massive scale. Letting them federate with an instance opens floodgates on that and for the stealing and selling of Fediverse participants’ data.
So, on one hand, yes. absolutely agreed, on all counts.
On the other hand, the point of social media is to engage with people. What if your mom has an account on meta’s new activitypub platform? Is the interoperability of these platforms not also a huge feature? What if I want to follow my mom on mastodon when she’s on facebook whatever, but not give meta my data? These all work best when we can protect ourselves and engage responsibly, and defederation/blocking at a server level, while a WONDERFUL emergency button, also rejects a lot of the funcitonality and beauty of the fediverse. And I think there’s probably room to find a middle ground that protects users well while still letting them have that sort of engagement?
Mom doesn’t necessarily have to be on Meta. If she wants her son to engage with her on a platform, she can be on kbin, lemmy or any other FOSS alternative once it reaches maturity.
Not being tied to a giant corporation should not mean “obscure” or “unusable for normal people”.
People figured out email, they can figure out the fediverse, it just needs time.
Two things–
- I agree it’s going to take time, but do you think meta will wait for that time to come?
- people did figure out email, but how many lay people are only using Gmail and not protonmail or whatever else?
Their prequel on Meta: A few thoughts about #Meta’s #ActivityPub project (and whether we should instantly block it)
To recap: I’m also very, very suspicious of Meta and I know they don’t have good intentions - I’m not suggesting that maybe they’ve changed and they will do things differently, to “give them a chance” first. I just don’t think that declaring to block them makes much sense at this point in time. Maybe they will give us real reasons to block them once they launch their platform. But I’m not by principle against interacting with Meta users, as long as I can avoid Meta’s ads, black box algorithm and data mining.
I guess you do need to know the domain name first to block it.
That’s about all I’m interested in. They took a huge shit in the kiddy pool last time, and then sold vials of the water for one million dollars each.
I don’t want to swim with meta.
I really don’t like Meta but we should encourage the use of open protocols, regardless of who uses them.
I agree, but the issue I see is that if we don’t de-federate them, they could aggregate and sell our data, which is something many people explicitly switched off of Meta services for.
Can’t they aggregate and sell your Lemmy activity data without federating with your sever? It’s all in the public Internet.
I guess it won’t be that useful if they have no way of tracking you. But, I don’t think federation makes it easier to track users. Do sever admins get ip addresses of every user on a server they federate with?