70
[Discussion] Pre-emptive Threads defederation
(sh.itjust.works)
Should sh.itjust.works preemptively defederate from Threads?
Threads is the not-so-new ~~reddit-like~~ twitter-like public forum platform by Meta, the same commercial company behind internet behemoths like Facebook, Instagram and Whatsapp. They're working on ActivityPub integration so that they can bridge (federate?) with the fediverse. As far as I know, the focus is on Mastodon instances, but in the future that could include Lemmy instances too.
Some have raised the question, worried about the future of the fediverse or even claiming that it goes against its definition.
What do you think should be done?
EDIT: correction
EDIT.2: The Vote is on! Go make your voice heard. You have until Friday the 29th.
Serious question - I’m not up to speed on what kind of effort goes into defederating/refederating. If it’s easy, then is “wait and see” an option?
Alternatively, is preemptively defederating now and refederating later if we want also an option?
If both options are on the table, I say wait and see just because it’s unexplored territory though I agree that it probably won’t be great. If it’s a big pain in the ass to change our stance, also wait and see; that way we only have to do the PITA thing once.
It is, there's basically a setting on the server to block certain instances. So the admin would just update that setting to enable/disable federation. However, defederation doesn't delete any data on either side, it just stops the flow of new data.
So if you care about Meta having a copy of all content for some period, the better option is to defederate now and refederate later if they somehow play nice. There is a risk that Threads could defederate from any instances that blocked it, but I'm guessing they're not going to bother, I think they're just looking for data to scrape.
It's actually relatively trivial to scrape the public data of any instance. If Meta really cared (which I highly doubt), there's nothing stopping them spinning up a temporary instance with a bit that auto-subscribes to all communities it knows about.
Defederating from threads doesn't change the fact that all information on any instance is pseudonymous, but very public.
Sure, but they have to actively do that, whereas federation just gives it to them. I want them to have to put in the work to scrape the various instances, not get it for free.
Couldn't they just get an intern to spend an afternoon actively doing that? It can't be that hard. I'm not a programmer, but it seems to me that if someone has the proficiency to set up an instance, they would have the proficiency to set up crawlers.
Probably, but they'll want to integrate it somehow as well. One of the main points of federation is to get access to more content, which means your users have more stuff to look at, which means you get more data to link interests to ads to improve click-through. Scraping it would just give them the data, but they'd have to recreate it for their users to consume.
So scraping is only half the battle here.
Both are an option and easy enough.
Honestly, and that's my personal take as a user not as an admin... I'm in the "wait and see" boat, but I'm nor particularly full of trust about this so my goto would be: