I used to think that there would be 1, main ‘Fediverse’ with all of the ‘big instances’ connected to each other. The recent Threads debacle has shown me otherwise.
The point of the Fediverse is that there is no one single entity, or group of entities, dominating it all.
Right now it feels like whatever the big instances do, we kind of have to go along with to be a part of anything. As the Fediverse grows, there will be more options to suit different types of users.
I think it’s fine if big instances federate with Threads and it’s fine if they don’t. People can just join instances that align with what they want. It’s not like defederating means being cut out of the Fediverse, that’s not possible.
Great design. I’m eager to see how it plays out.
The most reasonable solution I’ve seen so far, from the pixelfed and pixelfed.social creator
https://mastodon.social/@dansup/111617703110836835
Not the solution I was hoping for but it’s an extremely reasonable compromise. I’ve never heard of selective authorized fetch. Pretty sure he just invented it.
FOSS ingenuity at work. All it needs is adaptation and adoption.
Authorised fetch has been a thing on Mastodon and I believe Akkoma too. I don’t know if Pleroma, Soapbox or Misskey have it though.
Authorized Fetch has been a thing for a bit on Mastodon at least - but as far as I can see it’s a global toggle rather than saying “If you present as a domain on the blocklist then you must be authorized to fetch this resource” (the selective authorized fetch I assume they’re talking about).
Never used Akkoma though, so I can’t speak for it.
Yeah the selective part I think is new. I believe Akkoma’s authorised fetch is similar to Mastodon, though I’ve also heard it came at the cost of breaking MRFs (essentially policies to handle incoming messages, that can be custom-written if needed)
Unless I’m wrong, the unique thing here is that auth fetch is always off for the server. It’s on only at the user level and it’s only on at that level if a user has an active domain block.
That could actually solve a lot of problems for people. Admins are reluctant to enable it server-wide because it causes a bunch of problems. The biggest being that it breaks federation with servers running older software (Mastodon v <3.0 I think) and with other services (Pleroma, maybe others). It also uses more server resources. But there are always people who think it’s worth it.
It’s available on Rebased/Soapbox
For the lazy:
That’s a good solution. Keeps the all feed clear of threads content while allowing users to opt in
I think most Mastodon instances also simply silence Threads: Mastodon users can follow Threads users, but Threads won’t fill every algorithmic timeline and doesn’t get suggested.
Unfortunately, neither Lemmy nor Kbin have such a feature, though Lemmy’s user side server blocking introduced in 0.19.0 should help a little.
I’m pro federate, but honestly, this seems fair. However Lemmy wouldn’t need it, as to see a threads post on lemmy, the person would have to @ the lemmy community in their post.
I think limiting/silencing is an excellent solution. It gives the people who want to interact with Threads a chance, while not disturbing the people who don’t.
In regards to Lemmy: I’ve seen a few Mastodon users tag Lemmy communities and creating posts here (generating some hilariously broken Markdown titles in the process), so we may see a few posts here or there. I don’t think anyone over at Threads knows what the Threadiverse is, though, so it shouldn’t cause any problems.
Yep. And the ones that do know and want to post to our communities would probably have the right intentions anyway
All true, and making this a feature would simply be implementing the inverse of the new capability… overriding an instance level block instead of imposing one not already at the server level.
I don’t think instance level blocks should be overridden. Some blocks are the result of moderation disagreements (beehaw vs lemmy.world vs exploding-heads vs hexbear), others are because of NSFW content, but there are a bunch of Fediverse servers full of pornography that would be illegal to host in many countries around the world (lolicon porn, for instance). If an administrator doesn’t want that trash on their server, they should have the ability to block it.
I think there’s a significant difference between completely blocking off a remote instance and making all interactions with said instance opt-in, and the server administrator should always have the final word. I think leaving the moderation on a separate layer like Nostr and Bluesky do it (not that Bluesky is federating at the moment) was a mistake, the result of some laudable free speech ideals that just don’t work out in the real world.
I think you only made a case for having two or more levels of instance block, that already exist. One due to objectionable/illegal material that cannot be overridden, and another for something like threads where a significant number of users may not want to be opted in automatically, or want to block it due to purely ideological, non-illegal reasons, which would effectively be put in place by automatically adding the instance block to user accounts that can be removed at any time, which arguably can already be done with minor changes. That’s essentially what dansup is doing, complete with including a command for Pixelfed instance admins to apply the optional block to all user accounts.
That’s not how Bluesky does it. Sure it empowers users which is great but if there is a PDS for example with lolicon, one I don’t see it making past the content filters but if it did the BGS/Relay would blacklist that PDS
Jerry Bell did it first!
https://infosec.exchange/@jerry/111591446262881196
Good to see more reasonable people in all of this.
Did Dan ever get the messaging service Sup going? Tried to look it up, but his name being Dansup is throwing a wrench in my Googlefu.
@[email protected]
Releasing in beta soon!
Also searching for #sup in Mastodon has been a good way to find information about developments. Not so necessary now that there’s an official account I guess. :)
Thanks for the tip!
I am unreasonably excited about this. Where I live, there’s no decent options for Internet or cell signal. Which means normal calling/texting doesn’t work, and regular Wi-Fi calling/texting is choppy at the best of times. My whole family uses WhatsApp for everything. I’m hoping I can get them to switch to something like this once it’s stable.