Gregory's Server@polotek all other things aside, what matters isn’t the technical architecture but the social norms. Mastodon is full of folks who are extremists (by modern tech standards) about consent, and who want defaults to be opt-in for nearly everything. In particular, Bluesky has both a different economic model and a different privacy model than the rest of the fediverse, so it makes sense to start with consent because the decision to federate is irrevocable in terms of data leakage.
|
57 comments
  @anildash I don't know whose standards you refer to, but that is one of the reasons I like it here. But that was a good analysis of the contrasts between Fedi and BS and why decisions to-federate-or-not are important. So thanks.  @anildash @polotek the social norms are worthless the first time a bad actor joins the scene. I find it irresponsible by those folks to claim mastodon/fedi has safety properties that it clearly doesn’t have. And then yell at folks for pointing that out as if that is a sustainable safety practice. ¯\_(ツ)_/¯  @anildash @janl I'm not arguing that people shouldn't be upset about this. I am arguing that being mad at this one person for doing what devs always do with open and available tech is not actually reasonable. Especially when these same people haven't taken advantage of any of the tools actually available to them to protect themselves from unwanted federation.  @polotek @anildash @janl  @adriano @jakelazaroff @polotek @anildash @janl I don’t think the dev is pushing work to you, I think the protocol is. It seems like federation is default opt-out, so every new instance is work for an admin @adriano @jakelazaroff @anildash @janl right protecting your own personal privacy should definitely be somebody else's work. @polotek @jakelazaroff @anildash @janl I shouldn't have to explain why somebody coming from a different network with the intent to monetize what I write without my permission and making me do extra work to avoid that is, yes, somebody else pushing work on me that I didn't want. But here we are. If they wanted to not have a problem with the protocol, they could have made something *with* activityPub. Granted, people wouldn't have liked the monetization bit either, but. @adriano @jakelazaroff @anildash @janl um, it is made with ActivityPub. The amount of people who don't actually understand what's happening is pretty wild. Which is entirely my point. You are creating work for other people by not understanding anything about how the system you chose actually works. And then getting mad at them for using it as intended.  @polotek @adriano @jakelazaroff @anildash The irony is that Ryan is asking permission, not forgiveness. He hasn't started running a bridge yet. If he had, likely no-one would notice anyway - did you notice that his original post was bridged from his website? @adriano @polotek @jakelazaroff @anildash I’m okay to be dried from this. @janl @adriano @jakelazaroff @anildash sorry Jan. I should stop provoking people. Nobody actually understands my point anyway. @adriano @polotek i mean, yeah, that’s the *entire point of activitypub*: anyone can spin up a server that can communicate with yours. it would be great if mastodon had a way to proactively block servers that identified as bridges or search services or something. but getting mad at this one dude for building a server you don’t like is not gonna fix anything. @adriano @jakelazaroff @polotek I was under the impression that our admins could block BS (I assume by blocking the bridge domain). Everyone forgets how steep the original learning curves were for every single step of the chaotic history of modern tech... imo nobody was 'extremist' ENOUGH about privacy, starting decades ago, and that's just the problem but adjustments that seem impossible for people to make now can become second nature in a matter of months. @anildash what we disagree on is the definition of "community". Many people I've talked to about this don't give a fuck about community. They care about themselves. And they didn't know anything about what the norms were. They just hate bluesky. You're projecting the norms of your own circles onto a much wider group of people. Also there is a strong argument to be made that the actual community norms are to federate with everybody. Because that's what happens in almost all cases today. @anildash you may get to the rest of my threads about this at some point. But I'll say it here as well. My argument was never to say "people don't get to be mad about this". That's not even a discussion I care about. My discussion was "being mad at this one random guy does not solve your safety problem. So now what?"  @polotek @anildash i think people getting mad about it are trying to create and enforce cultural norms, they don't want the tools to be needed to protect people, they want it to be taboo, and harshly punished, to do the things they think are bad for privacy. @tshirtman @anildash right. Make one random guy's life hell as a warning to everyone else. Makes sense. @tshirtman @polotek it’s worked at chasing off lots of devs and even more casuals. For sure. @tshirtman This is getting thrown around a fair bit, but regulation is a legal/technical solution to social problems. The thing missing in a lot of what’s happening that bothers folks so much today (and arguably the reason Mastodon exists) is a reluctance/incompetence of gvts to regulate social media. I’m not suggesting I know how that should work but it’s still wild west and with ‘AI’ going to get worse pretty fast imo @tshirtman It seems a bit like the reason for the adage of being unable to solve social problems with technical solutions is that engineers are typically not trained in the social sciences. If they were (or if social scientists were welcome and encouraged to work in tandem with engineering teams) there’s a pretty good chance that technical solutions would be a lot more effective @tshirtman @polotek @anildash “you can’t fix social problems with technical solutions” is a phrase I see with basically no support whatsoever. Encryption, for example, seems to be a very effective solution to “don’t snoop on private conversations.” There’s even varying levels of technical security based on how much you care. @agocke @tshirtman @polotek this is only tech solving a social problem if every conversation everywhere is encrypted. You can't fix social problems with technical solutions. @anildash @tshirtman @polotek if it has to be solved everywhere to be considered solved, nothing can be solved. In some cases it looks like only tech can solve social problems, e.g. gun control.  @polotek @anildash someone here today spoke of “the” social contract of Mastodon and… no? I love me some social contracts, but this person’s idea of the network’s social contract was very different than mine (and I had an account on the first ActivityPub server). The network is already too large to have one set of norms, or one social contract. That fragmentation is perhaps not ideal, but it is reality.  @luis_in_brief @polotek @anildash I think Luis is probably right. At this point it's not conceivable to talk about the entire federated model in any cohesive sense of norms. That ship, if it ever even really existed sailed too long ago. Different servers can be loosely aligned and that's cool but there's no one broad consensus.  there is definitely a cohort of people who use mastodon, who create publicly-available posts on publicly-available servers, using an open source protocol designed for federation and propagation of their content, who seem to get offended when something (new) comes along to federate it? I'm all for privacy preservation and an aversion to big tech/Threads creeping into this space, but it feels like people broadly have a different notion of control of Their Stuff online. @repeattofade @polotek @janl yep, it’s an evolving social norm. It’s unfamiliar to people who came online in a context where protocols or formats or platforms were ultimate authority about consent. But it’s changing, and that’s fine, imo.  @anildash @janl @polotek He kind of does have to market well to Fediverse users though, or at least to Fediverse *admins*, because a failure to do that just gets you broadly defederated and then your project is dead before it even gets off the ground which is exactly what has happened (and has happened multiple times in that past when someone has tried to set up similar opt-out bridges).
they've already raised seed funding from "angel investors", they will doubtless continue to raise money from marks. given the founders' known propensities, they're probably using the content posted there for LLM datasets, which they'll sell. economic models don't require ads.  Technically burning through someone's capital on the way to starting the enshittification cycle *is* an economic model.  @anildash @polotek Hey Anil. Allow me to (politely!!🙂) question the statement “…who want defaults to be opt-in for nearly everything”. The Fediverse is mostly opt-out by default -- and by design. Your instance automatically federates with all other instances, unless you (a) block/mute them or (b) operate your instance in an opt-in (whitelist) model, which very few do. So the *federation* part of the Fediverse has, until now, mostly been opt-out. This bridge is just another AP endpoint, afterall. |
@anildash @polotek
"different economic model" different or much worse?