I got a lot of backlash for a thread on Lemmy the other day about how the “federation” system is fundamentally broken by its lack of Tor support.
To recap: the lack of Tor support seems like part of why Lemmy’s existence so far hasn’t brought the world’s “overton window” any further from capitalist fascism than it was at reddit’s peak. It just isn’t currently designed for the task.
If Lemmy got big enough to make too many people think it’s potentially on the cusp of triggering a serious organized resistance, it would just be fractured in a crackdown by the authorities, like reddit was when it got too big.
With the “federation system” not supporting Tor, Lemmy isn’t designed to resist corporate (ICANN) control. It isn’t designed to help or attract users looking for that ability.
I can’t find any solid clarification for whether Piefed/mbin have the same issue. Do they? Can instances / servers be “federated” using Onion service addresses instead of DNS / IP addresses?
Before posting, I’ll address some replies I expect certain people might post instead of addressing my question (based on my experience the other day)
“This is a possible future problem, not a today problem.”
Incorrect. The server admins / devs / users on Lemmy today are limited by this already. If PieFed / mbin are the same way, this impacts everyone.
You’re not seeing posts from people who decide there is nowhere worth posting. You’re also not seeing posts from people who post where the authorities aren’t in control, if you’re only using DNS and IP addresses to fetch posts. You also can’t reply to them, if I understand correctly.
This also might have deep psychological impact, where it’s not only that you can’t see posts from those people, but that you’re driven closer to the middle of the “overton window” by fear of becoming one of those people yourself.
It’s extremely far from a “tomorrow, not today problem.”
“OK, so you think it’s a today problem, but that’s a lie. Every instance has an owner deciding their own rules freely, without Tor.”
This seems impossible, since without Tor you’d be relying on DNS / IP addresses that can be conveniently blocked by the authorities at any time.
By “Occam’s razor,” the lack of Tor “federation” also seems to explain why every “federated” instance I can find has rules other than simply don’t spam.
“Very few people want to see content banned by the authorities; mainly just pedophiles and Nazis, so that’s who you’d have as users in a place the authorities don’t control.”
As examples of large groups of people, that would want to share content “banned by the authorities” - I believe Nazi groups would be dwarfed by Luigi Mangione supporters, and pedophile groups would be dwarfed by “internet pirates.”
However, it is true that these groups wouldn’t behave exactly how I want. They might splinter into separate clusters, where all the pedophile/Nazi groups are sent to their own echo chamber, all the Luigi Mangione supporters settle in theirs, the pirates only venture into the pirate bubble for piracy purposes, etc.
What I would want is instead for everyone to recognize that you can’t have a serious place for political discussion in a democracy if it’s full of adults who aren’t willing to listen or talk to each other. It should be one supercluster of people who agree on that, with communities for all topics, where everyone can see and reply to each other if they want.
I would also think it’s pretty reasonable if we had a “Tor cluster” where the standard for instances is that they block nothing except spam, and a “mainstream cluster” where the standard is to remove nothing except spam and whatever else the authorities require removal of. No bans for anyone that doesn’t spam, or bring obviously banned material into the “mainstream cluster.”
That might be a lot to ask of the human race.
But it’s still a good idea to take control away from “the authorities” and have people share that control more equally, even if they might never do what I want.
Personally, I appreciate progress towards zero censorship, even if we never reach actual zero.
“OK, so you disagree with me even if you’re outnumbered, but Nazis will take over if you let them talk, so you’re a bad person for disagreeing. Paradox of tolerance.”
The Wikipedia page says “the paradox of tolerance is a philosophical concept suggesting that if a society extends tolerance to those who are intolerant, it risks enabling the eventual dominance of intolerance, thereby undermining the very principle of tolerance.”
I don’t get how that would be a paradox. It sounds like what’s called “irony,” not a “paradox.” I definitely don’t see how it’s supposed to make the concept of tolerance itself inherently paradoxical.
But if it is, what do you do about it? Do you somehow tolerate nothing in a fair, unbiased way? Calling tolerance “paradoxical” seems like an insane premise that can’t go anywhere logically.
If you’re using that paradox to justify censorship, you’re saying we should be intolerant of free speech because if we try to tolerate free speech, we could end up with people who are intolerant of free speech anyway? That sounds like you being a defeatist, not tolerance being paradoxical.
I don’t see tolerance as paradoxical. I just tolerate stuff I find tolerable, like free speech - and don’t tolerate stuff I find intolerable, like racist vitriol.
So, when a racist uses racial slurs, the problem for me isn’t that they’re allowed to have their own views or use their own words. The problem I see there is, they’re racist, they’re promoting science denialism, and it also brings to mind other problems, like how the racists I encounter are usually also climate science deniers who are still eating quite well while climate change is starting to get to the stage where it causes famine for people in other parts of the world.
I don’t care if we stop racists from using their phones to type dumb shit. I wish we were busy stopping them from using their guns and drones to keep food crops away from people who never promoted science denialism like they did.
It seems like the underlying issue isn’t even really about any “paradox.” The undertone is that some of you think censoring Nazis makes them go away. How? It seems to me like internet censorship and Nazism have increased together in recent years.
That’s anecdotal, but the science is also not in favor of echo chambers. There’s plenty of “actual research” backing me up.
“In line with our expectations, we find that partisan echo chambers increase both policy and affective polarization compared to mixed discussion groups.”
“The researchers found that when people preferentially connect to people with similar opinions, they create an echo chamber that increasingly polarizes the views of everyone in the network. On the other hand, people who are part of a network consisting of a variety of viewpoints tend to moderate one another. Understanding that social networks influence polarization — rather than merely reflect it — could be crucial in developing interventions to curb polarization online and the spread of political extremism, the researchers report.”
“Our analysis reveals that while these algorithms are designed to enhance user engagement and satisfaction, they inadvertently foster digital polarization, diminish exposure to diverse viewpoints, and contribute to the spread of misinformation.”
Of course, to base your opinion purely on the consensus of scientists would be the “appeal to authority” fallacy. When we have anecdotal evidence as widespread as there is, we should be able to discuss this issue without linking to studies.
However, the main way I know racism is wrong is because it’s science denialism. So, after talking about the example of racism, I would feel remiss not pointing out that to promote echo-chambering also seems like science denialism. So it seems like projection when you suggest anyone that won’t ban a racist from online discussion is then somehow akin to a racist.
“Whatever, it doesn’t matter if you disagree while being so outnumbered. Too many people think you’re wrong; that makes you wrong and the app we’re using right.”
This is a logical fallacy called the “bandwagon” or “argumentum ad populum.”
“I wish you wouldn’t post about this without fixing it yourself.”
I am fine with posting about it being my only involvement in it possibly being fixed. I don’t see how it’s really my problem if that upsets you.



Yeah, let’s hope you can find what you’re looking for.
Tl;dr: Sorry for the following long comment… I think your vision isn’t easy to implement in practice.
To share a bit of my perspective: You’re not the only one demanding (or wishing for) the Fediverse to be without “censorship”. That discussion sub-surfaces every now and then. But the Fediverse isn’t really something like this. At least not currently and it’s not meant to be like that. Behind the scenes we rely a lot on trust (also on a personal level). Between instance admins, moderators… And we’d need to entirely re-invent that to work anonymously. Then the protocol isn’t meant for anonymity. It leaks metadata everywhere and lots of things get forwarded publicly. You want your upvote to federate to my instance, so as an unfortunate consequence, I as an admin will know what you like and if you were doomscrolling and handing out upvotes on your phone 3 weeks ago at 2am. And people want to use it as social media and talk about personal stuff… I don’t think the Fediverse can properly leverage the anonymity aspect TOR offers, it’s not a good fit and that’s baked in at the core of our design.
And then the users don’t want it to be censorship-free. I suppose you noticed that in the other discussion. The vast majority of the userbase don’t like being yelled at, being trolled or spammed or generally: share the place they hang out in their leisure time with assholes. That’s just who we are as a community. There are other places on the internet which are more lenient. And in consequence they attract more people of the other kind. For example I remember 4chan to have a lot of trolls, people who post not-okay “memes” about jews and yell the N-word all day and generally behave like 12-year-olds with behavioral problems. Those places exist and people can just go there. But I think we deliberately chose to not be like that, and everyone is allowed to arrange their place however they like. And it’s more a psychology problem how people tend to form groups and migle with similar-minded people.
But with that said… The Fediverse is far from perfect. We get people disagree and step on each others nerves all the time. Moderators will make wrong decisions or abuse their power… And generally, even with some information available about users, we already have enough to do to deal with spam. People will do ban-evasion and make countless accounts, use it to annoy people. Or come to vent or troll in a way that is abusive to other people. And people have done worse stuff than just that. And the idea of moderation is to keep that somewhat contained so all the other people can do whatever we’re here for. The majority of it is hard work and legit.
The Fediverse is a nice idea, though. In theory it’s supposed to deal with some of this. The instances are loosely coupled, and our means of dealing with for example political censorship is to allow everyone to take the software and run their own instance. And they can connect to whomever they like. People made left-wing instances and right-wing instances… And a lot of them are silly for first demanding “free speech” but then it seems they always go ahead to censor the other side. People open new communities all the time because some other mod has (or doesn’t have) some stance towards Israel. We can create an account on a different instance worldwide once the UK or USA adopts new laws to impede with free-speech. We have one instance dedicated to pornography… Which is notoriously difficult to host in most jurisdictions. And I guess liability for the admin is an important aspect. I can’t have illegal stuff being forwarded to my instance or I might get in trouble. So I’d need a solution with 100% anonymity for me as an admin or I can’t take any responsibility. And that kind of clashes with the stuff I said before… I think it requires a different platform with all of that baked into the design and with a userbase who want that in the first place.
And by the way, I’m also like that. I’m all for tolerance and free-speech. But I also hate fascists, I don’t want to see gore pictures or be attacked on a personal level. And ultimately I’m somewhat in line with what legal restrictions my own government put into place. I wouldn’t want to host hate-speech and toxic stuff, even if it was allowed to me. I’d go even further and remove misinformation, because it’s only meant to get people riled up and make them stupid. So I’m willing to give those people the boot and remove the content. And I’ll agree with you that this is an issue and a design-problem. I’d love the Fediverse to change the way power gets delegated. We could assign power by a democratic process. Or maybe enable the users to subscribe to moderation, so they can mix and match and pick what kind of moderation they like on an individual level. Nomadic identity would be nice. And try to keep metadata and personal information local so it doesn’t spread through the network. Add more encryption and unbreakable pseudonymity. There are a lot of ideas and proposals out there (some wishful thinking), for how the utopic perfect platform would look like… I think for example Nostr was invented because the author didn’t like how other platforms (like us) are designed. It’s just a massive multifaceted problem. You got to deal with the people, find a way to keep the assholes out so the place doesn’t immediately turn into a shithole, then we need the vision and ideas on how to pull it off. And then deal with physics and how some of the technological aspects conflict with each other come up with solutions or trade-offs and then implement all of it. And I guess some parts of the Fediverse strive to be a nice place for arbitrary people and we factored in how everyone likes different things with the federation idea. But it’s really difficult and I suppose there is no easy and straightforward solution do just devise a perfect platform for everyone… I have a hunch it’s impossible. We have total “freedom” on one exteme and we can see how these attempts traditionally became shitholes filled with trolls or one type of people and everbody else got displaced/left. And we have heavily-moderated platforms on the other side and I guess with them it just depends on if you personally like that specific filter. So I don’t think any of the extremes is viable, we might just have to make ends meed with some compromise. Should be a bit more clever than the Fediverse is today, but I really don’t think we can do without compromising on some aspects.
You have a lot of good points here.
One of the main problems with my suggestion is the privacy issue you mentioned. Even using Tor doesn’t actually come with a promise of anonymity. That would stop me from running an instance myself, even if federation over Tor works.
Enjoying PieFed overall so far, will keep investigating how it works
Yes. I enjoy it as well. By the way since that last discussion was a bit heated and oftentimes not very constructive… What’s your motivation behind asking for zero-censorship? Is that more an ideological argument or do you have certain specific topics in mind you can’t say here, and is that because they’re illegal or because you live in a repressive country or because people get offended and block you? Feel free not to answer the question.
I’ve been banned from Facebook for years and reddit has also repeatedly banned me. Among many other bans, those are the ones that have had the most impact on my life and made me committed to censorship-resistant internet tech. I also believe the more open source tech can resist censorship, the more corporate platforms might be pressured to offer freedom of speech / unbiased enforcement in order to compete