I dont like tankies or tories any more than the next person, but breaking federation is just wrong.
I dont want to have to make a separate account just to get around that, mainly because this is actually already my account for getting around that!
Its quite easy to block a community at user level, if needed, and we are not the target of any spam, but now we users have lost the option of the ability to interact forever with a corner of the threadiverse, which i think is not cool.
If its just me thinking this way, fine, i’ll just maintain several accounts, but i would hope its not, because its feeling like instances are gettinh pretty triggerhappy with the block button https://github.com/maltfield/awesome-lemmy-instances/tree/main
After a second thought, I’ll keep the blocklist free of political instances, even though most I personally disagree with. In terms of any large ones which are grey areas of legality (loli and stuff like that), yeh I’m protecting my own back by blocking them.
That does mean the blocklist is only 1 atm, I’ll keep an eye on what other big instances are blocked by other instances and keep the blocklist updated by doing that. Realistically a lot of the ones I would want to block will be so small I wouldn’t notice them so that list will be kept fairly thin I imagine. (I’m not gonna be like beehaw or even feddit.de who have fairly large blocklists)
I get the frustration of having stuff blocked which might feel restrictive, so I’ll try and keep the blocklist as small as possible but I do still think some should be blocked.
What is the blocked one :P
Blocked instances can be found here https://feddit.uk/instances
The one blocked atm is https://burggit.moe/
Okay! It is a bit weird so I can see why. I do ask that you DO NOT defederate from anything meta does as a kneejerk reaction as many want to do that. Give it a chance at least 🙏
breaking federation
Federation doesn’t imply that everything is open to everything. It can’t do, there are too many people that just can’t be civilised to each other.
If federation is to work and build enjoyable and productive discussions, then people’s behaviour cannot be completely free. And the work to notice and block individuals will largely be done on an instance level. You can’t have every instances maintaining a blocklist of every undesirable account across the entire ‘fediverse’
If instances are not able to keep on top of bad users then unfederating those instances is the built-in check. Hopefully that will not be required too often.
There are plenty of instances whose entire purpose is to facilitate discussion that is fundamentally distasteful. They are free to do so, but we should feel no obligation to allow that to happen through our site.
I’ve no problem with maintaining a minimal blocklist in as far as we are a small site and are less likely to face problems, but we should be prepared to block where not doing so makes the experience worse.
Hi, chiming in as a person from Mastodon from the past 7 years.
The rake in the grass that Feddit UK is going to run into is Transphobia.
FedditUK says in its rules that it wont allow Transphobia in its communities, fair enough, if thats truly what they mean great.
But this is the same rake in the grass that many of the newer instances over on mastodon since Musk took over twitter have run into, many of the journalism instances for example had strong “No Transphobia” rules, right up until a Telegraph journalist that was a user on their instance posted a transphobic article at which point the mods suddenly felt they simply couldnt restrict the free speech of a journalist even though it directly contradicted their own server rules.
The firm “No racism, no transphobia, no homophobia, no casteism” of the federation is pretty strict on that and many new comers to the fediverse find it shocking how “i just want to debate your right to use the toilet” isnt allowed on federated instances. And remaining connected to the greater federation requires that social contract and showing you’re actually moderating and removing bad actors from your instance and blocking bad faith instances to protect your users and users of other federated instances who may use your communities/magazines/subfeddits for discussion.
The giant federation that agrees to these rules thrive as a big hub network, but theres thousands of tiny instances out there that dont that are defederated for very good reason (loli, gore, terrorism material that is illegal to own/view in the UK), a dark side to the fediverse ocean which is often joking referred to by mods as the “c*m” side of the federation (due to the sheer staggering amount of tiny 1-5 people instances with that word in its url), that includes places like gab, truthsocial, spinsterdotxyz (anti-trans subreddit), and so on.
lemmy federates with mastodon and pleroma, which means its open to all the above instances, theres gonna be a ton of instances that lemmy instances need to block to keep their users safe from spam and illegal content.
But after that, the question still remains, what are the admin or moderators of any of the FedditUK communities going to do when the transphobia rake appears in front of them?
This makes sense. But the biggest issue is that the word “transphobia” means different things to different people, which then makes it a nightmare for anyone trying to police a space with free speech.
There is a similar problem with Homophobia but less marked and controversial perhaps? I’m a gay guy. If someone says “I hate gay people” thats obviously homophobia. But if someone says “I don’t think gay people should marry, because marriage is a religious thing” is that homophobia? I don’t agree with it but it’s an opinion. Someone can be hateful and hold that opinion, but that opinion itself doesn’t mean the person is hateful - at least to me. But another gay person may say “no that is hateful in itself”. Where do you draw the line?
Also to my understanding the fediverse doesn’t have any written rules on what is allowed. Each server sets their rules, and there has been consensus around certain rules (such as those you mention). The complexity comes in enforcing the intepretation of rules by one server on another, and the risk is fragmentation with some places defederated between others and people getting confused what is or isn’t interconnected. A may defederate from B, but both may stay federated with C. The content on A and B is visible from C but users interacting with content from C on A and B won’t see each other in C’s communities or interect.
I have no answers for this. It’s just going to be a big challenge that is inherent to this model. But it is still way better than what exists in old social media.
Yes, people used to twitter and reddit do often think thats a problem that cant be solved, but quickly find out it really is actually pretty quickly solved on the federated side of the fediverse.
“I dont think gay people should be allowed to get married” is a homophobic statement, so yeah no you cant post that publicly online on a space that doesnt allow homophobia. It gets flagged as homophobic, the person who flagged it’s moderators immediately talk to the person who posted it’s moderators, and they’re asked to remove it and either warn the poster from ever doing it again or if they’re seen to be a bad faith account thats just trying to spread hatespeech, they’re given the boot from the instance. If the instance moderators refuse to accept it as homophobia, then a fediblock post is made and over the next few hours most instance admins read the situation that an instance is refusing to uphold the pretty clear cut no homophobia stance and say hey if you dont start enforcing your own rules that you list on your site, we’re going to have to assume you’re a badfaith instance, a danger to our communities and defederate from you. It really is that simple. Same with transphobia, casteism or racism. People are welcome to hold hateful viewpoints, but you cant start trying to debate them or push them on a site that doesnt allow it in its rules.
The Fediverse, ActivityPub, was built by transgender people, the moderators of many of the key instances, are transgender, the federation is deeply transgender and the “No Transphobia” rule is pretty concrete in the same level of rock hard as “No Racism” regarding what most good faith instances accept those things to mean.
Thats why its a rake in the grass for new instances, especially ones that are UK based due to their abysmal degrees of transphobia in their media and how far anti-trans activists have sealioned the idea of “yes but what really is transphobia” over the last 8 years.
The good faith federation doesnt have free speech, the federation isnt a place of free speech. The federation bans outright most places that include the phrase “free speech” in their community rules pages because its 100% only ever used as a dogwhistle by far right instances.
if someone says “I don’t think gay people should marry, because marriage is a religious thing”
I’d say that’s pretty homophobic. You don’t need to be an outright homophobe to say or do something homophobic, or support homophobic systems.
If someone says, “I hate black people” that’s racism, but if someone says “race based slavery should have never been ended and black people should lose their rights to live as they see fit” that’s an opinion. /s
You see how plain off that sounds?
trans main character syndrome strikes again.
there is sadly lot of hate out there, but why on earth do you think it all about trans ?
It’s not as easy as that.
The problem at the moment with lemmy is that the moderation options for communities are pretty basic. So a community moderator can only delete posts or comments, block individual users or block any posts apart form moderator posts. So if a community is the victim of a coordinated attack from another instance, there’s not a lot that the mods can do to defend it.
There will be instances created just with the intention of disrupting communities that they disagree with. If community mods can’t defend their communities, defederation may be the only option.
Also defederation is a two way street. An instance might decide to have a no censorship policy, but that might lead to it being defederated by other large instances.
I think it should be fair if a community is illegal or starts to attack our communities with spam and brigading, but that’s it
This will very quickly go the same way as IRC. The spirit of Eris lives again on ActivityPub.
https://en.wikipedia.org/wiki/EFnet
IRC is a plaintext chat system which was originally decentralised and federated. Technically decentralisation and federation is still perfectly possible in the IRC protocol. Nobody does it so freely any more, ever since the Eris Free Net decided to go its own way because it made sense to do so.
It achieved this by blocking any IRC server that is connected to the Eris server.
In this case it wasn’t political - it was due to spam, flamewars, and organisational problems (what I just called ‘the spirit of Eris’). If you throw politics in there (or rather the fact that these days moderators are accustomed to seeing political views they already agree with on the Internet), I think it just will accelerate things towards defederation, or federation with only people we seem to agree with (read: easy to moderate).
I tend to agree that blocking some political subs etc is perhaps not the best, since every user on the instance will suffer, but please take a skim-read of the banned Beehaw instances - you will find something you do not want on here. Tom as a server owner also will have a vested interest in blocking some of this content that passes through his infrastructure.
Because the content is actually hosted on the server, I definitely think there’s a legal and moral cause to block some of these.
Sure, let Lemmygrad in, that’s fine, but a lot of those links are staying unclicked for me and I’d probably want my server clean of it.
From my cursory inspection of how things are working, all the the content is replicated between instances but things like images are just links, the UI and the back-end provide a means to upload an image but what that is doing is uploading an image to the user’s instance and then putting the link in the post. Pretty standard stuff for a link aggregator. The important thing here is that the image itself is not replicated around the fediverse, but stays where it was uploaded and is linked to. Of course, thumbnails seem to still be locally generated as you’d expect.
So the legal issue is diminished by that, but not illuminated. Something like loli hentai that is illegal in the UK won’t be hosted on these servers even if posted on a federated server, though I’m not saying it wouldn’t be reasonable to defederate from that instance on other grounds than legal liability. However the thumbnail is a very low resolution image of illegal content, and I don’t think lemmy has some kind of “grey list” where content is still available through your instance but nothing of consequence gets replicated locally, thumbnails not generated, etc. I’m not sure what the legal situation is surrounding outlinks to illegal content not thumbnails thereof. Furthermore I don’t know what kind of protections platforms have when it come to liability for user-generated content, like the US Section 230, but I don’t think there is much.
It might be better to host the actual servers for this instance outside of the UK as a result, that could very well be the case anyway.
Anyway IANAL and I’ve only done some cursory poking around into how lemmy handles the replication of content between instances so there are almost certainly things I’m overlooking.
Lemmygrad are fine as long as they don’t start brigading my Northern Ireland community.
I’m generally against defederation as much as possible, but would personally prefer that the proper NSFW stuff wasn’t on my feed.
It can be a bit jarring to be scrolling through content and there’s suddenly an anus or something on my screen…
I don’t think it’s too much for people to have a separate account for porn and such if they want to see that, and it helps protect the feddit.uk instance from legal issues.
It appears this instance straight-up has NSFW disabled, you can’t opt-in to seeing NSFW posts on your feed nor make posts marked as such. I might be wrong about what it being disabled means. So that issue may be moot and not require defederation.
One the flip side I do wonder if a blanket disabling of “NSFW” is too much, there are plenty of things that one might reasonably wish to mark or be marked as NSFW that is not porn, gore, or other such content. This could have the effect of either stifling some of the communities on this instance or people posting things in the grey area without marking them NSFW, even if they would have if the button to do so we’re present and people scrolling may have preferred that they did.
Food for thought.
deleted by creator
