cross-posted from: https://beehaw.org/post/567170
We’ve been defederated. Were there that many trolls/assholes on our server? What on earth happened while I was asleep?
hey folks, we’ll be quick and to the point with this one:
we have made the decision to defederate from lemmy.world and sh.itjust.works. we recognize this is hugely inconvenient for a wide variety of reasons, but we think this is a decision we need to take immediately. the remainder of the post details our thoughts and decision-making on why this is necessary.
we have been concerned with how sustainable the explosion of new users on Lemmy is–particularly with federation in mind–basically since it began. i have already related how difficult dealing with the explosion has been just constrained to this instance for us four Admins, and increasingly we’re being confronted with external vectors we have to deal with that have further stressed our capabilities (elaborated on below).
an unfortunate reality we’ve also found is we just don’t have the tools or the time here to parse out all the good from all the bad. all we have is a nuke and some pretty rudimentary mod powers that don’t scale well. we have a list of improvements we’d like to see both on the moderation side of Lemmy and federation if at all possible–but we’re unanimous in the belief that we can’t wait on what we want to be developed here. separately, we want to do this now, while the band-aid can be ripped off with substantially less pain.
aside from/complementary to what’s mentioned above, our reason for defederating, by and large, boils down to:
- these two instances’ open registration policy, which is extremely problematic for us given how federation works and how trivial it makes trolling, harassment, and other undesirable behavior;
- the disproportionate number of moderator actions we take against users of these two instances, and the general amount of time we have to dedicate to bad actors on those two instances;
- our need to preserve not only a moderated community but a vibe and general feeling this is actually a safe space for our users to participate in;
- and the reality that fulfilling our ethos is simply not possible when we not only have to account for our own users but have to account for literally tens of thousands of new, completely unvetted users, some of whom explicitly see spaces like this as desirable to troll and disrupt and others of whom simply don’t care about what our instance stands for
as Gaywallet puts it, in our discussion of whether to do this:
There’s a lot of soft moderating that happens, where people step in to diffuse tense situations. But it’s not just that, there’s a vibe that comes along with it. Most people need a lot of trust and support to open up, and it’s really hard to trust and support who’s around you when there are bad actors. People shut themselves off in various ways when there’s more hostility around them. They’ll even shut themselves off when there’s fake nice behavior around. There’s a lot of nuance in modding a community like this and it’s not just where we take moderator actions- sometimes people need to step in to diffuse, to negotiate, to help people grow. This only works when everyone is on the same page about our ethos and right now we can’t even assess that for people who aren’t from our instance, so we’re walking a tightrope by trying to give everyone the benefit of the doubt. That isn’t sustainable forever and especially not in the face of massive growth on such a short timeframe.
Explicitly safe spaces in real life typically aren’t open to having strangers walk in off the street, even if they have a bouncer to throw problematic people out. A single negative interaction might require a lot of energy to undo.
and, to reiterate: we understand that a lot of people legitimately and fairly use these instances, and this is going to be painful while it’s in effect. but we hope you can understand why we’re doing this. our words, when we talk about building something better here, are not idle platitudes, and we are not out to build a space that grows at any cost. we want a better space, and we think this is necessary to do that right now. if you disagree we understand that, but we hope you can if nothing else come away with the understanding it was an informed decision.
this is also not a permanent judgement (or a moral one on the part of either community’s owner, i should add–we just have differing interests here and that’s fine). in the future as tools develop, cultures settle, attitudes and interest change, and the wave of newcomers settles down, we’ll reassess whether we feel capable of refederating with these communities.
thanks for using our site folks.
But they can ban those users, just like they get banned individually in their own communities.
There is pretty much no difference, unless organised brigading was going on from lemmy.world ,etc. which I doubt.
I’m not sure why you doubt it. People have corroborated that slurs were being mass spammed. I myself witnessed a troll on lemmy.world yesterday who dropped an f-bomb and told a lascivious tale about dessalines. They can ban them for 5 minutes until they make another account and start spamming again. It’s pretty easy for a handful of trolls to cause chaos, especially if they had only 4 mods. The mods need to sleep sometimes also.
Yeah, I saw a spammer on lemmy.world spamming scat porn images (unfortunately there is still no way of blocking all images). I reported them and they were banned.
But they can still do this from any federated instance. Meanwhile they just banned 20k+ contributing users.
It’ll be interesting to see if they block kbin.social, fed.ia and startrek.website soon. All these same arguments work for them.
When federation wasn’t working for kbin, the community felt like a safe space. It was calm and interesting. With federation in place the main negative thing I have seen so far is downvotes everywhere. Yes, the Lemmy community feels a lot more Wild West too than kbin did.
I obviously can’t talk for every single sh.itjust.works user, but it doesn’t feel like that here to me, even with federation on. It really depends on the instance and the subscriber base.
Barbarian, you’re a gentleman and scholar. Your patience and understanding of the fediverse has been crucial to the success and growth of this instance, and indeed Lemmy as a whole. So thankful you joined up with sh.itjust.works. Keep fighting the good fight!
I appreciate that, thank you.
They didn’t ban us, they temporarily defederated from our instances. Totally different things.
And no, I don’t think there are very many instances which are as quick and easy to make an account as us and lemmy.world. That’s how we grew so quickly, and we obviously picked up a few trolls along the way.
“temporarily”
It just reminds me of all the Mastodon drama, like them de-federating the Swedish one over not applying shared blocklists, etc.
Why not focus on blocking actual abusive users, and contributing to shared tooling, rather than trying to pre-emptively ban users?
There is nothing more permanent than a temporary solution.
// This is meant to be a temporary solution
I found that in a SQL script from 2001…in 2018.
haha. I’m still rocking a temp solution I put in place on my media server about a year ago.
You’re not wrong. I just think you’re choosing a silly hill to die on. What is the purpose of your line of questioning? Do you want to be federated with them or not?
Do you have any experience moderating a Lemmy instance of that size? Presumably they would have been focusing on blocking abusive users for the past week and ultimately decided they had to choose nuclear option.
Neither of us actually know firsthand what went down, but I’m choosing to sympathize and you’re choosing to criticize. Neither of us is necessarily wrong, but I feel it’s wiser to be supportive right now, given that we don’t know what the future might hold and the platform is in a fragile state.
The issue is that it affects the growth of Lemmy as a whole. Like the outcome of all the Mastodon drama was just that I gave up using it, but I was never a big Twitter use anyway.
Imagine the average Redditor’s experience now - they come to browse some Lemmyverse posts. See that Beehaw has active Gaming and Technology communities like Reddit, so try to sign up there. Wait 2-3 hours with no response to their registration and then give up and go back to Reddit.
Ok, I see where you’re coming from.
I guess I just don’t feel the same because I never expected the original servers to be anything but an afterthought. From my perspective, Beehaw was never going to play a large role in the growth of this platform. It was kinda obvious that they were going to have a problem with the influx of users. So getting this out of the way early is actually one of the best ways for it to play out
You seem very concerned about controlling what other people are doing for someone who is experiencing angst over those other people controlling what you can do in their space.
That wasn’t their point at all
Isn’t the issue that this instance (and some others) don’t have any controls on creating new accounts?
Neither does Reddit.
What is your proposed solution? Private invites like lobste.rs ? Or the current tiny trickle of allowed users on Beehaw ?
I think open growth and banning spammers while developing analysis tools to automate it is the best way.
I was just explaining what their issue was, I wasn’t defending it or proposing any solutions.
They can do this from any instance with open registrations. Instances with closed or moderated registrations are a different story. So, could just as easily say, why doesn’t sh.itjust.works restrict new user creation, which is a common practice on smaller fediverse instances including beehaw.
(Answer: then you can’t onboard new users in a mass migration. But that’s what creates the tension.)
Doesn’t have to be organized when there’s 30,000 people on the sites they’re blocking. It can be a marginal percentage of users and still be hundreds of people they need to ban. Potentially repeatedly if the banned trolls just start using sockpuppets.
A small number of mods + dedicated bad actors = blocking the tools those bad actors use.