Post AIuRDlZDX05o7dl3vE by vandys@mst.vsta.org
(DIR) More posts by vandys@mst.vsta.org
(DIR) Post #AIuRDl0paqaSOzfbKC by tomasino@tilde.zone
2022-04-28T13:22:09Z
0 likes, 0 repeats
Today I've blocked 3 or 4 whole instances from federating with our server. That's a major step in combatting abuse which is available to instance admins and moderators. Rather than blocking individual accounts we can look and say "this instance is not moderating its own users appropriately and will continue to be a source of problems". Then we block them from communicating with our server, sort of cutting them out of the community.Normally if you run into someone behaving poorly you should report the individual account. Their moderation team will review and should take action if needed. Your instance moderation team will also get notified and can act for your local instance. If a moderator sees repeated problems from a single instance they may look deeper and decide to instance block. In some cases moderators will preemptively block instances too (such as Gab or Truth.Social in the past).This is one of the ways we keep our communities healthy! If you find yourself on an instance that has been blocked by other instances you can move to a healthier one with better moderation. If you host an instance for your community, pick good moderators and keep it healthy to keep it connected.
(DIR) Post #AIuRDlZDX05o7dl3vE by vandys@mst.vsta.org
2022-04-28T13:29:25Z
0 likes, 0 repeats
Hi @tomasino thanks for the primer.I would add that you should embrace censorship thoughtfully. While our societal ills no doubt result from many interacting trends, people ending up in echo chambers is certainly one of them.Blocking out everything except a match for your own current conclusion is a great way to get there.
(DIR) Post #AIwKQL4v13i6N2YPey by __h2__@mastodon.social
2022-04-29T10:56:17Z
1 likes, 0 repeats
@tomasino I never understood this practice. Why is it not up to the user to decide which other users and/or instances to block?I always explain fediverse by saying that it is "similar to e-mail". But none of us would advocate for e-mail providers to start rejecting incoming mail from certain domains, would we?
(DIR) Post #AIwZoQRYDeUXGx5Evo by tomasino@tilde.zone
2022-04-29T11:10:28Z
0 likes, 0 repeats
@__h2__ well the analogy isn't perfect, but we do block mail from certain domains quite regularly. If a domain is shown to be a source of bad behavior (in emails' case, spam, phishing, etc) it will be added to blocklists, like https://www.spamhaus.org/dbl/. Then other email clients and services can use that information to block it, defederating.And while it's possible to run your own email server that does not block anything it's also possible with the fediverse. Some users may choose to do that, but in community instances there's often an expectation that moderators will protect the interests of that community by acting to block sources of problems. For instance, instances that repeatedly allow users who use racist language or demonstrate transphobia. The presence of that sort of traffic in a users timeline can be damaging, and leaving it to the user alone to block each instance can dehumanize them.For all these things there's an element of personal judgement, and members of fedi communities talk about it frequently. Good moderators listen. Bad ones lose their members to other instances.
(DIR) Post #AIwZoR3TwcphAapX3Q by __h2__@mastodon.social
2022-04-29T14:12:39Z
1 likes, 0 repeats
@tomasino Thanks for the explanation. I guess I have always viewed fedi as less of "a specific community" and more as a means to connect to specific people that I am interested in (who are spread over multiple servers).I realise I am the kind of person less likely to be harassed online, but couldn't there be other ways than instance-block by the admin? E.g. users opting into blocklists or just allowing people you follow to @ you. Or just your instance.
(DIR) Post #AIwm9M038OBz1QxwRc by __h2__@mastodon.social
2022-04-29T14:15:44Z
0 likes, 0 repeats
@tomasino Thing is we all agree on some things, but we also disagree on other things.For example, I might actually be interested in what Donald Trump has to say. Not because I like him, but because what he says has an impact on the world. When I don't feel like reading the guy's posts, I can always not look at them / filter / block.
(DIR) Post #AIwm9MfscreX7AXLe4 by __h2__@mastodon.social
2022-04-29T14:25:00Z
0 likes, 0 repeats
@tomasino Another example: quite clearly German and EU government institutions are allowed to federate with most of fedi (even Frontex is mirrored!).But would we allow an instance run by the Russian gov to federate?Who gets to decide?And is (dis)trust transitive, i.e. if I run my own server, would I be blocked for not blocking others?
(DIR) Post #AIwm9NFgTkICuDHwS8 by tomasino@tilde.zone
2022-04-29T15:56:39Z
0 likes, 0 repeats
@__h2__ instance admins and moderators determine their own filtering, though there are some bulk lists available. As someone pointed out earlier, those lists are only as trustworthy as whomever put them together.In the case of a Russian state instance it would again be up to the individual instances, but once enough had blocked it, that would hamper its ability to reach others. I like to think of instance blocks like so (I'm not a federation expert, so someone with more technical knowledge, please correct me on this):My instance gets an incoming package of activity pub data. My server says, "whee! let me just trim out the icky bits" and filters things I have blocked. It then passes it along to other servers I federate with, but trimmed down. If that second instance has no other connections that chain their way back to Russian Instance and I was blocking Russian Instance, then they wouldn't see it either. If one of their users decided to follow a user on Russian Instance, though, it would begin directly federating with them.
(DIR) Post #AIwm9NomNGMif3hy9g by AIaYYAle4i1uKmKpqy.gme@bofh.social
2022-04-29T16:33:16.392284Z
0 likes, 0 repeats
I have run several instances in the fediverse and of course I've been part of several more.Here's my take:Just about every microblogging that supports ActivityPub supports the ability for end users to control what accounts can and can't communicate with them. Some even allow the user to control what domains can and can't communicate with them.This is how it should be in my opinion.If a user is being harassed, they first should tackle it and take care of it themselves.But there are legitimate reasons for an admin to block an entire instance. If the admin feels that instance as a whole is unhealthy for the network or their instance they are entirely within their right and have an obligation.The email analogy continues to hold up here.If a domain is constantly flooding your domain with spam, it makes more sense to add that domain to your blocklist rather than requiring every one of your users to block mail separately.But, most postmasters will ignore requests from their users to block email from a domain just because they don't like receiving email from it. They've got filters. Use them.