#ActivityPub carries posts, follows, likes and so on, but not moderation.
-
But what about situations where someone is blocked/banned for no reason other than someone doesn’t like someone else? Let’s say, a difference in personality, nothing more. Is it fair that one person/instance can choose to block that person throughout the Fediverse?
What solution is in place for this situation?
@ercanbrack
@apps not to mention that not everybody agrees on what constitutes racism, sexism etc etc - different people view these things with different lenses. Not saying that is good or bad, but the entire concept of the Fediverse is precisely the power to choose who you engage with or not.With great power, comes great responsibility.
-
#ActivityPub carries posts, follows, likes and so on, but not moderation. A ban stays trapped on the instance that issued it. So, for instance, an account banned for racism on one instance stays unknown to the next. Moderation is activity too, and it should travel like the rest: signed and verifiable. This is a real problem, and others have explored it already. I will look into what already exists and see what we could build on for #HolosSocial.
@apps I think this is interesting. I don't think they should be automatically applied however? It might be useful to have some UI for showing all the remote e.g. ban notifications for a certain target. So maybe you'd see that N instances banned some user, all the reasoning if any, and some button for applying/rejecting that ban. Maybe you could also see the total instances that applied/rejected that action as well, so it's easier to form consensus? Maybe you can as a moderator reply to a mod action with a note or whatever.
Instances could Create<ModAction> and remote moderators could reply to, Like, whatever to that ModAction. Remotes could subscribe (standard Follow/Accept) to your instances ModActions collection (on the instance actor??) and get the above UI?
Above is just stream of consciousness and poorly thought out so forgive me
-
@apps I think this is interesting. I don't think they should be automatically applied however? It might be useful to have some UI for showing all the remote e.g. ban notifications for a certain target. So maybe you'd see that N instances banned some user, all the reasoning if any, and some button for applying/rejecting that ban. Maybe you could also see the total instances that applied/rejected that action as well, so it's easier to form consensus? Maybe you can as a moderator reply to a mod action with a note or whatever.
Instances could Create<ModAction> and remote moderators could reply to, Like, whatever to that ModAction. Remotes could subscribe (standard Follow/Accept) to your instances ModActions collection (on the instance actor??) and get the above UI?
Above is just stream of consciousness and poorly thought out so forgive me
@apps I guess rather than being a 'mod actions' collection it'd just be the instance actors outbox, if it was to be that. but you get the idea
-
#ActivityPub carries posts, follows, likes and so on, but not moderation. A ban stays trapped on the instance that issued it. So, for instance, an account banned for racism on one instance stays unknown to the next. Moderation is activity too, and it should travel like the rest: signed and verifiable. This is a real problem, and others have explored it already. I will look into what already exists and see what we could build on for #HolosSocial.
I have become increasing irritated when external websites recommended by posters use privacy/firewall/proxy detection to effectively Block access to that site but gain a site click benefit under false pretense of providing at least *some* useful information from the advertisement.
Telling me that I must turn it off.
Likewise the most aggressive that make the page useless with modal popovers unless you submit to coercion to add an email address and REGISTER first.
I am asking for a "Reputation Link" that
feeds a federated reputation database when readers get conned into visiting the worst offenders.after a threshold of down votes has been exceeded:
A quantified warning is issued with that external site link. It does not get removed until *they* issue a published apology with a statement that sets acceptable behavior limits towards open access.
It prevents them from gaining ANY benefit from misleading advertisement
implying that at least a summary will be readable. -
#ActivityPub carries posts, follows, likes and so on, but not moderation. A ban stays trapped on the instance that issued it. So, for instance, an account banned for racism on one instance stays unknown to the next. Moderation is activity too, and it should travel like the rest: signed and verifiable. This is a real problem, and others have explored it already. I will look into what already exists and see what we could build on for #HolosSocial.
@apps
Sounds like asking for a fediblockchain.
Most mod logs are better kept and discussed behind closed doors with cold heads in my opinion. Public access to what would be looked at like criminal records would promote toxic behavior. Not everyone agrees on what constitutes an issue or triages these equally, and public dogpiling always escalates them. -
#ActivityPub carries posts, follows, likes and so on, but not moderation. A ban stays trapped on the instance that issued it. So, for instance, an account banned for racism on one instance stays unknown to the next. Moderation is activity too, and it should travel like the rest: signed and verifiable. This is a real problem, and others have explored it already. I will look into what already exists and see what we could build on for #HolosSocial.
@apps ActivityPub already has the vocabulary to express this: a "Ban" should just be a "Delete" activity for an Actor in reply to one or multiple "Flag" activities for the same Actor.
-
But what about situations where someone is blocked/banned for no reason other than someone doesn’t like someone else? Let’s say, a difference in personality, nothing more. Is it fair that one person/instance can choose to block that person throughout the Fediverse?
What solution is in place for this situation?
@ercanbrack
No one blocks anyone across the Fediverse. A moderator only has the option to share what they did, and sharing is never automatic. The receiving instance gets the information and makes its own decision. It can rely on trusted instances, or more generally subscribe to the moderation of an instance it chooses. So a personal dislike does not spread on its own. It only travels if another instance decides to trust that source. -
But what about situations where someone is blocked/banned for no reason other than someone doesn’t like someone else? Let’s say, a difference in personality, nothing more. Is it fair that one person/instance can choose to block that person throughout the Fediverse?
What solution is in place for this situation?
An age ago I wrote on Federated moderation to start a bit of a brainstorm and collect links to relevant resources. FYI:
Nowadays @Bonfire have a Federated moderation toolkit, written about on their website, which may be worth checking out:
https://bonfirenetworks.org/moderation-toolkit/
@ercanbrack you bring up a valid point. Moderation is a power tool as @bloodaxe mentions too. Imho it should be reserved to dealing with bad actors. There's a distinction between moderating and *curating* ones social graph (or in case of an instance admin, the social graph of the people on that instance).
Currently blocking and muting are the only tools we have, and people use them liberally to curate their feeds. But a Block is like an ostracization, and muting is akin to shadowbanning. If done by mistake (a simple misunderstanding, which happens often) or on a whim, there's no recourse for the ostracised person. They become inhumane features, that have no equivalence in the offline world.
-
#ActivityPub carries posts, follows, likes and so on, but not moderation. A ban stays trapped on the instance that issued it. So, for instance, an account banned for racism on one instance stays unknown to the next. Moderation is activity too, and it should travel like the rest: signed and verifiable. This is a real problem, and others have explored it already. I will look into what already exists and see what we could build on for #HolosSocial.
@apps this is 100% working as intended. I don't want the shrillest of canceldon influencing moderation where I hang out, and they don't want poast influencing their moderation.
-
#ActivityPub carries posts, follows, likes and so on, but not moderation. A ban stays trapped on the instance that issued it. So, for instance, an account banned for racism on one instance stays unknown to the next. Moderation is activity too, and it should travel like the rest: signed and verifiable. This is a real problem, and others have explored it already. I will look into what already exists and see what we could build on for #HolosSocial.
@apps I am not sure if this is a good idea. But moderation mechanisms that already exist in federated groups can be applied to servers as well.
See https://codeberg.org/fediverse/fep/src/branch/main/fep/5219/fep-5219.md
-
? [[global:guest]] crossposted this topic to General Discussion
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login