In mastodon, your instance can stop you from seeing content from other instances and ban users from other instances. While this moderation might be nice sometimes, I’d rather it be opt-in than mandatory. Nostr relays don’t have this power. Likewise, Mastodon instances can stop their followers from following you. Nostr doesn’t allow this.
This is exactly why I won't use Nostr. What you're describing here isn't ideal for many folk that are part of marginalised groups. When each individual has to individually block every bigot only after being exposed to their bigotry, then the vulnerable folk don't hang around. This is doubly the case when there is nothing stopping the bigots from just creating another account after burning their first one.
One question that fediverse needs to solve is: how are we going to fund hosting costs for instances and more broadly, development?
This is also something that activitypub communities do better, because they are communities not relays.
This is exactly why I won’t use Nostr. What you’re describing here isn’t ideal for many folk that are part of marginalised groups. When each individual has to individually block every bigot only after being exposed to their bigotry, then the vulnerable folk don’t hang around. This is doubly the case when there is nothing stopping the bigots from just creating another account after burning their first one.
In Nostr, each relay can set its own policies. Relays can and do establish policies for acceptable behaviours. If you want a strict content policy, connect to relays with strict policies. You won't have to individually block any users or relays/instances. This is essentially the same as mastodon. The difference with nostr is that you normally connect to multiple relays, so a single relay, where your identity is tied to, cannot block you from following who you want and seeing whatever content you choose. Let's say Relay A blocks a user you want to follow. No problem, you are connected to relay B and C that don't. And, of course, if for some reason you only want to connect to a single relay, you can.
This is also something that activitypub communities do better, because they are communities not relays.
"Hope our commmunity of users donate" didn't work out well for the previous iteration of P2P discussion spaces: forums. The fact is, hosting online discussion forums gets costly quickly, especially if you want them to be reliable. Hell, even IRC servers which serve only text can get expensive to host. I'm not saying there's no way to convince users to donate to valuable instances, just saying that as a general strategy for FOSS it hasn't worked particularly well.
In Nostr, each relay can set its own policies. Relays can and do establish policies for acceptable behaviours.
Sure, but it's still each relay admin playing whack a mole with bigots as they pop up with new accounts.
On AP, because identities are tied to instances, the instance admin can kick the bigot and they're gone for everyone, before many folk ever see it. And if there is an instance with admins that don't deal with bigots, the admins can defederate from the entire instance.
Nostr doesn't give you any of those options. An instance is just a generic relay. No community, no differentiation, no protection for vulnerable folk. Which is fine if your goal is "free speech" but not so good if you're a member of a vulnerable minority just trying to connect with people/communities without having to be super on guard.
I left twitter to get away from an environment like that. I'm not going to head back to a federated version of the same thing
The fact is, hosting online discussion forums gets costly quickly, especially if you want them to be reliable.
Absolutely. I admin several AP instances, including the one I'm posting from now. We have crowd funding to help, but we are still out of pocket running it, but that's fine, because the reason we run it is for the whole community aspect I've been talking about.
I have zero interest in paying out of pocket to run a generic relay that will probably end up being used by the very people I'm trying to avoid.
On AP, because identities are tied to instances, the instance admin can kick the bigot and they’re gone for everyone, before many folk ever see it. And if there is an instance with admins that don’t deal with bigots, the admins can defederate from the entire instance.
Same with Nostr. Relay admins can ban users and relays who don't have good moderation. The difference is, if you don't agree with that ban, as a user you can: connect to other relays and route around it (so you can still follow/be followed by/DM your person/relay of choice) AND keep using that original relay because you like the content on it (if they banned another user, not if they banned you). And there's no need to make a whole new account at another instance, login to it separately, etc.
In Nostr, your relay gets to make and enforce it's own content policies just as easily as it does on AP. They are literally the same in this regard.
In Nostr, your relay gets to make and enforce it’s own content policies just as easily as it does on AP
And yet they don't.
If you as a user can connect to multiple relays, then most people will do that, making the act of an admin banning a bigot on one relay pointless, because the bigot will still get through on other relays, and once they're blocked on enough relays, they'll just make another throw away account. Which means most admins won't bother acting except in the most egregious cases, leaving it up to the users to deal with their own blocks.
I agree, there needs to be an answer here that lets people keep their identities tied to them rather than an instance, but the nostr approach isn't it. It just leads to everyone for themselves, which is fine for some people, but it's exactly what many other folk were trying to escape by coming to the fediverse in the first place.
If you as a user can connect to multiple relays, then most people will do that, making the act of an admin banning a bigot on one relay pointless, because the bigot will still get through on other relays
Just like AP, the bigot can join another instance/relay, if an instance has bad moderation policy, your relay can block that instance/relay entirely.
Which means most admins won’t bother acting except in the most egregious cases, leaving it up to the users to deal with their own blocks.
Admins can block along any spectrum of severity they want, just like AP.
once they’re blocked on enough relays, they’ll just make another throw away account.
They can do this in AP too. There's no way to solve this unless we start requiring a passport photo with every account or a payment or something. Blocking naughty instances/relays that have weak moderation is the best solution atm.
Nostr does not reduce an instance admin's ability to block bigots or the relays that host bigots. None of that is different than AP.
Just like AP, the bigot can join another instance/relay, if an instance has bad moderation policy, your relay can block that instance/relay entirely.
By your own admission, this doesn't happen though. One relay is the same as the other, and that's because the bigot can just use multiple relays as well, making the effort of an admin blocking them largely a waste of time.
They can do this in AP too.
They can't though, because on AP, an instance that constantly spawns bigots with throw away accounts gets defederated, and that means that bigots have a barrier that doesn't exist on Nostr. On nostr they can create unlimited accounts and use unlimited relays to broadcast their content, relying on the admins to block each and every account manually (which they generally won't do, because what's the point?), which ultimately leads to the end users having to play whack a mole with the bigots.
There’s no way to solve this unless we start requiring a passport photo with every account or a payment or something
I don't have any answers for you. But I can tell you that I actively prefer APs implementation to Nostr's because APs current implementation is safer.
I'm sorry Ada, but you're just wrong here. I don't know how you are missing my bolded sections explaining how nostr and AP work identically here.
By your own admission, this doesn’t happen though. One relay is the same as the other, and that’s because the bigot can just use multiple relays as well, making the effort of an admin blocking them largely a waste of time
Every relay sets their own moderation policy. They can block users, they can delete posts, they can filter based on keywords. They can block other relays and "de-federate" from them. Relays can share lists of common servers to de-federate from because they host bigots. Same as AP. Early in mastodon/lemmy/etc relays didn't differentiate themselves much on moderation policies. As problems came up on some servers, they started to do that more, now there is a big difference between different instances. Same can and will happen on nostr, there is already some differentiation, just not as much as it's a smaller platform.
They can’t though, because on AP, an instance that constantly spawns bigots with throw away accounts gets defederated, and that means that bigots have a barrier that doesn’t exist on Nostr.
Same thing happens in Nostr, that relay will get defederated. A bigot can still post to that relay, but their posts won't be propagated to other relays. All the bigots will filter into the few relays that allow them to post. Same as AP.
Very true, ActivityPub is for building communities.
Nostr is for free speech and user choice but also sucks for building communities because the loudest and most mean people are the ones who dominate the space.
That’s not true. Bigots can go into web view and still see posts, they have alts on other instances. I’ve literally had friends that this happened to on Mastodon and the perp was making posts about still being able to see their content even while blocked. You run AP, it is not a secure protocol it was never intended to be.
That’s not true. Bigots can go into web view and still see posts, they have alts on other instances.
Absolutely, but nothing we're talking about here will stop bigots targetting specific people. What we can do though is decrease the drive by bigots, the ones who just spew hate at whoever they happen to stumble across, or whoever pops up in a hashtag search etc.
You run AP, it is not a secure protocol it was never intended to be.
I'm aware. However the options aren't "free for all" or "impenetrable security". There is a whole spectrum of approaches between those two extremes, and that's what we're talking about here. Insisting that the issue is all or nothing doesn't help
By being Federated that means data is being sent to remote servers. Sometimes that data doesn't always make it, like a delete request. So someone on their own home-server deletes their post, but on some remote server where that post they made is cached, it's not deleted, because the delete request never federated. For example, say you made a post on your own box, which you clearly have, and you delete a post, but it doesn't get deleted over on say, Lemmy.world. That's not purposeful, that's something the developers also trying to fix, so I think it's disingenuous to say they don't care.
This is literally a consequence of how federation works. It's not a purposeful violation of GDPR.
You may not be directly using it, but this is part and parcel of the entire point of federated social media. Other software will be accessing the pool.
Depends on the app. Mlem is native, and Voyager is a web view app wrapper. But Voyager definitely doesn’t feel like a web view. It was written to feel like a native app, I’m really impressed. And that’s coming from someone who generally hates web apps.
Did they actually re-code the whole thing to make it native? You can ship a web app via the app store just fine. Only making changes to take advantage of the better system access that provides.
A web app doesn’t magically turn into a native app when you put it on the app store. The way it gets installed, and the system access is different, and better for stuff like notifications…
But the word “native” refers to applications made using the native SDK, which in Apple’s case is Swift. I’d be extremely impressed if the devs re-wrote the entire app in new code. But that would be unnecessary.
You can have non-native apps on the app store. A lot of your apps probably aren’t native. SDKs like flutter, used by liftoff and thunder, make development much easier in exchange for the resulting application running a bit less efficiently. Flutter applications can also run on both Android and iOS, and even desktop, with little additional work.
At least hardly anyone on Lemmy believes the Fediverse was invented by Eugen Rochko in 2022 as a reaction upon Elon Musk's announcement to buy Twitter.
@fossilesque This is really neat! I really like being able to be on a completely different instance and commenting/favoriting from other federated instances.
Honestly, the mastodon obsession seems like a you problem. Talk about what you want on the fediverse. Maybe mastodon users see it, maybe they don't, who cares.
Maybe you've overlooked that, but: I'm mainly on Hubzilla, not on Lemmy.
By far most of my connections on Hubzilla are on Mastodon. This means that my posts show up on a) Mastodon users' personal timelines and b) the federated timelines of lots of Mastodon instances. This, in turn, means my content has to fulfill at least some of Mastodon's cultural standards.
Also, I'm one of the few non-Mastodon users who do care for their reach on Mastodon. That's because I'm probably one of the few Fediverse users to explain to Mastodon users the Fediverse outside of Mastodon. This is not the primary topic of my Hubzilla channel, but someone has to do that.
But if even more Mastodon users or entire Mastodon instances mute, block or shadow-block me for repeatedly flipping the bird at Mastodon's etiquette and being unabashedly ableist, this becomes impossible because my explanations can't reach their target audience anymore.
Even when I post about my primary topic, 3-D virtual worlds, I rely on being read on Mastodon. For it is there where the chances are the best for there being someone who is interested in that topic.
Late, but still: I dare say that what Mike Macgirvin has done.
Mistpark/Friendika/Friendica looks like and is marketed as a Facebook alternative. But it comes with extra features on top like a built-in file storage, and its actual killer feature has always been that it federates with everything that moves.
Red a.k.a. the Red Matrix used to handle much like Friendica on the surface, but it introduced nomadic identity and permissions as early as 2012.
Hubzilla, into which the Red Matrix was turned in 2015, is probably the most powerful of all Fediverse projects. It was the first Fediverse project to implement ActivityPub, two months before Mastodon. And it was the first nomadic one to actually kind of take off.
Finally, the latest offspring of 14 years (plus two days) of development since Mistpark is the streams repository which isn't as feature-heavy as Hubzilla, but the most innovative one, and it's constantly evolving. It will be there first that nomadic identity and even permissions beyond what Hubzilla has to offer will be implemented in ActivityPub. And it's likely that this will happen this year.
Fediverse
Hot