The content itself is harder to be deleted, because federation means that every post comment gets duplicated on all instances.
You do have a point regarding identity, and this is something that bluesky has solved already in a more elegant way. But this is also fixable with activitypub: as Takahe already showed it is possible to efficiently serve different domains with the same server. And on the extreme case, you can run your instance.
Did you see the scramble when feddit.de went offline for weeks and all its content became unavailable?
If there’s going to be duplicates anyway, why not do as I said (decentralize the hosting separately from the front end and make it available to all) and just really duplicate everything so there’s always a real backup and no one can wipe anything by shutting down their server?
Like I said, the content did not become unavailable. My instance still has the data from every community being followed.
The only unrecoverable problem with feddit.de is that the domain was lost. If the owner had given the domain to someone else, one could (theoretically) get all the identities back. They would need new keys, but the accounts would still be salvageable.
As for “separate frontend”: this is already possible and like I said it is a matter of improving the existing clients. We don’t need a fundamental change in the protocols to get what you want, we just need to get more resources available to developers so that they can continue working and improving on what we have.
But if a new instance is created after one was deleted, the new instance users will never have access to what was on that instance that got deleted.
We have “separate front ends” at the moment (guessing you’re referring to apps, otherwise people log in through their instance’s website), but the content the users have access to and the people they can interact with still depends on the instance they sign up on, I’m talking about eliminating that completely and letting the users be the ones that decide who and what they can interact with.
I’ll never be able to check what’s going on on beehaw or hexbear as long as my instance is the one I’m on, but no one should have the power to decide that for me or the other users I’m interacting with.
You are always free to run your own instance, and this is absolutely no different than “decentralizing” everything. The federation model where all users distrust each other degenerates into a fully p2p network.
Yeah, that’s exactly the point! How do you think that a decentralized system is any different?!
If everything is “decentralized”, you still must have a way to get rid of bad actors. Even nostr is set up in a way that you can not force your node into anyone else’s relay.
Forgive my bluntness, but the more you try to argue you point the more it seems you have no clue what you are talking about. There are plenty of things to criticize about Lemmy and ActivityPub in general, but you are missing the mark on all of them.
Getting rid of bad actors is the job of the users (from their feed) and the mods (from the communities they moderate), no one should have the authority the admins have.
If you follow that logic, people should never be able to block or ban you? That makes no sense. Of course anyone should be allowed to block anyone else for whatever reason they choose. That’s what defederation is as well. If you don’t have the option of blocking or banning, stuff degenerates really badly and really quickly.
No, by my logic only users should be able to decide to ban me entirely and only mods should be able to ban me from specific communities, admins shouldn’t exist at all, that’s real decentralization, Lemmy is an half-measure.
I’ll never be able to check what’s going on on beehaw or hexbear as long as my instance is the one I’m on, but no one should have the power to decide that for me or the other users I’m interacting with.
Well, that’s a choice Beehaw made. Shouldn’t they be allowed to defederate?
Quite a few people left Beehaw because of that, which is a sign that the decentralized model is working.
In your model, how do you deal with spammers, CSAM, trolls etc. ? Should every user do their own moderation for the 47k Lemmy monthly active users? Or should people create shared moderation lists? But then you still come back to the trust issues: do you trust someone else to add a user to a block list?
Allow NSFW content at your own risk, same for users and hosts.
Block users and communities as you see fit, why should a centralized authority decide for the users? It’s the same thing as Reddit except that there’s a bunch of centralized authorities instead of one.
I can create my own instance but other instances can decide to not federate with it.
If admins were the problem on Reddit we should work on making a platform where admins don’t exist at all, not one where there’s just more of them.
Allow NSFW content at your own risk, same for users and hosts.
I am not talking about NSFW, I’m talking about CSAM. There were a few CSAM attacks last year, some mods had to see some disturbing pictures of pedo pornography, that’s probably not something you want your average user to have to deal with.
It’s the same thing as Reddit except that there’s a bunch of centralized authorities instead of one.
Then it’s not the same. You have communities like [email protected] or [email protected] used to document abuse from admins and mods, and modlogs are public, it’s a drastic change from Reddit.
Have you ever had a look at Nostr? It only has moderation at the user level, so that might be what you are looking for.
Or, as I keep suggesting, you make the authority figures have as little power as possible, i.e. the only people with authority are mods so they only have control over communities and don’t have the power to prevent tens of thousands of people from communicating with each other.
That’s 5 admins out of how many users?
In the end Lemmy is centralized, just in a different way, someone can wipe out a huge part of the content in a single click.
The content itself is harder to be deleted, because federation means that every post comment gets duplicated on all instances.
You do have a point regarding identity, and this is something that bluesky has solved already in a more elegant way. But this is also fixable with activitypub: as Takahe already showed it is possible to efficiently serve different domains with the same server. And on the extreme case, you can run your instance.
Did you see the scramble when feddit.de went offline for weeks and all its content became unavailable?
If there’s going to be duplicates anyway, why not do as I said (decentralize the hosting separately from the front end and make it available to all) and just really duplicate everything so there’s always a real backup and no one can wipe anything by shutting down their server?
Like I said, the content did not become unavailable. My instance still has the data from every community being followed.
The only unrecoverable problem with feddit.de is that the domain was lost. If the owner had given the domain to someone else, one could (theoretically) get all the identities back. They would need new keys, but the accounts would still be salvageable.
As for “separate frontend”: this is already possible and like I said it is a matter of improving the existing clients. We don’t need a fundamental change in the protocols to get what you want, we just need to get more resources available to developers so that they can continue working and improving on what we have.
But if a new instance is created after one was deleted, the new instance users will never have access to what was on that instance that got deleted.
We have “separate front ends” at the moment (guessing you’re referring to apps, otherwise people log in through their instance’s website), but the content the users have access to and the people they can interact with still depends on the instance they sign up on, I’m talking about eliminating that completely and letting the users be the ones that decide who and what they can interact with.
I’ll never be able to check what’s going on on beehaw or hexbear as long as my instance is the one I’m on, but no one should have the power to decide that for me or the other users I’m interacting with.
You are always free to run your own instance, and this is absolutely no different than “decentralizing” everything. The federation model where all users distrust each other degenerates into a fully p2p network.
And then admins from other instances can decide they don’t want to federate with my instance, see how it doesn’t solve anything?
Yeah, that’s exactly the point! How do you think that a decentralized system is any different?!
If everything is “decentralized”, you still must have a way to get rid of bad actors. Even nostr is set up in a way that you can not force your node into anyone else’s relay.
Forgive my bluntness, but the more you try to argue you point the more it seems you have no clue what you are talking about. There are plenty of things to criticize about Lemmy and ActivityPub in general, but you are missing the mark on all of them.
Getting rid of bad actors is the job of the users (from their feed) and the mods (from the communities they moderate), no one should have the authority the admins have.
If you follow that logic, people should never be able to block or ban you? That makes no sense. Of course anyone should be allowed to block anyone else for whatever reason they choose. That’s what defederation is as well. If you don’t have the option of blocking or banning, stuff degenerates really badly and really quickly.
No, by my logic only users should be able to decide to ban me entirely and only mods should be able to ban me from specific communities, admins shouldn’t exist at all, that’s real decentralization, Lemmy is an half-measure.
Well, that’s a choice Beehaw made. Shouldn’t they be allowed to defederate?
Quite a few people left Beehaw because of that, which is a sign that the decentralized model is working.
In your model, how do you deal with spammers, CSAM, trolls etc. ? Should every user do their own moderation for the 47k Lemmy monthly active users? Or should people create shared moderation lists? But then you still come back to the trust issues: do you trust someone else to add a user to a block list?
Allow NSFW content at your own risk, same for users and hosts.
Block users and communities as you see fit, why should a centralized authority decide for the users? It’s the same thing as Reddit except that there’s a bunch of centralized authorities instead of one.
I can create my own instance but other instances can decide to not federate with it.
If admins were the problem on Reddit we should work on making a platform where admins don’t exist at all, not one where there’s just more of them.
I am not talking about NSFW, I’m talking about CSAM. There were a few CSAM attacks last year, some mods had to see some disturbing pictures of pedo pornography, that’s probably not something you want your average user to have to deal with.
Then it’s not the same. You have communities like [email protected] or [email protected] used to document abuse from admins and mods, and modlogs are public, it’s a drastic change from Reddit.
Have you ever had a look at Nostr? It only has moderation at the user level, so that might be what you are looking for.
And those CSAM attacks weren’t prevented by the way it works at the moment so that point is moot…
I mean sorry but that’s just what decentralization is, unless you want a fully peer-to-peer protocol which is not realistic at all.
Or, as I keep suggesting, you make the authority figures have as little power as possible, i.e. the only people with authority are mods so they only have control over communities and don’t have the power to prevent tens of thousands of people from communicating with each other.
All the feddit.de content is still available on https://lemmy.world/c/[email protected] and all the other instances which where federated.