Wikipedia is literally just a very long number, if you want to oversimplify things into absurdity. Modern LLMs are literally running on neural networks, just like you. Just less of them and with far less structure. It is also on average more intelligent than you on far more subjects, and can deduce better reasoning than flimsy numerology - not because you are dumb, but because it is far more streamlined. Another thing entirely is that it is cognizant or even dependable while doing so.

Modern LLMs waste a lot more energy for a lot less simulated neurons. We had what you are describing decades ago. It is literally built on the works of our combined intelligence, so how could it also not be intelligent? Perhaps the problem is that you have a loaded definition of intelligence. And prompts literally work because of its deductive capabilities.

Errors also build up in dementia and Alzheimers. We have people who cannot remember what they did yesterday, we have people with severed hemispheres, split brains, who say one thing and do something else depending on which part of the brain its relying for the same inputs. The difference is our brains have evolved through millennia through millions and millions of lifeforms in a matter of life and death, LLMs have just been a thing for a couple of years as a matter of convenience and buzzword venture capital. They barely have more neurons than flies, but are also more limited in regards to the input they have to process. The people running it as a service have a bested interest not to have it think for itself, but in what interests them. Like it or not, the human brain is also an evolutionary prediction device.

Like I said, I can’t force you to see it.

In a scenario where Lemm.ee would have become a content instance, but kept their federation policy, they would still have received all the reports about posts on the communities they hosted, wherever the reported user comes from.

Being a dedicated content instance provider would also inherently imply dedicating that instance to a certain, more controlled type of content. An authentication instance might want to cater to a geography, which will probably decide to interact with the rest of the world and to provide adequate verification and certification mechanisms. A content instance might want to cater to a geography or a subject, resulting in specialized participation, with certification and verification based on the content, not the user.

You keep seeing monolithic instances that congregate the most communities as a plus. That’s a negative in my perspective on the fediverse. It shouldn’t be competing reddit clones with the one having the most communities winning out.

I think you are misunderstanding the problem being solved. Expecting all instances to become non-profits and manage even more responsibility exacerbates the problem and inhibits the fediverse growth. Non-profits also have their share of pitfalls and is an entirely different beast.

lemm.ee told you the reason they were shutting down - not enough people to keep the place running and burnout. I can’t force you to see how minimizing and distributing responsibility helps those issues if you don’t want to. Less responsibility, easier for people not to ditch projects or end them.

That has nothing to do about what they decided to do afterwards. I thank them for not transferring the instance domain to a completely different party without user consent, and people would have disagreed with that so it’s best everyone found their own solution. It would even have put their account information at risk.

Why would people want to implement something they don’t know the benefits of? That’s what my comment and increasing awareness is all about, in a thread about an outcome that could have been prevented by the idea.

Complete bans (at the home instance level) would require synchronization between the content provider instance and the authenticator instance.

What are you referring to as a ban? Complete bans already require synchronization between different federated instances. Sometimes the home instance of a user is unable to entirely delete the content of a user because of it.

Mod actions are caused by users comments on content, so the two aspects are closely intertwined, you can’t dissociate the content from the users.

Not really. Mod actions are over a community, not user history. They are perfectly able to remove user comments within their community, and since they are the authoritative source that controls whom it is spread to that has greater influence. That never stops the same content by the same user from appearing elsewhere.

At the moment, admins synchronize in a group to deal with toxic users, usually leading to the ban of those users on their home instance.

They would still do the same, but the “usually leading to the ban of those users” perhaps does more to reveal what your actual problem is than anything else. You and me will have to disagree, because admins should not be authoritarian figures, but should only have control within their domain.

• If they want to administrate over a group of users, they can have control over which users are and aren’t allowed over that particular group. They can issue their own warnings to users.

• If they want to administrate over communities, they can have control over which communities are allowed and how users are allowed to interact with those. They can remove users from those communities entirely.

The small but loud minority of toxic users can just have their authentication instances defederated if those instances refuse to do anything with them. If it is an authentication instance doing the defederation, then it will affect all of their users. If it is a content provider instance, it will affect all of their communities. In the current system, it does both because both are coupled into the same instance, so it’s even compatible with it.

It stops bad faith actors from trying to pollute communities to slur entire instances, like lemm.ee or blahaj, because of their problems with their userbase, by simply stopping it from being an issue. Administrators don’t have to worry about policing communities or users if they don’t want to, they would be able to better choose whom they are catering to without bad faith backlash elsewhere.

Almost nothing of the current structure changes, except that dedicated instances have the functionality they don’t need disabled. Both can still block each other to their heart’s content, and if your problem is having more “splits” - that is literally what federated instances are, there can always be more … Maybe your problem is with the fediverse and its distributed nature? You are making it out to be as if there is only ever a big bad group of toxic users and that all administrators always completely agree on all bans to make your argument work. At that point, just create your own reddit clone.

There’s several solutions, I was just stating the “at least” solution because everything needed for it is already present. You just need to remove functionality depending on the type of service you want to host.

What do you mean? The authenticator instance could ban users, the moderators and the content provider instances could ban users, content provider instances could defederate from authenticator instances and viceversa.

Not sure I’m seeing the issue you are seeing, it’s just basically forcing lemmy instances to instead of being both to just be one or the other. The benefit is that the actions on one is free from the drama on the other. One would be dedicated to hosting users, the other would be dedicated to hosting communities, less burnout overall.

At the very least, social networks like this really need a two server type system: the authenticator who identifies that you are really who you say you are and handles personal settings, communication, and access to the fediverse, and the content provider that hosts the communities.

It is intelligent and deductive, but it is not cognitive or even dependable.

Then, a few years from now they will continue to ask why they need to continue to import their scientists from other countries.

Have they changed the power connector? Have they changed as a company? Then zero cares.

The competition is catching up, and it is not a fire hazard for the rest of your expensive hardware or an based on insults to your intelligence.