The model that isn't sustainable is a permanently investor funded one.
Often when these products launch - they have zero advertising, zero features to make money - they are totally aimed at growing market share - totally focused on the user.
Then it's inevitably downhill from there - whether it's people not being prepared to pay ( eg Blue ticks ) or complaining about advertising or scrapping of data.
One of the damaging effects of the large amount of investor funded products is that it makes it very difficult for the user supported ( paid ) ones to thrive.
Perhaps the current state is because we are not at equilibrium - and as regulation, and the market matures it will become a bit more sane.
The government can. These companies could be nationalized, or the government could make their own competing sites if they were interested in doing so. But they're just not.
It takes more than just the means. It takes a will to do it.
https://dynamorando.pages.dev/blog/the-public-web/
Though I am not sure the political will exists.
But I do quite like your proposal for how it could be accomplished technically through federation. It at least sidesteps the problem of the government building it through their fundamentally broken technology "procurement" processes.
In essence that's a user ( indirectly through taxes ) supported model.
In the UK, the BBC follows that model ( the license fee is in effect a tax in all but name ) - and it's model is constantly under attack from certain quarters.
Indeed the BBC has being doing pilots in the social space - with SOLID and data pods.
See https://www.bbc.co.uk/taster/pilots/together-pod
The key question here is any of this stuff actually important enough for centralized funded.
Let's not forget, there is also the spectre of government control - you know the government security services will have full access.
I found myself struggling with Linked Data for quite some time now, and I have struggled with the idea that perhaps part of the problem with adoption of Linked Data is existing mental interia of RMDBS or other systems.
I like the idea - kinda back to a decentralised web of peers like it was in the early days.
However the complexity of the technology is certainty off-putting - I'm not enough of an expert to tell how much of that complexity is adherence to a tech stack ( like RDF, SPARQL ) and how much is simply the complexity of the underlying problem.
I would say the guys behind it seem fairly pragmatic. As an example while the technology allows you to self host, they acknowledge that most people won't be able/want to - and are looking to enable providers as well.
I think one of the problems with the LD/RDF community is it can attract the type of person that things 'we just need a single well defined data model for the universe'.
I think trying to get one schema to rule them all is doomed to failure - for two reasons
- for ontologies to be effective all the users of that ontology have to have a shared understanding of the ontology - simple a written down definition isn't enough.
- the world can be viewed from multiple angles - even if you could agree one view, it's not going to be optimal for all use cases.
However as I said, the SOLID project doesn't appear to be falling into that trap - it appears very focused and pragmatic.
I’ve been trying to work on an alternative solution to SOLID, but I’m always double checking myself just to make sure that I’m not just simply failing to grasp the concepts.
I had half an idea yesterday on this from HN:
Or maybe to protect against that, there's a strict moderation governance document for that to the effect of "If it's not illegal, it flies". That could still get rough since there's lots of non-illegal content you could put on there which people wouldn't want to see.
Independently hosted nodes still feel like the way to go to me. No hard authority, just people talking with people and moderating to set the tone they want for their community. If the moderators are jerks, moving to a different node should be easy (thus putting the power into the hands of the users to abide by the governance they find most agreeable) and not cause you to exile yourself from your civic community.
You know how community meetings and town halls are awful and fruitless because they're filled to the brim with the noisiest cranks and they can't be kicked out because they're still members of the public after all, and nobody else participates because it's maddening to be around all those noisy cranks? Like that, but web scale.
I'm sure the noisy cranks would love this, but I'm not interested in it.
As a follow question: is this a solvable problem between host moderation and user self moderation?
In other words: Let’s pretend NPR hosts a Reddit-like site whose primary objective is to facilitate discussion on topics shared by NPR.
NPR doesn’t outright ban everything unless it violates some terrible things.
Could user moderation NPR Reddit also expand on this? So long as they fall under the same guidelines?
I ask this question because it seems to me that there does exist some useful moderation: there are well moderated Reddits and for the most part Wikipedia is also pretty well moderated.
Wikipedia is self moderated and there are well moderated Reddits. The host of the ActivityPub site doesn’t have to do all the moderation, and there doesn’t seem to be a reason that users themselves couldn’t “mute” troublesome posters from their own feeds, right?
The person who set up the server doesn't, no, they can appoint moderators that lack administrative privileges. I don't recall at the moment if moderators have the ability to federate/defederate with peers, and that might be an implementation detail anyway. This is probably the way a large well-run node should work - developing a team of moderators from within the community.
> there doesn’t seem to be a reason that users themselves couldn’t “mute” troublesome posters from their own feeds, right?
This is true as far as it goes, but if a node federates with peers that dump a high volume of content onto the network the user doesn't want to see, then the user will find it a headache to manually filter just what they want to see (approximately the feeling of manually filtering out email spam). So some filtering should be done by the host.
I think the killer feature of federation in general (particularly in the context of social media) is that it decouples user account administration (and by extension, data collection) from the network effect. I run my own node that just has my account on it, but I can still benefit from the network by federating with other nodes.
The closest to an alternative model that works here is wikipedia. They achieve it by having a different default for user generated content - that it doesn't remain unless it is strictly on topic (you can't just write your random thoughts and shitposts in an article) - and by having a social mission that attracts donations. (They also run incredibly lean, which other mass market sites could definitely learn from.)
Another alternative model being pursued is the federated approach. I think this approach in principle but I don't think it will ever reach the mass market hundreds of millions of active users position of the mainstream social media products. Its solution to the problem is that some subset of power users will self-host and absorb the costs, either via individual altruism or something, or by developing some other model. I think this will end up centralizing, with a small number of nodes hosting most of the usage, and probably eventually falling back to advertising to cover costs. But I dunno, we'll see, maybe this will work.
Another more successful model is to just not be mass market at all, like HN and other various message boards. This reduces costs both for hosting and moderation and can then be justified by some non-financial benefit (like tacit advertisement for YC in the case of HN, or by tight community camaraderie for niche message boards).
Then another model that works is subscriptions. This drops the active user count massively and generally makes it harder to get traction, but I think it works the best of any of these when the stars align. I'd rather be Netflix or the NYT than Twitter or Reddit.