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.
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.
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.