Very interesting comment. It took me time to unwrap it all, there are lots of great insights.
About "incentives", you might wanna check my other reply to IAmEveryone in this thread[1].
Basically the idea is that we probably should try to win not against but with other platforms, in a way that creates synergies for everyone.
That's how open-source thrives the most, that's how the best contracts are negociated, that's how capitalism works exponentially the best if we seek sustainability. IMHO. And it's damn more clever and efficient —if you can pull it off!; which is no small feat, but definitely possible. You probably need a master plan, a grand strategy within which fit several concurrent paths in subsequent steps, each with a strategy. This to address multiple dimensions / domains of the problem (think: UX/selling it to users; economics/sustainability, technical realization and maintenance, politics/PR/legal framework/compliance; competition/cooperation/synergies/open-source; etc.)
I've long thought about it and it's truly at that level, like a board is essentially the "meta" of a corporation, or like the IETF is the "meta" of internet, that this battle is fought. And it's not so much a battle against other actors —which, if you think about it, rose to prominence of their own virtue, be it National Post offices or Disney or internet or Google or Twitter, Insta, Black Mirror-esque dystopian companies and projects. It's more of a battle for communication, for social sanity, for whatever it is you seek. I.e., from this reality (the one we're in), how to best move the needle towards that ideal (the one you're seeking, if possible in your lifetime thus actionable). So we take what happened, and wasn't 'forced' on anyone, and we try to implement the 'why' it should move somewhere next. We open the gates wide towards this 'next'.
____
Crypto, definitely part of it, but I mean the blockchain, the technology. I truly think that its "killer-app" will firmly stand in a source of value (like most technologies emerge historically); my intuition is that a totally "free" social network is one such killer-app, and thus not implementing any shape or form of "currency". That is a blockchain worth having, for humanity, for civilization, like internet, browsers, encryption, etc.
The blockchain paradigm is the perfect candidate for a secured repository of personal data, with various rules (contracts, algorithms, etc) for read/writability by other personal blockchains, other users. When you aggregate the whole, you obtain a massive collection of billions of personal blockchains; and that is the "network" of blockchains, the neutral communication space. Upon which we build "pipes" between us as we see fit — maybe I'll let you see my "friends" posts, but not my "private" documents (which my wife however can read because we share some stuff), and I also have a "work" collection of visibility to allow colleagues' blockchains to see such content; up to a "public" level which is essentially twitter / insta etc. The one-to-one peer-to-peer unitary link is basically chat / email.
____
About distributed, and general implementation.
When you go back to first principles, a blockchain's content is plain text, like an http request/response, or a chat message, an email, a twit... It's just the simplest CRUD implementation — yes, including delete, if you consider that each personal blockchain is controlled by their user, who has the keys and control of their personal/local network, and thus can edit their chain. It's the whole point of control at a granular level, I can delete my own files if I so wish, I am root.
A connection to another blockchain is merely a right that their public key is OK to do something to some blocks; for instance we allow "friends" (a mere array of public keys) to "read" such and such posts. you may set expiry, you may allow deeper than n=1 ("friends of friends of friends..."; public is just about n=3.6 on average iirc).
Corporations, any other party, fit in this landscape as any other actor: "hello, can I read this data?", with the 'server' (my blockchain) having authority by my rules to allow/deny. In terms of UX, it needs not be much more complicated than permissions on a phone, ideally with a few nice smart rules to ease the cognitive effort.
The beauty of such an implementation, imho, is that any service, public or private, whatever its intent or set of rules, can emerge freely, just like internet websites. Each actor, each block of content is merely a proposition to other parties. If accepted, it can grow massively, virally, without any gatekeeping nor impediment towards the free flow — it's peer-to-peer at the lowest level. Like bittorrent, conceptually.
And it needs not be much more complicated for the user than a "my own chat / twitter / blog / email / facebook / insta app" (a simple http server, under the hood, which implements your preferences of visibility, presentation, etc.)
Such services could be default open-source modules, mere options for the user to install locally and use among many such free or paid software; or for-profit service/cloud actors who offer some value in the exchange of giving them 'read' rights on some of your data. Make it an auto-renew daily contract (or whatever term) and you've got revokability.
There is much more to say, but that's what I'd suggest experimenting with.
____
Edit: a word on scaling
Approaching human communication in a holistic "top-down" way tends to promote a fallacy in perception: it's not because there are 3.456 gazillion emails or chat messages or twits each day that each and everyone one of us needs or reads that many.
At an organic level, you only need to send whatever content you produce, and receive whatever content you read. It's between 0 and 100 emails, however much you chat, it's not that much for your "node" (collection of computing power in your devices) to serve as http packets.
Those who serve massively more content (influencers, corporations, marketers, etc) generally have the material and financial means to upgrade their infrastructure to meet the needs — rent cloud capacity. The fundamentally public nature of viral content makes it perfectly suited to remain in the cloud-caching space (the CloudFlares and AWS of this world), it's also often for-profit at the end of the day, whether as dollars in sales or indirectly dollars as reputation.
So I think such a peer-to-peer "master" neutral network basically self-hosted to a large degree can work in terms of resources, the economics of idling CPU on consumer devices leave much, much room to grow before we even have to consider selling $100 "power packs" to plug into your residential modem for extra uupmh.
That's my vision, and it's basically what Satoshi did but implemented not for currency but rather for communication.
In that world, hyperscalers becomes either resource providers for content distribution (AWS, ideally 'transparent' ISPs, etc) or 'organizers of information'.
Insta doesn't host the content anymore, it merely reads it from "adequately scaled servers" like my blockchain for my friends or Dwayne Johnson's AWS instance for his fans; so Instagram as a company becomes this pure information dealer, providing customer insight to marketers based on what data it has been allowed to obtain from users.
It hopefully creates a situation where they're incentivized to "play it fair" and offer some real value to users in exchange of their information. And I think that's one aspect of this we mostly like, having relevant recommendations, valuable suggestions, very refined services for all our stuff — because we have a lot of stuff and it's great when a cool company does a great job, we very much want to pay for that and we're willing to share with them, to help them help us.
[1]: https://news.ycombinator.com/item?id=21330417