The issue is the cost of the service. You have to pay for those servers somehow and for updating the site to the latest versions of iOS or Windows or whatever.
There have been MANY attempts at implementing a 'paid FB' or just a straight attempt to tackle FB outright (Ello comes to mind). None have thus far worked. Not saying you can't try, but talk to the other creators out there first to learn from their mistakes.
Again, the issue is that servers cost money and you have to find the funding somehow. Ads are good at that. Subscription services aren't thus far, but Patreon may be changing things.
Still, with NN out the window now, I'd chill and wait to see how the dust settles.
Seriously though, I know monetisation is tricky, but there must be a better alternative to ads. I'm thinking something which can perhaps enrich the quality of a user's experience for a fee, rather than earning via ads while annoying the user.
But I think you really need to re-think the core concept of a social platform. Can't find the comment on HN, but someone a few days ago mentioned: social platform vs. social technology. A social platform is likely to be closed; Facebook only wants you to communicate via Facebook, same with Twitter, G+, etc. If you decide to email your friends instead of using FB, they lose out. A social technology, however, doesn't care about other providers as much. Think about a telephone company. They obviously want you to use their service, but they don't care if you call someone who has a different carrier. Likewise, they won't listen in to the content of your calls to tailor their service to you, because them providing you phone data and the contents of your calls are completely orthogonal.
Try thinking along the lines of a social technology, rather than a social platform. E.g. my friends want to put together a movie/pizza night so they go to Facebook to create a poll to decide on a movie and pizza toppings. Why can't that be part of an SMS standard?
It seems to me like SMS has a lot of potential for future iterations. Right now you can send texts and images to one or more people. Why not bake in other features like polls, appointments, etc. It truly has frustrated me that my friends will use Facebook to organize something that could easily be done over SMS if it just had a few extra features.
I have some other ideas that I was going to formally write up and thinking of submitting it to Mozilla. Feel free to email me (email in my profile) to discuss more if interested.
IMHO online social networks have a 5-7 year half life, after that they stagnate and die because the company tries to chase ad dollars (the easiest dollars to get) and step by step degrades their usefulness.
For some reason we don't model business on natural systems (birth -> growth -> reproduction -> death). Seems to me that'd be a good idea if we want to be around in a million years.
Harvesting value from people's lives seems to be working really well for them at the moment.
In the real world we traditionally rely on other factors to keep out identities separate. A dirty joke told to some friends doesn't get archived and indexed and re-shared with metadata back to your boss. It gets forgotten or attributed to "a guy I know." So it's relatively safe to give your real name in both contexts, and maintain the separation between studious-worker in one context and bawdy-clown in another.
But online, maintaining that "freedom to be a different kind of person" isn't reliably safe when both things come from the same user-ID.
Thst would be pointless as people do not pay to be on Facebook. And people aren't paid to be on Facebook. People are the product being sold.
Proof: this last move tries to improve the product (people's time) for its actual customers (advertisers).