I'm not sure what you are describing is "the internet". You seem to be describing a particular way society can structure its use of part of the internet. To me, it sounds like saying we're going to re-invent North America. No we're not! The continents are going to stay right where they are (give or take a few cm per decade). We might choose to use them differently.

The thing about the internet--save for impingement on content neutrality--is that nothing about the fundamental architecture prevents any of your proposals. Yes, some require some extra metadata, but hey, if you can do HTTPS, you can do that. You don't even need to "fix" the internet! You just need to use it. It's fine! It's distributed, it's content-unaware, it's widely accessible, and if someone comes up with funding somewhere, you can get more people on it--it's totally a solved problem.

The only problem arises when, to comprehensively "build" an intersectional feminist internet, you want to "dismantle the infrastructure that facilitates the surveillance industrial complex". That's not building. That's very explicitly dismantling. It's quite true that the internet, because of its distributed and content-neutral nature, can indeed be used for some degree of surveillance.. (It can also be used to thwart surveillance: encryption, Tor, etc..)

So I think your title is a bit disingenuous. To the extent that there's anything wrong with the internet itself (as opposed to how we use it), you want to destroy, not build. This doesn't necessarily mean the goals are suspect, but the way you're packaging them is misleading.

However, the good news is that, again, because of the distributed content-neutral nature of the internet, nothing is stopping you from building anything you want and can. The internet already runs almost entirely on GNU/Linux so yeah, the open source thing is there. If you want to have community-owned infrastructure, nobody's stopping you. Find a community, buy some infrastructure, connect to a Tier 1 ISP or whatever, and you're good.

Create sites and services with community-centered design? Sure. Just do it! Nothing's actively stopping you. Of course, there's the whole "things don't happen for free" issue, but at any scale where you can translate interest into either labor or funding (which you then convert to labor) you can create content and services with these principles in mind. If they're in demand, you will either gradually take over, or other sites will adopt the same principles in order to compete. If they're not, well, maybe your conceptualization didn't actually match what people wanted in practice.

But I'm quite skeptical about your anti-capitalist rhetoric. It sounds to me like "I want to build something that sounds amazing but actually is worse for people on average, and force them to use it". If you want anything else, you can just embrace capitalism and win. There are some things that capitalism makes very difficult. Building something better is where it excels. On the internet, as it stands, there are no barriers to participation in something better. Just build it, and if it's actually better enough in ways that matter to people, it will win.

That's why Google exists. That's why Facebook exists. That's why Twitter exists. They solved a problem that mattered to people substantially better than anyone else--that's the core of why they exist. Yes, there are some other issues with scale and vendor lock-in and stuff now. But more than pretty much anywhere else, if you have a genuinely good idea (not a "eat your vegetables, and nobody else can give you anything else so you'll eat them or starve" idea), you can just go make it happen.