Honestly, I thought about the very same problems a lot. And not only from a user's privacy perspective but also from a browser's perspective and from the perspe

Honestly, I thought about the very same problems a lot. And not only from a user's privacy perspective but also from a browser's perspective and from the perspective of a "god's eye's view" that big tech companies have. For the sake of argument, Google's or Cloudflare's DNS alone probably have so much data that any country or ISP pales in comparison. The web is broken and it gets more and more rotten over time. Google found out they don't need cookies to track people, so they invented FLoC and decided to actively don't give a damn about any privacy regulation on the planet - activating the tracking mechanism for everyone without any real influence or decision on the user's side. I think the web3 idea will never exist when we keep trying to reinvent the wheel of network protocols. New protocols and new ports can easily be blocked. At some point they're just portrayed like TOR, as a tool for evil rather than good; because the politics will get involved when power shifts to quickly for them to compensate. DNS over TLS just gets its 853 port blocked by most ISPs already so it's useless as a tool to shift the means of power here, and only DNS over HTTPs actually makes a real difference because it's way harder to block HTTPS and tell customer's that their internet still works. Given the nature how the web works, how it's transported and how dynamically it's changing in its resources over time - I think it's more important to focus on the idea of statistical truth in the moment, when the specific end user was using it. Some moments afterwards the same URLs change anyways, so it's important to preserve the resource's timeline - but not back until the dark age. Most parts of the world don't have the bandwidth to download gigabytes and still have crappy 2G slow and even lower network speeds. So a blockchain for a single website isn't a good idea to begin with. As a user I also don't know how hashes work, but I know how URLs work. URLs are predictable, hashes are not. The real strength of a peer to peer system is not only ledging, it's also offloading of network traffic. Every cached download that's offloaded via a closeby peer's cache is another tracking prevented. Every request that looks and behaves like Chrome's rendering engine on Windows/MacOS is another user saved from network fingerprint identification. Privacy is not keeping things to yourself. Real Privacy is looking like everybody else. For my own Browser I've decided to push the boundaries a little by implementing those features with the primary goal of being able to make the web work when you're offline or just connected to trusted peers (that have the URLs already cached). [1] And I'm honest here, it still needs a lot of work. Not only is JS malicious, but HTML/CSS has to be assumed to be malicious as well due to all specifications making it possible to identify people just by using CSS and background images or by simply making the website unusable with tricks like overflow:hidden or opacity:0, so these malicious parts have to be filtered out as well. Also: You don't need custom network protocols there. You don't need to roll your own crypto. HTTP, WebSockets and TLS are already peer to peer. The only thing missing is a decentralized way of cross-signing certificates and the ideas behind TLS notary could make the real difference in peer to peer networks. [2] [1] https://github.com/tholian-network/stealth https://github.com/tholian-network/stealth (highly alpha) [2] https://tlsnotary.org https://tlsnotary.org