Independent Internet (Web 4.0)

The web most of us use every day answers to someone else. Your data is the product. Your attention is for sale. And when a platform changes its mind, the access you depend on can vanish overnight. It was not supposed to be this way.

The Independent Internet — what we call Web 4.0 — is the web as it was meant to be: one that people own, run, and control themselves. Resilient enough to keep working when the grid goes down. Private by design, not by policy. Built for everyone, and answerable to the communities who use it.

Built for people

Web 4.0 is the Internet as it was intended to be — made possible by hardware that is finally cheap and powerful enough, and software smart enough, to put a whole network in ordinary hands. No corporate landlord. No surveillance. An internet of, by, and for the people.

Every era of the web so far was built for someone else — universities, then data brokers, then AI. This one is built for you.

What’s at stake

For thirty years the web has drifted from its founding promise. A handful of platforms now sit between people and each other, and they pay for themselves by watching. Personal life became raw material; the feed became a market for attention. Whole communities — rural, remote, low-income, disaster-struck — get the thinnest, most fragile connection, or none at all. And the devices we buy are built to expire, not to last.

None of this is inevitable. Every piece of it is a choice about who holds control — and control can be handed back. That is the work.

How it works — key technologies

• Self-hosting. Be your own service provider, and keep complete control over your data and infrastructure.
• Mesh networking. Peer-to-peer connections (LoRaWAN / Meshtastic) form resilient networks with no central point of control or failure.
• Post-quantum security. Future-proof cryptography that survives quantum attacks, inspired by the US DoD’s NIPRNet / SIPRNet.
• Satellite & community connectivity. Low-earth-orbit satellite and self-healing community mesh reach places traditional infrastructure never did — keeping the unconnected connected.
• Energy-efficient & offline-first. Low-power systems that run on renewable energy and cache content locally, so the network keeps working on limited power — or no connection at all.

The web, era by era

Control of the compute and data underneath is the axis every prior era left open — and each of the eras before it still hands something forward:

• Web 1.0 — the static web (built for universities). Stable, verifiable records with a small attack surface.
• Web 2.0 — the dynamic web (built for data brokers). Participation and user-generated content — but returned to operators, without the intermediaries or the surveillance.
• Web 3.0 — the semantic web (built for AI). Blockchain and machine-readable data, pointed toward decentralized, autonomous infrastructure instead of extraction.

The protocol surface

Control is not a slogan; it needs concrete artifacts. The agent-readable web standards — from robots.txt to llms.txt — are the first of them: operator-authored, advisory files by which a site governs its own machine readers. Interpretable Context Methodology carries the same file-native, plain-text discipline into the workspace.

The ideas behind it

None of this is new — the argument is older than computing: legitimate power derives from the consent of the governed, and is deliberately divided so no one holds it all. In computing, it is the continuation of a sixty-year arc of people who believed computing should serve human beings — from Licklider and Engelbart, through Berners-Lee and the Semantic Web, to Nakamoto and decentralization — the long answer to surveillance capitalism. Web 4.0 is where that arc points next.

Explore the research

This page is a synthesis of open research anyone can read and check. Trace it back through the source library, follow how the ideas connect across the wiki, or read our story to see why we are building it.