There’s an irony in Tim Berners-Lee’s position that I can’t stop thinking about. He invented the World Wide Web as an open, decentralized system for sharing information. Thirty-two years later, he’s building tools to reclaim it from the companies that centralized it.

The web was supposed to be ours. Berners-Lee designed it that way — open protocols, no gatekeepers, anyone can publish, anyone can read. In the early years, that’s roughly how it worked. Personal websites, independent forums, distributed communities linked together by hypertext.

Then scale happened. And convenience. And advertising. Now five companies control what most people think of as “the internet.”

Berners-Lee isn’t taking that quietly.

The Contract for the Web #

In November 2019, Berners-Lee and the Web Foundation launched the Contract for the Web. Nine principles organized around three groups: governments, companies, and citizens.

Governments should ensure everyone can connect, keep the internet available, and respect privacy. Companies should make it affordable, respect user data, and develop tech that supports the public good. Citizens should create, collaborate, build communities, and — this one’s interesting — “fight for the web.”

Over 160 organizations signed on. Google, Microsoft, Facebook, Twitter. The cynic in me can’t help but notice: several signatories are the same companies Berners-Lee argues captured the web in the first place. But the Contract is more aspirational than enforceable. It’s a statement of values, not a legal framework.

The principles themselves aren’t controversial. Nobody argues against internet access or privacy in public. The question is whether principles matter when they conflict with ad revenue.

Solid: Data Pods and Decentralized Identity #

The Contract for the Web is the philosophical layer. Solid is the technical one.

Solid (Social Linked Data) started at MIT in 2016. The concept: everyone gets a “Pod” — a personal data store they own and control. Apps request access to your Pod to read or write data. You decide which apps get what. Your data stays in your Pod; apps can’t hoard it in their own databases.

It’s an inversion of the current model. Today, you create an account on Twitter, and Twitter owns your posts. You sign up for Spotify, and Spotify owns your listening history. You use Google Maps, and Google owns your location data.

With Solid Pods, your posts, your listening history, your location data — all of it lives in your Pod. Twitter, Spotify, and Maps read from it with your permission. They can’t retain it without your consent.

Berners-Lee founded Inrupt in 2017 to commercialize Solid. The company builds enterprise infrastructure: Pod hosting, identity management, developer tools. The BBC built a prototype Solid-based recommendation app this year — one of the more visible real-world experiments.

The technical architecture uses linked data and web standards: WebID for identity, WAC for permissions, LDP for resource management. RDF, URIs, HTTP — the semantic web vision, repackaged for personal data sovereignty.

The Appeal #

I find the Solid vision compelling for one specific reason: it separates data from applications.

Right now, if I want to switch photo services, I can export my photos (maybe) but I lose the metadata, the organization, the sharing permissions, the comments. Data portability exists technically but fails practically because each app stores data in its own format, in its own silo.

In a Solid world, my photos live in my Pod. Photo App A and Photo App B both access the same data store. Switching apps becomes trivial because the data doesn’t move — only the interface changes.

The same logic applies to health records, financial data, social connections, professional history. Any domain where your data is trapped in platform silos could benefit from a Pod-based architecture.

For someone who works at TaskRabbit — a platform connecting people for real-world tasks — this raises interesting questions. Our platform has profiles, reviews, booking history, payment records. In a Solid model, a Tasker’s reviews and work history would live in their Pod, portable across any marketplace. Great for the Tasker. Less great for the platform’s competitive moat.

The Skepticism #

And here’s where I get uncomfortable, because the gap between Solid’s vision and its adoption is enormous.

Network effects are the fundamental problem. Facebook isn’t valuable because it has nice software. It’s valuable because 2.8 billion people are on it. Moving personal data to a Pod doesn’t solve the coordination problem — everyone you communicate with also needs Pod-compatible apps. That’s a bootstrapping challenge of staggering proportions.

Developer adoption is the second problem. Building on Solid means learning a new data model (linked data, RDF), a new identity system (WebID), and a new permission model (WAC). The tooling is improving but it’s nowhere near the convenience of Firebase or Rails with Devise. For Solid to win, building on it needs to be as easy as building on centralized platforms.

Performance and reliability are the third concern. Centralized platforms invest billions in infrastructure for low-latency, high-availability experiences. A decentralized model with data in millions of individual Pods introduces latency, availability, and consistency challenges that centralized architectures handle by definition. Who keeps your Pod online? What happens when it goes down?

Then there’s the incentive problem. The companies that benefit most from centralized data — Google, Facebook, Amazon — have zero motivation to support a model that liberates user data. Signing the Contract for the Web is easy. Rebuilding your product to work with Solid Pods would cannibalize your core business model.

Where I Land #

I want Solid to succeed. The principle — that people should own their data and grant revocable access to applications — is ethically correct and architecturally sound. Having the web’s inventor lead this gives it credibility no startup could manufacture.

But wanting something to succeed and believing it will succeed are different things. The centralized web isn’t centralized because of bad technology. It’s centralized because centralization is convenient for users, profitable for companies, and reinforced by network effects that grow stronger over time.

Berners-Lee is fighting economic gravity. The Contract for the Web articulates what the web should be. Solid provides a technical foundation for getting there. But the missing piece — the one that neither principles nor protocols can supply — is a compelling reason for billions of users to change their behavior.

The BBC prototype is encouraging. Enterprise adoption through Inrupt could create institutional momentum. Government regulation (GDPR already pushes in this direction) might force the issue. But I’d be dishonest if I said I was optimistic about the timeline.

The web’s inventor thinks the web went wrong. He’s probably right. Whether his solution can compete with the entrenched power of the platforms that captured it — that’s a question I don’t have an answer to. And I suspect Berners-Lee, for all his conviction, doesn’t either.