Your browser is the most important program on your computer. More than your text editor, more than your terminal, more than any other tool you use throughout the day. Everything flows through it. Your email, your banking, your reading, your communication, your work, your entertainment. The browser is the operating system that matters, and the actual operating system underneath has been reduced to a bootloader for it. This makes the question of who controls the browser not a technical question but a political one. And right now, the answer is Google.

Google’s Chromium engine powers Chrome, Edge, Brave, Opera, Vivaldi, Arc, and every other browser that is not Firefox or Safari. Depending on which measurement you trust, Chromium-based browsers account for somewhere between seventy-five and ninety percent of all web traffic. Safari, restricted to Apple devices, accounts for most of the remainder. Firefox accounts for roughly two to three percent, and that number has been declining for over a decade. The browser engine market has collapsed to a monoculture, and the last surviving alternative outside of Apple’s walled garden is slowly suffocating.

Firefox was not always an underdog. It rose from the ashes of Netscape in 2004 and, within five years, commanded nearly a third of the browser market. Internet Explorer, the dominant browser of the early web, was bloated, insecure, and stagnant. Firefox was fast, extensible, and built by a non-profit foundation that openly declared its mission to keep the web open and accessible to everyone. Tabbed browsing, pop-up blocking, and a thriving extension ecosystem were not standard features. Firefox made them standard. For a brief window, it looked like the open web had won.

Then Chrome arrived. Google launched it in 2008 with an engineering budget that Mozilla could not match and a distribution channel that no one else on earth possessed. Every time you visited Google Search, you saw a prompt to download Chrome. Every time you opened Gmail or YouTube, the same prompt appeared. Google took the browser that was already installed on every Android phone and made it excellent. Chrome was faster than Firefox at JavaScript execution, sandboxed each tab into its own process so that one crashed page did not bring down the entire browser, and shipped with a minimalist interface that made the web feel larger. Within four years, Chrome overtook Firefox. Within eight, it overtook Internet Explorer. Today, it is the browser.

Mozilla, to its credit, did not give up. It rebuilt Firefox’s engine from scratch with Project Quantum, bringing multi-process architecture and a significant speed improvement. It launched Firefox Focus, a privacy-oriented mobile browser. It fought tracking with Enhanced Tracking Protection. These were genuine engineering achievements from an organisation with a fraction of Google’s resources. But none of them reversed the market share trend. The problem was never that Firefox was bad. The problem was that Chrome was everywhere.

And here we arrive at the uncomfortable core of the story. Mozilla’s revenue comes overwhelmingly from Google. Every year, Google pays Mozilla hundreds of millions of dollars to remain the default search engine in Firefox. This arrangement accounts for roughly eighty to ninety percent of Mozilla’s total income. Google does not do this out of charity. It does this because Firefox, even at two percent market share, is the only thing Google can point to when regulators ask whether the company has a browser monopoly. If Firefox disappeared, Google would face antitrust action across multiple jurisdictions with no remaining defence. Firefox is worth more to Google alive than dead, and so Google funds its continued existence.

The arrangement keeps the lights on at Mozilla, but it also traps Firefox in a cage. You cannot compete aggressively with the company that signs your paychecks. Every feature Firefox ships that genuinely threatens Google’s business model — aggressive ad blocking by default, for example, or deep integration with non-Google services — risks the revenue that keeps Mozilla afloat. Firefox must remain just competitive enough to serve as an antitrust fig leaf, but not so disruptive that Google questions the return on its investment. This is not a conspiracy theory. It is the structural logic of the relationship.

The money has also distorted Mozilla’s priorities. As Firefox’s market share dwindled, the organisation expanded into adjacent businesses. It acquired Pocket, a read-it-later service. It launched a VPN. It started an AI division. It experimented with advertising technology. The CEO’s compensation has risen substantially even as Firefox’s user base has shrunk. A growing share of Mozilla’s engineering talent is now working on things that have nothing to do with the browser. The foundation that exists to build an open web browser is increasingly not a browser company, and the money that Google provides for the default search deal is being spent on diversification into markets where Mozilla has no particular advantage and no obvious path to sustainability.

This is the context in which Firefox’s various community forks must be understood. LibreWolf, Waterfox, Pale Moon, Basilisk, and others all promise independence from Mozilla’s corporate compromises. They strip out telemetry, disable Pocket integration, tighten privacy defaults, and generally present a version of Firefox that is closer to what the original Mozilla mission statement described. These forks are genuine projects built by genuine people who care about the open web. Using them is better than using Chrome, and using them is better than using mainstream Firefox.

But they are not independent in the way that matters. Every single one of them depends on Mozilla’s Gecko engine, and maintaining a modern browser engine is not like maintaining a text editor. It is a continuous, grinding war against an ever-expanding set of web standards. Each new CSS feature, each JavaScript API, each WebAssembly proposal, each security patch for a speculative execution vulnerability must be implemented, tested, and shipped. The web platform specification is now larger than most operating systems. Google employs thousands of engineers working full-time on Chromium. Mozilla employs hundreds working on Gecko. No Firefox fork employs more than a handful of part-time maintainers. If Mozilla stopped maintaining Gecko tomorrow, every Firefox fork would be dead within a year — not because the forks are poorly run, but because no fork has anything close to the engineering capacity required to keep a browser engine current with the web.

This is the structural fragility that most discussions of browser choice miss. The question is not which browser you use. The question is which browser engines exist. Everything else is a skin. Brave is a skin on Chromium. Edge is a skin on Chromium. LibreWolf is a skin on Firefox, which is increasingly a skin on Google’s money. The number of organisations capable of maintaining a competitive browser engine is, for all practical purposes, two: Google and Apple. And Apple only does it for its own hardware.

What happens when that number drops to one? We already know, because it is already happening. When Google’s Chrome team decided that the original extension API was too permissive — that it gave ad blockers too much power — they designed Manifest V3 to restrict it. Every Chromium-based browser inherited this restriction by default because they all depend on Google’s engine. Brave and Opera could tweak the implementation slightly, but they could not fundamentally diverge without maintaining their own fork of Chromium, which would require Google-scale engineering resources. The web’s most popular ad blocker, uBlock Origin, no longer works in Chrome. It works in Firefox because Firefox has its own engine and its own extension API. But if Firefox disappears, so does the last cross-platform browser where users can block ads without restrictions.

This pattern extends far beyond ad blocking. Google decides how cookies work. Google decides how DRM is implemented in the browser. Google decides which video codecs the web supports. Google proposes a new standard, implements it in Chromium, and by the time the standards body has finished debating it, the feature is already deployed to billions of devices. The web is not a neutral platform with multiple independent implementations. It is whatever Chromium does, plus a Firefox that struggles to keep up, plus a Safari that Apple deliberately holds back to protect its App Store revenue.

The tragedy is that there is no obvious solution. You cannot build a browser engine with donations and privacy-respecting subscriptions. The engineering demands are too vast and they grow every year. Servo, the experimental engine that Mozilla started and later abandoned, showed genuine promise but proved that writing a new engine from scratch is a decade-long project even with a funded team. The Ladybird browser project is attempting the same thing with a small group of developers, and while their work is impressive, they are years away from something you could use as a daily driver. A browser engine is one of the most complex pieces of software humanity has ever built, and the number of people who understand how to build one is shockingly small.

The economics do not help. A browser is infrastructure. Like roads, power grids, and water systems, it benefits everyone but generates no direct profit for its builder unless you attach a business model to it. Google’s business model is advertising, so Chrome exists to funnel data into Google’s ad system and to ensure that the web remains a platform where Google’s ads can reach users. Apple’s business model is hardware sales, so Safari exists to make Apple devices more attractive and to ensure that web apps do not become good enough to threaten the App Store. Mozilla’s business model is Google’s money, which is not a business model at all but a temporary reprieve that can be revoked at any time.

A browser engine as public infrastructure — funded by governments, maintained by a non-profit, governed by the open standards process — is the idea that makes sense on paper. But the web moves too fast for government procurement cycles, and the political will to fund a global public good at the scale required has never materialised. The open web that we all depend on rests on a foundation that nobody is paying to maintain, except incidentally, as a side effect of one advertising company’s antitrust strategy.

I use Firefox. I have used it for years. I will continue to use it because the alternative is handing Google a complete and uncontested monopoly over the web. But I use it without the illusion that my choice alone can change the outcome. The forces that are killing Firefox are not the kind that consumer preference can reverse. They are structural, economic, and deeply embedded in the way the internet is funded. The open web is dying not because users do not want it, but because nobody has figured out how to pay for it without selling it.

When Firefox eventually stops being viable — and every trend points in that direction, however slowly — there will be no fork that saves us. There will only be Chromium, and Safari for those willing to pay the Apple tax, and the long quiet end of the idea that the web could be anything other than what Google decides it should be. That day is not here yet. But if you look at the market share graphs, the revenue reports, and the engineering capacity estimates, you can see it coming. The question is what, if anything, we are going to do before it arrives.