Gabriel's Blog

Chromium Foundation

We need to divorce Chromium from Google

The world of browsers is pretty bleak. We essentially have one viable player, Chromium, flanked by the smaller players of Firefox, spiraling in a slow painful self destruction, and Safari, a browser well optimized for Apple products.

note: We're specifically discussing browsers, but you can roughly equate the arguments I make later with the browser's respective engines.

The current state of browsers is a difficult one. Chromium has the backing of notable megacorps and is under the stewardship of Google, which come with a number of perks (great featureset, performance, talented minds working on a singular project, etc) and a number of downsides as well (Google has a vested interested in advertisements, for example). Firefox is off in the corner with a declining marketshare as it alienates its users in a flailing attempt to gain users from the Chromium market, including dumping XUL addons in favour of WebExtensions, some rather uneccesary UI refreshes (subjective, I suppose), and various other unsavoury moves that leave a bad taste in everyone's mouth. And Safari is in the other corner, resisting the web-first movement as applications move to web technologies and APIs in the name of control privacy and effeciency. While I don't think Safari neccesarily holds back the web, I think it could make a more concerted effort to steer it.

With all that said, it's easy to come to the conclusion that the web has a monoculture browser problem; over time, Chromium will emerge the obvious victor. And that's not great, but not because there would be only one browser engine.

The web in 2021 is a complex place - due to a number of factors it's no longer simple documents located on webservers, but we now have what are ostensibly desktop applications loaded in a nested operating system. For better or worse, this is where we've ended up, and that brings a lot of hard problems to solve for a new browser. This is why I believe there really hasn't been any new mainstream (key word!) browser engines - the web is simply too complex. The browser pushing this "forward" (somewhat subjective) is Chromium, but Chromium is controlled by Google. While there are individuals and other corporations contributing code, Google still controls Chromium, and this makes a fair few people uneasy given Google's primary revenue source - ads, and in turn tracking. Logically, we want a more diverse set of browsers to choose from, free of Google's influence! Forget V8/Blink, we need independant engines! Full backing for Gecko and WebKit! Well, yes, but actually, no. We need to throw effort behind one engine free of Google's clutches, but it should be Chromium/V8/Blink.

Hear me out (credit to @Slackwise for planting the seed of this in my head) - we should really opt to tear the most successful engine from Google's clutches and spin it off into its own entity. One with a nonprofit structure similar to how Linux manages itself (a good example of a large scale effort in a similar vein). The web is simply too complex at this point for new engines to thrive (see: Servo), and the other two options, Gecko from Mozilla/Firefox and WebKit from Safari/Apple, are having a really hard time evolving and playing catch up. With a foundation dedicated to the engine, and a licensing or sponsorship model built out, I genuinely believe that it would be better in the long run for the health and trust of the internet. We can still have Chromium derivatives, with their unique takes or spins, so it would not reduce the choice available (besides, people choose based on features, not engine). Concentrating effort into a single browser engine rather than fragmenting the effort across a handful might allow for some really great changes to the core of the engine, whether it be performance, better APIs, more privacy respecting approaches, and so on. It also finally eliminates the problem of cross browser incompatibilities.

Would it stagnate? Maybe. It's entirely possible this is a terrible idea that would stiffle innovation. But given the success and evolution of other projects with a matching scale (Linux), and the constant demands for new "things" for the web, I feel confident that we could maintain a healthy internet with a single engine. And remember, it's okay for something to be "done". Constantly shipping new features isn't neccesarily a plus - while we see new features shipping as a sign of activity and life, it's perfectly fine for us to take a step back and work on bugs and speed improvements. And if something isn't satisfactory, I'm pretty confident that the project could be properly forked with improvements or changes made later upstreamed, in the spirit of open source and collaboration.

There are calls for breaking up the large technology companies, but I don't really want to delve much into that here, or even consider this a call to action. Instead, I want this to serve as mild musings and hopefully get the seed of an idea out there, an idea discussed a few times in a private Discord guild. I don't expect this to ever become a reality without some strongarming from some government body, but I hold out some hope.