The Browser Engine Wars: WebKit Loses Its Last Friend

Ten years ago we had just escaped the trauma of the IE6 monoculture. In 2026 we are looking at a strangely similar landscape. Chromium and Blink own over 70% of the desktop market and over 65% when mobile is included. Safari and WebKit effectively monopolize iOS, while Firefox and Gecko have dropped to around 3% (StatCounter, 2026). The latest blow came from Microsoft, whose research prototype of Edge built with Apple's BrowserEngineKit framework scored 28.6% higher than Safari on Speedometer 3.1, proving that WebKit isn't even the fastest engine on its own platform anymore.

Why does WebKit matter?

Apple requires all browsers on iOS to use WebKit, the engine that powers Safari. That means Chrome, Firefox, and Edge on iPhone are effectively reskinned Safari instances running the same layout and JavaScript engine under the hood. The EU's Digital Markets Act theoretically changed that in March 2024, requiring Apple to allow alternative engines through BrowserEngineKit, yet more than two years later no browser maker has shipped an alternative engine on iOS for most users worldwide (MacRumors, 2026).

The reason is structural. Apple's compliance framework requires any browser shipping a non-WebKit engine to publish it as a completely separate app from the existing WebKit-based version. That doubles development cost, fragments the user base, and forces companies to maintain two codebases for one platform. The combination of regulatory uncertainty, technical barriers, and distribution complexity has proven sufficient to deter everyone so far.

The three-engine world

In 2013 there were five major browser engines: WebKit, Blink, Gecko, Trident, and EdgeHTML. In 2026 there are only three left: Apple's WebKit, Google's Blink, and Mozilla's Gecko. Microsoft killed EdgeHTML in favor of Chromium in 2020. Opera abandoned its own Presto engine back in 2013. The consolidation has been brutal and steady. Gecko is the only remaining independent browser engine, and it powers Firefox, Tor Browser, LibreWolf, and Mullvad Browser (Mozilla, 2026).

The numbers tell the story. Worldwide in July 2026, Chrome holds 68.22%, Safari sits at 16.47%, Edge takes 5.37%, and Firefox manages just 3.34% (StatCounter, 2026). On desktop specifically, Chrome dominates at 70.58%, Edge has grown to 11.15%, Firefox holds 6.51%, and Safari shrinks to 5.66%. Google's Blink engine powers Chrome, Edge, Opera, Brave, Vivaldi, Arc, and Samsung Internet. That single codebase renders the overwhelming majority of web pages on earth.

What Microsoft's Blink prototype actually proved

Kyle Pflug, group product manager for Microsoft Edge Web Platform, published benchmark results in June 2026 comparing a Chromium-based Edge prototype running on Apple's BrowserEngineKit framework against Safari on iOS 26.5.1. The Blink-based prototype scored 49.27 versus Safari's 38.3 on Speedometer 3.1, a 28.6% gap. It also beat Safari by 13.1% on the JetStream 3 JavaScript benchmark (306.35 vs. 270.9) and by 2.1% on MotionMark 1.3.1 for graphics rendering (4,773.52 vs. 4,673.68) (MacRumors, 2026).

Pflug described the work as a research prototype rather than a finished product, and the numbers as preliminary results from his own device rather than controlled lab conditions. But the implications are hard to ignore. For 17 years Apple has argued that its WebKit mandate protects iOS users by ensuring a single, optimized, secure browsing engine. Now a rival engine runs measurably faster on Apple's own hardware, inside Apple's own BrowserEngineKit sandbox.

Open Web Advocacy told The Register that these results illustrate a 17-year cost to consumers. The group called on the European Commission to open a specification proceeding instructing Apple precisely how it must remove barriers to alternative engines.

— The Register (2026)

Why hasn't anyone shipped an alternative engine on iOS?

Companies cite three overlapping barriers. First, Apple requires any browser shipping a non-WebKit engine to publish it as an entirely separate app from the existing WebKit-based version. Second, BrowserEngineKit is only available to browsers distributed in the EU, leaving users in the US, Japan, Korea, and everywhere else locked to WebKit. Third, the regulatory picture remains uncertain, with Open Web Advocacy calling on the European Commission to open a specification proceeding instructing Apple on exactly what barriers must be removed (MacRumors, 2026).

Mozilla's Andrew Overholt put the stakes plainly in a March 2026 blog post: if a small number of vertically integrated companies completely control browser engines, then competition, transparency, and user choice on the open web become much harder to achieve. Those companies will have strong incentives to favor their own services, limit interoperability, and steer defaults and standards to their advantage (Mozilla, 2026).

Ladybird and Servo: new hope for engine diversity

Ladybird stands as the most ambitious effort to build a completely new browser engine from scratch in over two decades. Started as part of Andreas Kling's SerenityOS project, it was spun off as a standalone project in 2024 and is targeting an alpha release for Linux and macOS in 2026. It is written entirely in C++ with no legacy code from Blink, WebKit, or Gecko. The project is backed by a non-profit foundation, has over ten full-time maintainers, and has received a million-dollar donation from GitHub co-founder Chris Wanstrath along with support from Shopify CEO Tobi Lutke (Ladybird, 2026).

Servo, originally developed at Mozilla Research starting in 2012 and transferred to the Linux Foundation in 2023, is being revived as a standard web engine for automotive infotainment, set-top boxes, and embedded devices. Several of its innovations, including the Stylo CSS engine and WebRender GPU renderer, have already been absorbed into Gecko and Firefox. Servo's true future is less likely as a standalone browser and more as the embedded web runtime that brings engine diversity to devices that currently rely on Chromium's WebView or Apple's WKWebView (youngju.dev, 2026).

The hidden engine economy: embedded systems and app frameworks

Most people never see the browser engine powering their apps, but it shapes their experience every day. Spotify's desktop client, the Steam store, OBS Studio widgets, and countless game launchers all embed Chromium through the Chromium Embedded Framework. Every Windows app using WebView2, since 2024 the default system component, renders through Edge's Chromium engine. On iOS, every in-app browser from Slack to Discord to Twitter/X goes through Apple's WKWebView. On Android, WebView is Chrome-based through the Trichrome shared package. The web isn't just browsers anymore. It's the UI layer for most desktop and mobile applications.

This hidden economy means the engine wars don't just affect what you see in Safari or Chrome. They affect how fast your music app loads, whether your game launcher supports modern CSS features, and how secure the webviews inside your banking app actually are. When one engine dominates, developers optimize for it. When engines diverge, developers test on all of them. The practical question is whether enough engines survive to keep that diversity meaningful.

What the EU DMA actually changed

The Digital Markets Act took effect in March 2024 and required Apple to permit third-party browser engines on iOS within the EU. It also mandated a browser choice screen for new device setups. Mozilla reported that the choice screen generated over six million selections in its first year, giving Firefox and other browsers a measurable bump in EU users. Brave and Mozilla gained the technical ability to ship their own engines on EU iOS builds. But the requirement remains: non-WebKit engines are only allowed in the EU, and only as a separate app (Mozilla, 2026).

The lasting significance isn't what happened in 2024 or 2025. It's the precedent. The EU proved that regulation can crack open a platform lock-in that had held for 17 years. The UK's CMA reviewed similar rules in 2025. Japan's 2024 transparency law on digital platforms moves in the same direction. Korea's Fair Trade Commission has been studying preloaded mobile OS apps since 2025. The iOS WebKit monopoly may not survive the decade, even if it persists for now outside Europe (youngju.dev, 2026).

The web standards governance problem and what comes next

Browser engines don't just render pages. They determine how web standards evolve. The W3C handles CSS and accessibility, WHATWG manages HTML and DOM, TC39 standardizes JavaScript, and IETF governs HTTP and TLS. Roughly half of active participants in major W3C and WHATWG decisions are Google employees. When the spec and the implementation come from the same company, the spec tends to follow the implementation rather than leading it (youngju.dev, 2026).

This matters practically. Manifest V3, Google's replacement for the Manifest V2 extension API, restricted the webRequest blocking that powers most ad blockers. Firefox still supports MV2, making it the last stronghold for blocking extensions. Privacy features like anti-fingerprinting, enhanced tracking protection, and telemetry transparency vary dramatically across engines. When one company controls both the dominant engine and the dominant search engine, the incentive structure for privacy-first defaults becomes complicated.

Several developments could reshape the engine landscape over the next two years. Ladybird 1.0, if it reaches a stable release good enough for everyday use, would be the first new major browser engine in 25 years. Servo's integration into automotive and embedded systems could create a new zone of engine diversity outside the browser market. The EU DMA model spreading to other regions widens the cracks in iOS WebKit lock-in. And Firefox's ad-blocker user base continues to grow, giving Gecko a loyal constituency even as its overall share shrinks.

For developers, the smallest meaningful practice for keeping the web diverse is straightforward: open pages in Safari and Firefox alongside Chrome. Test against all three engines. Try the Ladybird alpha when it ships. The monoculture isn't inevitable, but preventing it requires conscious effort from both the people who build the web and the people who use it.

• MacRumors — Apple's WebKit Rules Reportedly Cost iOS Users Almost 30% Browser Performance
• Mozilla — Competition, Innovation, and the Future of the Web
• youngju.dev — Browser Engines 2026: A Deep Dive
• StatCounter — Browser Market Share Worldwide, July 2026
• Ladybird Browser Initiative
• The Ad Blocker Crackdown Backfires. Again.
• The Self-Hosting Boom: Why Everyone Is Building Their Own Stack
• Dead Internet Proof: How to Tell What's Real Online
• Open-Source AI Goes Locally First