Seismic Changes To The Browser Landscape: Where We’re Headed Now

The last few weeks have witnessed a series of major web technology announcements from Google, Microsoft and Opera. Understanding what these changes means for developers requires a little background on how browsers are made.

“Closed” development is associated with corporate culture: hire hundreds of paid programmers, lock them in cubicles for years, and release proprietary software versions. Microsoft’s Internet Explorer is a product of this process: a monolithic black box. The Opera browser was developed under similar structures.

Open source, on the other hand, accepts code from anyone who is willing to contribute to a project. The results tend to be far more changeable and unpredictable. Open source products evolve in a manner akin to software ecology: projects swap code liberally, often becoming so specialized that they branch out into their own species. Firefox, Chrome and Safari are, to various degrees, open source projects.

Well-run and active open source projects issue new versions nightly; closed-source typically releases working code every few months.

What’s happened in the last few weeks is a dramatic re-alignment of both open and closed-source browser technologies. A simple evolutionary tree would look like this:

Some highlights of this process:

• Announced last week, Google’s “Blink” is simply another fork in the long evolution of the Webkit layout engine, which was itself derived from an earlier open source project. It’s important to understand that forking is a transitional process: the most recent version of Chrome has Blink components already, and its code will continue to be open to adoption by the Webkit team if they choose to use it. However, the nature of open source implies that that Blink and Webkit will drift apart as their codebases become more diverse.

• The Blink development team have stated that they do not intend to create a new vendor prefix for the browser. In fact, there is a general consensus that they will use as few CSS prefixes as possible in the rendering engine. Where necessary, Blink will use –webkit prefixes.

• It’s assumed that Chrome will carry on using the same branding, look and feel: it’s the engine under the hood that is changing, not the bodywork.

• Opera technologies will be merged into Blink, and any future browser operating under the Opera badge will use the Blink rendering engine, rather than Webkit.

• Meanwhile the Mozilla Foundation and Samsung are working on Servo, a completely new layout engine designed for multi-core processors built on the Rust programming language.  While not yet integrated into Firefox, the suite of technologies would seem to be a natural fit for future products.

• Microsoft deserves credit for delivering a standards-aware browser in Internet Explorer 10. The new browser is being pushed out to Windows 7 users as an update while the company works on a sequel that will embrace even more technologies. IE 11 looks very promising: the beta has the ability to run WebGL right now. It’s also interesting that the browser will deliberately advertise itself to servers as “Netscape compatible” rather than as pure IE to avoid browser-sniffing pages that attempt to deliver a broken, non-standards compliant “IE experience”.

• Microsoft has also back-tracked on its initial offering of IE in Metro mode not running Flash to enable the plugin, in response to the use of the technology on popular sites such as Netflix. I regard this as a temporary capitulation: more of those sites will be shifting to HTML5, especially as it now supports DRM-encoded media (a controversy in itself).

There are many reasons behind these titanic changes in the browser landscape, but the outcomes appear positive: greater browser diversity, more standards-aware code, a better experience on mobile devices, and an increasing emphasis on speed, security, and efficiency.