Could a Google-designed SoC finally end Android fragmentation?
Long ago, when Google was a rising star giving “everything” away for free, Apple was busy revolutionizing the mobile industry with the iPhone. Google fought back, not by creating the “Google Phone”, but by acquiring the Android operating system, spinning up a consortium of carriers, OEMs, and technology providers, and rolling out a massive initiative that has finally surpassed Apple in number of handsets in use. That strategy, and the momentum behind it, has one flaw that self-professed pundits continue to proclaim: Android fragmentation.
- Skins: Various “branches” of the main Android “trunk” which are customized by OEMs and distributed “fragment” the core operating system. These branches (sometimes erroneously called “skins”) eventually become out-of-date due to the manpower required to keep the code patched, which takes us into the second (and more commonly considered) form of “fragmentation”;
- Versions: Android OS fragmentation is the perceived abandonment of particular devices to get patches and updated versions of the Android operating system, and is usually what comes to mind when people refer to “Android fragmentation”.
Apple’s iOS has it’s only kind of fragmentation: “feature fragmentation”. As new Apple devices are released with new and updated hardware, iOS is updated to take advantage of the new hardware. Devices running older hardware may not be able to use these newer features, so Apple either doesn’t include the (allegedly) incompatible features, or limits their functionality on the older hardware.
Both forms of fragmentation can be frustrating and confusing, but Google’s is probably moreso. The very benefit that made Android so successful – virtually limitless customizability by OEMs – is the same limitation which makes keeping those devices up-t0-date so terribly difficult. To address that, Google had a plan.
Google teamed up with local telcos to offer low-cost Android-powered handsets. Unlike the major carriers in the States, Europe, and Asia who can solicit the major hardware OEMs to create a custom handset for their markets, carriers in emerging markets are much smaller, and can’t support custom handsets like the others can. The solution? Android One.
Under the Android One plan, Google offered a hardware framework and the promise to develop and support the operating system running on those handsets. It’s the unified hardware framework and the lack of customizations to each type of handset that makes this possible.
Unfortunately, Android One reportedly isn’t doing so well, and Google is rumored to be working on revising the plan. Android Two, perhaps? Naming notwithstanding, Android One is very telling in terms of maintenance on devices – and much more in line with what Apple has been doing with iOS all along.
Unlike desktop computers, smartphones and tablets don’t use discrete chips, customized specifically for the job they’re tasked with doing. Instead they use a concept called a “system on a chip”, or SoC for short. There are still some specialized chips in today’s phones, but most of the heavy-lifting is done by that SoC.
There are lots of SoC makers out there – and each has a pretty respectable selection of SoCs from which manufacturers can base their newest models after. Qualcomm, Samsung, MediaTek, Intel, Nvidia, and others make SoCs. Each requires its own drivers – and that’s where the problems begin. The chip that powers iPhones is designed by Apple. It’s the core that drives the phones and tablets upon which iOS runs.
Why doesn’t Google’s Nexus One run Android 6.0 Marshmallow? The “old” Snapdragon SoC at the core is simply “too limited” to provide the experience that Google wants – and the company would have to invest quite a bit of time updating the drivers and dependencies of the “old” hardware to make it work.
That’s where a Google-designed SoC could come into play. With a common platform upon which all Android-powered devices could run, Google would be able to streamline driver updates (including better backwards-compatibility), theoretically extending the support window for those devices, but more importantly, making it much easier to deploy patches and updates than the scattered, disorganized way in which Android-powered devices get updates today.
Could a Google-designed SoC finally end Android fragmentation? Maybe not overnight, but it’s certainly a step in the right direction.
Now that you’ve got some background into the issue, and have heard our thoughts, we’re curious to hear what you have to say! Head down to the comments and sound off!