An Excess of Operating Systems

by Jean-Louis Gassée

Received wisdom tells us that developing a new OS is technically feasible but ultimately doomed by the need to create the type of large-scale ecosystem that protects incumbent platforms. Why, then, undertake the creation of a new OS universe?

Over the past few decades, operating systems settled into a seemingly inevitable duopoly: There’s an OS for low-end OEM devices, and another for more expensive, vertically integrated contrivances. In the PC era, the players were Windows and macOS; for mobile devices, we have Android and iOS (although this time, Apple has a much nicer market share). It’s a familiar, comfortable pattern. Order reigns.

Not quite.

It seems there are new operating systems sprouting up everywhere:

  • Samsung is busy developing the Tizen operations system.
  • Google publicly discusses its new Fuchsia OS.
  • Huawei is at work on an engine called HarmonyOS.
  • Amazon has its Android “hard fork” (I’ll explain) called FireOS.
  • is reportedly at work on its own OS..

And there may be more.

Why such a plethora? I can see two reasons.

Most important, the tech world has changed…again. Just as the playing field was redrawn when smartphones dethroned PCs, so now we have a swarm of devices that want to populate — some say invade — our lives, homes, and bodies. We wear smartwatches and headphones, and, someday, smartglasses. The “modern” home contains smart speakers, lightbulbs, and window shades; we prepare our food and wash our clothes in IoT appliances; we protect our property with connected security (?) cameras…

And just as it became clear that smartphones couldn’t be powered by desktop PC operating systems, it’s become obvious that smartphone era OSes are a bad fit for these newest devices.

(As an aside: One might object that iOS was initially presented by Steve Jobs as a version of OS X running inside the iPhone, and that there are still some source code bits common to macOS and iOS. True, but they really are two different software engines, something made clearer by the struggle to make iOS apps run on macOS. Apple plays by rules that don’t quite apply to the rest of the industry, but , even for Apple, a single OS that works on “everything” isn’t realistic.)

The requirements of this new breed explain Google’s drive to create the new Fuchsia OS. Of particular note, Google’s wearable business would benefit from a modern, agile software engine that isn’t encumbered by Android’s size and the layers upon layers of patches it has acquired in its more than ten years’ existence.

All the other OS developments fall into a second type of reason: Defensive moves, protecting one’s business from an OS vendor that ultimately captures one’s business strategy.

Consider Amazon’s FireOS, a “hard fork” (meaning “Google-incompatible”) of the Android OS. Why the incompatibility? A normal Android distribution includes Google Mobile Services (GMS), a collection of apps and programming interfaces that end up providing search advertising revenue to Google. Funneling money to Google isn’t in Amazon’s self interest (to put it mildly), so the company stripped GMS from their Android derivative when it built the Fire Phone. Unfortunately for Amazon, the stripped down OS doomed their phone…but that same GMS incompatibility turns into an advantage for Alexa devices. Smarter, faster, and no revenue stream to Google.

Still on the defensive side, we have Samsung’s Tizen. The Korean company doesn’t want Google to dictate the future of its smartphone (and other products) business, so it’s building its own OS. Critics of the strategy point to the enormity of the task: How will the company recreate the enormous Google Play universe of apps? Writing a new OS kernel is a manageable task, but designing and delivering the frameworks and APIs (Applications Programming Interfaces) that third-party developers will use to create their products is a much larger, much more involved effort. It’s a mission deemed impossible by most critics.

Huawei is another company that doesn’t want its future tied to a US supplier, although for slightly different reasons: Given the current administration’s tense relationship with China, Huawei’s motive is as much geopolitical as it is a desire to break free from the Google ecosystem.

As with Samsung, critics have declared Huawei’s efforts “doomed from the start”, that building an entire new ecosystem simply isn’t credible. But there’s a difference between the two companies: Samsung doesn’t have the breadth and skill level of Huawei’s engineering resources, as attested by the latter’s supremacy in the telecom infrastructure sector. Critics also fail to give weight to the determination of Huawei’s CEO, the size and attitude of the company’s home market, and the support and patience — some say complicity — of the Chinese government.

(Personally, I think Huawei is Apple’s most dangerous competitor. I see how successful Huawei already is against Apple in Europe, a set of markets where Apple is clearly less competent than one would have hoped.)

Last but not least, Facebook’s drive to build a homegrown OS is easily understood: While the company’s current hardware business is small (Portal devices and Oculus VR helmets) and isn’t dependent on an aging OS, it doesn’t want Google or another OS vendor to meddle with its strategy as it broadens its already huge reach. Whether Facebook will create its own swarm of devices — and maybe backfill its portfolio with smartphones and tablets — is anyone’s guess, but looking at the company’s determined, no holds-barred culture, and its financial and engineering resources, anything is possible.

As Benedict Evans remarks in one of his must-read essays, big industry fortresses are not so much overrun by barbarians as they are made less relevant because the trade routes that they once controlled are redirected away by new entrants. Thus, we saw the PC hegemony give way to the worldwide smartphone generation; now we’re seeing a new, pervasive, “fermenting” breed of smart, connected devices that threaten to obsolete the Android/iOS duopoly.

As one who enjoyed living through the first two generations, I look forward to the good — and bad — innovations that the swarm of smart devices will bring.


To Monday Note readers, my wishes for an “interesting” 2020 and my thanks for reading and commenting my weekly essays.


An Excess of Operating Systems was originally published in Monday Note on Medium, where people are continuing the conversation by highlighting and responding to this story.