Apple might achieve what Windows 10 never could
Putting the same chips into MacBooks and iPhones opens up exciting possibilities
When Windows 10 was first revealed back in 2014, I was naturally interested in its new features but what intrigued me most was its promise to bring the desktop and mobile app experiences into unity.
At the time, hopping from phone to laptop to browser had become a fact of life, and I was getting decidedly fed up of dealing with multiple versions of everyday apps and services, all with different experiences and feature sets. I even thought this could be the feature that persuaded me to ditch Android and switch to Windows Phone.
Well, we all know how that turned out. But it's lately become apparent that the dream might not be dead - and its fulfilment could come not from Microsoft, but from the company's long-term antagonist, Apple.
To explain, this all stems from the recent revelation that, in the next few years, Apple plans to stop buying chips from Intel and start designing its own laptop and desktop CPUs instead. As it stands, this makes a lot of sense: Intel's processors once helped Apple lead the market, but today it's no longer the prestige brand it was.
Indeed, being tied to Intel has become a liability for Apple. The capabilities and specs of its computer designs are greatly delimited by what Intel's CPUs and chipsets can do. In recent years, customers have complained about the 2016 MacBook Pro being limited to 16GB of RAM, and about the 12in MacBook lacking Thunderbolt 3. Apple has no control over this nor over Intel's manufacturing timetables, which means Apple doesn't fully control its own supply lines either.
Of course, this has long been the norm for computer manufacturers, whether they've sourced their processors from Intel, AMD, Motorola or whoever with the exception of one upstart company that, nine years ago, decided to create its own A4 processor for the original iPad.
There's no debate to be had about whether that was the right decision: Apple's CPUs have since made the iPhone and iPad consistently the fastest devices in their respective classes. What's more, Apple has gained the ability to expand the design as it sees fit, and to build in new components such as the "taptic engine" coprocessor.
In short, switching to in-house chips seems like a no-brainer. Unfortunately, migrating macOS onto own-brand CPUs isn't going to be quite as simple as it was for iOS. With its mobile devices, Apple was easily able to license the ARM architecture that the iPhone was already using, and customise it to its needs. ARM's business is based on companies doing precisely that.
Intel does things differently. Back in the 1990s, the company did authorise a few third parties to make their own x86 processors - a right that AMD and VIA retain to this day. There's an argument to be made that, as the company seemingly struggles to advance its own architecture, it might be smart to once again let someone else take on the challenge, and sit back and reap the licensing fees. But, for better or worse, the Intel of today isn't in the licensing business.
Would such an arrangement even make sense for Apple? On the plus side, it would ensure that everything that works on macOS today would continue to work. But consider the alternative: if Apple instead migrated macOS onto its existing ARM architecture, it would have only a single hardware platform to support, rather than two.
That's a hugely attractive idea, and not just because it greatly streamlines Apple's engineering challenges. It's a fantastic opportunity for developers too, making the old dream of "write once, run anywhere" shockingly achievable. Best of all, it would finally deliver on the promise of Windows 10: a standard platform for apps that would run identically on tablets and laptops, or on smartphones and workstations.
Am I too optimistic? Frequently. But the fact that Microsoft blew it doesn't necessarily mean that Apple can't make it work. For a start, Windows' Universal app framework was effectively an attempt to foist a brand-new platform on an unconvinced world.
Apple will be starting with a market of a billion iPhones, a mature set of tools and an enormous developer community.
What's more, this sort of radical migration isn't new to Apple. In 2001, it aggressively transitioned its entire computing family from OS 9 onto an entirely incompatible successor. In 2005, it did it again, shunting the whole shebang from the PowerPC architecture onto Intel. By comparison, getting the modern macOS platform to work on the same hardware as iOS should be a breeze. After all, the two already use the same Darwin kernel, and have many core components in common.
I'm not the first person to predict that macOS will end up being merged into iOS. I remember Barry Collins confidently making the call back around the launch of the original MacBook Air; you can see exactly what he thinks about that possibility opposite.
While most analysts have focused on the global opportunities for Apple, I'm thinking closer to home. I ditched the iPhone nearly a decade ago, swearing I'd never return to that walled-off, overpriced, back-buttonless platform. Yet if anything can woo me back, it could be the once-denied dream of the top-to-bottom application stack. After all, I use my MacBook Pro daily, and I don't see myself ditching the smartphone any time soon.
In short, Apple could be gearing up for its greatest coup yet. Moving 100 million macOS users onto a whole new platform - heck, that's been done before. But getting me to consider ditching Android and moving back to the iPhone? I never thought I'd see that.
Unlocking collaboration: Making software work better together
How to improve collaboration and agility with the right techDownload now
Four steps to field service excellence
How to thrive in the experience economyDownload now
Six things a developer should know about Postgres
Why enterprises are choosing PostgreSQLDownload now
The path to CX excellence for B2B services
The four stages to thrive in the experience economyDownload now