You can't discount the confused product strategy re: old software in this failure.

People want desktops to run desktop software. Not WinRT/universal/modern/UWP. They did not understand this. So for multiple iterations, they shipped tablets, phones, etc., under the "Windows" brand that could not do what people think of when they hear "Windows".

They should have shipped a device circa 2012 that let you recompile desktop Windows apps for arm. They had the technology.

Just like with the Ubuntu Phone: we just wanted a regular X11 on the phone, not a copy of Android app space on Android base.

Nokia N900 had regular X and a forked Gtk+. I liked that thing. They didn't really support it well. Then Elop came.

We should not forget the weight of the board on his decisions.

It's not that Continuum failed. It's that Windows Mobile failed as a platform (for other unrelated reasons), so we didn't even get to the point where Continuum was even a realistic consideration.

The Lumia 950/XL were the last hurrah for the platform, so one can hardly blame Continuum for the demise of Windows on phones.

we still need the pc revolution on mobile devices.

we are still in the mainframe/timeshare stage.

> we still need the pc revolution on mobile devices.

The PC revolution happened because of very open hardware specifications and a non-monopoly on the firmware (the IBM BIOS was reverse-engineered by Compaq).

That revolution was due to a mistake on IBM's part.

"Microsoft tried this with Continuum for Windows 8/10 and Windows Mobile, it failed."

There are plenty of reasons for that, chief among them being the staggeringly-uphill battle of Windows regaining any significant marketshare in the mobile space and (presumably) the lack of software compatible with Windows on ARM.

Meanwhile, Android (and Samsung's phones in particular) are quite popular, and the vast majority (to my knowledge) of Linux desktop programs work on ARM just as well as they do on x86(-64).

It wasn't that no one wanted them - they just weren't powerful or portable enough.

Snapdragon's 835 changed that. So will the A12X from Apple.

I think you’re underestimating Apple’s chip performance. Going by https://browser.geekbench.com/ the best performing Android devices(Samsung’s special thing for single-core, Snapdragon 845 in a OnePlus for multi-core) appear around even with an A10 for single-core and an A10X with multi-core. If this/last year’s Android chips were enough to make Dex viable, you’d expect last generation’s iPad Pro and the iPhone 7 to be capable of running Apple’s solution.

Oh not at all. Apple chips are beasts and surpass anything out there.

But it's likely a greater chance to see a snapdragon head into a convertible desktop format than an iPhone or iPad, neither even allow a mouse. Still have hope for an A12X powering a MacBook tho.

The Snapdragon 835 is not the enabler here. A mid-range 5-year-old phone has 4 cores at 2+ GHz, more powerful than a 10-year-old computer. Even back then this was powerful enough to run plenty of desktop workloads. The thing that's enabling usage of phones as desktops is USB Type-C which is fast enough and versatile enough to carry video output and a USB singal while powering the phone.

Absolutely. The power of the 835 was not the processing power alone, but pairing the 835 with greater than 3 gb of RAM. This helps push it into a desktop capable experience.

I bought a Sentio Superbook, and my pixel is the only non-apple device I have so I can try to use it in this manner.