The End of the Windows Era? (Premium)

Open-source advocate Eric S. Raymond believes that Windows will evolve into an emulation layer on top of Linux. He’s nuts.

“If you think this is fantasy, think again,” he writes. “The best evidence that it’s already the plan is that Microsoft has already ported Edge to run under Linux. There is only one way that makes any sense, and that is as a trial run for freeing the rest of the Windows utility suite from depending on any emulation layer.”

Here’s the thinking: Microsoft makes most of its money today, Raymond says, on Azure. That’s not actually true, of course, though Microsoft does make a good chunk of its revenues—30 to 35 percent, based on its most recent quarterly earnings—directly from Azure. And it’s fair to say that some of its other revenues are also implicitly derived in some way via Azure as well.

Anyway, Windows is now a “sideshow” at Microsoft, Raymond says. That’s also not strictly true: I estimate that about 26 percent of Microsoft’s earnings came from Windows, same quarter as noted above. But let’s go with it.

With PC sales falling, he argues, the return on investment on Windows development spending is falling and it will inevitably stop being a profit center and become a drag on Microsoft’s overall business. And sure, all things must end. But Windows today is a huge profit center for Microsoft.

“[Microsoft would] do better putting more capital investment into Azure—which is widely rumored to be running more Linux instances than Windows these days,” he argues. “What’s the profit-maximizing path forward? It’s this: Microsoft Windows becomes a[n] emulation layer over a Linux kernel, with the layer getting thinner over time as more of the support lands in the mainline kernel sources. The economic motive is that Microsoft sheds an ever-larger fraction of its development costs as less and less has to be done in-house.”

In Raymond’s view, the “new” Windows is mostly just Linux, but with that “old” Windows emulation layer. But the key to this fantasy—sorry, this possibility—is that Microsoft’s own software, increasingly, will not use the “old” Windows emulation (which one might equate to how Windows 10 on ARM runs Win32 software today, I guess). Instead, it will run on something called Proton that’s used to help Windows-based games distributed via Steam run on Linux.

“It’s not perfect yet, but it’s getting close,” he says of Proton. “Games … are the most demanding possible stress test for a Windows emulation layer, much more so than business software. We may already be at the point where Proton-like technology is entirely good enough to run Windows business software over Linux. If not, we will be soon.”

Over time, Raymond believes that Microsoft will incrementally converge Windows with Linux and then deprecate and remote the emulation layer from the “new” Windows.

“The OS itself , and its userland tools, has for some time already been Linux underneath a carefully preserved old-Windows UI,” Raymond writes of this alternate reality. “Third-party software providers stop shipping Windows binaries in favor of ELF binaries with a pure Linux API … and Linux finally wins the desktop wars, not by displacing Windows but by co-opting it.”

This is a fantasy.

And not for the obvious knee-jerk reasons, but because Microsoft has now spent several years trying to get Windows 10 on ARM to work, and that platform still can’t run 64-bit Win32 apps or use the millions of Intel-compatible device drivers floating around out there. If it can’t even make a new Windows that can run all Windows apps, how is it going to make a version of Linux that does so?

Raymond is also not considering the very real costs of this imagined transition. Right now, there are many thousands of developers at Microsoft that understand the architecture of Windows and how to create apps that run well on this platform. Moving to Linux would require all of them to be trained on this new platform and then move their products forward as well. And that’s just inside Microsoft. It’s hard to even contemplate the number of person-years such a transition would require for the entire industry.

And even then, what’s the point? If this happens and then succeeds as much as it possibly could succeed, all Microsoft would accomplish at the end of the transition would be to have created something that looks and works exactly the same as the Windows we already have. A saner approach would be to slow down Windows development, slowly siphon off resources as needed, and move it into a long-term servicing schedule for everyone.

But even that would require Microsoft to believe that there’s no point in pushing Windows forward. And everything we’ve seen this year suggests that’s not happening at all. Not right now, anyway.

So, yeah. Sorry, Eric. But you’re nuts.