There have been two successful OS transformations on the desktop. One was Mac OS Classic to Mac OS X, which was implemented using virtualization. The other was DOS to Windows, which was a slightly weirder affair (initially, Windows was a DOS application, then Windows NT ran DOS under virtualization). You might argue Windows was a horrible kludge, but its more elegant step-sibling (OS/2) handled DOS compatibility by virtualization and failed miserably in the marketplace.

It seems pretty clear, especially given the power of current hardware, that virtualization is the way to handle an OS transformation. Indeed, many commentators have suggested that Microsoft should replace Windows with a brand new modern, lightweight OS, and manage compatibility by virtualization.

Right now, iPhone OS runs under Mac OS X via virtualization. Multitouch is not well-supported (for obvious reasons), but that’s simply a hardware issue (Macs don’t have touchscreens).

Of Apple’s two operating systems, one generates over two-thirds of its revenue, and an even larger proportion of its profits. And that OS isn’t Mac OS X. Apple is notorious (I might say famous, but chose not to) for doing a lot with a little — there are probably fewer people working on Mac OS X right now than on Microsoft Word. But we haven’t even heard a whisper about Mac OS X 10.7.

So, the question is, whither Mac OS X?

Merging it with iPhone OS is impractical for numerous reasons, not least of which is that iPhone OS runs very lean and mean and Mac OS X conspicuously does not. A virtualization solution would allow iPhone OS to continue working beautifully on low-powered devices (by not providing the compatibility box) while allowing higher-powered devices to offer full backwards compatibility.

Of course, Apple already has an iPhone virtualization box for Mac OS X, so a “unification OS” could be released tomorrow if Apple wanted to make Mac OS X that OS, but I think a Tablet computer that boots instantly into iPhone OS and lets you run Mac OS X in a virtual box as needed is far more desirable than a Tablet computer that boots in 30s into Mac OS X and lets you run iPhone OS in a virtual box. Either would be pretty compelling, though.

The other question is, what benefits are there to keeping the two platforms separate? I would argue there are none. iPhone OS devices with a Mac compatibility box would, in essence, answer all the “closed platform” criticisms — the Mac platform is rich and open, and running it on a virtual machine would sandbox it from the managed world of iPhone. Indeed, virtualization affords Apple the option of opening iPhone OS devices without adding risk for users who don’t want it. The only real reason not to go down this route right now is hardware.

There’s your Mac App Store, by the way. It’s the App Store, and iPhone OS running on your Mac.

So, I predict that iPhone OS will subsume Mac OS X within three years. Obviously, it will long since have ceased being iPhone OS, of course. Hence, the title of this post.