Each week Ross Rubin contributes Switched On, a column about consumer technology.

DNP Switched On Surface tension

If you hadn't been paying too much attention to the wide swath of tablets and clamshells (and mashups of the two) that Microsoft is targeting with Windows 8 and Windows RT, then the announcement of Surface contained many surprises. If you have, then there were probably far fewer surprises. But even if you knew just about everything about Windows 8 and Windows RT, you were probably a bit surprised to see Microsoft jump in with its own pair of devices to support the flavors of the imminent Windows upgrades supporting Intel and ARM processors.

After all, the fundamental business model of Windows has relied upon licensing to third parties. Ballmer himself has noted in the past that products that ship in the hundreds of millions (or more) of units per year lend themselves well to the licensing model as opposed to the vertical integration model most notably pursued by Apple among Microsoft's main competitors.

Windows. In a role that will surprise you

Given a high enough screen resolution, Windows 8 can run on everything from Kindle Fire-like mini-tablets to 27-inch all-in-ones and beyond. But the Surface product focuses on what has historically been a sparsely supplied form factor for Windows despite slates being the initial favored design for Tablet PCs. With Surface, Microsoft is leaving today's high-volume Windows clamshell and desktop form factors to its existing OEM partners. It makes a statement that Windows is ready for tablets that trade off some thinness and battery life for backward compatibility and vice versa.

In creating Surface, Microsoft reminds us of several key assumptions that helped to drive the development of Windows 8. That, unlike Apple, it firmly believes that the tablet is just another PC and that touchscreens on PCs will become nearly ubiquitous. By leveraging its own brand, design and limited, but growing, direct distribution, Surface shows that Windows is at home on a tablet, at least as long as one uses Metro-style apps.

This is not to say that Windows 8 needs Surface to prove its point. And Microsoft could have stayed even further out of the path of its partners (and created a device with sleeker dimensions) by creating only a Windows RT version of Surface. But Microsoft -- both directly via Surface and via the many other Metrofied Windows tablets that will appear in the coming months -- needs to establish more credibility in the modern tablet market to prevent further encroachment from the iPad. Microsoft's largest OEM partners may not be excited by the prospect of Surface, but they will all derive at least some benefit from it if it succeeds. A win for Surface is a win for Windows 8.

The world is no longer horizontal

If you think that having a company use an operating system for its own devices while trying to license them to others causes conflicts, you bet your Live Tiles that it does. There is a track record of failures in this regard, including Apple's brief foray into licensing the Mac operating system that occurred before the return of Steve Jobs to the company and Palm's attempt to license the Palm OS, which ultimately split the company into the PalmSource software company and the palmOne device company. Microsoft itself recently felt the sting of trying to eat its own dog food when it released the Kin smartphones in advance of Windows Phone 7.

Notably, though, none of these ventures failed for the reason one would think would be most likely to befall a licensee, that is, that the originating company got such preferential treatment that competing with it was impossible. In Apple's case, in fact, it was just the opposite. Clone makers were growing so quickly that they were eating quickly into Apple's share. In Palm's case, as noted, the company split in two to avoid any conflict of interest, but PalmSource folded first as the Palm OS became increasingly uncompetitive and development sputtered. And Kin was so fundamentally flawed from a business model perspective that any Windows Phone licensee would have had no problem trouncing it had it survived.

In fact, there are increasing instances of companies dabbling in both the "license" and "leverage" models. The Google Nexus line began in very limited distribution, but has now expanded to be a part of the portfolio proper at two major U.S. carriers. And that involvement in hardware will be dwarfed by the company's recent purchase of Motorola Mobility. RIM, long a company that developed its own hardware and software, has acquired with QNX a company that earns its revenue by licensing software to other companies such as automobile makers. Once a licensed platform achieves enough momentum, its main developer also has a license: to step into the market itself.

Primed and coated

Since it announced its intention to acquire Motorola's handset business, Google has been adamant that it will maintain a firewall between Android software engineers and Motorola hardware engineers. Despite vague public statements of it valuing Motorola's device business, Google wants to allay the concerns of other Android licensees that its new hardware subsidiary will have an unfair advantage as part of the company developing Android.

But we see no such reassuring pronouncements from Microsoft. Why? First, while Motorola has a small chunk of the smartphone market in the U.S., Microsoft has no market share in the PC hardware business, even among tablets, so it is less of a threat for the moment. Second, while Android licensees can play that operating system against Windows Phone (as Nokia did when it was negotiating), major Microsoft licensees have no practical licensable alternative for the vast majority of PC users, Not competing with your partners was a central rule in an unstated code. These days, though, the company that makes the code makes the rules.

Ross Rubin (@rossrubin) is executive director and principal analyst of the NPD Connected Intelligence service at The NPD Group. Views expressed in Switched On are his own.