TightWind is written by Kyle Baxter. Stay Hungry. Stay Foolish.

From Mathew Honan’s Surface review:

The versions of Word, Excel, and PowerPoint are, well, Word, Excel and PowerPoint. I’ll admit to not trying anything complex with Excel during my testing period (or ever, for that matter) but they did everything I expected. Unlike apps that run in Metro — excuse me, Modern UI-style apps — you can view several Windows at once on the Desktop. Also, all have application chrome — buttons and menus and other onscreen elements banished by Microsoft’s Modern UI style. You can use touch with these apps, but they really demand a keyboard and trackpad to work well.

Overall, Honan liked the Surface. He loves the hardware, loves the gestures Microsoft’s built into the Metro interface, and thinks that it could absolutely be someone’s only computer.

The above section gets at something that I find perplexing about the Surface, though. Surface RT gains the Metro interface, which uses touch as its primary interface and is designed as such, but still includes the traditional Windows desktop. The included Office suite only runs on the desktop, and though it’s designed to be more touch-friendly, Honan says the suite all but requires a keyboard and trackpad, which appears accurate from what I’ve seen of PowerPoint.

If that’s the case, Metro applications are for more casual, less intensive kind of activities—e.g., “consuming”—while the traditional desktop remains for “serious” applications like Word, PowerPoint and Excel, where a mouse and keyboard can be used. In this way, Microsoft’s reasoning seems to go, the Surface can switch between being a regular, productive PC (multiple windows, keyboard and mouse) and a casual, easy-to-use tablet (full-screen applications, touch).

That doesn’t position the traditional desktop as a legacy feature so existing applications still work; it positions it as an integral element of Windows’s future, the place where real work gets done. Unless I’m missing something, this encourages developers to build real applications for the desktop, and at minimum doesn’t encourage them to build fully-capable Metro applications. In other words, it constructs a firm division between real applications and tablet applications, and Microsoft is encouraging it.

The idea of having one relatively small device that can be both a good PC and a good tablet is exciting, but Microsoft’s approach appears to intentionally limit Metro’s possibility to protect traditional Windows. Maybe that’s the right approach for converging the PC and tablet, but at least right now, it seems like they’re purposefully limiting how much Metro’s tablet concepts can grow and develop to keep the desktop involved.