When I first saw this image of Bill Flora, a key leader on the team that created Microsoft's Windows 8 Metro interface, I almost laughed myself silly. Notice what Flora, who had left Microsoft after 19-years as a creative director in September to start TECTONIC, a user experience design firm, uses for his main computer? That sure looks a MacBook Pro to me!
Seriously? One of Microsoft's go-to Metro guys left the company ahead of Windows 8's launch and now uses a Mac? The picture says it all. Of course, Floria's not the only Metro developer to abandon ship. Brandon Watson, head of developer experiences for Windows Phone, is the latest executive to leave the Microsoft’s phone unit. Between Flora and Watson's departure, Matt Bencke, General Manager for Windows Phone Developer and Marketplace, left the Windows Phone team, but he did, at least, stay in Microsoft. He's now over with the to Xbox Live crew.
OK, so Metro, while it's getting closer to beta, is still too wet behind the ears for serious work. That said, I can't be the only one who thinks it's a little odd that Metro developers are running out the door so late in the game and with so much Metro development and design work still to be done. I wonder what they know that Microsoft isn't telling us. Mind you, Microsoft isn't telling users or developers that much about Metro applications and programming.
We do know, however, that it seems Microsoft may just be supporting both Metro-style and Desktop apps, aka classic Windows, on Windows 8 on ARM. That tells me Microsoft is starting to hedge its bets that dyed in the wool Windows users are really going to take up a radically new interface to do the same old things.
That said, I also doubt that Microsoft can bring its old x86 Windows look and applications to ARM in a way that will work. I think the Metro interface will be a total failure on the desktop, but I think that “Classic” Windows on a tablet or a smartphone will be even more of a design flop.
Leaving aside the foolishness of trying to use a Windows, Icons, Menus, and Pointer (WIMP) interface on a tablet, there's the engineering problems behind the display. It's not that easy to port old x86 code and recompile it for ARM.
You could, I suppose use an Apple-style Rosetta x86 binary emulator for ARM, but that approach has three major problems: 1) It's slow; 2) It eats power like an elephant eats peanuts; and 3) you're still stuck with very different screen architectures.
Sure, I can see why the idea of using Desktop on ARM sounds attractive at first glance: “Give the people the interface and apps they already know,” but practically speaking it comes with a host of problems. Not least of these is that it makes me wonder, yet again just how really ready Metro is for any users on any platforms.