Microsoft may be creating confusion among their customers if they abandon the long-held practice of operating system backward compatibility in Windows 8.
There’s been a measure of uncertainty whether legacy software, such as Word or Quicken or Photoshop, will run on Windows 8-based tablet equipment. That is, on hardware sporting an ARM processor. Despite a “firm” denial by Steve Sinofsky, Microsoft’s chief of Windows development, rumors of backward compatibility with the vast body of legacy Windows software have continued.
(By “legacy,” I mean software designed to run on Windows-based desktop and laptop computers on top of previous Windows operating systems.)
Every major version of Windows, from 2.0 onward, has been able to run application software written for previous versions of that operating system. In order to bring along the corporate computing world, Microsoft has made it possible for customers to stick with older software for a very long time while they continue to purchase and use ever-newer versions of Windows. It's been key to that company’s fortunes. While application incompatibilities pop up each time a new OS version emerges, they are a minority case.
This practice has made life easier for Windows-based computer users. The operating system upgrade cycle remains a decision separate and unconnected from application purchases and upgrades. There’s no need to worry about buying new applications when it’s time for a new Windows computer, either. All the user’s old applications can be installed on the new device.
Consider the case of 2004-era Microsoft Office for Windows, or Quicken 2007 for Windows. Both still function as expected on today’s Windows 7 platform.
Contrast that with Apple’s practice, where backward compatibility is a secondary concern. Rosetta, a Power PC compatibility layer for Intel-based Macs, was removed from OS X Lion. Customers who had enjoyed the use of Microsoft Office 2004 for Mac, or Intuit’s Quicken 2007 for Mac, both of which relied on Rosetta to run, are now looking for replacements. Though customers had plenty of warning, and in the case of Office two newer versions to pick from, the end did come for those applications.
Apple puts the benefits of innovative changes in the underlying operating system ahead of application longevity. Application developers are left with the choice of occasionally needing to make significant product code changes in order to keep up, or abandoning the Mac platform. Customers receive a significantly improving computing product as a result, despite the occasional abandoned application. Quicken for Mac users are casting about for a replacement right now.
Windows 8 will roll out on a variety of hardware platforms next year. It will present a unified appearance across those platforms by including a legacy Windows desktop (with a Start button, etc.) as well as the sleek, new Metro user interface. Metro can be seen in a younger form on today’s Windows Phone 7 wireless handsets, where it has been well-received by users and technology critics.
Trouble might come when Windows users, long accustomed to Microsoft’s backward compatibility policy, run up against Sinofsky’s dictate that legacy Windows applications will not run as-is on ARM-based tablets despite the inclusion of a legacy Windows desktop (which would appear to allow a place for that to happen). Imagine the customer who, after plonking down the price of a shiny, new tablet, docking station, external display and keyboard, finds to her dismay that none of her commonly used software will run on her new device. Every application will need a new, ARM-compatible version, if one is even available.
Welcome to the world iOS created, the one where Mac users have no expectation of using their desktop software on their iPad tablet. It provides a clean break from the legacy OS environment and, as Gruber argues, that opens up a wide horizon for innovation.
Geeks will have no trouble navigating the new divide between Intel-based Windows 8 laptops and desktops on the one hand, and ARM-based tablets on the other. Normal customers, though, who don’t know and don’t care about the fine details of operating systems and processor technologies, will.
I raise this issue now because of the confusion that has already surrounded Windows 8’s legacy application capability. Microsoft has a significant education effort ahead if they hope to avert a PR mess next year.
Are there alternatives to the coming divide? Intel will roll out their next-generation processors, called Ivy Bridge, before Microsoft declares Windows 8 “gold master.” The least power-consuming chip in that family has a Thermal Design Power of 15 watts, which is more than seven times that of today’s Apple A5 processor. Yet the chips slated to arrive in 2013, the year following the Windows 8 debut, promise even lower TDP numbers. There is the possibility of using next year’s power-sipping Ivy Bridge chip in a fully capable tablet product, then following it with an upgrade the following year that is competitive with today’s tablets on battery power. But that puts Microsoft fully three years behind Apple and the Android clones.
The choice for Microsoft, playing catch-up in the tablet space, is a poor one. By providing twenty-five years of backward compatibility they’ve primed their customers to expect it anywhere the Windows desktop appears. Yet by insisting that some Windows 8 platforms (desktop, laptop) will retain that capability while others (tablets) will not, they risk alienating those same customers. Next year looks like a period of catch-up “innovation” riven with compromise for Microsoft customers.