According to a ScriptLogic study, 60 percent of all companies surveyed said that they will not be moving to Windows 7 any time soon. Thirty-four percent said they'd probably deploy by the end of 2010, but even that number may be optimistic. This means that by 2011, for the first time ever, a 10-year-old operating system will still be the most-used desktop OS.
Of course, Microsoft's licensing means that this unfortunate fact won't cut too deeply into the company's bottom line. While Microsoft's OS market may be stagnating, hardware is hardware, and it will fail and need replacing. That's when they'll manage to sell you yet another license that can be downgraded to XP.
[ Randall C. Kennedy says Windows 7 is shaping up quite nicely. | Preparing for Windows 7? Get the overview you need in the Windows 7 PDF Report from InfoWorld's J. Peter Bruzzese. | Can your PCs run Windows 7? Find out with the compatibility checker in InfoWorld's free Windows Sentinel monitoring tool. ]
As the recent support extension for XP shows, Microsoft does see that users aren't falling all over themselves to upgrade to Windows 7, just as they weren't for Vista. The fact that many seem to hail Windows 7 as a far better OS than Vista doesn't really make a difference -- the real problem isn't that Vista or Windows 7 aren't ready for the enterprise, it's that for the vast majority of business cases, Microsoft XP with Microsoft Office 2000 is all that's necessary -- possibly for quite some time.
After all, why do you think that Office 2007 had a massive UI change? Because that was one of the only ways to differentiate it from Office 2003. The back-end stuff, like support for the OpenDocument Format, could have been added to Office 2003 as it was to Office 2007. Office 2007 was basically a "New and improved!" sticker on Office 2003.
As far as business desktop computing goes, that's a novel idea. For the past 15 years, companies have been upgrading constantly, moving from Windows NT 4 to Windows Server 2000 to Windows Server 2003 or, on the desktop, from Windows 95 to 98 to 2000 to XP. And that's where they sit.
As a consultant during those fiery days, it was upgrade or die, and I was on the front lines -- the inherent problems in NT and 2000, and with Windows 95, 98, and 2000, made yearly upgrades essentially a requirement. If I had a dollar for every Windows NT-to-2000 migration, or Windows 2000-to-2003 migration I ever did -- actually, I have more than a dollar for each one. Never mind.
The reality is that in any industry that grows as fast as business computing has, there will come a point of Good Enough. That's where we are right now. The vast majority of ISV applications in use support XP and still don't officially support Vista. Nine-year-old XP is still the sweet spot.
I recently spoke with an IT manager who was budgeting for an Office 2010 upgrade from Office 2003. I casually asked him what features he had deemed important enough to justify a $100,000 budget item. He thought for a minute and admitted that he couldn't think of a single one. So I asked the logical follow-up: Why are you buying it? He had no answer for that either. The $100,000 line item disappeared. He's also sticking with XP.
This isn't Microsoft's fault, necessarily, unless you believe that they caused this by finally coming up with a somewhat stable and secure desktop OS (that statement includes a huge grain of salt). The company certainly did delay far too long in releasing Vista, which was late, slow, buggy, expensive, and essentially DOA, and then compounded the issue by hyping Windows 7 a year before its release. This provided some cover for the Vista debacle, but also ensured that several more years would pass before most companies would move beyond XP -- a modern-day example of the Osborne Effect. The economic downturn just solidified this situation.
The past 15 years have been a whirlwind of innovation, expansion, invention, and production. The next 15 will be the same -- but not in the same places. The corporate desktop is mature in both hardware and software. Same for servers and network architecture -- very few companies actually need 10G. The new frontiers are portable productivity, interconnectivity, and virtualization. Unfortunately for Microsoft, they're far behind in those categories. After all, it's hard to quickly move in new directions when your saddlebags are full.