I really enjoy analysts' predictions, especially when they foretell a major shift in the market. It's not that often a major analyst firm like Gartner steps out of the "$xx billion in 5 years" mold and actually puts it on the line with a whopper statement. (Don't worry Richard, I won't bring up the "death of IPS" prediction.) Gartner's Michael Silver and Neil MacDonald have done just that, calling Microsoft's gluttonous collection of legacy code untenable, expecting Windows to cave in on itself by 2011. The "comet" striking Microsoft is the rise of the web application to dominance. Most of their argument rests on the failures of Vista and the rise of the OS-agnostic web app. I can't say that I disagree with either arguments, but they forgot to predict how Ray Ozzie and his mesh strategy would either make or break Microsoft. That's the real question, isn't it? I think so.
Will Microsoft be able to establish critical mass in the Live world of web apps and services? Will Microsoft be able to keep pace with the likes of Google, Salesforce, Amazon and the growing number of SaaS application ISVs and cloud infrastructure providers? Can Microsoft Mesh, Microsoft Sync Framework and Silverlight become mature enough to take hold as a viable platform to attract web applications away from Linux, Java, Ruby, Python and open source based technologies? Can Microsoft succeed in its extreme corporate makeover? It's a very tall order to ask of Steve Ballmer and Ray Ozzie. A VERY tall order.
Microsoft's task at hand is the equivalent of IBM remaking itself from the dominant mainframe hardware company into a global services company and Linux advocate. Hey wait a minute... IBM did just that. Sure, there are still mainframes around. You know the saying - old Cobol programmers never die, they've got to stick around to fix the next compiler syntax error. (You just had a flashback of spelling DATA DIVISION wrong again didn't you.) IBM isn't the 800 pound monopoly they once were but they certainly continue as a going concern and a major industry player.
VIsta's failures may have been the division bell that woke Microsoft up to a new set of software realities, but Microsoft has yet to learn some very hard lessons - lessons they don't appear to have glommed onto yet. In the world of On Demand software, user experience is king. If software is confusing, hard to use or just plain user hostile, the user can just as easily move on as put up with lousy software - and they will. That means no more Office 2007 ribbon menu fiasco crap, or "redesigning" Windows features by hiding them behind some new GUI widget. Software must be easy to learn and use. Better yet, software's got to be a pleasure to use.
It also means no more bloatware. Outlook is only second to Apple iTunes in the bloatware winner category. It's huge, kludgy, slow to respond and loves to sport that all too familiar "Not Responding" message. That won't cut it in a browser window running over a network with virtualized components and data synchronization everywhere. The bloatware thinking's gotta go. And lastly, Microsoft needs to learn that it doesn't have to have aproduct in every conceivable software category. How about focusing on fewer areas and doing them exceptionally well.
Does Ozzie have the clout, stamina and moxie to pull off this kind of major shift? Is Ballmer fully behind it? Those are questions I'd like to hear Gartner and others lay a prediction on.