Why Google Gears Is Good News Bad News For Microsoft

Google Gears, Google's new JavaScript API to help Web apps work offline, is certainly an intruiging concept. As recognition that people aren't always online, it will also surely be paraded by Redmond as validation of Microsoft's vision that the future of software covers the desktop, the Internet, and particularly, a combination of the two. But it's not all gravy. Google Gears gives browser apps persistance, long the sole bailiwick of the desktop client.

Don't think the primacy of the browser is suddenly here. Out-of-browser-experiences, or OOBE's, as I will call them to be cute, are still more responsive than your typical AJAX app, bound by tricky code and a lack of access to hardware resources like graphics acceleration. The browser also constrains the look and feel of the user interface, hence Microsoft pushes Windows Presentation Foundation for Internet-connected apps. Plus, Google Gears could take away one of the browser's greatest benefits: the ability to run applications without installation headaches.

And yet, Google Gears was an inevitable development Microsoft has neglected. Offline capability for heretofore purely online apps, like we now find from Zoho to Salesforce.com, is a necessary thing. Microsoft has repeatedly brought that to my attention over the past few months, as they well should since software plus services is the company's future. Microsoft itself has had things like Office Online, which searches the Web for templates and returns them, all inside of Office apps like Word, for years. But they've all come from the direction of the desktop toward the Web, rather than the other direction.

The value of Windows is that it is a relatively exclusive and powerful sandbox. Well, that, and that it is the dominant operating system by a mile, but that's beside the point. The value of the browser-based Web, on the other hand, is lightweight openness, as Google demonstrated with Gears, itself a relatively small download that's cross-platform and cross-browser. How to reconcile those two worlds is a bigger predicament than I am able to unravel. Adobe's Apollo run-time made an attempt to do so, now so does Google Gears and sometime soon will Mozilla Firefox 3.0. I know I'm comparing offline (and/or out-of-browser) apples to online (and/or in-browser) oranges, but the middle ground is sparse. Where's Microsoft's platform entry?

Silverlight, Microsoft's new cross-platform, cross-browser Internet platform, actually carries some potential, but it has no native offline capabilities and isn't likely to start working with Google Gears anytime soon. And the Novell-run Mono project may actually beat Microsoft to the punch of making Silverlight work outside a browser. Even then, will it run offline? No. Will it be as open as Gears? I'd say no.

Needless to add, Microsoft is far from the kind of openness that would have had the company build Google Gears itself. There's as much, if not much more, downside to Microsoft putting out an API that would enable people to develop offline apps for competitive Web and desktop platforms as there is upside. Rock, meet hard place. Move too fast, you could cannibalize Windows and Microsoft's desktop apps business; move too slow, and get left behind in a new world.