There are a few ways to see Microsoft's plans for a free web-based version of Office. One, it's self-competition; two, it's competition with open source software; three, it's competition with other web services. Which one matters most?

The first one's not hard to understand. Why would someone pay for a full-blown version of Office when they can get a sizeable chunk of the functionality for free in a web browser? They could, but the web browser is still a pretty lousy application-delivery platform. My own experience has been that when people have a choice between a "fat" local client (which may not be all that "fat", actually) and an in-browser application, they go for the local app for the sake of not being constrained by the browser's quirks and modalities. New browsers don't so much solve the problems as push them off into a corner.

That brings me to item #2: OpenOffice (dot org). Rather than compete with OpenOffice by releasing a slimmed-down free version of Office -- say, just Word, without the dictionaries or spell checker / grammar tools and maybe some of the other pro-level, server-needed features -- Microsoft is providing hosted apps. And here, again, I'd rather choose the local client than the web-based one. There's a lot about OpenOffice that I don't like, but I'd rather put up with those limitations than the ones I get in a browser-hosted app.

#3 is fairly obvious: with stuff like Zoho and Google Docs popping up, Microsoft's obliged to compete and not lose too much market share to them.

As you can guess, it's the degree of competition from OpenOffice that I'm most concerned with. My own experience has been that I was able to replace most, but not all, of Office 2007 with OpenOffice. I still use Word and Outlook; OpenOffice Writer still lags badly behind Word in a lot of respects and doesn't show any signs of catching up soon, and there is no replacement for Outlook in that suite anyway. But I haven't bothered with Excel or PowerPoint or all the rest of them, because the OO editions of same more than cover my needs. It's the stuff specifically in Word and Outlook -- where I spent a lot of my workday -- that I paid to have access to, like a robust grammar checker.

This, then, is how I see Office losing to OO or other such programs: not in big strides, but in tiny nibbles. A free web-based version of Office won't stave off that kind of lossage, since free-on-the-web simply isn't the same as free-on-the-desktop by any measure.

But the door revolves in both directions. OO stands to lose users to for-pay programs if the price is reasonable enough and the prospect of paying for things you can't have at any price in OO (like a decent grammar checker) is strong enough. Microsoft's not taking all this lying down; there's no reason OpenOffice should assume just delivering free is enough. They need to market themselves as intelligently and broadly, by providing the core for free and asking a modest price for a selection of things that truly matter to people. I don't think they want to become the WordPerfect of 2010, but that's where they're headed with their current if-you-build-it-they-will-come approach.

InformationWeek Analytics has published an independent analysis of the next-generation Web applications. Download the report here (registration required).

Follow me and the rest of InformationWeek on Twitter.