In Microsoft's battle against Linux, Project Green deserves the green light

Redmond should stick to its strengths: strong ties between its popular suites and its OS

It's strange. For years, geeks, sys admins, and nerds in general have been praying for someone to come along and knock Microsoft off its near-monopoly perch. I've been one of them: The healthiest thing that could happen to Windows is competition. But now that it may actually be happening, I find myself strangely concerned about Microsoft's reactions. Redmond is just going about the Linux war all wrong.

First the company announces what sounds very much like its own version of Beowulf . It's called Windows Server 2003 HPC (High Performance Computing), which is intended to provide high-performance computing across multiple small servers and CPUs. Sound familiar? Given all the anti-Linux hype we're hearing out of Redmond these days, you've got to wonder why this rather sudden announcement and why now?

Worse, Microsoft seems to be doing it at the expense of its application business. We've heard of Project Green, which is Microsoft's ongoing development effort to take the disparate CRM, HR, and finance applications it acquired from Great Plains and Navision and move that functionality (presumably along with all its current customers) to an entirely new, proprietary, and unified code base.

But it looks like Redmond is having a bit of a reaction to the sudden positive light that Linux has garnered in government sectors. Microsoft has re-assigned roughly 130 developers from Project Green's stable of 200 to work on "core product lines." And the results of Project Green, which were originally supposed to be viewable by the end of this year, now have been bumped out to a probable ship date of 2008. Since the Office development pool is pretty much full and most of Microsoft's new announcements have been OS-related, we can guess where these reassigned coders are heading.

But why? Think about it for a minute and it's easy to see that the biggest hurdle in knocking Microsoft off the OS supremacy platform isn't really a push into OS technology. If that were the case, Linux would have been kicking Redmond booty for the last four years. Compared solely on features, Windows and Linux have some differences, but for the vast mass of desktop users, they provide all the essential tools any typical user needs except one: Linux can't run Office, which has so far kept Linux in the magenta minor-threat category back in Microsoft's war room.

But with CodeWeavers's implementation of Wine, CrossOver Office, that perception has changed -- fast. You still can't run all your Windows applications on the Linux desktop, but you get the important ones, namely your favorite MS Office productivity applications. Then there's OpenOffice , which is a solid piece of coding, but it's still a bit behind Microsoft Office in terms of intuitive user interface, polish, and even specific feature sets, especially Excel. Plus, the bulk of America's office work force has no desire to relearn a new productivity suite -- and most business managers have no desire to pay for that training either. That's a hurdle that OpenOffice needs to cross in the long term.

No, the Linux community's best immediate bet is to ensure Office compatibility on its desktop. Now users can run any productivity suite they like, imparting that safe and fuzzy feeling with their Linux desktop choice. But that means Microsoft's best weapon isn't some new high-performance Beowulf bombshell but precisely the kind of integrated and advanced functionality that Project Green was going to give them: all the advanced stuff you can't get out of a standard productivity suite neatly tied to a new OS platform.

Let's face it: Whether or not Linux is better technology than Windows, Microsoft has one big advantage over the Linux community: more than a decade of experience in what corporate users want. I mean specific experience, such as business processes and UI designs, as well as reams of corporate, ISV, and consultant input. For corporate customers, the OS is just a platform; it's the application tools they're buying. OpenOffice or not, Linux still has a ways to go in that category, and Microsoft should be exploiting that weakness, not trying to go head-to-head on operating system capabilities -- a fight that many argue they've already lost.

Of course, now I feel like I just gave Russia tips on how to beat the U.S. hockey team.

Copyright © 2004 IDG Communications, Inc.

How to choose a low-code development platform