Two developments threaten Microsoft's position in the post-PC, Internet-centric world. The first threat is to Microsoft's core product: the Windows OS. Upstart OS Linux, whose hardware compatibility list (HCL) fit on the back of a business card just 5 years ago, is now Windows NT's number-one server competitor and a potential competitor in the desktop space.

The second threat comes from the US government, which began investigating Microsoft as long ago as January 1990, when a joint IBM/Microsoft announcement about Windows and OS/2 drew the attention of federal regulators at the Federal Trade Commission (FTC). Microsoft's relationship with IBM quickly fell apart, but a short FTC investigation turned up numerous complaints against Microsoft, and the commission spent the better part of 4 years trying to bring charges against the company. When those efforts deadlocked, the Department of Justice (DOJ) got involved. In 1995, the DOJ forced Microsoft to sign a consent decree designed to limit its questionable business practices. When the consent decree failed because of lenient wording in the agreement, the DOJ and 19 US states sued Microsoft for Sherman Act antitrust violations, winning a guilty verdict 2 years later. That case, of course, is just entering the appeals process. But the antitrust verdict threatens to break up Microsoft or—at the very least—seriously curb its business practices as the company develops its next-generation Web services strategy. The legal problems come at a critical time for the company, which is increasingly vulnerable to competitors as it attempts its next great Internet embrace.

Given this recent history, you might assume that Microsoft is in trouble. But I don't think it is. Although Linux and the federal government are credible threats that must be dealt with, Microsoft will come out just fine, thank you very much. And it almost doesn't matter what the company does—because Microsoft will overcome its foes without doing much to make it happen.

First up is Linux, which International Data Corporation (IDC) says captured 24 percent of the server market in 1999, a dramatic increase over the previous year. Of course the increase was dramatic: It's easy to make big gains when you're starting from scratch. Linux advocates note that the market share for NT remained steady at 38 percent last year, the first time its share didn't rise. But does that figure really represent a loss for Windows, or does it simply indicate market maturation? Consider that Linux's gains came almost solely at the expense of Netware, which slipped to 19 percent of the market in the same time period. What Linux backers won't tell you is that IDC, which produced these figures, believes that Windows 2000 (the next version of NT), not Linux, will dominate the server market by 2003. IDC's view is based on factors such as Microsoft's awesome support infrastructure, the company's billions of dollars in annual sales, and its clear market strategy. In 1999, worldwide Linux revenues totaled just $63 million. In that year, Microsoft revenues on its USB mice products alone were more than that.

Sorry, but the Linux bubble has burst. Now that the IPO heyday has passed and Linux companies will actually have to make money, we will see a dramatic consolidation of the Linux market, a consolidation that is already taking place among equally overvalued Internet companies. Linux companies lack a cohesive strategy to beat Windows; they share just a desire to be anything but Microsoft. And that's not enough. Attempts to duplicate Microsoft's success—Red Hat Linux tried, for example, as did Corel and Inprise when the two companies tried (and failed) to merge—demonstrate that these companies just don't get it: No one is going to out-Microsoft Microsoft. And it's not enough for Linux to be like Windows. Linux needs to be different from Windows. It needs to be better than Windows. And I don't see that happening in a market that seems hell-bent on duplicating features rather than innovating and discovering what real customers with real dollars want. Like Netscape before it, Linux will simply fail to surpass Windows on its own, without any interference from Redmond. The Linux companies will try, of course, but their attempts will only widen the gap between Win2K and Linux.

As for the desktop, forget it. Just this week, the Linux community was abuzz over screenshots of a prototype file manager program that looks sickly next to Windows' 5-year-old Explorer. Linux has a long way to go to catch up with Windows 95's ease of use, let alone Win2K's riches and the advances Microsoft is making with Whistler. Windows is a moving target, and it's going to take more than "almost as good" for the Linux desktop experience to match Windows.