It was a preternaturally warm day in February. I think it was Pogo who said, "What is so rare as a warm day in June?" to which I'd reply, "When it comes in February." Of course the heating system still knew it was winter and continued to pump hot air into the computer center offices. Shirt sleeves rapidly appeared, then rolled shirt sleeves. Calls to the maintenance people produced nothing but anguish on all sides. While we waited for computers to melt, we decided to haul out the oldest server on the bench and install Windows 2000.

We called in a student intern, handed him a box of Windows 2000 Server, and said, "Go to it." The intern turned to go with a disgusted look on his face and said, "Windows 2000, yuck!" It was an interesting reply. I know for a fact that he had never installed, used, or even seen Windows 2000. My client, the intern's boss, is only slightly more experienced with Windows 2000, since he's seen it running on a computer (mine). When I first made the suggestion to install the newly released Windows 2000, his reaction was to make a face as though he had sucked on an old lemon. I do not consider these folks to be representative. I do think it's fair to say they represent shades and hues of the anti-Microsoft attitudes that are easy enough to find these days among the computer literati.

In most of the previous two decades, the appearance of a new PC operating system was a big deal. Stimulated by a full-court media press, people eagerly awaited the products. I remember the introduction of Windows 95, when we saw network television coverage of people in New Zealand waiting for the doors of a computer store to open. As I've often written, things are different now. The Windows 2000 event, arguably one of the most significant software releases in several years, occurred without so much as a fluttering heartbeat even in the computer industry.

Of course, Windows 2000 barely qualifies as a new product. It's been in widespread beta testing for several years, and Microsoft even sold a beta version. It's been delayed so many times that all anticipatory energy has long since bled away.

That's a pity. Windows 2000 is the operating system Microsoft promoted for nearly 20 years: Elegant graphical user interface, (nearly) bulletproof, fully featured, (relatively) fast, secure and graceful. Yes, graceful. Using Windows 2000 can be as smooth and comfortable as the best Apple Macintosh system, but covers a lot more territory with compatibility and functionality. This is a heckuva piece of work, perhaps a once-in-a-decade accomplishment.

Yet, reading that paragraph, how much do you believe? Scarcely a mind out there didn't also say (at least), "Yeah, but ... " At the top of the "buts" list would be bugs. Aren't Microsoft products always buggy? Didn't somebody say they admitted to 63,000 bugs in this first release of Windows 2000? I'm sure this issue is enough to drive the people at Microsoft, well, buggy. "Of course there are bugs in Windows 2000," they would probably say. "There are 40 million lines of code. But what's a bug, anyway? That so-called list of bugs was really a compilation of things to change, anomalous behavior, and yes, some real bugs. That does not mean Windows 2000 is not reliable."

Ironically, stability is one of the hallmarks of Windows 2000. Based on trial experience and the testing done by independent labs, this is by far the most stable operating system Microsoft has ever produced. Windows 2000 Server edition (along with the two upscale versions, Advanced Server and Datacenter) will give Unix and even Linux a run for their money in that department.

Of course, there are problems in Windows 2000 that people won't like. One problem certainly will be incompatibility with hardware and software. The odds are fairly good that computers older than three years will have some piece of hardware that doesn't plug and play with Windows 2000. The question is, how responsible should Microsoft be for this kind of problem? One of the main reasons it took so long to produce Windows 2000 was the necessity of making it compatible with a lot of important hardware and software. At some point, however, Microsoft had to pull the plug on compatibility. Some things work, some don't. They publish a list of what is known to work. End of story?

There's always security to flare about. Windows 2000 has so many major components, no wonder there are areas in and between these components that beg for cracking. I won't be surprised if there is a steady stream of discoveries concerning flaws in Windows 2000 security--this despite the fact that Windows 2000 implements Kerberos encryption and numerous other security features that make it much more secure than Windows NT 4.0.

I guess Windows 2000 must fit the bell curve of notoriety: 10 percent will think it's just great, 10 percent will think it stinks, and most people will fall in the middle. Still, I expect harsh criticism and gushing praise for Windows 2000, and I expect the media to dutifully record both extremes while largely ignoring the cautious middle. In a product this big, there will be some truth in most claims.

Perhaps the most interesting commentary on Windows 2000 is that it's old-fashioned. In this line of punditry, Windows 2000 represents the high-water mark of the monolithic (do-it-all) operating system. In a world of small Internet devices and handheld computers, it's the last big operating system--maybe forever. I'm not sure what's really significant about this commentary. Microsoft has long recognized that operating systems of the future need to be composed of numerous relatively independent components that can be reconfigured quickly and effectively.

I wouldn't recommend being fooled by the "what's new and hot" form of hype, like the Compaq commercial that proclaims: "Desktop computers ... that is so last century." If I were a gambler or a stockbroker, I'd keep my eye on the long run. Ten years from now, Windows 2000 (or whatever it's called then) will still be a staple of the computer industry. It will be by default since the growth of computer sales--still largely for the desktop--ensures it. Even if desktop and full-sized notebook computers represent a declining portion of the computer market, a very large number of Windows users will still be running them.

Windows 2000 is a good operating system in its own right, and Microsoft knows how to market it. Its server versions will eventually be accepted by the large corporations in ways they have, up to now, been reluctant to embrace--especially for large-scale enterprise server-based applications. For individuals using personal computers, Windows 2000 Professional has enough inherent value to make it worth the cost of software and additional hardware. Eventually the numbers for Windows 2000 will be very big, as that large group of people in the middle decides it's not a bad thing--even if it comes from Microsoft.

I suppose my conclusion that Windows 2000 is a good product that will succeed quite well for a number of years will not sit well with some people. It sounds like some sort of apologia for Microsoft. Can't be helped. I happen to believe that in this case, bias against Microsoft--even one based on its track record in software--shouldn't be the sole influence in evaluating Windows 2000. It would be a mistake. The best thing I can do is to encourage you--as corporate representative or individual--to evaluate Windows 2000. Read the scads of (mostly good) analysis of the product; try it if possible, and draw your own conclusions. It may be stretching a metaphor, but I think you'll find Windows 2000 like the unprecedented warm winter: An exceptionally good thing while it happened; and in the long run, like global warming, perhaps a sign of big changes.

Editor at large Nelson King is ComputerUser's leading industry analyst.