Windows 2000 is here, and you don't care. And, says Microsoft, you probably shouldn't — unless you have a few major servers to worry about. Most industry analysts advise caution even then; yet oddly, both conventional wisdom and the Windows 2000 marketing materials seem to be telling average users the same thing: skip this one.
After Windows 2000, the next major release for the Windows platform will reportedly be known as Windows Millennium Edition, or by its cutesy acronym, Windows ME. The core of this OS will bear little similarity to Windows 2000, instead functioning more like Windows 98. For most users, this is probably the one you want.
Confusing? To the consumer, maybe — though historians and lexicographers may be pleased to note Microsoft's apparent acceptance that the new millennium doesn't begin in the year 2000.
But then, Windows 2000 isn't really targeted at consumers. Windows ME will be the consumer OS, sharing the lineage of Windows 98 dating back to the days of the MS-DOS command prompt. Windows 2000, on the other hand, is actually the long-awaited successor to Microsoft's server OS, Windows NT 4.0.
The numbering confusion arose back when someone at Microsoft had the bright idea to unify Windows NT and Windows 95 into a single operating system, marketable to every computer user. It was a good plan. Apple had the same idea, when it recently abandoned its dual-OS strategy and repositioned the upcoming Mac OS X as the sole solution for Macintosh computers. Unfortunately, it was also a plan that Microsoft ultimately couldn't pull off.
For one thing, Windows 2000 was hopelessly behind schedule, and had been for some time. A new Windows NT release hadn't appeared since 1996, and new code builds were notoriously plagued with bugs. As late as December, Microsoft was claiming the product would ship before the start of the New Year. Its actual release date was February 17.
In September, the marketing boys in Redmond pushed to change the name of the new product from Windows NT 5.0 to Windows 2000, in anticipation of a unified Windows platform. But by then, to the engineers the writing was already on the wall: they weren't going to make it.
There was too much development left to be done, and already too many bugs to squash. The decision was made to abandon hopes that Windows 2000 would be an upgrade for Windows 98 users as well as Windows NT users, and the development teams refocused on making Windows 2000 foremost a world-class server OS.
But why did they give up on their more ambitious goals? Windows NT 4.0 had held up Microsoft's enterprise strategy successfully for four years — what would another six months matter? Or even another year? But then, if you were the world's leading software vendor, and a rival operating system — one that was readily available and charged no licensing fees — appeared out of nowhere to challenge your flagship product, that might have a way of lighting a match under your ass.
For Microsoft, though they don't like to admit it, that challenger is Linux. While analysts estimate Windows NT owns about a 32 percent share of the server market, Linux has rocketed into second place in recent months, capturing about 24 percent.
Its growth is incredible, especially compared to more venerable Unix-based OS offerings from companies like Sun Microsystems, Silicon Graphics and IBM, each of which holds a loyal, but more modest market share. And this is without any marketing campaign to speak of, very little bundling with PC systems and certainly no activities that would attract the attention of the Department of Justice.
To answer the threat, Microsoft needed a new OS product aimed at the server market, one that would grab headlines away from Linux and rekindle brand loyalty amongst those IT managers who might be considering straying from Redmond's fold. Windows 2000 is meant to be that product. But the reaction from the industry so far has hardly been uproarious.
Sure, Windows 2000 was kicked off in San Francisco with an event of true deep-pockets Microsoft proportion. Bill Gates addressed the crowd at the Bill Graham Civic Auditorium with a mixture of marketing hype and flashy, staged demonstrations of the new OS.
Actor John O'Hurley of Seinfeld participated in a skit that took place in a mocked-up set of an airplane cabin. Carlos Santana made an appearance. Even Star Trek's Patrick Stewart was on hand to crack wise about "enterprise" computing. It was an event worthy of a Fox TV special.
But very little of the buzz seemed to carry to the floor of the Windows 2000 Expo at Moscone Center. While some booths touted their owners as "Windows 2000 Experts" — I was tempted to ask them how long they'd held that distinction — few seemed truly excited about the new release.
At least one vendor demonstrated their products for Windows NT 4.0, promising a Windows 2000-compatible release sometime later this year. To be fair, most software developers probably hadn't had enough hands-on experience with Beta releases of Windows 2000 to have products ready by the time the OS launched. But even many IT managers are skeptical about deploying the new release.
The most basic version of Windows 2000 contains some 29 million lines of code, and while it's meant to be compatible with NT 4.0, much of the older OS has been completely rewritten. To the people responsible for deploying NT for mission-critical business servers, that's a red flag.
NT 4.0 itself is on its sixth maintenance "service pack," and most new software for NT requires that administrators install at minimum Service Pack 3. For many, a "bleeding edge" Microsoft OS release with this much new code is completely off limits; they'll wait until the first few rounds of bug fixes appear, hopefully by this summer.
Another challenge the new OS faces is the traditional reluctance of Fortune 500 IT managers to upgrade their operating system unless it's absolutely necessary. Unlike the fast-paced, super-hyped consumer market, with so much riding on their servers corporate managers typically subscribe to the "if it ain't broke, don't fix it" philosophy. For example, as Sun Microsystems rolls out version 8 of its Solaris operating system, many new servers are still deployed on the older Solaris 2.6.
So the question remains, what will be the fate of Windows 2000? The average home Windows user won't want it. And the corporate users seem to be shying away from it. Chances are, the new release will become more widely adopted as the bugs get ironed out and IT departments become convinced of its reliability.
But will that happen before Microsoft realizes its unified OS strategy? Or will Windows 2000 become something like Intel's Pentium Pro chip, relegated to a niche market and then largely forgotten, once the Pentium II was released?
Any way you slice it, the next stretch of the road for Windows is going to be a difficult one. If, as Microsoft claims, Windows 2000 is a "make or break" product for the company, then it's only bound to add more troubles to what has already been a tough time for the software company you love to hate.
