|
BUSINESSWEEK ONLINE: DAILY BRIEFING | |||||||||||
| |||||||||||
|
| |||||||||||
What Windows 2000 May Mean for Investors It'll affect nearly every computer-industry player as corporate customers gradually adopt it
By the end of this year, if all goes according to plan, Microsoft (MSFT) will launch its new operating system, called Windows 2000, and another chapter will open in the software king's history. With that event only months away, it's time for investors to join computer-industry savants in trying to piece together its implications. But it's not only Microsoft shareholders who should pay close attention: Over the next two years, nearly every technology stock, from Internet startups to motherboard makers to game designers will be affected by the new OS. For that matter, most companies in the world will eventually end up using Windows 2000 -- or a competitor that by necessity will have been heavily influenced by it.
The most notable thing about Windows 2000, though, may be its heft. It may be the largest consumer-market computer program ever written, with 30 million lines of code. More than 4,000 Microsoft programmers are designing it, and thousands of beta-testers are running the latest iterations and stress-testing the OS to ferret out as many bugs as possible. It is, declares one Microsoft executive, "the most important program ever written." Of course, until Windows 2000 is released, it won't be clear whether it's the next Corvette or the next Edsel. Even then, you won't be able to tell right away: Most corporations may wait a year or two before they upgrade. In fact, no major company is likely to mess with it until Y2K D-day has come and gone. So if you can't count the number of shrink-wrapped boxes to tell if Windows 2000 is a success, what should you watch for? As its release gets closer, here are some questions to track: When will the baby actually be delivered? Among the many brilliant ideas that have sustained Microsoft is vaporware. That's when you stifle an opponent's new-product release by leaking news that you're about to release a similar product that's even better. Maybe you never release it, or do so only years later. What's important is that you've killed the buzz about your competitor's product. The problem with that strategy, of course, is that it can lead to the Boy-Who-Cried-Wolf syndrome. By no means is Windows 2000 vaporware. But promised release dates have come and gone so many times that it's hard to know when it'll be ready. Right now, Microsoft says that the program will be launched before the end of 1999. That's a believable date. But if it slips, the product's reputation will begin to suffer while it's still in the womb. Has Microsoft solved the "blue screen of death"? Nothing bothers information technology professionals more than the tendency of Microsoft products to freeze or crash. Ed Muth, group product manager for Windows 2000, claims that the most recent iteration of its NT operating system is crash-proof in 99.9% of computing situations. (Feel free to be skeptical...just ask your friendly network administrator.) A major goal of Windows 2000's designers was to make it more robust, which is computer-speak for a program that keeps running through thick and thin. If they've succeeded in making Windows 2000 as robust as some Unix operating systems -- which power most Web servers today -- it would be a major leap for Microsoft. So far, the signs aren't all that encouraging. One computer consultant for a company testing a beta version of Windows 2000 -- who wishes to remain anonymous -- says that it, too, frequently crashes. "We spent most of our time rebooting the whole system, even more so than with NT 4." To be fair, though, every computer program is unstable during beta testing. How large a network can Windows 2000 handle? One of the biggest marketing problems that NT faced after its launch was that it was a great OS for servers with a limited number of processors, usually less than 16, but not so great when you try to run it on truly industrial-strength servers that run giant corporate networks. In a phrase, it wasn't scalable. Probably Microsoft's top goal with Windows 2000 is to make it scalable. Still, Muth doesn't pretend that it's for everyone: "We understand that a lot of enterprise-wide systems are going to need a different operating system, so we designed it to be interoperable with other operating systems in that area." But Windows 2000 is also designed to run on much larger machines than NT 4 can manage. In fact, it should be able to handle servers with 32 processors. If Windows 2000 is truly scalable, it will have overcome one of the biggest problems that hurt its predecessor. Is the dark horse Linux really catching up? One of the amazing things about Windows 2000 is that it has no established competitor. Apple (AAPL) doesn't have a strong entry in the corporate server OS market. Most other server operating systems are variations of Unix, but that family has become so fragmented that it's not really fair to call it a single system anymore. Of course, the Unix version that has gained so much attention lately is Linux. Partly because of its robust performance and partly because it's free, Linux has quickly gained a foothold in the server OS market, with a 17% share, according to International Data Corp. Linux is also a favorite among techies because its so-called source code is open to anyone and everyone, which makes it easy to fix when problems occur or modify to meet specific needs. Will Linux blow out the candles on Windows 2000's birthday cake? Right now, it's hard to say. Linux has grown quickly, mostly at the expense of other Unix operating systems, but it's a far cry from the feature-rich Windows 2000. One promising sign for Linux proponents is how quickly it seems to be able to adapt. Just six months ago it had no widely accepted graphical user interface, but it now has two very popular ones, KDE and Gnome, both of which have quickly evolved into workable computing environments. It's even close to gaining an important technical component, called a journal filing system. The JFS, which helps reassemble a program after the system crashes, is one of the key components of Windows. Linux' JFS, donated by Silicon Graphics (SGI), will probably become an integral part of that operating system soon. If Linux can make changes like that to its core structure in such a short time, then it could become a legitimate competitor to Windows 2000. Its greatest advantage is that it has tens of thousands of programmers throughout the world who can adjust and improve it in their free time. Poor Microsoft, by contrast, can afford only 4,000 programmers to work on the code for Windows 2000. Jaffe writes about the markets for Business Week Online _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ BACK TO TOP |
Assistive Technology barker.online Byte of the Apple Eye on Japan Hers.online Inside Wall Street Not-So-Neutral Corner Online Asia Power Lunch Privacy Matters Sector Scope Sound Money Street Wise Washington Watch News Flash Archive | ||||||||||