Petreley's column: Down to the Wire

Here's how both Microsoft and Novell can avert their impending disasters.

Last week I finally jumped on the disaster bandwagon and predicted a stock crash caused by an earthquake on December 27, 2000, that damages areas where high-tech companies are concentrated. Many readers from all over the world wrote in to tell me they were fascinated by the prediction, and both of them wanted to know how I arrived at this date. I used a highly scientific technique called quarterly deterministic randometrics. Heads equaled 2000, tails equaled 2001. Heads equaled January through June, tails was July through December. And so on.

This week I'm going to predict a couple of minor disasters. Disaster No. 1: Novell will pass the point of no return and wither into a legacy-support company for the evaporating NetWare installed base. No 2.: Microsoft will make a valiant attempt to dominate home entertainment and electronic-commerce appliances but will ultimately fail and lose interest in the market.

But this time I'm prepared to offer some advice on how to avert these disasters.

Novell is in a pickle. Anyone who knows or cares about directory services seems to think that Novell Directory Services (NDS) is the ultimate solution. NDS has a distinct advantage over Active Directory: It is a relatively platform-neutral solution in a heterogeneous world.

Novell's problems are numerous. Few people understand directory services, and even fewer grasp how important they are. For those who do get the big picture, Novell is often off the radar screen. You can blame some of that on inadequate marketing.

Novell has to be willing to bet the company on its last chance to establish NDS as the foundation for Internet directory services. The company has to launch a marketing campaign to end all marketing campaigns with its most intense concentration on large Internet and application service providers. Novell should seed the market making NDS free to anyone who manages a directory with 50 users or fewer. Finally, Novell should open the source code to ride the Linux wave.

Even then I'm not confident that Novell can regain its former stature. Novell doesn't have a chance unless it takes one. A big one.

Now on to my second minor disaster. Microsoft will release a magnificent specification for a home computing appliance. As expected, it will include the groundbreaking Nvidia GeForce 256 3D graphics processor. After a brief flirtation with success, the Windows CE-based machine will flop. The effort will fail for two reasons: royalties and Windows 2000.

Sega is losing money despite strong sales of the Windows CE-based Dreamcast machine. Royalties may not be the pivotal factor, but I suspect Sega would be in the black if it didn't have to pay Microsoft for every unit out the door.

Royalties are also the reason why I predict Microsoft will enjoy initial success. Although I don't think that Redmond is unaware of its competition with free software, I do think that Microsoft will misjudge it. Here's how: In an attempt to persuade partners to eschew embedded versions of Unix for Windows CE, Microsoft will cut one or more deals so companies can ship a limited number of these units without paying royalties. But companies will have little reason to continue supporting the OS once the investment dries up.

Microsoft will make another very predictable mistake. It will design the software so the appliances require Windows 2000 at the server to provide one or more crucial e-commerce services. But the attempt to leverage one market to expand in another will fail because Microsoft hasn't figured out that it can't leverage a monopoly in a market where it doesn't have one.

The solution for Microsoft is simple. First, Microsoft should charge no royalties for Windows CE. It should plan instead to make its money by selling Windows CE development tools and support services. And, of course, Microsoft shouldn't try to wedge Windows 2000 into the deal. I stand by my July 27, 1998 prediction that Windows 2000 (then called Windows NT 5.0) will lose to Linux at the server. So tying the success of Windows CE to Windows 2000 or vice versa would be a mistake.

What would you do if you were responsible for making NDS the foundation for Internet directory services or making Windows CE the e-commerce platform of the future?

Nicholas Petreley is editorial director of LinuxWorld ( Reach him at, and visit his forum at

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

More about AvertDreamcastMicrosoftNDSNovellNvidiaSegaStock crash

Show Comments