Questions about the future monopolise predictions for 1998

I recently put together a list of predictions for the new year and beyond for an upcoming column. I found I have more questions than predictions. I have never felt more sympathy for IT decision-makers than now, because the future of critical products is so uncertain.

First, I wondered about the consequences of the battle between Netscape Navigator and Microsoft Internet Explorer. Nowhere was this conflict more evident than on the floor at the Internet World show in New York. A number of vendors were showing tools that allow you to create animated, interactive Web pages without doing any HTML editing or programming. These pages end up using a combination of Dynamic HTML, JavaScript, JScript, VBScript, Java, JavaBeans and other bags of tricks.

During each demonstration, the presenters repeatedly pointed out: "if you had to code this page manually, you'd have to design the page one way for Navigator, and another way for Internet Explorer."

It is a marvellous thing that these programs can insulate you from the differences between Explorer and Navigator. But I have to wonder: how long will the market tolerate two major sets of HTML, scripting, and Java standards in Web browsers?

I can't blame either company for reaching beyond the available standards to provide more competitive features. But the differences create a dilemma for Web design, and it's a problem that shows no signs of going away. I sincerely doubt either company will acquiesce to the other in the coming year, if ever.

I also wondered how anyone can realistically predict or plan any meaningful IT development strategies regarding Windows NT or Windows 98, other than just choosing to upgrade. This uncertainty results from the two priorities that drive the evolution of Windows and override all other considerations.

Priority number 1. Microsoft must always look out for any emerging threat to its desktop monopoly and be prepared to adapt its Windows strategy to squash that threat.

That gives rise to the moving target called "The Next Version". You wanted an object file system? Multiuser capabilities? A scalable, manageable directory system similar to Novell Directory Services? Telnet-based remote administration? A shell scripting language with open APIs for easy network management?

Sorry, Microsoft programmers will get to those features right after they are finished putting the final touches on their plans to eliminate Navigator, Java, network computers, and Unix.

Priority number 2. Microsoft must keep the US Department of Justice off its back while it attends to priority number 1.

This means Microsoft must continually repackage, rename, redirect and restate its Windows 98 and Windows NT product strategies in ways that will keep it out of trouble. One day, Windows 95 and Windows NT are parallel products. The next day, they're merging technologies.

Whichever it is today, and whatever gets bundled tomorrow depends more on the legal loophole that will allow Microsoft to attend to Priority number 1 than on what the customer needs.

This maze of twisty passages for Windows hasn't ended yet, nor does it look as if it ever will. The rise of the Internet, the World Wide Web, and network computing (generically speaking) are opening new doors to potential threats to Microsoft's monopoly. And each one could potentially redirect Windows and distract Microsoft from delivering that one feature you were counting on.

Finally, I wonder if Web servers are evolving into application servers, and what that means for the future. This question is relevant regardless of which technology dominates Web development.

Help me with these questions.

Are you planning to build intranets that provide application services? Why have you chosen that strategy? How do you make strategic plans regarding promised features of Windows 98 and Windows NT?

And if maintaining a public Web site is or will be a crucial part of your business, how are you affected by the differences between Navigator and Internet Explorer?

Former consultant and programmer, Nicholas Petreley may be reached at