What will be the dominant user interface of the corporate intranet -- the Windows operating system or a light-footed browser such as Netscape's Navigator? Neither.
IS can't independently select a user interface for the intranet because the charge toward intranets is being led by end users, not IS. IS is scrambling to get in front of the intranet parade, just as it did with the desktop revolution. But this time, IS managers must acknowledge that the marchers know their own minds.
Yes, that's right: minds, plural. As the computing revolution continues, end users are gaining increasing sophistication at the top, even as their numbers continue to spread out rapidly at the base. In the long run, this mix of users will require a desktop environment that can address many skill levels. But the shape of that environment isn't visible yet. In the short run, the challenge to IS is to leverage the existing infrastructure to get users quickly on the intranet.
More likely than not, that means Windows will be a part of it. And by the time end users have been weaned from their Windows applications, who knows what a desktop operating system, or for that matter, a browser, will look like?
When Delphi Consulting Group in Cambridge, Massachusetts, surveyed 400 organisations about their intranet plans, it found that 20% expected Microsoft to provide the dominant intranet environment and that 10% expected Netscape to set the standard.
Seventy percent expected some kind of Web browser to constitute the user interface but were unwilling to say that either Microsoft or Netscape would dominate. The IS challenge, then, isn't to focus on Microsoft versus Netscape but to build an infrastructure that delivers intranet computing.
Windows critics argue that it's time to execute a paradigm shift to network computing and dump the desktop operating system. If you're an IS manager, don't tell that to end users, not if you want to keep your head.
Users are saying, "Give us Internet-style capabilities." They aren't demanding that their Windows applications be purged. Microsoft, in its own way, understands what's needed and is hell-bent on giving end users the metaphor -- inside the Windows operating system. That approach will work in the short run, given the status of Windows on desktops. But it may not be viable in the long run.
I see Windows 3.x, Windows 95 and Windows NT Workstation merging into a high-end user environment with many capabilities for taking advantage of the intranet/Internet. Part of this user environment will be sophisticated applications that put mainframe-style database, financial analysis and document management on the desktop.
Organisations with many sophisticated knowledge workers will be heavy Windows NT users. But many end users won't need that much power. They'll need only a few applications with general information-retrieval capabilities. For a user interface, they'll be satisfied with a simple browser.
In the long run, the intranet will become an environment that can sustain a wide variety of user skill levels while allowing a few information-retrieval patterns to be shared throughout. If IS managers can assimilate this end-user demand, they'll have a lot of leeway in how they implement it.
(Charles Babcock is US Computerworld's technical editor. His Internet address is email@example.com.)