Charles B. Kreitzberg, Ph.D.
President
Cognetics Corporation

Net-centric computing is a new technical architecture in which programs are stored on a server and loaded on the user's (diskless) computer when needed. For many IS departments, net-centric computing is a way to reduce the cost and hassel of maintaining applications software. But net-centric computing is seen by many users as a ruse to take back control of technology and limit user's choices. This article was published in Communication's Week, December 1996.

Serious issues need to be resolved before businesses can capitalize on the opportunities promised by net-centric computing, where business users access all internal software applications through the corporate intranet.

This model puts the IS department in control of everyone’s work stations, and, by extension, largely in charge of how people go about their jobs. Business and technical management must take steps to ensure that IS does not unilaterally change the organization, but instead involves end users extensively in the development of the computing tools that are increasingly central to their jobs. Otherwise, we could end up compromising the productivity and corporate competitiveness that intranets--the linchpin of net-centric computing--are intended to enhance.

To be sure, the net-centric model has some great advantages. In the net-centric vision, all corporate computing will be facilitated by the intranet. The browser, enhanced with collaborative communications modules, will become the new desktop. Software will be housed on distributed client-server networks. Applications will either run on a server or be delivered to a diskless network computer (or a "thin client") and will be cost-effectively concstructed from components.

In concept, net-centric computing streamlines the delivery of software applications to business users. Change an application on one server, and the entire organization is effectively upgraded. Users who shift computers can (in principle) carry their environment with them on a removable disk or perhaps a smart card. And a library of cooperative software components simplifies development.

But beware. To begin with, there is no standard for net-centric computing. If anything, Microsoft, Lotus, Netscape and Sun are promulgating incomplete, if not competing, standards (pick from the master list of: HTML,Java, Javascript, J++, ActiveX, OLE, COM, and Corba). Yet net-centric models will only work if the software components fit together seamlessly. The IT industry must agree on a stable, open, comprehensive, and flexible architecture before corporations can move forward powerfully.

IS management also needs to clearly outline how net-centric computing will provide value to business. With dedicated applications that fit the "airlines reservation model," net-centric computing makes sense. But when it comes to the corporate desktop, the user may suffer. The most important lesson of the PC era is that corporations only benefit when information technology is linked to corporate strategy, responsive to user needs, and tailored to organizational workflow.

Such synergy has not come easily. Frankly, as IT professionals, we have a mediocre track record for developing software applications that enhance productivity and provide strategic advantage. Some 25% of development projects fail outright, and another 35% fall short of their initial goals.

The principal reason is that the development process is dominated by programmers who have little or no connection to the end users; the resulting software applications are technically sound but a poor fit to workflow and users’ competence.

Most companies are just beginning to grasp the importance of user-centered design, which can ensure productivity. If net-centric computing is used to dictate the organization’s software choices, however, IS will effectively forfeit years of hard-won progress in closing the gap between its technical focus and the business it serves.

Users are just achieving competence with their office suites, and they are largely distrustful of IS. Instead of rushing into net-centric models that will force them to adapt again, IS needs to take advantage of today’s relative technological stability and begin correcting some past mistakes. Now is the time to raise the technological proficiency of the organization, solidify the strategic relationship between technology and business management, and partner with users to learn how they actually work. Only when IS knows users’ jobs, procedures, and preferences can they determine what software they really need or what process they're willing to follow.

As we elevate corporate intranets from bulletin boards to delivery vehicles for core information technology, we have an opportunity to fix some of the design errors that have plagued software. We can reduce user frustration, improve business productivity, and, perhaps, even become heroes in the process. That is, if we're willing to change our own perspective.