CIOs looking at the wireless data scene today might get so daunted that they'll begin to think their current tangle of cables and wires looks pretty good. Where do you begin?
On the most general level you have short-range intraoffice wireless technology (like Bluetooth), meant to replace local cabling; fixed wireless, to replace or supplement LAN, extranet and "last-mile" broadband cabling; and mobile wireless, pushing data communications to laptops, cell phones and PDAs. Each of these raises questions of reliability, scalability, return on investment, security, coverage, application design, interface and peripheral availability, integration with the corporate network, and compatibility with whatever other signals are in the neighborhood.
That dizzying complexity helps account for the rather minor incursion that wireless technology has made into the corporate world. At the moment less than 5 percent of all corporations have a wireless application in place, according to Jack Gold, a Westborough, Mass.-based vice president with analyst firm Meta Group Inc. In part this complexity comes as a price for living at a time when wireless is the technology du jour. But the landscape was not that different eight years ago, when CIO ran its first articles on the medium. "Today wireless is a consultant's dream of complex, rapidly changing technologies and rate structures," we wrote in October 1992. Sound familiar?
In 2000 we at least know which products made it to adulthood. For example, we referred in passing to setting up a radio network to handle dispatch data, specifically routing and scheduling service calls. In retrospect, this idea was a win. Support and maintenance costs were high, but the return--more calls per day--was quantifiable and real enough to attract large, technically competent clients like IBM Corp. and General Electric Co. As technology (and antenna coverage) improved, non high-tech companies, such as Otis Elevator Co., picked up the application. Large retailers (including Sears) figured out how to extend the system into deliveries and build it into the heart of same-day delivery policies.
These big companies wrote their applications themselves, but startups eventually emerged to take on this job. In the mid-1990s, dispatch wireless spread to ever-smaller trucking fleets and out to sectors like manufacturing (materials handling), retail (inventory) and health care (hospital rounds). Today, hosting companies, such as Broadbeam Corp. of Princeton, N.J. (until recently Nettech Systems), make it almost as easy to set up a dispatching application as choosing a news feed on Yahoo. You register with the hosting service, take delivery of the data terminals, mount them on your trucks, and you're done. All you need to manage your fleet is a browser. This is certainly a CIO-friendly operation.
The lesson is an old one: When innovations first appear, they are usually rough and full of unknowns. To take root, a technology has to attract users prepared to spend the time and money needed to make it practical for the next generation. As these users hammer away, experience accumulates, standards solidify and employees trained in the technology quit to become consultants. Costs come down and reliability goes up. A similar story could be told about two-way paging, the other great wireless data success story of the 1990s.
This history suggests that a CIO currently trying to pick the best entry in the world of wireless data apps should look at domains where the returns are easy to define, the technical issues are well-posed--if not necessarily few or simple--and the early contenders are likely to have the resources and the skills to deal with the unavoidable problems. In such sectors, prices will fall and reliability will rise more rapidly than in others.
Of the three categories of wireless mentioned, fixed wireless probably clears the bar with the most room to spare. In fixed wireless all the features of the system, from signal strength and signal direction to the nature of the equipment, are known and controllable. This provides a huge technical advantage over mobile data wireless, which has to handle signals of various strengths coming from anywhere and going anywhere. The infrastructure supporting mobile wireless data has to be built out considerably before it can be relied on for a business purpose, which raises costs and imposes long development schedules. Fixed wireless needs only two stations to start putting in a day's work. Mobile devices, with their small screens and tiny memories, can require completely new suites of interfaces. Fixed wireless links are invisible to the rest of the system.
Finally, the heart of the fixed wireless market--supplying last-mile broadband in areas where there is no ideal wired alternative--is well-defined and enormous. The overseas 30 "fiberless optics" (TeraBeam Networks Inc.), they seem fewer and simpler than penetrating the dust and noise swirling around mobile wireless data. Mobile and short-range wireless might be more important in the long run. They might launch more apps down the road, but if you are looking to make your life simple, history suggests fixed wireless should go to the top of your agenda.