At Comdex last fall in Las Vegas, national Semiconductor President and CEO Brian Halla confidently predicted that the growth rate of semiconductor spending would hit a new peak this year. I regret to note that the date he mentioned was five weeks ago. It looks as if Hallas projection, even though based on an impressive modeling effort, was premature by at least three quarterspossibly five or six.
More interesting than mere timing, though, is the question of what the next chip boom will mean in terms of enterprise IT capability, as well as the skills required to turn potential into fact. Like the six blind men examining the elephant, its easy for people to walk around a beast as large as worldwide chip manufacturing and perceive things in terms of their past habits and future hopes. Whats needed now is a fresh perspective that reveals outdated assumptions masquerading as unavoidable problems.
Hallas forecast relied, in part, on a boost in demand from what he called the "China effect" of exploding Chinese demand for semiconductors. He could not have predicted SARS, with its brief but chilling effect on Asian trade, or the worldwide hesitations due to uncertainty in Iraq. All in all, he did rather well to be off by only a year, with Semiconductor Equipment and Materials International now forecasting sales of chip-making equipment growing by 24 percent next year compared with only 4 percent this year.
That said, we come to the question of what those chips will do. Halla was not predicting a repeat of any previous cycle. Rather, he observed that each of the past three chip booms represented successive quantum jumps toward smaller but more highly interconnected devices: mainframe computers in 1974; first- generation PCs in 1984; Internet- connected, PC-based devices in 2000; and embedded connectivityemphasizing highly integrated wireless technologiesin the boom thats soon to come.
Its entertaining to think about the next leap in general-purpose, relatively high-priced microprocessors that push the edge of the envelope in desktop, workstation and server computing. Looking at the volumes of chips, though, and their real impact on how we do business, its clear that the more important action is at the other end of the bell curve. What will make the next boom genuinely different will be developments surrounding much smaller and more numerous devices, such as those involved in radio-frequency ID technology. We will also see rapid growth in highly integrated "system on a chip" (often abbreviated SOC) processors of the kind that are destined for use in environmental control systems, point-of-sale devices, flexible manufacturing systems and personal communication tools, among other enterprise applications.
Not only the hardware costs of these devices but also their software costs will plummet as Linux operating systems and Linux-based vertical applications make up a rapidly growing share of the software on those machines. This double-barreled cost reduction will doubly propel new chips proliferation.
Next years enterprise IT to-do list does not feature Windows upgrades or yet another wave of the latest in beige boxes. Its a much more challenging slate of opportunities to devise and deploy constellations of devices, integrated by wireless links and Web services protocols. Entertainment venues such as theme parks, for example, are already finding it cost-effective to get ticket sellers out of their booths by giving them wireless terminals with integral printers. This approach makes it much easier to respond to surges in attendance. Any logistics-driven environment can explore similar options for taking transactions to where customers and goods are, instead of incurring the extra costs of moving themor the paper that represents themto clumsy, fixed-location workstations.
The 04 expansion of IT will move us decisively beyond the PC and its immediate descendants. It will call for a wide variety of hardware knowledge, a close relationship with a larger array of providers and continual conversation with field personnel about whats making their jobs hard today.
The key technologies, and the standards that make them useful business tools, are ready. Whats needed now is an eye for the opportunities to make them strategic.
Peter Coffee can be reached at email@example.com.
Peter Coffee is Director of Platform Research at salesforce.com, where he serves as a liaison with the developer community to define the opportunity and clarify developers' technical requirements on the company's evolving Apex Platform. Peter previously spent 18 years with eWEEK (formerly PC Week), the national news magazine of enterprise technology practice, where he reviewed software development tools and methods and wrote regular columns on emerging technologies and professional community issues.Before he began writing full-time in 1989, Peter spent eleven years in technical and management positions at Exxon and The Aerospace Corporation, including management of the latter company's first desktop computing planning team and applied research in applications of artificial intelligence techniques. He holds an engineering degree from MIT and an MBA from Pepperdine University, he has held teaching appointments in computer science, business analytics and information systems management at Pepperdine, UCLA, and Chapman College.