The PC has maintained a dominant position on the desktop and achieved a level of nonsubstitutable infrastructure, making it difficult, if not impossible, to replace. Wintel systems benefit from the image of being the safe choice in many market sectors, with the result being a powerful positive feedback loop. Despite this, it's inevitable that the PC will be replaced, and IT departments need to start thinking about what might come tomorrow. To understand the shift that will occur, understand that nearly all technology shifts take place over time and go through the following five phases:

1. Competing standards are introduced.

2. Market forces lead to one standard emerging as nonsubstitutable infrastructure.

3. The prevailing standard is attacked by pseudochallengers but isn't displaced because the challengers don't offer enough meaningful differentiation or have flaws different from the technology they try to displace. (However, pseudochallengers often have enough value to remain viable alternatives.)4. The standard re-emerges stronger than before and appears invincible.

5. A new and better technology emerges and displaces the existing standard.

This cycle has held true for a variety of technical innovations, from the development of refrigeration to audio technologies. The Wintel PC is now going through stages 3 and 4. It has achieved dominance and faces a host of pseudochallengers. Throughout the years, these challenges have included the Mac OS, OS/2, Linux, Java and network computing. But none was able to cross the gap that leads into the fifth stage by providing a clear and demonstrable difference. Despite being unable to displace the Wintel PC, many of the challengers provide enough value to remain viable alternatives in certain markets.

To supplant existing technology, a new technology must meet the following three criteria:

1. It must offer visible and demonstrable value and differentiation that end users can directly exploit. One reason the CD replaced the LP so quickly was that end users were able to clearly hear the difference between the two technologies.

2. It must offer economic benefit to software vendors. In the CD market, software vendors in this case music companies had a clear benefit to shift to the new format since the retail price they could charge for a CD was much higher than that of an LP or cassette and has remained so.

3. It must offer clear economic benefit to hardware vendors. If conditions 1 and 2 are met, hardware vendors have a strong incentive to build new systems that will take advantage of the new technology and drive upgrades. CD hardware vendors were behind the new audio technology and had no fears of systems cannibalizing sales.

Today, we're seeing the rise of a technology that meets these criteria: ubiquitous computing. Ubiquitous computing is the ability to access useful digital content, both personal and business-related, through a variety of digital information devices. By 2005, it will evolve and ultimately displace the PC on the desktop.

Three core components will enable ubiquitous computing: a multitude of information devices; connectivity between personal-area networks, wireless LANs and WANs; and Web-based services that provide the software infrastructure. In future columns, I'll discuss each of these components and explore how the PC will evolve.

Michael Gartenberg, a former vice president and research area director at Gartner Inc., is an independent technology analyst and consultant. Contact him at michael.gartenberg@mindspring.com.