As we approach the end of 2001, I want to look back at one of the most popular and interesting topics we addressed this year—wireless networking. Today, with 11Mbps Wi-Fi—the 802.11b wireless standard—becoming a common networking scheme for businesses and homes, you'd think the future was clear. But a lack of standards and a set of competing technologies that functionally overlap point to a future in flux. If you've already implemented Wi-Fi, you're probably in good shape, but if you haven't yet implemented wireless networking, this might be a good time to assess the state of the technology.

Wi-Fi is an easy and relatively inexpensive solution for tasks such as email, Web browsing, and light file transfer. Wi-Fi's greatest problem is scalability: Unlike most wired networks, each 11Mbps wireless connection is shared, and available bandwidth can drop quickly as you add users and increase distances between base stations. But newer OSs such as Windows XP automatically select the fastest available connection, and adding base stations, although sometimes expensive, is straightforward.

An increasing number of devices are becoming Wi-Fi enabled. In addition to desktop PCs and laptops, you can now add wireless connectivity to Pocket PC devices through PC card expansion or even CompactFlash cards. And new residential gateways and routers from companies such as Actiontec are supplying empty PC card slots so you can easily add Wi-Fi capabilities on the fly. Other wireless-enabled devices are coming soon, including home appliances, toys, car stereos, and stereo and video home entertainment devices.

Given the increasing pervasiveness of Wi-Fi and a general understanding of the way the PC industry works, you can assume that speedier solutions are coming down the pike. In the first few months of 2002, 54Mbps solutions will begin appearing (theoretically at first). The problem, however, is that the next-generation Wi-Fi solutions are divided into two incompatible camps, and choosing one solution over the other is risky.

The safe choice seems to be 802.11g, which is backwards compatible with Wi-Fi. You can use Wi-Fi-based devices on an 802.11g network, and they'll continue to operate at the slower 11Mbps speed. Likewise, you can use 802.11g-enabled devices on a Wi-Fi network, and they'll ratchet down to 11Mbps as well. The problems with 802.11g, however, might doom this technology. First, 802.11g signals don't travel as far as Wi-Fi signals, thus requiring more base stations to cover the same area. This limitation might be prohibitively expensive in a corporation. Second, the 802.11g solution operates in the same crowded frequency range as Wi-Fi, which can cause interference problems. Third, 802.11g appears to have some of the same security concerns that currently plague Wi-Fi. Fourth, although 802.11g offers the potential of 54Mbps speeds, the first generations of this technology will likely operate at 20Mbps to 22Mbps, with short "turbo" bursts of speed operating near the 54Mbps limit.

Another high-end contender that will apparently hit closer to the 54Mbps target on a regular basis is 802.11a (Wi-Fi5). The main difference between Wi-Fi5 and 802.11g is that Wi-Fi5 operates in a less-crowded frequency range, and is therefore less likely to cause any interference problems. However, Wi-Fi5 doesn't interoperate with Wi-Fi; the two technologies were standardized at roughly the same time, and developers expected users to implement them in different scenarios.

Despite the impending arrival of two higher-speed alternatives, I expect the original Wi-Fi to be the best seller in 2002—partially because of falling prices, but more because of compatibility and interoperability. In most cases, 11Mbps is fast enough, and users with bigger bandwidth requirements will likely require wired connections anyway.

Before moving on, I want to address some responses I received about last week's column. My point about Dean Kamen's Segway isn't that the device is a perfect solution, although I do expect it to find success in numerous scenarios, including third-world nations, delivery services, large factories and warehouses, and such. I simply tried to point out that the Segway represents the sort of solution that arises when an individual or company looks at a problem and develops an answer that isn't based on pre-existing product dominance or market conditions. In other words, the inventor didn't already have a monopoly in transportation devices and attempt to meld his existing products to meet a new need. Instead, he created a unique product that wasn't influenced by a need to extend his existing dominance.

If we look at this phenomenon from the Windows perspective, Microsoft's software development will never solely address users' needs because the company always thinks in terms of its dominant product. So the question isn't really "Where do you want to go today?" Instead, it's "Where would you like Windows to take you today?" It's a not-so-subtle difference.