Maybe our industry is too used to instant gratification, but it does seem like broadband wireless is slow in rolling out. There's a lot of talk about WiMax, free municipal Wi-Fi services, 3G, wireless/wireline integration and so forth. But where's the action, and when?

The main factor slowing broadband wireless deployment is questionable ROIs (returns on investment). Wi-Fi is unlicensed so it's hard to base commercial service on it; 3G licenses can run to $1,000 per customer. WiMax chipsets and terminals also are expensive, costing hundreds of dollars. High terminal cost means low customer adoption rates, low revenue generation and generally dismal financial returns. Is there no hope?

WiMax integration is a way off
For WiMax there is, but not right away. Intel's recent announcement of its "Rosedale" schedule for Centrino-like WiMax support suggests the critical combination of standards support (in particular, for WiMax mobility) and integration with CPU chips won't come until mid-2007.

This doesn't mean there can't be WiMax deployment before then, but it does mean WiMax equipment won't be particularly cost effective.

Wi-Fi hangs on convergence
What about Wi-Fi? For all the hype, the real hope for Wi-Fi is the potential for wireless/wireline convergence using Wi-Fi in the home and 3G outside. The problem is Wi-Fi is a power-hungry standard and wasn't designed to support portable/mobile devices.

As an alternative to the older cordless phone technology, where the home instrument is in a cradle most of the time, the Wi-Fi phone makes near-term sense. Most of the vendors and carriers I've talked to think phones that converge home networking and 3G could promote broadband fixed wireless and 3G, but most don't think Wi-Fi will get much penetration in that space; it's going to be overhung by the more power-efficient WiMax.

In citywide hot-spot-like deployments, such as the one proposed in Philadelphia, Wi-Fi's power limitations aren't much of a factor. Still, there are a couple reasons not to go crazy about that model, either. First, unlicensed spectrum invites a lot of interference, and it's not clear how reliable such a service would be. Second, Wi-Fi isn't really a great protocol for outdoor use because it's susceptible to noise from atmospheric sources.

3G still needs applications
3G seems to be catching some wind, but on closer examination it's clear that there's more 2.5G lower-speed equipment out there. The mobile operators are looking for applications that will justify the high cost of network modernisation to 3G. Video, a la Verizon's V Cast, is a hopeful start, but how broad the portable video market is has yet to be proven. Still, this area seems to be showing signs of near-term growth and might be the first market to develop.

Licensing is the key
But does all this adequately explain the seeming snail's pace of broadband wireless deployment? Maybe there's another key factor: spectrum licensing. The high cost of licensing, created in part by the bidding concept of license allocation, favors big incumbents. In the US, for example, Sprint holds most of the licensed spectrum for WiMax.

With a small number of "competitors" that are really established players in a different broadband technology area, it's possible that there's not enough incentive. Players who might want to move and shake the market with wireless broadband just don't have much of a shot at the licenses needed.

It might be that the Wi-Fi/WiMax dynamic is a key to understanding the wireless market. Wi-Fi, unlicensed and competitive, is generating a lot of buzz and support at the equipment level, but its value in really changing the broadband market is limited. WiMax, with better power management, licensed spectrum and range, is a potential disruptor, which might be why exploiting it seems to be in the hands of the establishment.

That might mean that before we get too concerned about the state of wireless technology, we need to look at how licensing wireless spectrum affects the way that technology is used. [As it happens, new models of spectrum management are evolving, and the UK has already started to adopt them - Editor].

Nolle is president of CIMI, a technology assessment firm in Voorhees, New Jersey. This article appeared in Network World.