I recently spoke with folks at InterDigital, a company that has quietly been one of the forces behind the development of the IEEE 802.21 standard, which is all about converging various types of fixed and mobile technologies.

The result, soon to be an official standard, is an elegantly designed piece of technology that could form the basis for many future convergence products. However, as is normally the case with new standards, this standard will need a good deal of marketing to really get rolling, and it also will need to get the best of competing convergence efforts.

First, a few words about 802.21. With the exception of 802.1, the charter of the IEEE's 802 LAN/MAN Standards Committee is to deal with the MAC layer and below. This means that 802's take on Ethernet (802.3), wireless LANs (802.11), and wireless PANs (802.15), just to name three examples, specify the standard up through the MAC layer. Any other upper-layer functionality can be specified in 802.1.

The 802.21 convergence effort, however, takes something of a "Layer 2.5" approach, allowing networks already spec'd under the 802 framework to hand off connections among themselves with no changes to the underlying technology. 802.21 might be defined as "shimware" or "wedgeware," to apply terms that sometimes have been used to describe this approach. It creates a transparent layer of functionality that takes advantage of the client/server model that otherwise embodies most upper-layer services today.

This means we would install an 802.21 client on the mobile device, and it communicates via IP with an 802.21 server. After a few exchanges of information, handoffs between otherwise incompatible PHYs, or physical layers, become easy. (Click here for a good technical tutorial on 802.21.)

This is good technology, and it has the support of the likes of InterDigital and even Intel. But, as we all know, good technology alone is of relatively little value, so now is the time to toot the horn a little. There are two reasons for this.

First, there's a lot of competition here. This includes IP Multimedia Subsystem (IMS), Unlicensed Mobile Access (UMA), the CTIA/Wi-Fi joint effort to explore what's possible and required in converged solutions, and 802.11u.

This level of technical interest and competition is a strong indicator of the importance of converged solutions as a basis for future voice telephony. It also is important as the core of the future IT strategies for many large organisations.

Of course, not all of these approaches will survive. It pains me deeply as an engineer to note this, but the secret to success, not just in convergence but in all of high tech, relies on more than developing great standards and building great products. Two other essential keys to success are marketing, which raises awareness and creates demand, and listening to customers so that the technology can actually be used for truly useful features. But, in this case, customers aren't quite ready to raise their voices on convergence - they don't know what they need at this point.

The obvious conclusion, then, is that a Wi-Fi Alliance-like effort is needed. A collaborative group should specify what the production (not just standardised) protocols will look like, perform interoperability testing, and generally promote the technology. The group that is most successful at marketing its convergence standard likely will emerge as the winner. It's easy to steal features from another camp, but not so easy to steal mind-share once it's established. And it's mind-share that will determine the approach to convergence, not technical excellence alone.

Craig J. Mathias is a principal with Farpoint Group, an advisory firm specialising in wireless networking and mobile computing. He can be reached at [email protected].