One of the classic techniques for marketing high-tech products is to claim to be first with something new. First to produce a new category of products. First to market with a new technology. "First-mover advantage" is a phrase seen in many business plans over the past few years. It's a great strategy-provided there is some substance behind the claim.
So, in an industry where new firsts occur almost daily, it's not surprising to already be hearing about 4G wireless networks. Now, it should be noted that we're really not even in the 3G era yet, with 3G deployments caught up in a poor economy, battles over standards, ridiculous spectrum-auction expenses, technical challenges, and politics-always politics!
We won't see meaningful deployment of 3G technologies until around 2003 in Japan, 2004 in Europe, and 2005 in the U.S. I also doubt the viability of combined voice/data systems as embodied in most 3G technologies, and we could be disappointed again by the actual data throughput provided in 3G systems.
So, in one way it's not at all surprising that we're beginning to hear discussion of 4G cellular systems. I think it's now evident that 3G is going to have a limited impact on the availability of cost-effective wireless broadband data and that a different architecture will be required. Now, while some are defining 4G as throughput of well above 10 Mbits/s (I've seen 20, 50 and even 100 Mbits/s mentioned more than once), I think we should be looking at 4G technologies in a slightly different light.
If you follow what's happened in wired communications in recent years, you'll notice what I like to call the "inversion" of communications networks. Once, networks were based on telephony technology, and a brilliant hack, the modem, made bits into sound. Data thus ran on top of networks designed for voice. With the invention of the LAN, and especially the router, we inverted the network . Now the bearer was fundamentally designed for data, and voice could become just another digital traffic stream on top of it.
Such it will be with 4G wireless. While some think 4G will be typified by ever-greater throughput, I think 4G will be most significant because it will fundamentally be a wireless IP network. Support for time-bounded traffic will be similar to that on the wired Net. And while I don't expect 4G systems to be a market factor before 2008, they will finally close the performance/capability gap between wired and wireless. Convergence at last.
Craig Mathias is an analyst with the Farpoint Group (Ashland, Mass.).
