New technology for the Stone Age - Tanner - Brief Article

Telecom Asia, May, 2002 by John C. Tanner

As a technology editor, there is not a weekday goes by that I don't receive at least five press releases from some company or other, be it a global giant or a start-up, announcing some new telecoms/IT product or enabling technology. If you discount OSS/CRM software products, the volume drops considerably, but not all that much. Some are hype, some are genuinely amazing, some are obvious vaporware, and some are so superfluous that they beg the question, "Why bother?"

One case of the latter--or so I thought at first--that recently landed in my email box was an announcement from semiconductor manufacturer Conexant that its V.92 modem chipsets were shipping .with Legend PCs in China. V.92, if you haven't heard, is the newest standard for V.90 56-kbps analog modems that was approved by the International Telecommunication Union late last year. V.92 doesn't make analog modems any faster--what it does do, however, is pretty cool.

Not only does it improve the handshake processes between modems to establish a connection in about half the time of a V.90 modem, it also supports Call Waiting, allowing users to temporarily put the Net on hold for incoming and outgoing calls without losing their connection.

Cool or not, my initial reaction to V.92 was: why bother? We have DSL and cable modems where no dial-up is required--and in cases where it is, it takes three seconds--and while download speeds aren't actually in the megabit range, they're fast enough to make 56-k a grueling experience to fall back on. Plus, there's no worry about sharing voice and data on the same line. So what's the big deal?

The same day, as it happens, I received another press release from the ADSL Forum with their latest projections for worldwide DSL take-up over the next few years--from 18.6 million users in 2001 to 200 million by 2005. The growth figures were so impressive that it took a few minutes for me to realize that broadband users would still be vastly outnumbered by analog dial-up users during that time. By many estimates, in fact, analog dialup users will still represent the majority of Netizens by the end of the decade.

The point is that for all the hype over next-gen technologies, from an end-user standpoint, such technologies will still in the minority in their fledgling years, and even when they become the majority, the older technologies will still be in use for some time.

Pockets of 3G

This is why, for example, telecoms vendors have been developing AODI (Always On Dynamic ISDN), an always-on version of ISDN that provision extra bandwidth on the fly as you need it via additional B channels--this at a time when corporate customers are being enticed with faster and sexier access technologies like, say, metro Ethernet.

The wireless arena offers perhaps the clearest assessment of the situation. 3G may be the future, but if every cellco in the world somehow managed to launch 3G services this weekend--while we're fantasizing, let's say that everyone's 3G networks all work perfectly from Day One--it would take years before their 3G users outnumbered their 2G or 2.5G users. Even the UMTS Forum--which is optimistic enough to reckon that 3G services will be a trillion-dollar business by the end of the decade--says that while there will be 2.25 billion mobile subscribers worldwide by 2010, 28% will be served by a 3G network. That leaves a noticeable 72% unaccounted for.

Meanwhile, since even in a fantasy setting cellcos will only be able to offer pockets of 3G service within their 2G coverage area in the short term, 3G users are likely going to want as much performance as they can get when outside the 3G zone. This should drive cellcos to use technologies and solutions that allow them to get the most out of their existing network as possible. This can mean EDGE for GSM cellcos (or GSM 1x, if you believe the CDG) and cdma2000 1x for CDMA operators, and also includes platforms like i-mode, whose secret to success hasn't been bandwidth but offering popular services that work within the network's data rate limitations.

Next-gen technology is undeniably cool and will be great when it gets here, but it's nice to know that some companies are still looking for ways to make things even incrementally better for the hundreds of millions of people over the next ten years who will still be using analog modems, 2G cell phones, ISDN, and other Stone Age technologies.