by Andy Oram
American Reporter Correspondent
December 30, 2003
The American Reporter Celebrates
Our 10th Anniversary Year
1995 - 2004
An A.R. Special Report
CAMBRIDGE, Mass., Dec. 31, 2003 -- Editor's Note: American Reporter Webmaster Andy Oram, an editor at O'Reilly & Associates, writes each year on the promise and peril of the Internet. Here are his reflections on developments in 2003, and their significance for the years ahead. For members of our information-rich stratum in Western society, it used to be the wealth of data - that is, the results of communication - that we drowned in. But 2003 took technology to another level. It threatened to drown us in a wealth of communications channels themselves! Voice over IP, WiFiaccess points, satellite radio, 3G cell phones - when will the cornucopia trickle to a stop?
This article takes a practical approach to sorting out the offerings, by asking what we really get from each one. Does a technology simply replace another technology in current use - perhaps with added convenience or at lower cost - or does it add something new?
At the end, I'll touch a bit on the complicated regulatory issues that came up in 2003, because a few of these debates - such as the regulation of voice over IP, the resale of phone lines by local phone companies, and access to cable networks by competing Internet providers - have excited enough vibrations to be heard by the general public.
It is typical for new technologies at first to function as replacements for older ways of doing things. Thus, satellite radio is marketed as just a better radio - wider reach and higher audio quality. And voice over IP (heretofore referred to as VoIP) simply replaces standard voice service (plain old telephone service, or POTS). VoIP is adopted because it costs less - although it's doubtful that VoIP really more efficient than POTS; for the most part it just cuts corners by playing tricks with regulations.
Similarly, WiFi(a convenient buzzword for the standards known as the IEEE 802.11 specifications) is a replacement for older local area networks (LANs) that one finds in every modern office building today. Therefore, a WiFinetwork is often called a Wireless LAN. Instead of some cable- or fiber-based technology such as Ethernet, WiFitakes over some underutilized radio spectrum that was lying around. Convenience, of course, is Wi-Fi's advantage over Ethernet.
Ah, but replacement is only the start. New technologies generally find new uses through the ever-present spark of human creativity, and often spawn social revolutions. VoIP and WiFihold this kind of potential, but speculating about it is easier than bringing it to fruition. A lot of pieces have to fall into place before the revolutionary wheel starts to turn.
Thus, VoIP could be tucked into entirely new multimedia applications, serving simply as their audio components. Imagine, for instance, that instead of editing a document and emailing it to coworkers for review, you set up a teleconference where you are all chatting in real-time and making simultaneous edits to a document - over long-distance connections. Several new technologies hint at this new way of communicating. Other possible revolutions in communication could occur from a combination of VoIP with the cutting-edge research topics of voice recognition and artificial intelligence. However, both of those cutting-edge topics are struggling to solve incredibly difficult problems, and will probably not cut through to their treasures for some time.
But there are many types of VoIP, and some lend themselves more to innovation than others. Some services let you use an ordinary hand-set and call other people who use standard POTS. This kind of service clearly offers a low barrier to entry and makes VoIP valuable - but it limits VoIP to its replacement function. Services that offer only computer-to-computer connections are less useful at the current moment, and any services that require you to buy a microphone and hook it up to your computer present more barriers to adoption. But because they are better integrated with other digital technologies, they present a base for exciting future applications.
The hard part is getting people to make the leap.
Similarly, WiFiis envisioned by wireless evangelists to be more than a replacement technology. They plan for it to open up a new freedom where people can stay constantly connected to one another and participate with their environment in a richer manner (such as finding local businesses to deal with as users move around). To reach this new level of functionality, WiFihas to move from laptop computers to even smaller devices that are easy to carry, and probably needs to be integrated with the Global Positioning System (GPS) so that the computer's location is known both to itself and to its potential correspondents. Both these things are happening now. Soon enough, the applications will follow.
Should WiFiachieve the vision of becoming a geographically aware, always-on service, it will move from being a replacement technology for wired LANs to being a replacement technology for 3G cell phones. (The term 3G, for "third generation," is mostly marketing. If it means anything at all, it refers to the use of a pure Internet-style kind of packet switching instead of sending Internet data over older cell phone technologies.) In other words, the vision itself has been around for years. But 3G costs too much for most people, partly because cell phone companies hope it will be hot stuff and partly because they sunk more money than they really should have into the infrastructure behind it.
The lag in 3G adoption lends an interesting new twist to the market. Cellular companies realize that WiFihot spots are formidable competitors and that the companies can't lick 'em, so they've decided to join 'em. Literally. Recent cell phones have come out with support for both WiFiand standard cell service.
The reason this may be a win is that WiFihas a limited range. It's normally effective over just one or two city blocks, although extraordinary distances have been achieved under special conditions. The general rule of thumb is that WiFihas enough range for somebody you don't know about to snoop on your communications.
Anyway, the cellular company strategy is that people can use WiFi(cheaply) when they're near a WiFiaccess point, and switch to 3G (at higher cost) when they move out of range of Wi-Fi.
But there's also a replacement technology for longer-range cell phone towers. This is an emerging standard that falls under the IEEE moniker 802.16 (as opposed to Wi-Fi's 802.11). While WiFiis often called a wireless LAN technology, the emerging standard is called wireless MAN (for municipal area technology).
And while Wi-Fi is very low-budget and simple to use (practically any computer user can buy and set up an access point) wireless MAN technology requires something more like a cell phone tower as a hub. And I expect that, once the technology is firmed up and takes off, cell phone towers will become wireless MAN towers.
The advantages to wireless MAN over cell phones basically involve flexibility. You can reserve the bandwidth you want, and use the connection in a variety of ways depending on the type of application you're running: voice, video, or basic best-effort (I am tempted to write "old-fashioned"!) Internet access. Thus, wireless MAN displays the same characteristics of other new technologies: it will probably start as a replacement technology and then find ground-breaking uses.
We mustn't forget the familiar Internet access most people have in their homes: dial-up, ADSL, and cable modems. The latter two serve mostly as replacement technologies for dial-up; their users are generally doing the same tasks of email and Web surfing. Some people manage to use ADSL or cable for real-time videoconferencing or to play games over the Internet, but the faster services can't be said to create a revolution in communications. They're not fast enough for that. And perhaps this is why the regulatory wrangles over these services are so frustrating.
Both standard phone lines and cable modems have become the focus of regulatory debate over the past few years. In the case of phone lines, incumbent local companies (the Baby Bells) have had to offer their facilities to competitors for a long time, but in 2003 they persuaded the FCC to release them from this requirement. In the case of cable modems, the situation has been reversed (for complicated historical and definitional reasons): the cable companies have been able to maintain a monopoly on their cable lines, while competitors have been asking the FCC and local governments to force competition.
I am sympathetic to the case made by competitors, but I wonder whether the concessions they want would help them very much. The reason is that competitors don't have much more to offer than incumbents. For years they've been trying to make a business (and sometimes succeeding, although barely) by offering various combinations of their own service with the service they lease from the Bells.
The new, small competitors have always had trouble because they can't compete well with the Bells on the basis of price. There are good reasons to think this is because the Bells kept their costs for ADSL low and subsidized the rates from profits they made on standard service. On the other hand, they are so good at hiding their costs that the task of unmasking any unfair competition is far beyond the abilities (not to mention the desires) of state and federal regulators.
But I also believe there is a deeper issue: competitors can't shatter the current desultory phone market by offering exciting new applications. And that's because ADSL isn't fast enough to support exciting new applications - especially in the upload direction.
Cable modems present an even more depressing case because they function like LANs. Everybody is sharing bandwidth. Therefore, some kind of limits must be placed on what cable modem customers do with their service; barriers to innovation are inherent in the architecture. Competition here has much less justification.
The one other regulatory issue that has seen widespread publicity concerning new communications technologies in 2003 is the question of whether VoIP should be treated as a standard voice service, and therefore be subject to all the regulations that long-distance and local phone companies have to deal with: offering 911 service, paying into the universal service fund that subsidizes poor and rural users, and so forth.
The proponents of VoIP hate to see themselves regulated, as do most businesses in our society. One could view their opposition cynically (as the older telephone companies do) asking whether VoIP is really better, and if it were better, why couldn't VoIP businesses make a good profit while meeting what the older telephone companies like to call their "obligations."
But the supporters of VoIP - which include all the computer experts and forward-looking technology analysts I've found - make good arguments against regulating it like older voice services. The friends of VoIP point out that the universal service fund is outdated and out of control.
More generally, they hammer out their appeal on the basis of the visionary future of VoIP. It looks superficially like a long-distance service now. But its future as an integrated digital application renders any talk of meeting current regulatory "obligations" absurd. Thus, if I'm running a whiteboard application over the Internet with you, and we talk intermittently over the same connection, how much of that connection pays into the universal service fund?
Regulatory questions are important, but ultimately the public cares about what it's getting from new technology. Is VoIP just a cheap way to make long-distance calls, or does it make the most ancient and familiar form of communication - the human voice - a part of our digital future? Will wireless MAN allow high-speed Internet applications to go where they could never otherwise reach, or is it just a way of saving the sunk investments of cellular phone companies?
Clearly, we don't have answers to all the questions. But they are what we have to ask as move forward, seeing new convergences, new specifications, and new businesses year by year.
Andy Oram is a member of Computer Professionals for Social Responsibility and an editor at O'Reilly & Associates.