Can You See Me Now?
Thanks to voice-over-IP, videoconferencing–the eternal technology of tomorrow–once again appears to be on the verge of success.
You dont hear much about videoconferencing any more, and for good reasondecades of unfulfilled promises have left most people a little jaded. Still, over the last year Ive seen some surprisingly high-quality performance from gear selling for under $1,000, and last week when Polycom, the worlds leading videoconferencing vendor, introduced its first software-based PC videoconferencing package, called PVX, I decided the topic deserved another look. If Polycom, which had long eschewed software-based videoconferencing systems, now believed that the installed base of computers was powerful enough to handle two-way video without a pricey dedicated box attached, then maybe a true consumer videophone market was ready to fly.
There has also been growing evidence on the services front. On September 28, Time Warner Cable of Northeastern Ohio launched a videoconferencing service based on Viseons standalone VisiFone for its Road Runner cable-modem customers. The VisiFone, which has earned rave reviews, is a standalone device, but at $599 its almost cheap enough to succeed in the consumer realm. Then last week, an AT&T executive reported that the company was working on integrating videoconferencing with its CallVantage voice-over-IP service. Earlier in the year, Skype announced it would introduce a VoIP-based videoconferencing package in 2005. And according to a study published in June by Wainhouse Research, the videoconferencing market is, after several years of slump, growing 27 percent a year, from $530 million in sales in 2003 to a projected $1.1 billion in 2008. In China, the market is surging at a scorching 83 percent per year.
Still, I remained skeptical. As one of those journalists who misled the public about the coming boom in videoconferencing during the 90s, I vowed several years ago that I would never again write about videoconferencing unless remarkable signs or omens appeared. Either my Mom would call me up asking me to get a Web camera, or Hollywood would produce a reality-TV version of Gilligans Island, or a Republican administration and Congress would run huge deficits and pass massive education and healthcare legislation. As I pondered my vow anew, my cell phone rang and a familiar number appeared. I went running from the room.
As it turned out, my Mom only wanted to knowwhat elsehow to get rid of all the adware clogging up her computer. (I suggested Spybot, but if that didnt work, abstinence.) Afterward, I returned to studying the recent videoconferencing boomlet and decided it might not be a fluke after all. First there were the basics mentioned in the Wainhouse report: falling equipment costs, improving quality, and tightening travel budgets. True, thats what I was writing ten years ago, but there are many other reasons why the segment is taking off. In fact, there are even signs that videoconferencing may finally move from the enterprise market–where most of the growth has occurred to date–into the potentially larger consumer realm.
Exhibit A: IP-based equipment has improved. In the late 90s the videoconferencing industry realized that the existing phone-system platform was a dead end, and that the Internet was the way to go. An IP based system would provide easier scheduling and better integration with conferencing software, the Web, and other applications. It could also potentially deliver much higher (and cheaper) bandwidth. The whole industry shifted toward IP, but in a sense, the technology had to be reinvented from scratch, and the quality suffered. The Internet was never designed for real-time two-way communications and it is particularly plagued by latencythe momentary delays that can turn video chat into a badly dubbed movie. Thanks to improvements in software and hardware, however, IP systems are now far more reliable, with better algorithms to handle latency and other glitches.
Exhibit B: Standards have matured. Much of the recent improvement in videoconferencing performance comes from adherence to mature standards. True, there are two of them: the high-end H.323 typically used in corporate conferencing systems and the simpler, but faster-growing SIP (Session Initiation Protocol), which is common to software-based VoIP. Yet, after much acrimony, both have remained standing, and the two sides have reached a grudging mutual acceptance. In fact, many new systems now support both. H.323 offers more comprehensive support for group conferencing and enterprise gateways, while the low-overhead SIP is the typical choice for desktop videophones, as well as the standard that is most likely to be used to give videoconferencing abilities to third-generation cellular phone services.
Okay, I lied. Theres actually a third standard you need to know about: the H.350 directory services standard that bridges the two other standards. Adopted last year by the International Telecommunications Union, which established H.323, H.350 defines the storage of dialing information for both video and VoIP networks, thus making it easier to find others with whom to hold a video chat (including SIP users).
Exhibit C. Broadband gets broader. Theres a good reason why videoconferencing has grown faster in businesses than in homes: most broadband networks found in the corporate world are symmetrical, offering as much bandwidth for uploading as for downloading. Consumer cable-modem links and most DSL connections, on the other hand, are asymmetrical, with far more bandwidth available for download than for upload. Yet upload speeds have been improving lately, with a growing number of cable-modem and DSL providers raising the minimum to 384 kilobits per second–the base standard for decent-quality videoconferencing. Starting next year, cable companies will start deploying new equipment that could double or triple these upload speeds.
Exhibit D: Computers are faster. It may seem that most of your software runs about as fast on your 3-gigahertz computer as it did on your 1-gigahertz machine, but videoconferencing can make use of much of that untapped processing power. If a large percentage of the computing population has Pentium 3 or faster computers, fast broadband connections, and good webcams (which can be had for about $50), well have the proper conditions for an explosion of software-based (i.e., cheap) videoconferencing.
Exhibit E: VoIP, Instant messaging, and Web conferencing are helping. Voice over IP may be coming on fast, but its a risky business. The market is getting crowded and prices are dropping fast, eating into profits. The best way for a VoIP vendor to succeed is to take advantage of IPs flexibility to build more profitable services on top of voice. Since video over IP is essentially an extension of VoIP, expect most major VoIP vendors to offer it in 2005. In fact, AT&T may gamble everything on establishing itself as the leader in the field.
Internet messaging, which like videoconferencing is real-time (well, nearly), is another growing springboard for the technology. Earlier this year, America Online introduced a 5.5 version of AOL Instant Messenger, which supports links to video chat. Meanwhile, Microsoft (Windows Messenger) and Apple (iChat AV) have also added webcam support. In all these cases, videoconferencing is not incorporated directly within the messaging interface, but users can be invited to a videoconference via IM, and a single click then opens up a SIP-based software package that activates the webcam. For videoconferencing vendors, IM offers a huge ready-made directory of real-time users. Along with the H.350 standard, IM could finally make videoconferencing a spontaneous act.
In similar ways, the steady spread of Web-based document conferencing packages such as Microsoft NetMeeting and WebEx offers another entry point. The software for switching between voice, video, and data conferencing (most of it associated with H.323) is becoming more sophisticated, allowing all three modes to intermingle in a single conference. In this way, video can be incorporated incrementally, easing individual concerns about privacy and quality. For example, its fairly typical now for two remote groups to be connected by a room-based videoconferencing system, both a PowerPoint presentation and video sharing a screen. In addition, however, with some systems individual remote users can be invited to join (perhaps via IM) and can participate via desktop videoconferencing, voice only, or voice combined with remote document conferencing.
Exhibit F: Travel is problematic. Even with signs of growth in consumer videoconferencing, the biggest market for some time to come will be for business meetings. (Vertical applications such as remote education and healthcare will also continue to grow as videoconferencing mainstays.) Although business travel has crept up lately, a recent survey by Smith Barney found that 54 percent of corporate travel managers said their companies are investing more in teleconferencing and videoconferencing equipment. The driving force here is that air travel is growing ever slower and more expensive. Rising gas prices could also put a dent in longer car trips that might be replaced with videoconferencing.
Although the evidence points to a rising tide for videoconferencing in the years ahead, some say the core problem of the technologylatency–will never really be fixed until we get a new Internet. Yet TV took off without color; the Web was a winner despite slow dial-up connections; and cell phones continue to be popular despite static and disconnects. Its possible that theres an innate psychological resistance to video telephony thats more stubborn than our tendency not to want to fly around in a tin can or answer a business call at the beach.
Yet if so, why has the videophone played such a central role in our vision of the future? Long before AT&T decided (a bit prematurely) to promote videoconferencing as the next big thing back in the early 1960s, our popular culture embraced the concept. In fact, Charlie Chaplin featured the technology in his 1936 film, Modern Times, more than a decade before video hit the scene. True, Chaplin revealed the potential privacy fears about video communications that still make some consumers hesitant today, but the idea stuck, and videoconferencing became an absolute assumption of sci-fi futurism. Once a concept becomes so deeply engrained, and once affordable technology arrives to fulfill its destiny, all thats required is a tipping point.
Become an MIT Technology Review Insider for in-depth analysis and unparalleled perspective.Subscribe today