Video conferencing has been around for a surprisingly long time. AT&T ran the first call in 1927. Since then, pundits have been consistently predicting that video conferencing was just about to take off. They have been wrong for eighty years. Why should we believe the techno-optimists now?

In the last year, several companies have launched video conferencing products that provide an experience similar to real meetings. The quality is surprising and even sceptics have begun to see the advantages of using a meeting room for an hour rather than spending three days going to Hong Kong and back. Cisco’s Telepresence product is generating enthusiasm that is tempered by the enormous costs of setting up the equipment and providing the bandwidth. But the company says that prices will fall dramatically over the next few years.

Is this going to be enough to get people out of planes? The signs are good. Even low bandwidth alternatives suitable for home use are getting praise from the experts. So if Cisco doesn’t make video conferencing work, Bay Area start-ups like VSee will probably start eating into the market for lower cost products.

***

Video conferencing: why we need it to work UK companies are increasingly promising ‘carbon neutrality’ to their stakeholders. Electricity consumption can be neutralised by the purchase of energy from renewable sources. Gas is more difficult to counter-balance, but is a small element in most firms’ carbon emissions.

Carbon produced by travel is an increasingly important part of the budget of most large companies. Amongst the very largest companies, business flights dominate the total emissions from employee travel. And as air travel is perceived to get more time-consuming, stressful and unproductive, some companies are beginning to investigate much more extensive use of video conferencing. Other than investing in dubious ‘offsetting’ projects, video conferencing may be the most plausible way of beginning to hold down the apparently inexorable rise in air travel.

The previous generation of video conferencing products are widely regarded as wholly unsuitable replacements for meetings. The experience seems to be that only groups well know to each other with similar professional backgrounds can work around the deficiencies in older conferencing products. The new generation of video conferencing, universally called ‘telepresence’, is clearly a huge advance on the old systems. First reports from companies that have installed the expensive technology are extremely favourable. The computing press reports a Wachovia bank executive saying that telepresence suites were already in use 45% of the time within two months of installation. The previous kit had never got above 20%. Tate and Lyle is quoted as saying that the new service makes good financial sense because the typical trip between the UK and US headquarters costs $25,000 and three days of senior executive time.

Video conferencing still has a long way to go. I have only found two large UK companies that report their hours of video conferencing use: Pearson and Reed Elsevier. Both companies are diversified publishing companies with widely dispersed operations and very high levels of air travel per employee. Pearson employees travel an average of 4,000 miles a year by air, down slightly last year but still rising at approximately 1% a year over a longer period. Air travel represents over two thirds of all business travel. The company used its video conferencing suites for a total of 9,000 hours last year, up significantly in 2006, but still a tiny fraction of the time spent travelling by air. The average employee spend less than 20 minutes in video conferencing last year. The time taken to travel the average 4,000 miles by air was probably the best part of a working week.

The air travel of Reed employees was up 5% last year, and accounted for 45,000 tonnes of emissions or about a tonne and a half per employee. Its advanced ‘collaboration’ suites saved about 323 tonnes, or less than 1% of the air travel figure.

We are not going to stop the need for international collaboration. The general quality of these collaborations is widely thought to be poor, largely because people don’t spend enough time together. The impulse to travel more and more will remain unless we can really get video conferencing to work.

The problems with video conferencing One source describes working with another person over a video conferencing link as being similar to collaborating with a ‘mentally defective foreigner’. What goes wrong?

Audio needs to be synched with video. But sound is easier to process and tends to arrive first. If the voice reaches the listener too early, the speaker tends to be perceived as untrustworthy and glib. If an adjustment is made to correct this and the video arrives first, the remote person is seen as stupid and slow-witted.
Social protocols demand that people look at each other directly. A conferencing system that gives the user an impression that his or her interlocutor is looking more than 3 degrees away from the eyes will make the user uncomfortable, and give an impression of disrespect.
Successful oral communication demands rapid and seamless switching between people in the conversation. Bad videoconferencing makes this more difficult than in an audio call.
Most clues to the speaker that he or she is boring the audience, confusing them or patronising them are non-verbal. For example, few people actually say that a conversation bores them; they give subtle and not so subtle clues to their conversation partner. Video conferencing prompts the bored person to offer these clues, but they are not received by the other person. The speaker does not adjust his or her communication style. Irritation ensues.
Similarly, people implicitly expect video conferencing technologies to make their speech persuasive (which is one of the reasons why one wants to speak face-to-face). It does not and everybody finds the interchange unsatisfactory.

In summary, bad video is worse than no video. The availability of a picture sets up an expectation that normal free-flowing conversation is going to take place. So both parties behave as if they were in a conventional face-to-face meeting, in which verbal and non-verbal clues are being unconsciously processed. A phone conversation would have been better because we would have adjusted to the well-understood restrictions on our communications ability.

What Cisco, Teliris, and others have done Cisco’s telepresence product has been on the market for a year or so. Most users speak glowingly of its usefulness. Other products such as the one from Teliris also get enthusiastic reviews. People say that these new products both reduce travel and increase the number of meetings across geographically remote teams. If this first flush of success is maintained, we can hope that video conferencing will eventually reduce the need for business travel.

What are the key features of these products?

Video conferencing takes place in a specially designed room. Each room around the world looks the same, even down to the wallpaper and light fittings. Each ‘side’ of a video conference has half an oval table – the other half is in the remote room.
High definition plasma screens fill the opposite wall. Very clear images of the people in the other room are shown.
Video and audio are precisely synched. People look directly at each other. The sound of speech comes from the direction of the other person. One set of user comments suggests that the Teliris product is better than the Cisco version because it gives equal visual ‘weight’ to people across the remote room while Cisco emphasises the centre of the screen.
People are life-size.
Video frame refresh rates are extremely high. Cisco mentions 30 new frames a second.
Unusually for Cisco, it developed the technology internally. Cisco sees the product as an enhancement to its existing VoIP product. It has simply added video. This seems a strange comment, but Cisco was clearly determined to avoid the poor image of conventional video conferencing. It decided to sell the product as in some sense an enhancement of its existing easy-to-use network VoIP offering.
Cisco’s product works over the corporation’s existing data network (powered by Cisco routers, of course). Teleris offers users its own data network.

The costs are intimidating. To prepare a full room, with several plasma screens on the remote wall, Cisco charge $300,000 (and the same for the other end). A much more restricted product with just one screen sells for $80,000. The monthly cost is said to be ‘$3-5,000’.

All the evidence so far points to successful introductions of the new range of telepresence products, whether from Cisco, Teliris, HP or Tandberg. Data is scarce, but Teliris says that its customers added 50 rooms in the second quarter of this year.

What is going to happen to the costs in the future? Cisco claims that good telepresence products will eventually become viable in the home office. (CEO John Chambers already has one at home.) We expect outrageous techno-optimism from Cisco, and this looks a little on the unrealistic side. Others say that the cost will never fall below $10,000 a room. At this level, it may not make sense for many home workers, but for a senior executive it seems perfectly possible that he or she will eventually equip a room. Bandwidth may be a more intractable issue.

Early results on the sociology of use One large bank I spoke to said that the telepresence room was in high demand, and much liked by those who used it. The bank was seeing demand from people who didn’t really need the video element of the conversation and was restricting its use to the most senior executives. A board meeting had taken place via telepresence with one senior member calling in from London, thus avoiding a long international flight. These signs look good – if the heavy hitters want to stop the middle ranks from using the rooms, there is clearly a high status attached to making a video call. If a board meeting can take place in the rooms, we can see this as a further endorsement of its acceptability for important discussions. But true success will only come when the senior investment bankers get a new deal mandate via telepresence. Then the barriers really will come down.

Other anecdotal material suggests that exposing junior team members to each other via telepresence has had a good effect on trust and morale. Many international collaborations run into the sand because team members do not trust the other parties to deliver on what they promised. The possibility of holding every participant to account via a telepresence conversation seems to be improving productivity and providing a sense that people are genuinely working together.

The cheaper alternatives Cisco and its competitors are focusing on offering a product that tries to replicate face-to-face meetings with as much fidelity as possible. Low latency life-size images are extremely expensive in bandwidth to transmit. Other companies such as VSee believe that we can solve the psychological problems with conventional video conferencing by understanding what really impedes communication. VSee says, for example, that we don’t need high frame rates to offer real improvements over existing products. Cisco gives us 30 a second, but VSee says that only 5 is perfectly satisfactory, provided we get good synch between audio and video. Similarly VSee says that we don’t need a life-size image to talk to, but we do need to offer the ability to look straight into the eye. By careful research into what really makes a video conversation work, VSee says it has hugely reduced bandwidth requirements. It has also incorporated true collaboration tools, such as ‘laser pens’ for PowerPoint presentations, that enable more immediate and compelling mutual understanding. VSee says it can operate effectively at conventional broadband speeds.

I don’t know whether VSee will work: I couldn’t find another user with whom to try out the product. But for close collaborators who know each other well, I suspect an intelligent product that works round the old problems with video conferencing can operate at surprisingly low data speeds. We won’t always need a glossy room costing $300,000 for our conversations across the world.