The following text is copyright 1997 by Network World, permission is hearby given for reproduction, as long as attribution is given and this notice is included.

Assumptions of video needs

It is certainly common folklore that we all need desktop video conferencing as soon as it can be developed and deployed. This has been the case for quite a while now, at least since the NY World's Fair. For the last dozen years or so it has been just around some corner or another. As someone who has an excessively "busy" office, I have always viewed the prospect as a mixed blessing at best but am now starting to wonder about the basic assumptions of usefulness.

Clearly, video conferencing has proven itself to be a valuable tool. Most large corporations, Harvard is no exception, have multiple video conference set ups. Currently most of these seem to be of room or at least multi party scale. The use of these facilities can save large amounts of money and avoid much of the wasted time that results from having to gather meeting participants into one location. They also increase the flexibility of scheduling meetings because the involvement time is limited to just about the actual meeting time. A number of businesses have sprung up to provide videoconferencing facilities for those organizations which can not make the investment on their own. The development of Internet-based video conferencing would lower the cost of these systems and permit more organizations to make use of them.

But I am not so sure that all that much of the corporate world has much current use for ubiquitous desktop video conferencing capabilities. (I could make the snide observation that, in much of the industrial sector, most of the production line employees don't have desks, making desktop services less useful than they might be but I won't.) Now note that I do not include playback of instructional films in the same category as video conferencing, the timing demands are far harder in real-time interactive operation as is needed for audio or video conferencing. Training films can have a playback latency in the multi-second range with no degradation in usefulness where tens of milliseconds count in interactive work.

My worry in this area is one of the need to reengineer existing procedures to make this technology all that useful in any broad scope and the return on the large investment in money, time and pain that reengineering requires. In some areas, those where group meetings are already the normal way of doing business, the ability to participate in these meetings without leaving the desk could be helpful, but it is another case altogether where group interaction (as opposed to exhortation or information dissemination) meetings are rare. I find it hard to accept that employees who currently do their job without frequent group interaction will have all that pressing a need for this capability.

Thus, I find it hard to understand the constant drumbeat in favor of all encompassing quality of service capable local networks that I see in much of the trade press and from consulting companies. This may seem like a strange way to proceed, but I'd rather figure out what functions are actually needed before deploying technology to meet some "strategic direction." There is a big difference between deploying RealVideo using the existing networks and putting in whole new networks. Do the latter only when a need, rather than an assumption of need, is present.

disclaimer: Harvard does assume that it is needed but the above worries are my own.