TechWatch: Rx for Recovery

I recently received an e-mail from a highly-stressed friend, an executive in the telecommunications business. He described what he was going through as “the meltdown in the telecom industry.”

This is indeed a tough time for all segments of that industry. How did we get here?

The usual explanations cover most of the reasons. Many of the “dot coms” that were buying huge amounts of computing and telecommunications equipment have gone away. “Fire sales” on their used equipment have become a strong force in the marketplace. A second reason given is Y2K. Many users of telecom equipment spent significant dollars in 1999 to get ready for Y2K. Their usual buying patterns have not yet returned. A third reason is the state of the overall economy. Companies pull back on their capital expenditures in times of slowdown, and the telecom industry got caught in this pullback.

All of these reasons would suggest that distance from Y2K, distance from the dot com crash, and an improving economy is all that it will take to get the industry back on top. Perhaps so. But there is one other reason for the slowdown that I would like to explore, because it is a bit more subtle and may indicate a longer recovery time. Maybe there is more telecommunications capability available now than we will need in the near future — like an overbuilt office market that has far outstripped the demand.

Kris Hudson in the Denver Post (August 12, 2001) put it this way:
“The question of whether the boom-to-bust telecom industry spawned an oversupply of fiber-optic communications lines over the past five years now hangs over the industry, and it threatens to drag a weary economy down the drain. The elusive answer: Technologically, there is no so-called bandwidth glut. Economically, there is. The difference lies in the strong possibility that much of the fiber-optic strands now underground eventually will be used, even if some are called into service only at peak times. That’s assuming new bandwidth-hungry uses emerge over the next twenty years.”

Twenty years is a long time in our fast-paced economy. What happens in the meantime and who cares other than the telecom community and those who own their stock?

As a technology, the value the telecommunications industry brings to the table is what it enables others to do. Telecommunications is an infrastructure element and the creative use of this infrastructure is what drives demand, and what generates value for the users. So the key value of this infrastructure depends on others using it creatively, and therein lies the bottleneck. Often, but not always, it takes a long time for others to adopt technology in their business.

Rich Kaplan of Microsoft created a graph of adoption rates for various technologies that tells a rather amazing story. Most technologies take a long time to achieve significant market penetration, whether the PC, the automobile, the VCR, or the cell phone. In fact the cell phone and the PC are the fastest in adoption rate on his chart, and both took longer than ten years to achieve a ten percent penetration in the U.S. market. Even the zipper took 32 years to become a commercial success after its invention, as pointed out by Malcolm Gladwell in The Tipping Point.

So, how does this apply to the telecom industry, which has been around for a long time? I would argue that in many ways the challenge in creating commercial success in the use of high bandwidth is a tougher problem than in the use of a totally new technology. With a totally new technology, people might be very aware that they don’t know its best application. With a technology that has simply gained dramatically in potential performance, people think they do know what it is good for — doing the old things faster. As I have said before in this column, that is almost never a winning strategy.

So users have to be creative, as in the case of a new technology, but more importantly they have to overcome the thought that they already know how to use it.

Let’s take a specific example of where bandwidth might bring value: the airplane mechanic. Diagnosing and fixing complex equipment involves having access to lots of information. In the mid 1990s, one airline reported a large percent of its mechanics time was consumed away from the airplane, looking up detailed information in the large collection of manuals.

Wireless communications is a technology that could play a big role in delivering information to that same mechanic. Wireless is needed because the mechanic needs to move around a large vehicle. So why don’t mechanics immediately get “hooked up” with this technology and save time and dollars?

There are several reasons and some are technical. The environment for delivering the wireless information is not always conducive to reliable service. We can perhaps live with the gaps in our cell phone service, but the mechanic cannot become more effective in his or her work without the reliable delivery of information. Cost performance of this approach vs. others is another factor.

But there is a competitor to the wireless solution for the mechanic, and it is the CD ROM. A mechanic does not need to download the information to do a task at the moment of the task, if all of this information can be available on a small portable device. Here is the key issue. Most of the data needed by a mechanic changes very slowly. The maintenance manual is generally written at the time of the product creation. Over the life of the product, there may be the occasional service bulletins that represent updates, but these are more likely on a quarterly basis, not a minute by minute basis. Updating the CD ROM would not require wireless capability at all.

Does this mean there is no room for wireless in the business scenario for the mechanic? No, wireless could play a big role, but it would require rethinking the problem. What data does change rapidly and require a “real time” update rather than a quarterly update? Recent repair records for that particular airplane might be one. Access to experts in another location for particularly tough problems might be another. Having achieved this access might then involve sharing new information between them that would require a great deal of bandwidth.

The successful development of reference tools for a mechanic would require, not simply automating the existing documentation but, restructuring the application, the data, and the way people are trained and do their work. I believe this is at the heart of the slow adoption rates for new technology, and at the heart of why it could take a long time for users to develop “bandwidth hungry applications.” Significant time is taken conceiving the application, and then additional time is taken implementing the solution.

If, in the end of this process, the new technology brings value to the user (the key question), why wouldn’t businesses be motivated to gain this value earlier? In particular, the time required to envision the potential application could be shortened. There are two things at work here, one a good reason and the second a poor but well entrenched reason.

The good reason for going slowly in adopting new technology is the high risk that it may not work as advertised, may not succeed in the marketplace, may have some bugs that take time to work out, or may be very expensive at first offering. All of these factors argue for not being an early adopter. In the case of taking advantage of higher bandwidth telecommunications, this is less of an issue.

But the second reason is that many companies believe they should not look at “technology for technology’s sake.” Figure out the business needs and then look for technology to address these needs, not the other way around. I believe this is very poor thinking, and is holding many companies back. Needs are usually formulated in terms of what a company believes is possible, and with technology changing at such a rapid rate, this strategy will almost always perpetuate the past.

This brings us back again to the telecommunications world. There is lots of bandwidth potentially available. Many will look at this and ignore it because they don’t want to think about technology for its own sake. But I believe some will look at this as an opportunity, and start to think about the business opportunity that could come from taking advantage of this bandwidth. It won’t come from doing today’s things faster, but it will come from doing new things.

I am not suggesting that we build product just because it can be built. There are market issues, mission issues, and ethical issues that must be brought to bear on this decision. I am suggesting at the product idea phase that it is a good idea to know what is available from technology, and use this to inform potential solutions.

In one sense this is not much different than developing applications for more powerful computers — not just to use the power of the computer but to bring value to the business. About fifteen years ago, a company I was working with brought in a powerful new Cray supercomputer (about as powerful as today’s PC!).

Our division president said it would take two years to fill up the capacity of the supercomputer. I countered that it would take only three months. We were both wrong — it took one month! Why? There were creative people who had looked at the potential of the supercomputer, and had applications ready to exploit it as soon as it was available. This use generated significant value for the company.

I believe there are many creative people who could help their companies gain value from the significant and idle bandwidth available today, whether wired or wireless. Though this would accelerate the return to profitability of the telecom industry, that is not the point. Done right, it would accelerate the competitiveness of the companies that develop and apply these new applications. Maybe there is an opportunity for partnership between the telecom industry and its users.

While speed has been the critical element in the recent past success of our economy, it is remarkable that adoption of technology, particularly increased capability of proven technology, remains slow. There is an incredible opportunity for companies to create new solutions with this technology. Leaders could exploit this opportunity, and it will be interesting to see who does.


Al Erisman is executive editor of Ethix, which he co-founded in 1998.
He spent 32 years at The Boeing Company, the last 11 as director of technology.
He was selected as a senior technical fellow of The Boeing Company in 1990,
and received his Ph.D. in applied mathematics from Iowa State University.