Oral-History:Richard Snelling

About Richard Snelling

Richard Snelling studied Industrial Engineering in the University of Florida, while serving as a reservist for the Korean War. He joined Southern Bell after graduating, where he specialized in transmission and special service engineering. He went back to school and received a master's degree in electrical engineering. His career at Southern Bell ended as Executive Vice President for Southern Bell and Bell South. Snelling joined the AIEE in 1956; all of Bell System was in the AIEE, as were most people in the power and telephone industries. He notes that IRE members tended to look down on the AIEE, while AIEE members dismissed the IRE. There were more IRE people in the space industry, so there was IRE dominance at the first Cape Canaveral region joint meeting after AIEE and IRE merged—an unusual occurrence.

Transformations in the telephone industry included, in transmission, the change from tiffin ring copper to electronics. Snelling notes engineering challenges posed by the attempt at reducing outside plant's investment, and that getting technology changes into the “last mile” was always the hardest part of transformation. He discusses the tensions and dynamic between Bell Labs and the operating companies (Network Heads) in the 1970s. He explains how fiber allowed development of capacity without much investment. He mentions frictions during the AIEE and IRE merger, joint meetings of Super Com and IEEE, and attempts at globalizing IEEE.

After Snelling retired from South Bell in 1991, he became involved with the Georgia Center for Advanced Telecommunications and Technology and the Olympics Committee. He describes his involvement in the development of intelligent network devices and analyzes the dynamic between telephone companies and equipment manufacturers, and the role of the Antitrust Divestiture Act. The interview covers Snelling's activities in various boards after retirement, including DCA, Boston Technology, VSI, and Digital Wireless. He talks about founding the Home Wireless Network company and Home Wireless Network' s new development of an interface platform. He discusses standardization, including the IEEE role in standardization, and speculates on IEEE’s potential role regarding the Internet, the professionalization of engineering, and the globalization of engineering. Snelling concludes with analysis of the impact of the Cold War on networks' design, computer hacking as a real threat to national security, and powering and backup power during emergencies.

About the Interview

RICHARD SNELLING: An Interview Conducted by David Morton, Center for the History of Electrical Engineering, 25 April 2000

Interview # 395 for the Center for the History of Electrical Engineering, The Institute of Electrical and Electronics Engineering, Inc.

Copyright Statement

This manuscript is being made available for research purposes only. All literary rights in the manuscript, including the right to publish, are reserved to the IEEE History Center. No part of the manuscript may be quoted for publication without the written permission of the Director of IEEE History Center.

Request for permission to quote for publication should be addressed to the IEEE History Center Oral History Program, IEEE History Center, 445 Hoes Lane, Piscataway, NJ 08854 USA or ieee-history@ieee.org. It should include identification of the specific passages to be quoted, anticipated use of the passages, and identification of the user.

It is recommended that this oral history be cited as follows:

Richard Snelling, an oral history conducted in 2000 by David Morton, IEEE History Center, Piscataway, NJ, USA.

Interview

Interview: Richard Snelling

Interviewer: David Morton

Date: 25 April 2000

Place: Atlanta, Georgia

Childhood, family, and educational background

Morton:

The purpose of this interview is to gather information about the history of the Communications Society during your involvement with it, to talk about your career as an engineer and as a businessman, and to have a biographical statement from you. The standard first question is could you tell me a little about where and when you were born, and something about your early education? Also, how did you become interested in technology and how did you end up in this field?

Snelling:

You have my bio, so you have a reasonably good framework to work with. I was born in St. Petersburg, Florida in 1931, which is generally conceded as being the bottom of the Depression. So, I grew up in that era. I had an older brother who was an engineer.

Morton:

Were your parents engineers?

Snelling:

No, my parents were not. Actually, the family business was women’s hats. Five boys in women’s hats.

Morton:

Manufacturing or selling them?

Snelling:

Selling them. We also made hats as well and subcontracted hats. Hats that are specifically made for a person. I grew up in that kind of a location and environment. I lived my first eighteen years in St. Petersburg, Florida, then attended the University of Florida. I received a bachelor's degree in Industrial Engineering. The reason for that particular curriculum was that they offered an engineering degree and I wanted to be an engineer. I had acquired an interest in matters mechanical. I was quite impressed with my older brother Bill who had become a naval pilot in the second World War, who was engineer.

Morton:

You must have just missed the War.

Snelling:

Yes. I was ten when the War broke out.

Morton:

So you would have been in college just after the War. Were the veterans still there or had they moved on?

Snelling:

When I got to college, it was 1950. The Korean War broke out in the summer of 1950, and the War went well until about October. That is when we got to the Yalu River and a million Chinese came across the Yalu in a wall of humanity that swamped the Marines, and just about emptied all the campuses in the United States because everyone started jumping for enlistment. I stuck it out at the University. Actually, I was in a Naval air squadron that my brother was the Executive Captain of, and basically did my military service in the Navy in various capacities as a reservist, going to college until I finished. That was about the same time that the Korean War finished.

My interest was always in electrical engineering. My life plan was to go work for a large corporation. In fact, my Senior Seminar professor, who was the Department Head, asked us to write down what we wanted to do with our lives, which I did. I put down is what I have actually done. Ironically, I wanted to go to work for a large corporation that had stability, but enough technology that I could use my skills and where I would have a chance to advance through it, and ultimately to become an administrator.

Morton:

Was there an electrical engineering component to industrial engineering program?

Snelling:

Yes. Industrial engineering at that point had shifted away from mechanical engineering, but was more related to the economics of engineering, but coupling the various disciplines. I took as many electrical courses as I did anything else, and I took business administration courses too. But it was an engineering degree.

Employment at Southern Bell; graduate studies

Snelling:

When I graduated, I spent the first few years of my career with Southern Bell in what they call outside plant. You would go out, take field notes, design the outside plant construction, shift it to a construction crew, and they would build the plant. I liked that. It did not have an awful lot to do with engineering or with electrical, for that matter. It had more to do with structural or civil engineering.

About eight years into my telephone career, Bell System had this notion that the half-life of an engineer was about ten years and that they needed to go back to college or receive some formal training. So I did that. By that time, I had begun to specialize in transmission and special service engineering, which was actually one of the few design components of engineering in the old Bell System in the operating company outside of Bell Laboratories and Western Electric.

I went back to college and took what would lead a master’s degree in electrical engineering. From there I came back and took my professional registration. I received my professional registration in the State of Florida in the field of electrical engineering, because that is what I really wanted to do. I continued to work my way through the ranks. There were some twenty-two steps at that time to go from student engineer, where I had started, to Chairman of the Board of AT&T, which I had declared to myself as my principle objective in business life.

Morton:

That seems reasonable.

Snelling:

Audio File

MP3 Audio
(395 - snelling - clip 1.mp3)

It is interesting. The person who had hired me for Southern Bell was the Chief Engineer for the State of Florida. I had been accepted but I had deferred my actual arrival at the Bell System for a year because of family reasons. My father passed away and my mother asked me to stay in St. Petersburg to help run the family business. I got married during that time and decided that I wanted to be an engineer. The man who interviewed me interviewed me personally for three hours. As I said, he was the Chief Engineer for Florida—a prestigious position. At the end of the interview he said, “The reason I spent so long with you is that I have hired many people and I have never had anyone quit. I think you might quit. I don’t want to break my record, and so I want to be sure.” I replied, “I promise you, I will not quit.” And I didn’t. That was really the way it was. You went to work for the Bell System and forty years later you retired, and had a happy life.

Contrast that to 1985. I had made a presentation to British Telecom senior executives in London because they were about to go through a competitive environment. I did about a three- or four-hour presentation of all the things that I thought it would take to get ready for competition. By that time, I was the Executive Vice President for Bell South, Southern Bell. I did my three-hour presentation about the architecture of the Network. At the end of that, they had several questions. The first question was, “How do you like competition?” And I said, “I love it.” The man asked, “Why?” I said, “Because the pay is better. When my competition came in, other job opportunities became available, and Bell South doubled my salary in two years. I like that.” And that contrasts the old engineering lifestyle with the non-competitive environment and the few opportunities that go with that.

Skipping back to the career process, my career shifted a considerable amount with going through the regional transmission school. That was a fast-track program. I was hired as a trainee; they called us college trainees. At that time, it meant that either you were promoted in four and a half years, or you were fired—one of the two. I was not promoted or fired in four and a half years and neither was anyone else that I ever saw who had been in the program. After that, I began to shift around and to do something interesting things.

AIEE activity; work with NASA

Snelling:

My first professional activity with the AIEE was in 1956. I actually joined the Orlando section of the AIEE.

Morton:

That is interesting. Why not the IRE? Or was the AIEE stronger?

Snelling:

All of the Bell System was AIEE. The only people who were in IRE were electronics types. The AIEE was principally composed of the power industry and the telephone industry. AT&T had been a very strong supporter of AIEE. In fact, there was quite a lot of rivalry between the two. Most of the IRE people looked down on the AIEE members as kind of dummies who had mundane issues. Conversely, the AIEE people looked on the IRE people as being flaky or goofy types of people. So the wedding of those two was not particularly easy, by any stretch of the imagination. There were some people who had been in both organizations, but they were rare.

I saw a good bit of that because one of my earlier opportunities was at Cape Canaveral. I became the Supervising Engineering for Cocoa Beach, which was where the Space Program was centered at the time. My role gradually shifted to being one that was increasingly oriented toward technical support for NASA and for the Air Force, and for the various space programs.

Morton:

Still as a telephone company employee?

Snelling:

Always. We will not change from that until thirty years later.

Morton:

That is what I thought. That’s why you were providing communications services for NASA?

Snelling:

There were a number of factors that were unique at the Space Program. Number one, NASA was a civilian agency. Up until then, only military agencies had done their own contracting. Otherwise, it was done through the Bell System. NASA wanted to have their own contracting capability, so there was a considerable amount of discussion between the Bell System and NASA about where that contract would lay. One of my jobs was to be the technical liaison for NASA at the Kennedy Space Center. There was a special team put together to do that. But I had other duties as well.

While I was at the Kennedy Space Center in Cocoa Beach for the Space Program, I had moved my way through the Chairs of the AIEE, and ultimately I became Chairman of the section. It was certainly the largest in the southeast at the time; it may have been the largest in the country outside of New York, perhaps. That was when the IRE and AIEE joined. It was interesting because most of the Space industry was located there: Lockheed, Boeing, Pan American, GE, IBM. They were Space Row contractors. Many of them were in fact IRE members. So, it was an unusual Section because there were more IRE people than there were AIEE.

Morton:

What was the first joint meeting, like?

Snelling:

It was interesting in that sense. Of course, my background was more technical than most of the AIEE members, so we seemed to get along fairly well. One of the interesting things about that tenure was that we entertained the Board of IEEE. It was one of the first times that the Board had met as a Board. They came down to Florida to meet. We had someone loan us Rockefeller’s yacht and had the meeting aboard the yacht.

After my Cape Canaveral Cocoa Beach experience, I went to Jacksonville as a planning extension engineer, which was fundamentally administering the construction program for the State of Florida. I went on to some other administrative jobs. I ultimately ended up in Atlanta as Engineering Director for Southern Bell. I did some cross training in Atlanta, in the plant, and became the Executive Vice President for Southern Bell and Bell South.

Transformations in the telephone industry

Snelling:

Some noteworthy things during that period of time was the transformation from what might be described in the telephone industry as a tip and ring to electronics from the transmission standpoint. All long distance, obviously, had been electronics for many years, but the actual local loops were in fact pretty much base pan upon tip and ring copper. The development of carrier on copper had some pretty pitiful results originally. By the way, those kind of results are reflective of the difficulty currently in trying to get carrier put on copper in the form of DSL and ADSL.

Morton:

That is interesting. To be sure I understand what you are saying, but can you explain that in more detail?

Snelling:

Well, you take one pair of copper wires and multiplex it with some sort of carrier system and derive twenty-four channels for a T1 line, or something different for other lines. Currently one of the main industry efforts and where a lot of money is going is trying to do the same things, trying to put carrier on copper wires with similar kinds of results.

Morton:

What are those results?

Snelling:

Copper plant in the distribution does not support carrier very well, for many pragmatic reasons. Water in the cable is one reason. Water in the cable is bad. Corrosion is another reason. Bridge tap, which means bridge wires onto the line that may not really on the record. Et cetera. So there is a variety of things which make that difficult. The point being that starting back around 1970 there were fairly extensive efforts made to put carrier on copper wires without a great deal of success. Which leads to some things that I think may be of some interest to you.

Morton:

That is of interest, actually. One of the things that makes this interview interesting to me is that usually I am sent to talk to people who worked at AT&T. More specifically, worked at Bell Laboratories. It is unusual that I get to talk with someone who worked out in the field. Now, the strong impression you get from the AT&T people or from reading the official corporate history is that one of the corporate goals was to supply the equipment that the local operating companies used. So it is interesting to hear that there continues to be engineering problems, even after these systems had supposedly been perfected in the laboratory. So if you have specific examples of that, that would be interesting.

Snelling:

Audio File

MP3 Audio
(395 - snelling - clip 2.mp3)

I probably need to go back a little bit. I would say beginning around the middle of the 1950s, there was a considerable effort made at Bell Laboratories, and around the world as well, to reduce the investment and the inflexibility of the outside plant. The outside plant typically was sixty percent of the investment of any telephone company in those days, twenty percent being the central office facilities, and the rest was spread around. So the big bad actor was the outside plant. There were many efforts to reduce that investment. One was ready access, where you just threw a bunch of wires into a terminal and then connected them willy-nilly and hoped that everything worked out all right. We used color-coding, spliced green to green, and blue to blue, et cetera. That did not work out so well. Then they developed a plan called Dedicated Plant, which was a modified way of putting the outside plant to the living unit and the building location. Dedicated Plant was a giant step backwards, actually, from an operations standpoint.

Then about 1970, there was a lot of field input and there was an effort to begin to put carrier on the outside plant, which began to take hold. Unfortunately, the first efforts were analog, whereas digital had been quite successful in the trunk and the toll plant. It had not been applied at all to the outside plant, and the first efforts were unsatisfactory.

What did happen, though, about that time was a permanent plant concept where you would take the plant to the neighborhood and you would perhaps have carrier on that. We began to use slick carrier—a digital line carrier—and applied more or less dedicated facilities from that carrier point to the main unit and provided extra pairs. This is where we are right now, basically with fiber to the permanent plant points. Feeder Distribution Interface is what they are called.

Now, the real challenge is what to do with what they call “the last mile.” It is actually more like 2500 feet. The notion being to put carrier on that last piece. A good bit of the industry effort, both by the carriers and by many of start ups right now, is centered around how to build using sophisticated algorithms to overcome those unpredictable irregularities on that plant.

So the problem has been a thirty-year problem that continues to thrive. My interest in all of that has always been first to do the best you could with what you had. But also to kind of look towards a more distant horizon and try to figure out how to get yourself out of this ditch that you are constantly in—this money pit that you are throwing your resources down.

One of the false efforts in this respect was the use of an aluminum plant. That had a disastrous effect because while copper is not a wonderful transmission medium, aluminum is worse. Aluminum is much more sensitive to corrosion and is not as good a conductor and handles considerably poorer. So a lot of aluminum plant was put in the ground. And many countries went whole hog for aluminum plant offshore. Great Britain being one of them with British Telecom.

Morton:

How long did it take them to start realizing that was wrong?

Snelling:

About five years. First of all, they knew it was hard to splice. But that also is one of the reasons why a good bit of the emphasis for wireless technology has come out of the UK, because they had this tremendous problem thirty years ago.

Communications theory; dynamic between Bell Labs and Network Heads

Snelling:

By the way, during the time that I spent in transmission, I began to focus professionally more on the theory of communications than just the pick-and-shovel work. It became clear to me that following some of Shannon’s information theory relative to digitalization of the network end to end. Also, other developments such as fiber optics (or the photonic development) was probably the longer-term solution to the dilemma of investment and bandwidth increasing. It was clear to me fiber around a photonic digital architecture would be the ultimate solution.

Morton:

When you say that you began to focus on that professionally, I am not sure how you did it. Was this a personal interest that you took in this, or were you involved in strategic planning?

Snelling:

Well, being at Cape Canaveral, I was exposed to high technology probably at its most advanced stage. At one point, about twenty-five percent of the available engineering talent in the country was directed to the Man-on-the-Moon project in some way or another. All of the new technology was centered around that. I was the technical liaison for the NASA involvement. We built many sophisticated facilities for them. We also began one of the first efforts for electronic switching commercially at Cape Canaveral on the first T1 carrier. It was put at Canaveral mainly because of the nature of the technology effort at that time. So I was exposed to it early.

But also, as I moved up the ladder and became Engineering Director, I had a two-billion-dollar-a-year budget to administer. More homes and businesses were under development to accommodate than resources. So, developing better ways to provide the service at a broader bandwidth was my job.

Morton:

I guess I am still unclear on that because I do not really understand the situation you were in. Again, coming from a perspective of someone who has only spoken with people from Bell Labs, the impression is that they developed the technology, and the operating companies bought it and installed it.

Snelling:

Let me give you a little perspective on that. The Network Heads were an organization within the Bell System as a network for the twenty-three operating companies. They typically met quarterly, or four times a year, unless there was some special reason to meet in between. They were a group of all men who took the material from Bell Laboratory and Western and AT&T and deployed it or didn’t. They had some latitude to do that because the old operating companies were profit centers. So, the main objective for Bell South/Southern Bell was to have the best rate of return on the assets in the Bell System. We had the best rate of return in the Bell System for ten straight years. The best rate of return was how it was measured.

There was a lot of latitude. That group was a very powerful group of people. In many respects, they funded the projects. There was a lot of interaction between Bell Laboratories and Western Electric and those Network Heads. Particularly towards the end of the 1970s, there was considerably more input from the field Network Heads. It was quite an unusual group of folks. As you can imagine, they were pretty rough and ready. You succeeded as a Network Head by being an aggressive, hard working, knowledgeable person who could adapt and handle most situations and keep the politics of the situation under control. By being responsible for a lot of money, they were all pretty heavily engaged. They began to break away from the Bell System products about the end of the 1970s, and I think Southern Bell put in the first Northern Telecom Central Office in Southern Bell in 1980. That was a real interesting experience.

So part of my job was become knowledgeable. I taught and gave a lot of speeches. My guess is that if somebody wanted me to give this interview, probably one of the reasons I gave a ton of speeches, first in the south, then regionally, then all over the country, and really all over the world. I think at one time I was averaging fifty or sixty external speeches a year and about that many internally. It was centered around making the network more flexible from a bandwidth standpoint; putting more intelligence into the network; using operation support systems to make it an automatic network, untouched by human hands; and to provide a high level of flexibility for customer access and control. That is quite a different network than had been built around the world. Of course, about that time there was an awful lot of Justice Department activity with Judge Green and the antitrust case. It was evident that something was going happen with the Bell System. In some ways, there were going to be fewer constraints on the part of the Network Heads to make design decisions from a cookbook, if you will.

There was an interesting meeting at Bell Labs about the middle of 1970s. We were in a service crunch in Chicago and Miami and New York and so forth. At that meeting, the head of Bell Laboratories cancelled the fiber optic project, the intelligent network project, and the automatic network project all in one fell swoop. This was a thing that drove a lot of the operating people up a wall, because we all wanted those things.

Fiber network

Snelling:

As competition began to develop from MCI and others, it became evident that we had to have a lot of capacity which did not involve a lot of investment. Fiber was the only way. We put in the first commercially viable fiber trunk route into Atlanta in 1979. GTE has a great ad about fiber optics in Chicago that they ran for years and years and years. But it was never commercial. Ours actually carried traffic. It was obvious to me that as soon as we put it in that we didn’t have any troubles. I figured we must have gold-plated it. But when we put some more in, we still didn’t have any troubles.

That is why I am interested in the characteristics of fiber. It not only has bandwidth almost to infinity, depending on quality and what carrier system you put on it. But by its very nature in photonic information theory development, if you will, coupled with digital deployment, then the trouble recurrence goes down enormously. What I was seeing in my fiber network was that it had only four percent of the problems of a conventional network. That has a big implication on cost, service, and just being able to sleep at night.

Morton:

That first line, was that a direct replacement for some section of trunk?

Snelling:

It was a direct replacement for a copper trunk that had T1 carrier on it.

Morton:

Integrating that into the network wasn’t really the problem. Once it was functional, it was invisible.

Snelling:

One of the criteria for integration of anything into the network—whether it was one hundred and twenty-five years ago or thirty years ago or today—is that it must be compatible with everything that has ever gone before in some way. And it must be compatible with everything that you are going to do afterwards, which is a big design problem. It is a problem which the computer industry has never addressed. I don’t know if they would have gotten where they have, and I don’t know that they should have, and I’m sure that compatibility was the right for the telephone industry to do it. Virtually everything in telecommunications works with everything else in the world. It works with everything that ever was and everything that will be.

So here I am, the telephone representative part of this operation. A lot of electronics people and consumer products people had a hard time with that. They figured the way to make a success in a business is to have a new product issue come out each year, obsolete the product you had last year and make another set of customers. Well, that’s not the history of telecommunication. So that was a fairly straightforward conflict.

IEEE activities and merger; SuperComm convention

Snelling:

Let’s go back to my professional activities with IEEE and some of the interesting aspects about that. The marriage of the two organizations, as I said, was not easy; it was contentious in many ways. In fact, many of the IRE people thought that they had been dragged down into the mud, and vice versa.

Morton:

You didn’t actually answer my question earlier about the first joint meeting. Were you active when the IEEE and regional activity started taking place? What was the atmosphere?

Snelling:

I was actually the Chairman of the Section. I ran on a joint platform of AIEE and IRE. I guess I became Chairman mainly because I had gone through the Chairs. But I knew many of the IRE people from my contact with them on the various committees. I remember that somebody ran against me, which was unheard of at the time. It was an IRE guy, and he didn’t get it. He was very upset. He wrote some nasty letters. But other than that, it was pretty professional. I would say it was not an easy marriage in many ways, although it sounded that way whenever you read anything about it.

Morton:

It strikes me as a hard thing to integrate, especially at the regional activities level, because you have to begin holding meetings that supposedly appeal to everyone. You have to find speakers and find conference topics that will appeal to a broader group all of a sudden. How did that work out?

Snelling:

Well, of the two organizations, my recollection was that the AIEE was much larger than the IRE in membership.

Morton:

At the time, they were comparable. Regionally I’m not sure, but overall, the IRE actually surpassed the AIEE.

Snelling:

Yes. I think regionally they probably were. But in our case, that was just the opposite. IRE was by far the larger at Cape Canaveral and Cocoa than the AIEE. So we quickly adapted the policy of having technical programs. For one thing, the AIEE people liked it better anyway. The AIEE members were going through the same sort of technical revolution in their businesses and they were able to get a lot out of it. You could only have so many highway right-of-way programs that were interesting.

My involvement with IEEE at this time was more centered around my duties as a Network Head for Southern Bell. I keep saying that because until 1990 Southern Bell and Southcentral Bell were separate companies under which Bell South was the holding company. It had separate organizations and staff. My last title was Executive Vice President for Bell South.

We had a series of Southeastern Conferences that became more and more interesting. I was Chairman during the next noteworthy thing that occurred in 1990. We had the first joint meeting of SuperComm and IEEE.

Morton:

What is SuperComm?

Snelling:

SuperComm is a large convention that occurs yearly. It is the TIA, the Manufacturers for Telecommunications, et cetera. It will typically have twenty to twenty-five thousand attendees. It is more of an exhibits-type conference, but they were beginning to develop technical programs as well. They were quite eager to have a joint conference with the IEEE and the Comsoc organization. So we elected to give it a try with the ICC, which is a spring IEEE conference. It coincided well with when SuperComm was normally held. We elected to do it in Atlanta, and I was the Chairman of it.

That was an interesting exercise. The ICC normally runs a thousand, maybe fifteen hundred people. SuperComm runs twenty to twenty-five thousand people. One is technical papers and the other is management/technical applications with a more political emphasis. So pulling those two together was challenging. Of course, SuperComm had its own constituency. At that time, it was TIA and other manufacturing organizations. They had a professional staff that did nothing but put on SuperComm, and of course the IEEE staff was all volunteer. They especially wanted to do it in Atlanta in conjunction with us because it gave them access to the Southern Bell, Bell South, AT&T and Nortel volunteers.

During that convention, we did some things that were exciting. One was the Super Home/Super Office, which was an illustration of new technology and applied technology for those two environments. We asked various manufacturers to contribute to that, and then we had our exhibits based on what manufacturers offered. We had about twenty different exhibits of different rooms. You would go into one room and see an ATM, then go into another room and see a photonics switch. Some of these devices were never developed, but we were demonstrating the prototypes. And the same thing was true for the Super Office.

The conference turned out to be a financial success for both parties. I have continued to do that ever since. We had another joint conference in 1998 in Atlanta of the ICC and SuperComm. I was not Chairman of that; Bill Smith was. He is the Executive Vice President of Bell South. But I did do the Super Village. Various manufacturers would provide different aspects of their technology that they wanted to demonstrate, which was a little bit different than just going to a booth. You could see a whole village with different applications.

We might shift gears a little bit and backtrack and go back to some of the developments in IEEE and my involvement in that. I served on many different boards, publication review boards, et cetera, just in the normal course of events. I did serve for about five years or so on the ComSoc Board of Governors or Directors. There I served on the Executive Committee to provide administrative direction for the Society.

During that period, Maurizio Decina was Chairman, not for the entire period, but for a good bit of the time. He’s a professor at the University of Milan. He was at Bell Labs and he did early work with ISDN. He is also at the Telecommunications Center at the University of Milan. He has done a great deal of activity with Telefonica Italiana over the years. His notion was to globalize IEEE to make it not just an American society that has different offices around the world, but to begin to diversify the actual administrative and technical thrust. Part of Maurizio’s notion is that Americans tend to be…less respectful of other points of view than they really should be. Part of his notion was that if you could globalize IEEE, you could go a long way toward getting better world standards, because the United States typically goes it alone on standards. That used to work fine when there was an AT&T that set the standards for the world, but right now it is not working too well for the United States because we lag in standards.

During this period of time with the IEEE, both professionally and intellectually there were many things that I wanted to be engaged in. One was the notion of expanding the bandwidth of the network and the use of fiber to do that. I am quite satisfied and gratified to see all that developing.

Georgia Center for Advanced Telecommunications and Technology; 1996 Olympic Committee

Snelling:

Audio File

MP3 Audio
(395 - snelling - clip 3.mp3)

I retired from the Bell South at the end of 1991. When I left my job at Bell South, I became the Senior Technology advisor for the 1996 Olympic Committee. I thought that was what I was going to do for the next five years. I didn’t because about that time a center that I was interested in at Georgia Tech, the Georgia Center for Advanced Telecommunications and Technology, was funded by the governor. The person who had accepted the job and was going to become the first Executive Director, but he went to Australia to become the Australian Telecom’s CEO instead. So I agreed to become the Executive Director and Founding Director for GCATT. I thought I would do that for three or four months until we could get another CEO. Two and a half years later, I finally stepped down as Chairman and let somebody else come in. It took two and a half years for that particular duty to be shed. In the meantime, I stayed with the Olympics for about a year, year and a half, until we could get all the specs written for the various products. I was really doing a lot more than I wanted to do in retirement.

One of the interesting things about that period was at GCATT the notion of dealing from an academic, governmental, and an industry standpoint with the conversion of the various industries, which could be defined as computing, telecommunications, CATV, content origination, and the fifth one is consumer electronics. We received funding from the state and from corporate organizations, Bell South, AT&T, IBM, Turner, and Cox. The idea was to throw those various industry representatives together into a pot and come out with collaborative research that would transcend the normal component or single item research that is typical, even in Bell Laboratories. It is actually a center at Georgia Tech.

At the Olympics, the technology committee or the technology group was under funded. When they submitted the bid for the 1996 Olympics, the committee left out funding for technology totally as a separate item, which turned out to be a one-hundred-and-twenty-five-million-dollar problem that caused a lot of grief. Well, one of the ways we got out of that was to get the corporate sponsors to designate people to come and be part of the technology groups. Their salaries, at least initially, were paid by the corporations. I did it on a volunteer basis.

We were exposed to the fact that everybody wanted everybody else’s businesses. Motorola wanted to do networking, IBM wanted to run a network, and Bell South wanted to do wireless. It was just a mish-mash of musical chairs. Consequently, the technology group was exposed to leading-edge technology from companies to use their technology for the Olympics and we were going to showcase technology for the 1996 Olympics, which in some ways we did, in some ways we didn’t. We also formed a technology advisory counsel of various technology sponsors. But also, interested parties like Australia who have bidden next, and Germany, et cetera. So, there was an opportunity to see the world of telecommunications from a development standpoint at the leading edge. At GCATT, I watched that same sort of integrated development in theory but in practicality at the Olympics, to watch those convergent industries begin to merge and smash through each other. It was exciting to me.

Intelligent network devices development

Snelling:

Part of what I had worked on diligently at Bell South over the years was the realization that so long as you had a black telephone interface, the interface device could be pretty simple, but if you were going to market sophisticated services that would include call forwarding and three-way calling and even not very sophisticated services, you needed some sort of intelligent device that would aid the user in accessing the complicated network with a complicated array of network services.

The current Chairman of Bell South is Duane Ackerman, a long-time friend and colleague. He was the staff person at Bell South for the two operating companies, and I was the Executive Vice President for the Southern Bell piece. We issued jointly two RFPs (request for proposal), one in 1987 and one in 1989, for a device that would be an intelligent network device.

Morton:

What is an RFP?

Snelling:

Actually, they were RFIs, Request for Information, because we weren’t ready to let bids. But you ask the manufacturers for what they can do to solve a particular problem. You write out this sponsor’s specs. The specs we were writing out were basically an integrated terminal that would provide voice, data, and video termination and let the user pick and choose bandwidth on demand and have an intelligent aspect to it. We issued those two in 1987 and 1989, and we had some forty-five or fifty responses the first time and maybe twenty the second time, none came close to accommodating what we wanted to do. One of my objectives all along was to someday have such a device that would do that.

Morton:

That’s very interesting, because that’s so early, in a sense. It is pre-Web, so you obviously aren’t talking about household consumers. Who were your imagined customers?

Snelling:

Actually, Southern Bell and AT&T and Knight-Ridder had a joint venture in 1982 in Miami on videotext. The device they had was basically just a clicker that you would use for a television set today. But it was a clicker for the TV set, actually, but it had a set-top box on the TV.

Morton:

Was this imagined as a cable-type of service? Or a twisted pair?

Snelling:

Twisted pair. The notion of videotext was to have slow motion video, for one thing, but mainly to have text and to do it inside a normal telephone channel which was three thousand-hertz, which is a fairly narrow bandwidth. The problem with three thousand hertz is that it was like a straightjacket for the industry. This is why I was so interested in getting out to carrier.

The project was ultimately discontinued in 1984, not because there wasn’t interest and not because it wasn’t technically feasible, because it was in both cases. By the way, that was when the first packet switch was put it. It was put in to service that particular application back in 1982. IP-1, as a matter of fact, at Western Electric. But the problem was the user interface. There wasn’t an adequate user interface, and it cost too much money. It started out at six hundred dollars and with volume went to eight hundred dollars. Ultimately, Knight-Ridder wrote it off. We had a chance at that point for the newspapers and the telecommunications industry to come together, and it fell apart and things became divisive between the two. The newspaper industry began to see the telecom industry as electronically supplanting their classified ad pages advertising. Which is a very real thing to worry about.

So that was the origin for this notion of a sophisticated device. But there are other reasons. Penetration of call forwarding and all the vertical services were in. If you read their annual reports, most of the telcos are getting their growth from the idea was that nobody could remember what the codes for call forwarding and three-way calling and so forth. It had to be simpler. So I always wanted to have that. But also, there was the notion of unlimited bandwidth, automatic network, intelligent network, voice/data/video bundled together, and some sort of accommodation of the CATV industries. I used to run around the country and talk about how we could accommodate with the CATV industry. I took a lot of cross-examination in the depreciation hearings from the CATV industry. It was a very hostile for that period. Their contention was that there was no place for fiber in the network. Of course now, they probably have a far higher percentage of fiber than in telecommunications industry.

But there again, all this time this device kept cropping up in my thinking. At GCATT we developed a National Science Foundation funded activity with Corning, Nortel, Bell South, Hughes, and some others who I forget, to develop a gigabyte network, an all fiber network, an all optical network, including the switches. What that would do for you and how you would do that, et cetera, and actually demonstrate it. But the more important part of that particular activity was the coming together of the above parties for the fiber and the sophistications of fiber and the use of a higher slot the fifteen hundred nanometer bandwidth then also to build the photonic components that would go with that.

So, we began to develop architectural notions of fiber rings around the world and fiber rings around Georgia and so forth. So a lot of the fiber ring activity that you see right now is quite satisfying to me, going back ten years ago.

Interactions of telephone companies and equipment manufacturers after Bell divestiture

Morton:

You mentioned the RFI procedure, which I have never really heard of before. It leads me to the question of how a telephone company would communicate its needs or its desires to equipment manufacturers, particularly after the breakup of AT&T, where you no longer had an easy corporate channel already in place. What was the relationship between the people developing new switches and the system?

Snelling:

It was a mixed bag, and still is. They would come in with a product that is not a Lucent or a Nortel or a Cisco. You have a new product—a small company trying to get into a major telecommunications major player like Southwestern Bell, they have a tough time. But you also have some constraints by the Fair Trade Administration and the Antitrust Laws, et cetera, in which you have to deal equally with all people. You also cannot have cross subsidization between product lines. There are many legalities involved.

One of the mechanisms to overcome those and let people have fair access to the organization is to have a Request for Information. With something like an integrated terminal, which is what we called it, there are only five companies in the world who could do it because of the need to have an interface with the transmission, linkages, central office switch, and the consumer electronics piece on the end. I would say there were five in 1987 there are probably fifty by now.

They would come in with price targets and you would give them price targets. You would tell them what you wanted to do, and they would come back and respond, “Well, we don’t have this but if you give us a purchase order, we will develop it.” You would go from there.

There were also problems during that period because under Judge Green’s constraints on the Antitrust Divestiture Act was some restrictions on customer premise equipment. He never actually said you couldn’t do that, but everyone was afraid to move into consumer electronics or the consumer activities on an integrated basis. In fact, he took the consumer telephone sets, basically, and turned them loose. It became a forest of problems relative to termination of the network. I think that was one of the major flaws in his ruling.

Bellcorp was considerably effected by that. I was the Chairman of the Applied Research piece of Bellcorp for five years. I was the first Chairman when it was formed because I was the Bell South representative that formed Bellcorp. I then stayed on as Applied Research Chairman for five years. We did many projects, but we never really did the one I wanted to do which was the Integrated terminal. The reason was the legal concern that we were getting too close to a violation of Judge Green's consent decree.

Morton:

I seem to recall seeing Bell South branded telephone sets. How did that get around the ruling?

Snelling:

They would just buy them from Western. I remember the Western Vice President came into my office after the divestiture was announced and put down a bid for our telephone set business. I had to tell him a couple of months later that we had given the business to someone else. I thought he was going to have a heart attack in front of me. It was quite a shock. And on the PBXs, I remember Western came in and said that they would be glad to give us their last year’s version of the PBX, but the new ones coming out were on reserve for AT&T. So we said no thank you, and we went to Nortel and bought our PBXs from them.

Morton:

I saw some statistics in the paper the other day about the number of people who are still renting telephones from AT&T. There are still quite a few of them.

Snelling:

Many people don’t like to change.

Board memberships: DCA, Boston Technology, VSI, Digital Wireless

Snelling:

So, I am working my way through the ‘90s. I had retired. First I was heavily engaged at GCATT, and then towards the end of that time I joined a lot boards and said, “Okay. I would like to be part of governance in the various industries.” So I was a board member of DCA, which manufactured the ERMA board and that basically took the PC to the mainframe. We took that company private for many interesting reasons.

But it exposed me to the consumer electronics world and computing world in a way that I had never been exposed before. Joined the Intergraph board, which was a mapping type company. Then I joined the Boston Technology board, which was a voice messaging company, quite similar. In fact we had used Boston technology equipment when I was at Bell South. It rather surprised me because we always figured that when the Judge let us do voice messaging, we would probably use Octel because that was our internal suppler of voice-messaging equipment. We had built up an internal network to get experience with it, and Octel was the only game in town. And then when the time came to deploy in Miami, out popped Boston Technology with a bid that beat them all over the place, so we deployed Boston Technology. John Taylor was the President and CEO of that particular organization. I joined that board. Gregg Carr was also the Chairman of the Board and founder of it.

I joined another board called VSI, which is a videoconferencing company. I was always intrigued with the notion that videoconferencing is an efficient substitute for transportation. VSI is a publicly traded company. One morning I woke up and was having breakfast with my wife in my pajamas. We were about to go out and play golf. I received a call from our VSI corporate attorney and he said, “I needed to come down here because all the other board members except you and one other have resigned. This company is out of business because we have to have three board members. By the way, we don’t have anybody to run it because the Chairman and CEO was one of those who resigned.”

So, I went down and we elected another board member through a legal ruling. I agreed to be the CEO and Chairman until we could find ourselves a CEO and Chairman.

Morton:

I have spoken to a lot of fellows who when they have reached a certain point in their career are asked to join the boards of various companies, many of which had nothing to do with their prior careers. I always wondered what exactly being a board member involves.

Snelling:

I’ll give you two contrasts. I was on the Southern Bell board for nine years. It was a very prestigious board. We would go to Washington and seventy-five percent of the legislative delegates, Senators and House members, would turn up. They were a prestigious board.

I never had a serious challenge to my construction program. I would present it in very great detail with all the acronyms. But everyone always voted yes and off we went.

When I went to DCA there must have been fifteen different motions in the first fifteen minutes. The board was running the company. So it varies a great deal. That was an advisory/governance board. So they vary quite a bit depending on the circumstances.

VSI was a troubled company. When I became Chairman, the stock was at nineteen cents a share and dropping. So, keeping VSI afloat, and there were a lot of good reasons for doing it, encompassed a lot of my activity for a long time.

Morton:

What had happened to that company?

Snelling:

It went through a period when it launched a new product with some intellectual properties that it had, patents, and simply ran out of cash. There were some delays in the product deployment, principally from software. A common story. It basically got some loans that put the assets of the company up for grabs. So making payroll was a challenge the first time. I learned all about going from a big company where you have a two billion dollar construction budget and a two-billion-dollar maintenance budget to a company that is fighting for next week’s payroll, which is maybe a hundred thousand dollars.

Morton:

What happened? Where are they now?

Snelling:

They are still operating. Six weeks is how long I thought I would be there. It turned out to be more like three years. I finally brought in a CEO that actually came from Bell South, a lady and a colleague of mine from Bell South days. She was CEO for about a year and a half. She is still on the board, but I’m no longer on that board. So I was very interested in videoconferencing. Still am. And still have a lot of their stock.

I also became intrigued with wireless. One of the notions that I had over the years—and it goes back to some of the original information theory—was that landline is good for bulk and reliability, and wireless is good for flexibility and portability. Just that simple. You need a combination of the two to do an overall telecommunication job. I joined a board called Digital Wireless, and that was a spread spectrum company, also a private company. It still is, although they’ll probably go public soon. The Chairman of that was Stucky MacIntosh.

I kept agitating for a wireless integrated terminal that we hadn’t been able to pull off at Bell South. Digital Wireless had a wireless spread spectrum telephone that was very good. In fact now it is licensed to go offshore and this is now the principal wireless portable telephone in the US with that technology. But the Chairman had concluded by that time that it wasn’t really the right way to do this wireless system that we were contemplating which was going to be a network. So he and I began to talk about how you would build an integrated terminal and how it might fit into a wireless world. If you turn around and look at the plaque back there, that is our first patent, and it is the first of the home wireless network patents. We now have eighty-seven. The notion there is that you would build an integrated terminal that would have user access to help the user interface with prompts et cetera, as well as a display device.

Home Wireless Network company

Snelling:

Audio File

MP3 Audio
(395 - snelling - clip 4.mp3)

I want to digress a bit and go back to the Nortel days. Nortel put out a display telephone back in 1982. A display telephone that had a very simplistic display on it. Unfortunately, it didn’t have any brains behind it. But that device coupled with the other CPE of products of Northern Telecom at the time had us shift, at least, from our purpose of predominantly Western Electric to Northern because of their CPE and their ability to support Centrex services, et cetera. The person who was originally working on that was John Taylor at Northern Telecom. He is also the originator of Fiber World, which is access known currently node for Northern Telecom. That is where Northern gets most of its revenue stream today.

The reason I mention that is Taylor showed up again later in my life at Boston Technology as CEO. Stucky MacIntosh being at Digital Wireless, so we filed for that patent under the Digital Wireless corporate structure. Of course, Digital Wireless was basically an industrial component company, not really a mass-market consumer company, so they were not going to develop that product. So Stucky, John Taylor, and I founded Home Wireless Network. Stucky was the RF specialist, particularly spread spectrum and TDMA. He naturally had a great deal of activity with the FCC in developing standards, doing licenses and auctions, et cetera. He has a lot of prestige in the FCC relative to RF. More importantly, he has a lot of knowledge. Taylor comes from a manufacturing background end that is optical, but which also has roots in consumer products. Then I come from a telecommunications background.

So we formed the Home Wireless Networks company. The first company that gave us a conditional PO was Bell South. The conditional purchase order (in letter form) basically stated if you do these things, (the things being pretty much what had been in the two RFIs) they would consider purchase. In other words they were interested but not committed.

With that, we went trotting around the world to Japan and Germany and the US. Lucent was our first investor, and still remains our principal investor. We formed the company and we now have a whole bunch of patents. The patents do the things that we set out to do.

I wrote this paper in college about what I wanted to do with my life—I wanted to found my own company that would do new, interesting, and intellectually challenging things that would be of benefit to others, but that would also be rewarding.

We have now launched this product and it interfaces with a network. So you can terminate a twisted copper to it. By June hopefully we will have a DSL or the digital interface incorporated. By September or October we will have it integrated.

Morton:

What is the product, other than the part that connects the network?

Snelling:

It is best to think of it as a platform. It is a platform that will interface the outside world to the inside world, and provide an intelligent node to the user for interface purposes. We terminate copper wire now. It can ultimately terminate optical. We have an architecture that we have applied for a patent for a fiber-wireless-wireless network.

All along in this process there has always been the cost dilemma of getting fiber to the home. Even in the middle of the 1980s, the reason I had all those unpleasant cross-examinations from the CATV attorneys (eight hours, ten hours, twelve hours in public under oath) was because it was very difficult to justify the cost of fiber versus copper if you are just doing voice. Video was not allowed by the consent decree. I had always contended that we could, in fact, just do voice on optical. With such a device as this, if you integrate sophisticated services such as home security and home automation array of services, you return to the notion of merging the five industries applied on an integrated basis to applications.

So this product is a platform that can interface to the outside world, whether it be fiber at the full photonic level, whether it be fiber to the block or neighborhood, and then wireless in that last one hundred and fifty feet. By the way, the current architecture is to cluster roughly six to twelve living units to an optical network unit. That is how you share it. That particular architecture was developed by the Executive Director of Bell Laboratories, Jess Chunac (who is now deceased), and myself. We presented papers all over the world. Bringing fiber to the curb was the only way to have fiber deeply integrated into the network, because it was economical.

Dick McCormack, the Chairman of US West, decided he would do fiber to the home in Kansas City. So he retrofitted fiber to the home. He spent two billion dollars at it and decided that wasn’t the way to do things. You have grass and water pipes and private property and trees and shrubs. It is just not a good way to do things. So they fell back to the next notion of fiber wireless from that Optical Network Unit, which would be a cluster of six to twelve living units, depending on the geography and also the usage of the neighborhood: upscale, midscale, or downscale.

So, you take and launch the signal to the device inside the living unit, which would be a controller or an integrated terminal. On the customer side of that, you could distribute it with either the existing wiring: coaxial, twisted copper, or optical. Or you could do it on a wireless basis, so you could move around and take your laptop home and you don’t have to sit upstairs in your office; you can sit down and watch TV with the family, if you want, but you could still print to the common printer with a local area network.

So it does a local area network function and it acts like a key system or PBX as far as sophisticated services. It also supports the high level of vertical services from the telecommunications industry at this point. It would accommodate a termination from the CATV set-top box like a USB. Our product does that sort of interfacing. It can migrate forward. The bandwidth that you would need to do that would be roughly thirty megabytes to each living unit. So it is a daunting challenge. Trucks driving through will disrupt the signal at that frequency level.

You need some intellectual property, which we have applied for. The notion there is that that the Integrated terminal becomes a network element. So the platform could be a network element for simple applications for plain old black telephones where the customer wants the sophisticated services without changing out the set. Or it could be as sophisticated as a four simultaneous video channels, three of which would do standard NTSC and one high definition in some sort of termination. But also with the notion that if you have a controller that actually terminates the signal in a true sense, either phontonically or electronically, no matter where you would go, you could have your service origination be at that point. The customer would pick the bandwidth that they want on a given instant basis and pay for it on a dollars-per-bit basis with the billing and the OSS functions incorporated. If you do that, you basically take the LEN (Line Equipment Network) out of the central office and put it in the home device. So you have extended the network to an intelligent node that is at the user. When I say “living unit,” it could be a small business, it could be a large business, but the application is a different problem with a large business.

Those are evolutionary things for our product. Every time the Chairman of Bell South, and this is my long-term friend Duane Ackerman and I get close to each other he asks me when we are going to do the integrated terminal at the voice/data/video. My response is always the same thing: twelve months after you give me a purchase order. We are at that stage.

The product, as you can probably tell by now, pulls together an awful lot of things that I have worked on in my life. I am now sixty-eight years old. This is probably my swan song; at least I hope so. The notion is to build this product and to market it either through a telco or through retail. We are currently distributing it through Office Depot. Our customer is MCI Worldcom. The notion is also to build a consortium of telcos around the world with input to the product in particular difficulties. We have a collaborative consortium currently with British Telecom where we have done joint development. They have paid for a portion of it and we paid for the rest. That is in beta trial now in England, and we are under review with Deutsche Telecom.

This goes back to some of the things I talked about earlier. The same person who was on the Advisory Council for the Olympics reviews our product currently. We have now been approved technically by Bell South, US West, and Southwestern Bell, and we are under review by TelMex.

Offshore there is a straightforward standard called DECT. It is a European standard that has now been accepted by forty-seven countries. Designing into that product is considerably easier because you can use off-the-shelf technology, and the standard is such that you have a known quantity to work with. It is not as flexible. You end up with a product that may not be quite as robust as you would like. But on the other hand, the trade-off is by far the more valuable. So that European product is ready for launch, pending approval by the European community technically as well.

By the way, Maurizio Decina was on the Advisory Council for Boston Technology. I still have a lot of contact with Maurizio. I have had contact with a number of the professional people on IEEE all along. One of our Home Wireless board members is Dick Jelkut, who was the president of NYNEX and a very strong IEEE member. He has his own company now having retired from Bell Atlantic.

We just finished our last round of financing before an IPO. We have about one hundred and twenty mouths to feed times three for the children and wives. We are probably somewhere between six and twelve months away from an IPO, hopefully. We will develop along the lines of what the market will dictate, which will vary from country to country. But it will also have the fundamental notion of an Integrated platform that terminates the network in its many varied forms. As a user interface, it provides a user ease of use and prompt information.

Morton:

You mentioned its use as connecting a portable device to a local network. I suppose it could also be a voice communication device.

Snelling:

Yes. It has all the functions of an intercom set. In fact, when we were writing down the specs for it originally, I just took a 1A key system and wrote it out because I had a 1A key system going. You probably don’t know, but the 1A key system was the workhorse for the old network. It was extraordinarily good. It has never been duplicated. But it also looks a lot like a key system like NorStar, which is the leading Nortel voice system. It has a PBX key system functions for voice. It also connects to LANs and it has an automatic connection to the Network interface. Which coupled with the Telco’s desire to integrate digital carrier DSL would probably be the trigger driver for the device. The price range is between five to six hundred dollars for a starter set.

Morton:

When I saw the name of the company, I was thinking of home automation, which many people are talking about these days.

Snelling:

Our notion on that is there are many people in that territory and a lot of activity going on with Smart appliances. As long as you follow a base ten standard, what we will do is simply to interface with all home automation. In fact, we have a point issue of our generic, which is due out in May or June, that would incorporate home automation. Bell South has what they call their all-in-one solution center in Jacksonville. Starting Monday of this last week, they will be pushing our product as an all-in-one solution. They are very much interested in home security as a home automation system. We can interface with that. Turn appliances on remotely. All you really need is a Radio Shack interruptible circuit breaker and you are in, if you want to be on a wireless basis. You will have the flexibility of putting in a jack for whatever appliance that you want, other than having to rewire the house.

Standardization and the IEEE

Morton:

You mentioned the standardization issue before.

Snelling:

Just some observations on standards in general. I have been watching standards for years and years. A person named Ted Erma is still the technical standards person for the United Nations telecommunications group. They have been struggling for years and years and years. It is very difficult to get participants who have vested interests in one approach versus another to come in and come to terms. So standardization has had, in many ways, a black name because it took so long. If you stand back and look at it on a broader basis, the standardizations have worked well, particularly if they can get in ahead of time. That is, before you have a large embedded base to deal with. For station sets companies like IBM or Motorola or Microsoft are pushing hard for a particular standard because they believe they are dominant. Then people come along and want to change it because there is a better way to do it. There almost always is. That is a real dilemma. The ATM user group is a good example of an organization that got in early, got some standards out quickly that were accepted before there was a large embedded base.

Currently in the world standards, there is DECT for wireless sets. There are a great deal of other initiatives, one of which is Bluetooth that is a combination of various organizations. Everybody wants to do a standard. That is not what we are after. We are not after component devices. What we want is a platform control device—we call it the transport—that interfaces with a wide variety of things, whether it is black telephones, Bluetooth architecture, or whatever.

My own personal view is that standards will gradually take place over time and become more efficient. But right now it is wide open. Even such things as spread spectrum, which may seem to be settled, like a Qualcomm spread spectrum architecture, you want to head on into GSM, which is a world standard as opposed to a North American standard.

The difficulty with GSM is that it does not have some of the values of say a Qualcomm spread spectrum capacity. GSM now has an architecture migration plan that would produce similar capacity and likeness. So you have to say to yourself, where will Qualcomm enter in on that? Qualcomm being almost the American product and GSM being the European product. When you can take a GSM telephone set and call anywhere in this country or anywhere in the world, but you cannot do that with a North American devices, then that is a distinct disadvantage.

One of my views is that Maurizio’s notion as applied to the IEEE specifically, but he really had a deeper feeling about internationalization standards bodies being a mechanism of how to do things. He would say how very inefficient and terrible they are, but he would also be supportive of the notion that once you have a standard, it needs to be a world standard as opposed to just being a European standard.

Morton:

Do you think an institution like the IEEE, even an internationalized IEEE, which has its own standard setting section, is the best way to do things? Or do you think a more industry-based group is more efficient?

Snelling:

Currently the IEEE has a standard of 802.11. Home Wireless Networks is actually using that, and it has become the industry standard worldwide. It is 2.4 Gigabit frequency. It is a very good example of a standard that was needed for data and for wireless. It seems to have been accepted by just about everybody now. It did have some competition at the start, but it all washed away.

So yes, I think it is one of the major functions of IEEE to assist in the construction of standards. Now, you have the Telecom Agency of the United Nations, which also has the same charge as well with worldwide funding. Then you have all the user groups that sit on top of that or underneath it, if you will, who have had a much better success rate. If you take the 802.11 model, and it has a whole series of successions to it, there is a tremendous opportunity there for a major contribution of IEEE.

Functions of professional organizations and the IEEE

Snelling:

I have not said much about IEEE. I am a fellow in the IEEE. I am also a fellow in the IEC, which is the International Engineering Consortium, generally for the same reasons because it promotes network architecture. I think that professional organizations, particularly the IEEE since it is probably the largest worldwide and certainly one of the most effective, have several functions that need to be focused on. One is the exchange of ideas.

The Internet certainly facilitates that. It is interesting that no one really pushed the Internet other than the Federal government with funding. It was basically an academic idea. But there were many opportunities for IEEE at one point or another to interject itself into the process in a welcome way. I worry a little bit about the Internet’s intermediate timeframe because sooner or later somebody has to pay for it. There has to be some commercialization there to get the quality level. My own personal view is that it will probably split at some point and the original academic notion of the Internet will probably be the non-charged portion, and then the portion with reliability and quality as a major factor for voice and video would be another commercial portion. They would not separate; they would just have dual billing charges perhaps.

I also worry about the Internet’s architecture relative to some of the tried and true formulas for network congestion. I think there is a role for IEEE there in assisting and gaining some sanity back into network congestion questions. I went into a Traffic Engineer’s conference once where I was giving a talk. I was throwing bigger and bigger software and hardware at these kinds of problems. My big thing at the time was timeslot allocation. Inside of the switch when you needed the timeslot, instead of it being dedicated to a particular port, you would have a reserve set of timeslots that you would allocate to high usage ports. This is going to happen, more or less. But the notion that you could throw hardware at an infinite problem is crazy. The infinite problem has to have some level mathematical congestion and guidance. I think the IEEE has a definite role to play there.

Another role for IEEE is the professionalization of the organization. I am a registered professional engineer and have been a registered professional engineer since I joined them in 1965. I have been politicking to get it so that if you practice public safety or welfare then you had to be registered. Well, the industrial exemption precludes that because all the engineers in all the industries are just never going to want that to happen. On the other hand, upgrading it for the non-registered engineers from a professional standpoint I think is a real role for the IEEE. It is an encouragement for the younger engineers particularly to see that it is a profession as opposed to just simply a place to work.

Atlanta is probably one of the more successful IEEE sections around. It has about thirteen to fourteen hundred people in it. I have not spoken a lot about that, but I have been active in the Atlanta section since I have been in Atlanta, obviously. That is how I became the Chairman of the SuperComm in 1990. I was in the local section.

The local sections have a hard time with funding. I think some of these surplus funds probably ought to be returned to the section to get a higher quality of speakers and better meeting locations. Just simple things like paying for the newsletters. I think that probably needs to be readdressed to provide more local financial support from the dues back to the local section. I think that would be quite helpful but very difficult to do politically because of the embedded administrative staff and the myriad of demands on the current funding structure.

This is a particularly good time because of the surpluses. The surpluses, at least the ComSoc surpluses, have come about because of the wildly successful conferences. There is some thought that maybe those have seen their heyday and are beginning to taper off. I know that the 1998 IEEE SuperComm Conference was not nearly as profitable to the local section as the 1990 was, for instance, for whatever reasons.

I am not much for job security through the profession; that is an individual thing. That is what profession is all about. Even salary protection I think is also an individual thing. It is a vehicle to enable the individual to move along professionally to go ahead and get their member status and move up through the ranks and take an active role in it. I wonder about the fragmentation of the various organizations like Comsoc. There are so many now. It seems to me that there is a place for five or six. Once you get to the fifteen or twenty, then no one has a great program unless they come together, which might be a good idea too. That would be a very difficult one to do politically. But perhaps with more joint activities, particularly conferences, can accomplish that.

The last thing for an IEEE role going forward is the further enhancement of this globalization notion. IEEE is not an American institution but in fact is a global institution.

Morton:

The various national engineering societies are, as far as I know, still very healthy. I have wondered whether the French engineering society or the Italian engineering society are particularly interested? They have not been in the past. How would the IEEE actually globalize engineering?

Snelling:

Part of Maurizio’s notion was not to try to compete with national engineering societies, but to join them on a joint venture basis. The problem with that is that they receive benefits but do not pay the dues. You are basically subsidizing these members. There is a balance that you have to strike. There is an Australian society, which I have spoken at several times. Certainly there is at least one Italian one there, and there is a fairly substantial one in Russia. They are interested in job security and job procurement. Everyone has things in which they are interested. The idea of simply amalgamating the various societies on some joint venture basis makes a lot of sense. But you have to do it in such a way that you do not subsidize the world with the IEEE American dues. This is not an easy task because that is the principal reason why they would want to join in with the IEEE on a shortsighted basis.

Impact of the Cold War on R & D; hacking and defense vulnerabilities

Morton:

I would like to ask you a question, getting back to the earlier part of your career when you were still down in Florida. At AT&T in the 1950s and the 1960s, I have heard that there were certain technologies that were developed with the Cold War in mind—survivability from attack and so forth. That language later passes out of being talked about. I wonder if at the more local level if there were things that the telephone companies did in installing or in maintaining or in planning their systems that reflected those times?

Snelling:

As the Network Head for Southern Bell, I carried a top-secret security rating, and that got me into a lot of things. Of course, part of that was the bombs’ falling scenario. There were a number of convened “bombs falling” committees operated for four or five years. I sat through many of their meetings; not because I was the Network Head of Southern Bell but because somebody wanted me to go there. My notion was that if the bombs fall and destroy the network, they ought to bypass the opportunities through the telco internal administrative networks. All that was needed was to connect at a crucial point. I think in some fashion that notion might have received some deployment, but not a lot. I think the Cold War ended before the committee was really able to do much.

After the fire where the three astronauts were killed, NASA developed what they called the life support philosophy: No system that had life support requirements would be a single component network. It could not fail with a single component failing. It had to have at least two. In some cases if it was very critical, multiplicity. So I did many things as far as our interface. For one thing, colleagues and I walked our way through the swamp and put in a bypass microwave route down to Green Swamp coming up from the south and going up to Canaveral from Lakeland. Thus, life support redundancy became a component of the design criteria for NASA.

At the Southern Bell network, we applied that with rings. The rings notion of failsafe—so that if you cut a ring it would still operate—comes out of that architectural philosophy. That is pretty much what the network is at this point. Because of the high reliability requirements, there are very few components. In fact, one of the things you do when you look at a network is to see if there are any points in that network (if they were to fail) would cause the network to fail? And if the answer is yes, then you have failed in the design. Whenever we read about somebody’s network going down because somebody put a backhoe through something somewhere, you say to yourself, “Yes. I can understand that. Somebody screwed up.” Not necessarily because they were stupid because they are certainly not stupid. But there is a lesson you have to learn relative to what you are willing to pay for redundancy and increased reliability.

So, the bombs falling scenario had a lot of impact on the material outcome of design. In fact, there is a strange thing that has to do with satellites, which is the most vulnerable of all mechanisms. At least they have the value of having a universal footprint but it would be the first thing you would go hit in any kind of a war scenario.

Morton:

Were there ever any kind of underground trunk lines to be hidden from satellite cameras?

Snelling:

Every operating company had its own bombs’ falling plans. Ours was nearby to go to. I don’t know what you were supposed to do when your wife came to the door and knocked on the door, but the idea was that you would go to this place that would be embedded in concrete. You could up and run the communications access.

Morton:

There were some underground facilities where people were expected to go and work?

Snelling:

Yes. I always wondered just how I would handle my wife and children at the door. I concluded that this was the stupidest thing I had ever heard of.

Morton:

What has happened to those? Is that system still in place?

Snelling:

No, no. First off, the food is rotted by now, I guess. When you get to the point of spending a lot of money—you would need to have redundant communications access to that site—generally people would back away from it. There was a limited amount of that. There was the notion that you would, in fact, produce some sort of administrative plan. We would have monthly or quarterly war room exercises.

The real danger turned out to be the hackers. They began to materialize about 1987 or 1988. I do not know if you have read the book, The Cuckoo’s Egg, it is a description of a person at the University of California at Berkeley who was in charge of telecommunications. He was defunded from his NSF grant, so he became the IP person for the campus. He couldn’t reconcile the billing. He was an astronomer by training and trade, and so he was very precise in his measurements. He kept finding a half-hour or so discrepancy in his billing. It turned out it was these hackers coming in from Germany. The book is about two or three years of activity and then tracing them down. It was about that time that the hackers started surfacing all over the network. A lot of money has been spent to begin to mitigate that. I gather that it is now worse than it was then because the hackers have progressed faster than the defense mechanisms.

So, I do think that we are probably most vulnerable at this point, from a national defense standpoint, by computer hacking and intrusion. That goes in two directions. One is the mechanical direction and the other is the human engineering direction.

We received an alert one Fourth of July that the hackers were going to attack. I called Bellcorp using a cellular phone. Irwin Doris, who was the Bell Corp EVP at the time said, “You are talking on a cell phone. That is not secure.” I said, “It is a whole lot more secure than the landline telephone you have in your office. We are about to get hit and we need to put out a national alarm.” Sure enough, we did get hit. They knocked out 4E in California. It was not as bad as we thought it was going to be. I shut my maintenance channels down, which was the principal source for hackers because those are generally dial-in. It is almost impossible to guard against a dial-in. You can defend the network for short periods of time. I was the security officer for Bell South at the time.

It was a false danger of bombs falling. The real danger is ongoing and forward going which is the likelihood of computer hacking, either economically or for other reasons such as terrorist.

Powering and backup power for emergencies

Morton:

As an analogy of the Cold War issue is still having some permanent effect on the telephone business, when you were down in Florida, was there anything about the physical plant that reflected the special environmental conditions that occasionally arise from tropical storms?

Snelling:

Audio File

MP3 Audio
(395 - snelling - clip 5.mp3)

Yes. One of the other big things in my life has been powering. It turns out that one of the limitations to fiber to the home has been the disproportionately high cost of power. Thirty percent of the cost in an all-fiber network is for backup power. A man named Jess Chernac, with some input from Southern Bell and myself, developed what was called a Controlled Environmental Vault, which really enabled electronics to be distributed in to the outside plant from the central office. CEVs look like giant concrete vaults. You lay them in the ground in two pieces and seal them. They are underground, and they have backup power, commercial power, and natural gas backup power. That is generally where you put your last electronics in the network; although now, we have extended that to the optical network devices in some cases.

In Florida and in the Carolinas (which has been hit more than Florida lately), you have to have a wide array of backup generators. When Hugo hit Charleston, Charleston was without power for six weeks because it knocked the electrical transmission lines leading to the city. Much of the telephone carrier network in Charleston worked for six weeks on tiny fifteen or twenty-five kilowatt generators. The Telco plant probably now has backup generators that could run forever as long as you put gas or diesel in them that flow to the batteries. That is why the telephone network continues to work.

When you get out in the field, if you have electronics but no backup power, you need some way to get it. We used to use a couple hundred gas generators. We would chain the generators to the pole, and we found that people would cut the cable and steal them. If you are cold, and your family is in the dark, and there is this great generator sitting out there, there is not a lot of morality concern connected with that. That was at best a half-decision.

My own personal view is that the longer term solution to that problem—which has many other solutions tied to it—is the notion for distributed power. The deregulation of the power industry would lend itself to doing that, if in fact you could have a reasonably low cost natural gas generator in a substantial number of homes. Number one, you would not have all the hits on your TV sets and computer, etc. You would have much more stable power. You wouldn’t have to reset your clocks all the time. More importantly, you would not have to buy all those power interruptors. You would also not have to build many new central generators, because if you can break the peak power at four to seven in the evening and ten to twelve in the morning, then you do not need to build anymore generators for many years to come. But that does involve a very substantial deployment of power.

At one time, I had the notion that at off-peak times you could raise this water that would be your storage water for the living unit under power failure conditions, hurricane or tornado conditions. Then in the daytime, use potential energy to generate power during the peak power. That is not economical, but gas power is. That is probably one of the longer-term solutions.

To go directly to your question, during the Cuban Crisis in the early 1960s, I was the transmission engineer for Central Florida, which extended from Hollywood to Saint Augustine and included Orlando. We were on invasion alert at the time. We thought the Russians were about to invade south Florida and the Keys. We built a backup route from Key West to Miami in about two weeks. Being on duty, we had twenty-four-hour coverage of our control centers, which would operate during a hurricane or other natural disaster. But this was an alert, and we put a telephone in the toilets for those taking a break. It was pretty serious stuff during that crisis. Of course, we had Patrick Air Force Base in the Cape. That was a military installation, and they a firing range that actually skirted the north end of Cuba, so there was particular vulnerability to that.

I think the Cold War had a big effect on the military. The Bell System did the Autovon network, which was a four-wire network. We had the astronaut network, which became a four-wire network. The idea there was that in the United States, the Bell System would provide the communications for the bombs falling scenario. That was, by the way, one of the two principal things for Bellcorp. It was to take over the national security functions of AT&T and to provide basic research for the organizations.