Oral-History:Klaus Gueldenpfennig

About Klaus Gueldenpfennig

Klaus Gueldenpfennig grew up in West Berlin, Germany. After studying telecommunications and electrical engineering (TFH Gauss schule), he began working for DeTeWe (Deutsche Telephonwerke) in 1954, later taking employment with Telefunken. Gueldenpfennig came to the United States in 1965, working in New York City for two years for DeTeWe and Telefunken. He then spent eleven years working on switching system development for Stromberg, a career which brought him to Rochester, New York, and which produced multiple circuit and switching patents. Gueldenpfennig received an M.S. in electrical engineering at the Rochester Institute of Technology in 1974 and received an M.B.A., again at RIT, in 1977. When Stromberg relocated to Florida in 1978, Gueldenpfennig stayed in New York and launched his company Redcom.

In this interview, Gueldenpfennig provides a comparative analysis of telecommunications industries in Europe, Asia, and the United States. He pays particular attention to government regulation, taxing, specifications, and market demands. In recounting his own education and career, Gueldenpfennig assesses the state of communications education at a time when switching was seldom included in formal academic training. He considers the emergence of digital switching, driven largely by U.S. industry and by the incorporation of new components into the design process. Gueldenpfennig also describes the design challenges that accompanied the introduction of computer controls.

In the section of the interview covering Redcom, Gueldenpfennig explains his rationale for starting the company and describes its early organization and contracts. Assessing the benefits and limitations of Redcom's niche-marketing model, he details Redcom's pursuit of rural customers as a successful business strategy. Traffic generator equipment designed to simulate customer use for telecommunications testing also became a successful niche market for Redcom.

At the conclusion of the interview, Gueldenpfennig describes the challenge of developing standards that can incorporate new technologies. He considers the role of the IEEE in international standards. Gueldenpfennig then identifies influential trends in the history of telecommunications, recounting the influence of transistors on industry and assessing the contemporary growth of bandwidth.

About the Interview

KLAUS GUELDENPFENNIG: An Interview Conducted by Frederik Nebeker, Center for the History of Electrical Engineering, 6 August 1993

Interview #162 for the Center for the History of Electrical Engineering, The Institute of Electrical and Electronics Engineers, 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 at Stevens Institute of Technology, Castle Point on Hudson, Hoboken, NJ 07030 USA. 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:

Klaus Gueldenpfennig, an oral history conducted in 1993 by Frederik Nebeker, IEEE History Center, Hoboken, NJ, USA.

Interview

INTERVIEW: Klaus Gueldenpfennig

INTERVIEWER: Rik Nebeker

PLACE: unknown

DATE: August 6, 1993

Leading figures in telecommunications

Nebeker:

This is the sixth of August 1993. I am talking with Klaus Gueldenpfenning. This is Rik Nebeker. Before asking you about your career, you were just mentioning some of the people who would be good to talk to in this field.

Gueldenpfennig:

Yes, those who spent, I think, more time on the historical events than others, you know, other people who've been in this business for a long time. And of course you know Amos Joel, Henry Abbott -- you can probably track him down since he's retired -- Walt Nolan in California is another one. He used to be President of the Comsoc Society. Then we have Rustowick in another one; he works as a consultant now, Charlie Pleasant I thought wrote something about it. If you are looking at the older stuff, you might find some literature from Jay Gordon Pierce who used to work at Stromberg and he's passed away, but I think he wrote some books...

Nebeker:

I see.

Gueldenpfennig:

Up to the mid-'70s that technology as it moved through usually was covered by those people. There were a whole bunch of others; a good source of information is the Switching Committee. Dick Dott. Bell Labs, I believe. I'm not so sure, but he is fairly much into this.

European switching industry and products

Nebeker:

We are trying to always take an international perspective on things.

Gueldenpfennig:

Ah.

Nebeker:

Have you been in touch with what is going on in Europe?

Gueldenpfennig:

Well, to some extent, I am sitting on the fringes. We've developed all our products and modified them so that they fit into the European market.

Nebeker:

I see.

Gueldenpfennig:

We've just received German approval and that took about a year and a half to go through the intricacies of putting unique inputs into that -- somebody would look for some patterns, you know. Totally irrelevant from the outside user point, yet the rules over there... you have to use that specific Siemens connector, and so forth.

Nebeker:

I see.

Gueldenpfennig:

You can find out very quickly who is making the rules and trying to set the stakes. For better or worse.

Nebeker:

Is there very clearly a European switching community and a North American switching community?

Gueldenpfennig:

There is in my opinion, a European switching community, probably centered to a large extent around the Siemens, Ericsson, and Alcatel type groups...

Nebeker:

Ericsson is another company?

Gueldenpfennig:

In conjunction with the CCITT, you know, they are all sitting on top of each other, and they don't have to spend the same amount of travel money we do. And so obviously you find them clustering around each other, and the various operating companies or government entities which run the telecommunications field are obviously all involved with each other.

Nebeker:

I see.

Gueldenpfennig:

So, it's not as deregulated and as free and as open as here.

Nebeker:

There is close cooperation within Europe.

Gueldenpfennig:

Yes, there is very close cooperation between the PTTs and the industry there. I want to be careful in what I'm saying, but you get the impression that there are a lot of sweetheart deals going on, and that there is a certain amount of, let's say, "engineering talent osmosis" going on there.

Nebeker:

And is there much common technology? I mean, are companies such as yours selling a lot in both markets?

Gueldenpfennig:

Well, we are not selling anything over there. We're just starting. We haven't done any marketing. We've just started to look at it in the first... We go a different way than probably a lot of companies. A lot of companies will flood the big markets and then see what happens and then try to fill in the blanks afterwards. We've done it the other way, you know. We get the product to a certain acceptance level, and then we go out and try to find a partner, try to see how we can get it into the marketplace.

Nebeker:

But are they pretty much separate markets? Have they been in the past?

Gueldenpfennig:

They are separate markets. It's costly, you know. You sometimes question why is it different? Why can't you do it this way or that way? Why does it have to be that way? Many, many times I've questioned the validity of the reasons why it has to be that way. It will work either way. And therefore it should be irrelevant to some extent, you know. If you look at the US, it is a wide open market. You come in, you take your product, you pass FCC Part-68 and Part-15 and so on, which are very loose and very general specifications, and you are in business. Yet, you get into the European market and you will end up with the products having to be ever so slightly different or that one... There are many, many factors. Maybe there would be a technology clash, or if there would be... But you find that in various countries there are certain unique things which are just there -- in my opinion -- to keep out the other guy. In order for you to get in there you'd have to invest a lot of money up front. And unless you get a piece of the market you may not be able to afford it. So it's some sort of control, you know. A similar situation exists in Japan.

Nebeker:

I see.

Japanese switching markets

Gueldenpfennig:

Even though they say it's wide open, they have changed things and have made them unique in such a way that unless you make your changes, you don't fit in the network. So the question here becomes, "How do you get into the Japanese market?", while if you try it all by yourself and you run in there, it's very difficult. Yet, if you have something they want, I found out, it's relatively easy.

Nebeker:

Let them do that!

Gueldenpfennig:

Well, not "Let them do that." We did it! But we basically ended up getting great assistance. You know, we sell a lot of equipment to NTT in Japan.

Nebeker:

I see.

Gueldenpfennig:

And NTT liked it and wanted it and hence there were no hurdles. There was kind of describing what is it they'd like and are we willing to modify that and they are willing to buy and of course we were more than happy to do that and hence, everything happened and went on. There wasn't a third party where you had to go through some government entity which is going to do your testing and your verifications and is going to run the difficult bureaucratic way, which is how it happens in Europe.

Nebeker:

I see.

Gueldenpfennig:

You know, theoretically, that wasn't the case.. You buy the specs, you make the thing meet exactly what's in the specs, you are going to submit your equipment to some testing facility just like here and that's it. Well, fine and dandy, until you get to the guy who starts interpreting things! In the engineering field, when you get into the interpretative mode, that can give you some discussion for the next half year or so!

Nebeker:

And to the extent that you think something meets the specifications and they say it doesn't.

Gueldenpfennig:

Yes! I came from there. My background is European. I know how picky we are and how much we pay attention seemingly to things which probably don't deserve that much attention. You know, because it's neither here nor there.

Switching specifications

Nebeker:

So, as a rule of thumb, the Europeans have more specifications?

Gueldenpfennig:

I think we have. In Europe there are more specifications for the sake of specifications, and not for the sake of accomplishing certain real value-type things. I questioned the value very often. I travel around the world. We have a lot of presence internationally and we are looking at the developing world. We have a lot of switching systems in the developing world and to many of those people it's totally irrelevant how you accomplished what. It is relevant to them that they can make reasonable communications. They wouldn't even understand all the other things to a large extent. Yet, we are looking at nickels and dimes and down to a level where they can get everything including the kitchen sink, but they can't talk! We need to come down to earth. We're going into China and we are telling the Chinese and maybe now the Russians and all these other countries how much they need the new special signaling No 5 [ESS] and ATM and anything else. These guys don't even have a telephone on their desk! That's the last thing on their agenda. What they want to do today is pick up the phone and reliably call somebody else! And there are many ways to do it.

Nebeker:

Oh, yes!

Gueldenpfennig:

You know, right now they've got to travel twenty miles before they get to the nearest phone. A lot of people have no idea of what it looks like out there. You know?

Nebeker:

Yes.

Gueldenpfennig:

I mean, I've been in areas where the best communications is a CB Channel or a little HF Radio, where they get on some frequency, during some time of the day, when the weather is right and they can get through. You know, there is no reliable communication in places like Western Samoa. I've seen that they have a switching board near the airport which is operative from eleven to one o'clock! When that is all done, they will take it all in trucks and move it to some other place, where they can work from one o'clock to three o'clock. And so on. You know, the local operator is a grocery clerk. Yet, here we are talking about all the satellite traffic and all the great things and it's a challenge to marry the very basic things with the sophistication of the high technology overlay network which we have throughout the Western world.

U.S. telecommunications industry and global trade

Nebeker:

Is the US market really more open to European companies than the other way round?

Gueldenpfennig:

Oh, definitely! There is no doubt in my mind. The US is wide open. You know, even if you look at the government financed operating companies, nothing prevents a foreign entity from bidding the highest and owning it. But if your price is significantly below the other, then they can go ahead and do it otherwise anyway. I hate to say it, but we as Americans are less competitive in the global marketplace. The cost of doing business in this country has gone exorbitantly high and it's hurting everyone and that's why the jobs are being exported.

Nebeker:

Yes.

Gueldenpfennig:

And that's why other countries are rising and we're going to sit on the sidelines and watch dollars flow that way. All the politicians can talk about it, but it's not going to make any difference unless they change the tax structures, unless they change the rules. This is me as a businessman talking -- and it's not just unique to telecommunications. They should get us to a level where the cost of doing business, the general terms, get somewhat lowered. The social burden is tremendous, the tax burden is tremendous. When you look at the whole structure, you know even when you don't make any money you still pay all but the minimum taxes. It's self-defeating. It'll ruin the industry, you know. No wonder people go off-shore. That's a tough thing to crack.

Nebeker:

What are the main factors in that? You said the taxing...

Gueldenpfennig:

The cost of doing business, the taxation and so on. We are not producing any more. That's the whole key. We can't produce when we become a service industry. Then you consume instead of producing. You don't generate products. You don't generate value. Any kind of service, what's the value? Once the service is completed the value is gone. I mean, you've got the value for that period of time but you are not building upon it. But when you are producing things, you are creating real value, something you can touch. If we are manufacturing products, that's value to everybody. Even though it gets depreciated after a period of time, it does create a value. The more you do service the less value you produce and we have lost a tremendous amount of manufacturing jobs.

Nebeker:

Yes.

Gueldenpfennig:

We have great ideas. And we're building the first ones, but when it gets to the real volume business where you have to compete out in the marketplace, and on the international scene, what's going to happen? Take a look: where are all the integrated circuits being manufactured? Where are all the cars being built? Where are the telecom... you know, where are the giants sitting there?

Nebeker:

I was just going to ask about Europe where they probably have taxes that are comparable to...

Gueldenpfennig:

True! But the European markets are basically closed. We as telecommunications had a hell of a time breaking in there. Obviously, I have been out of things too long, but you are not going to sell wholesale, let’s say, infrastructural equipment, telecommunications switching systems, to European entities which are run by the government. The government will fail if they don't protect those jobs. They'll get a vote of no confidence in Parliament and they are out because they didn't provide enough jobs for the Siemenses and the Alcatels. Yet, those guys can run all over the country, all over the world and appear anywhere else.

Nebeker:

I see.

Gueldenpfennig:

With the government support, with decent financing. I just got something out of Pakistan. How'd you like a financing package with twelve years of financing? The interest rate is zero. How do I compete with that?

Nebeker:

Right.

Gueldenpfennig:

You can't. You can't find a way. And a two-year abatement. Everything. I would love to do that and sell everything and anything to anybody. So, that's on the business side. From the technology side, we have great equipment. I think we have great ideas and a lot of things which are going into the marketplace. You know, we're finally well-developed, so unless we throw everything out and start all over again... Where is the wisdom? How much are we willing to spend for the increment? It's like an S-curve. You've got tremendous achievement at the low end. Very little cost will get you a lot of the basics.

Nebeker:

Yes.

Gueldenpfennig:

And as you start progressing, less and less is what you're getting for the increased cost. Then you have to spend a tremendous amount to get that last 5-10% up there. That's what's happening. So you go back to the structures. We don't get government financing or help to do research. Yet, most of the European companies do. It is standard in Europe that if you want to participate, you want to develop a new system, the companies that engage in it will be assured a section of the market in the business. There will be cross-licensing assured, so that they don't lose it. So they can invest. Here, you are on your own. You put your own money into it, and you don't get any assistance. Tax credits have largely disappeared.

Nebeker:

What about competition within Europe? Can German companies compete with French companies?

Gueldenpfennig:

Oh, I think within the European Community, they are quite competitive. But the prices are not that low; the prices are high. Yet, when these people appear on the international scene, the prices are very low. You can argue about the dumping laws, and all of that, but that's difficult to prove. So, from my point of view, they get around a small player, really. And a lot of good ideas come out of small companies. You know, a lot of good thoughts come out of it, but a small player cannot survive unless you do what we're doing -- going into the niche market. If there is some volume there, every time we appear on the scene, we will steal it from you. So you've got to be unique; you've got to have something other people don't have. It's our job, the way I see it, to always have something other people don't have and to be able to react faster than anybody else, to implement things faster than anybody else, to be able to do things which everybody else would have to say "Well, sorry, I can't do that", or "My system doesn't accommodate it," That's the utmost flexibility, high engineering content, tremendously high and costly. But, if you're the only game in town, you might have a chance to get it.

Education and early career

Nebeker:

Let me ask you, if I could, about your own career. How was your education, maybe your background, where you grew up?

Gueldenpfennig:

I grew up in Berlin. In Germany. I started working for DeTeWe in 1954, which was basically a Siemens company.

Nebeker:

I see.

Gueldenpfennig:

You know, the top management came from Siemens. And it was a great company.

Nebeker:

So, you grew up in West Berlin?

Gueldenpfennig:

Oh, yes. It was very interesting and I got into the telecommunications world, but everything was relays and selectors. The good, old electromechanical systems. And, boy, they worked, unless you sort of threw a monkey wrench into the works. It took a tremendous amount of power, but if you couldn't kill these things. Everything was distributed to the ultimate sense, you know. If you've lost a piece, okay, something else took over. It didn't degrade the service. I then went on, subsequently, to work for Telefunken.

Nebeker:

What was your education?

Gueldenpfennig:

I was in electrical engineering-- telecommunications, specifically.

Nebeker:

Where?

Evolution of telecommunications education and industry

Gueldenpfennig:

That was a time when you couldn't find US colleges that would teach telecommunications. You either went to the technical school system and you learned there, or...

Nebeker:

Amos Joel complained, you know, about teaching it.

Gueldenpfennig:

That's right.

Nebeker:

Not being an academic subject?

Gueldenpfennig:

Talking about telecommunications in any of the colleges until very recently, until the last few years, that meant radio communications. That meant television, that meant satellite traffic, that meant single item switching, Telecom switching-- that wasn't part of the agenda. So there they used to be the so-called... call them "brain drains" or whatever, you know, shiploads full of people who got picked up over in Europe and moved over here. Into Kellogg, into GTE, into AT&T, into Stromberg, for that matter.

Nebeker:

Is Kellogg still going strong?

Gueldenpfennig:

I don't think so. I haven't paid any attention. I don't even know if it still exists.

Nebeker:

Well, it's funny, because one of the...

Gueldenpfennig:

That's so old!

Nebeker:

One of the slides for my talk this evening is a radio advertisement. They are from Kellogg. I know they got out of the radio business in 1930.

Gueldenpfennig:

Yes.

Nebeker:

But, well they must have stayed on in switching or...

Gueldenpfennig:

I mean, except for Kellogg's Rice Krispies, which is a different thing, I think, we're done with that one. And you know, you got those rather large European telecommunications conglomerates. You've got Siemens; you've got Alcatel; which is basically the former ITT plus some additions; you've got Ericsson, you've got Bell in Belgium. These are all large development centers. They have a lot of good ideas, a lot of engineering staff. It's very difficult to compete with them.

Nebeker:

Yes.

Gueldenpfennig:

They have more difficulty in moving and responding quickly and so on. When you see technologies evolving, you know, we went from the electromechanical to the so-called wire-partially electronic and we had, I think, some rather questionable technologies in between. We built throughout the world switching systems with reed relay contacts. You know, that stuff used to stick all over the place; you had giant conferences on it, because contacts were sticking. Ultimately it evolved into the digital network.

Nebeker:

Yes.

Digital switching

Gueldenpfennig:

I think the US was driving the digital. The Europeans came on board much later.

Nebeker:

I see.

Gueldenpfennig:

Everything happened with the development of the transistors and the base technology. Really all the telecommunications people followed. We had inventions and good ideas on the systems level. And of course, Bell and others and AT&T on the component level. But most of the other companies were really following the component developments and then designed systems around the new components. As they became available, they started to get better and better and better. There were two things: one was the technology which related to transmission analog or digital, and so on, and the other one was the control technology. Once the computer got introduced, it became a centrally controlled type of environment.

Nebeker:

Right.

Gueldenpfennig:

That in itself raised all sorts of other problems again. Now if you lost one, you lost the whole thing, so then we went in to do all these redundancy schemes. My favorite one was a plan that had five computers controlling the johns on this space shuttle, you know, and it still didn't work. So much for redundancy. So there is no fail-safe operation.

Nebeker:

Yes.

Gueldenpfennig:

So, we went back and gave it a stab to use the new technology, the digital and the microprocessor technology, to create distributed-control architecture. It is related to the old step-by-step and individually controlled-type things. It has proven to be very, very good in terms of applications, in terms of reliability, in terms of availability of service. All these things are absolutely marvelous. Of course you don't share, like you would do in a common control environment, and therefore the price of the larger systems is more expensive. There comes a point when you say, "Is it worthwhile? Do I do it? Don't I do it?" And so on. It's a trade-off. I believe in it and I would rather not have the headaches. I'd rather not have to explain to people why the telephone doesn't work because I saved a buck here, a buck there. If I have so many thousand lines and I've lost three, who cares? Right? I have a great service index. Well, let me tell you, to the guy whose line, whose phone is out of service, to him it's a catastrophic failure. So, so, it's all relative, you know? Where am I sitting? Where am I coming from?

Education

Nebeker:

Are you a diplomaed engineer?

Gueldenpfennig:

Yes, basically. I went to what we call TFH Gauss schule in Germany.

Nebeker:

I see.

Gueldenpfennig:

When I got over here, I went to RIT and got my MSEE.

Nebeker:

Oh, I see. Oh, you went to RIT.

Gueldenpfennig:

And an MBA and all that. I added a few things.

U.S. employment; Stromberg

Nebeker:

When did you come to this country?

Gueldenpfennig:

In 1965.

Nebeker:

In '65?

Gueldenpfennig:

Doesn't seem to be that long ago. Yet, it's twenty-eight years ago. I went to work for Stromberg .Well, not right away.

Nebeker:

Oh, I see.

Gueldenpfennig:

I basically worked in New York City in conjunction with DeTeWe and Telefunken.

Nebeker:

So you first got an assignment in this country.

Gueldenpfennig:

Yes, for two years. And then I sort of figured there had to be more America than New York City. I didn't particularly care for it. So I got up here to Rochester, and I'm still here. Stromberg in '78 decided to locate everything down to Florida, so people often ask me, "Why did you leave Stromberg?" Well, I didn't leave. I'm still here.

Nebeker:

So you worked for Stromberg from...?

Gueldenpfennig:

Eleven years.

Nebeker:

Late sixties...

Gueldenpfennig:

Basically developed switching systems.

Nebeker:

I see. That was your...

Gueldenpfennig:

CO-stuff as well as PBXs. Those were the times I was really involved in circuit design and system design. You know, it gave me a good opportunity to look at some new things, new ideas. I ended up with a lot of patents during that time.

Nebeker:

What sort of patents?

Gueldenpfennig:

Circuit and systems patents.

Nebeker:

For?

Gueldenpfennig:

They're all related to switching systems.

Nebeker:

What sort of switching systems were you working on?

Gueldenpfennig:

Originally, electronic ones. Then I went to digital ones. There were certain ways to...

Nebeker:

And what was it that Stromberg was selling?

Gueldenpfennig:

Oh, Stromberg was developing small digital PAXs. First, analog PBXs, then digital PBXs, but all processor-controlled. We developed a unique type of interface circuit. So there was plenty of room to let your spirit run free and come up with a new way of doing it.

Leaders in the U.S. telecommunications industry

Nebeker:

You named many of the big players in Europe. Over your experience, what have been the big players in this country?

Gueldenpfennig:

Oh, the big players in this country really were ITT, GTE, and AT&T. You know where ITT is these days. It's basically disappeared and even Alcatel's presence in the telecommunications field is not very pronounced. AT&T is as strong as ever. Northern Telecom moved basically from out of Canada. It's very strong, but even there that technology has been warmed over and over and over again, and eventually something new will appear, I guess... Now I'm saying, "Well, guys, that's all nice and dandy and we're going to do it better and faster and everything else, but just let me remind you, you still want a top over it." One of the major ingredients, major features, is that I would like to reliably get from A to B and have a reasonably good transmission medium to convey my messages. And I don't want to forget this. So, every once in a while, we engineers have to get down from cloud nine and be put down on the ground and say "Here!" You know. "Let's sit down and talk."

Nebeker:

So, is Northern Telecom doing well in the United States?

Gueldenpfennig:

Well, I think right now , in this last year or so, nobody was really doing well. Everybody is doing adequately, but I don't think this is the time... I can only go by hearsay, but I see many resumes coming on my desk. A lot of people are being laid off and reduced, restructured, so it's a slow economy and it's world-wide. NEC in Japan is very strong, but except for some peak in the late seventies, early eighties, it's not really progressive here. Ericsson is not doing very much here yet, but I find them all over the world in other areas.

Nebeker:

Where are they based?

Gueldenpfennig:

I find them in China, I find them... Ericsson is a Swedish company. They have probably one of the smallest home markets and everything else is international. We find plants from Ericsson all over the world. You know, I go down in the South Pacific and there they are. My competitors are usually Ericsson. I think their equipment is maybe a little bit behind in terms of technology, but they are good old work horses and they have a good presence. No doubt about it, it's a very reliable, very viable operation.

Nebeker:

What about the next level of players? The ones not as influential?

Gueldenpfennig:

Well, there are a lot... For instance, if you take the PBX market, I would say that at any given time there are one hundred to one hundred and fifty different systems manufacturers and/or engineering companies in the US alone who make a stab at it.

Nebeker:

Is that right?

Gueldenpfennig:

Most of them last two to three years. Some of them get absorbed into some other operations, and disappear again. I would think that there are less than a dozen which ever amount to anything or manage to stay alive by themselves.

Nebeker:

Can you name some of those?

Gueldenpfennig:

Oh, it's kind of difficult for me to pull them out of my hat, but there are some papers written on them and SRS is now part of another company. Philco-Ford disappeared; it was a viable operation at the time. The ones which survived are still around as MITEL: but every time I read the newspapers, there is another hundred thousand dollar loss from that one. I don't quite know how the Canadian operations manage it. They went in the low options in terms of pricing and yet at the same time you see these tremendous losses and the government guarantees their survival by offering loans. How do I compete with the Canadian government? You know, that's one of my problems. When you look at the central office market there are really only four to five manufacturers viable in the US. There is Northern, there is AT&T, there is Redcom, there is Stromberg and to some extent, the private, I guess the MCI type. With the carrier market, you've got Digital Switch Corporation and that's about it. I've probably forgotten someone.

Redcom

Company formation and structure

Nebeker:

So you worked for Stromberg for eleven years, you said.

Gueldenpfennig:

Eleven years.

Nebeker:

And when they moved to Florida?

Gueldenpfennig:

Then we started Redcom.

Nebeker:

I see.

Gueldenpfennig:

A number of people worked with me there, designing the systems entry digital PBX at that time, and they moved it all to Florida. There were very few people who wanted to go. So that offered an opportunity to see if we could do something. Of course, it's the old thing; you never forgive yourself if you haven't tried it. On the other hand, it is very risky to try it. You don't have any blanket, you know. If you are an employee at some company what do you get? Money? You get some idea, you've gotten something you might be able to do, you sit down with a yellow pad and pencil and see what you can do. When I look back though, I have to say there is a golden opportunity which you only get at that point when you start new. You don't have any customers, you don't have any liabilities, you don't have any products, you don't have any money, and you don't need any sales guys 'cause you don't have anything to sell. Right? You need a bunch of engineers to dream up something which in the future might be able to play some place. There are always some niches where you can get it to do something. On the consulting end, and on helping other companies... We were very fortunate within the first few days with opening the doors of opportunity, because of some people we knew and reputation and so on. We ended up getting a software development contract for Philips in Holland to develop the software for their PBXs. That lasted about three years. It was a joint effort with them and us and some of our people were over there and some of their people were over here and went back and forth. And it was a very enjoyable type of work relationship. There were little bitty clash points: you didn't do this, you didn't do that, and so on. But by and large, it was very nice, very rewarding, and it gave our people a little international flavor.

Nebeker:

Tell me about the initial organization of Redcom.

Gueldenpfennig:

I can give you a lot of newspaper articles also, but it's basically three guys and a cat. The cat snored and these three guys were trying to figure out what they needed to do. As I phrased it once, I said, "Look, we have one golden advantage; we have no skeletons in the closet." Let me tell you what that really means. That means you've got your elbows clear except for the financial restraints, and you can really sit down and think. Your only goal at that point is putting something in the market if you can manage financially. Some time down there two years from now, you have to get a little bit of revenue, and that gives you a totally different way of thinking. You are not carrying that pack on that you always have to be concerned, that you're always trying to sell. I mean, we're now fifteen years old. We built our closet, and we put the skeletons in it. They rattle in there quite nicely. One is carrying fifteen years of commitment. You have equipment out there, a market presence, a technology which has to be supported. You have to make sure everything works; you've got to worry about your customers' investment. None of that is there initially. That is why it's so marvelous. I think it was the best time we had. We were all maybe ten, fifteen people, and all of that wasn't there. Then you don't constantly have to worry, "Oh gee, what did I do five years ago? Now I'd better do something that doesn't clash with it." Everything has to be compatible and so on. You don't have these constraints at that point. You're designing something brand new.

Nebeker:

Right, right. But certainly within a larger company there could be a team that...

Gueldenpfennig:

There could be a team, but there are usually in large companies too many constraints, and from what I've seen in all of the companies I worked with, they always put the cart before the horse. They always go out and market something they don't have. In return then they legislate the engineering and the invention. And invariably, you've got to go out there and sit with a product which is not ready, which was promised, playing catch-up. That relates to how you solve it. Compromising. It becomes a huge compromise and you've got to compromise forever. We have signs of that to some extent now. If you can't forget what you've done, with whatever you do new you can't just throw everything overboard and start new. You have to worry about whether the investment is out there, you have to worry about it as compatible, and you have to worry about "is it suitable?" This is highly capitalized stuff. You, from an operating point of view, from a customer's point of view, have you gotten your money's worth? Have you depreciated it yet? It's easy to say, "Well, technology changes every five years." Is the guy who spent some million dollars on this switching system prepared to throw this thing out in five years from now and put a new one in? I doubt it. Particularly when the Public Service Commission tells you how much you can charge in your rates. So it's not that easy.

Rural markets

Nebeker:

What was the niche that you saw?

Gueldenpfennig:

We saw a niche in the small rural -- and when I mean "rural" I don't mean rural in terms of backwards, I mean rural in terms of size. I do not make the distinction, say, well, because these customers are farmers they don't need call back, call-forwarding, or this or that other feature. They need everything you need in the city. There's absolutely no reason why the equipment shouldn't be as sophisticated and have the same capability as the metropolitan systems, except they are smaller systems because they are smaller communities. There are not as many people there.

Nebeker:

Were AT&T and the other major players marketing to that group so effectively?

Gueldenpfennig:

<flashmp3>162 - guldenpfennig - clip 1.mp3</flashmp3>

No, and that's because when you look into large companies they put huge amounts of resource dollars or research dollars into it. The cost of developing a system is high no matter who does it. We all cook with the same water. Their structure may be lower, or certain other factors may be lower, and if you built a smaller system and designed it from scratch and said, "Here's the upper limit of it," you can get your hands around it a little easier. So maybe there's some savings. But by and large, you know, software is software. If you've got to do everything you're going to end up with that much. And so, it is very very costly to do it. Well, if you can advertise your development costs for larger systems, that's a little easier. Your costs just die, so you're more competitive. The same thing is when you got to sell corporate. You sell a large system which maybe costs a million bucks, but the cost of sales and maintenance dollars on that large system is probably not much different than when you go out and sell something else for fifty thousand dollars. The large companies don't like to do these small works. Their main effort is in the bigger area. But being a smaller company and having a smaller structure you don't have that top-heavy type of organization. We were able to survive. But it becomes difficult. Now you go on to the niche markets where you have something unique, where you can modify it. With the way we design the systems they are very adaptable and very modifiable. So, I don't compete really in a general-type, mass-production market. I compete in the niche area where I have unique capabilities. and yes, it is more expensive. It has to be more expensive. There is no way in the world that I can compete manufacturing-wise with any of the other guys. It's probably going to cost me twice as much to do it, so when they want to they can put it on the market, but can the customer wait? Can he get what he wants? You know, there are areas where I can move it.

Nebeker:

I see.

Gueldenpfennig:

And that means there's a limit as to what a niche-market company can do, how big it can grow, how big are the niches. You know, you have to consider all that. From a company point of view, you have to say, "I'd rather not try to apply mass-production type techniques and business-type philosophy." In a niche-market company, you'd fail.

END OF SIDE ONE.

Gueldenpfennig:

If we just looked at all the markets and went in there... If you look at Alaska, this is a small community. Seventy percent of the installed central offices in Alaska are ours.

Nebeker:

Is that right?

Gueldenpfennig:

Oh, that's a market share. The market is so huge, but it's huge in different areas.

Nebeker:

That's what you identified in the beginning as...

Gueldenpfennig:

We did it in the beginning and we have stuck with and followed it; it has worked.

Nebeker:

And who were the other two who started with you?

Gueldenpfennig:

There really were seven or eight or nine [people], but they were partially in all our areas until we got to the point that we could afford to pay them. It's a bootstrap operation. You get one job, you can add another guy. You get another job, you get another person, and so on.

Test equipment; competitors

Nebeker:

Were you the main player in this?

Gueldenpfennig:

Yes. Oh, I drove it. That's what I wanted to do since I had to move anyway. I sort of had a nice offer from them. And I also had one from Vitar at that time.

Nebeker:

Who?

Gueldenpfennig:

T.R.W. -Vitar was a central office manufacturer who no longer exists. One was in California, and one was down there, Tennessee... I probably would have gone to Northern if I couldn't have made it. In fact, I called them up, it was over Memorial Day weekend, and said, "Well, the bad news is I am not coming. The good news is I found a little bit of backing and I am going to try it. If it fails, I'm going to come." Of course, a lot of people over at Stromberg subsequently moved to Northern and in some cases it became a good customer. We designed some test equipment and I've sold it to all the competitors.

Nebeker:

Is that right?

Gueldenpfennig:

Yes.

Nebeker:

And that's clearly a niche market? Test equipment?

Gueldenpfennig:

Yes, that is a niche. Traffic load systems. Basically simulate the subscriber.

Nebeker:

What was that?

Gueldenpfennig:

To simulate, you know, so many telephones. If you have 100 telephones, you can simulate them all off-hook at the same time. Try that with people. Impossible. And repetitively do the same thing, which helped to trouble-shoot and analyze the problems.

Nebeker:

What's that called? That kind of thing?

Gueldenpfennig:

We call it a traffic generator.

Nebeker:

Traffic generator?

Gueldenpfennig:

It's really a traffic simulator, a subscriber simulator, a load simulator. It comes in so many names, you know.

Nebeker:

I see.

Gueldenpfennig:

But there wasn't any decent equipment in the market. When we had our own switching systems designed, we were looking in the market to find one to buy, so we could test our equipment. Well, we couldn't find one to buy. And in my travels I met a lot of people who said, "Well, when you find one, let me know. I'm looking for them, too." So we got back home, and said, "I guess we have to do our own." So we sat down and made our own. We let some others know, "Well, this is what we're doing." Oh, one of the Northern guys came in and wanted to take it right away. "Wait a minute, wait a minute. This is my lab model. You can't have this!" "Wow, I want eight or nine of them!" "Well, okay." So we struggled to put something together, and it became a very good business relationship.

Nebeker:

Very similar to something I'll mention this evening. The standard signal generators for testing radios.

Gueldenpfennig:

Yes, right.

Nebeker:

General Radio got into that market of producing those. Some other companies had to produce their own in the early days because there wasn't such a thing.

Gueldenpfennig:

Everybody had his own traffic generator in the lab, but nobody built it as a product you could carry from anywhere to anywhere else and reliably use. Repeatedly.

Nebeker:

Was Redcom the first to market such a thing?

Gueldenpfennig:

Well, Siemens had a small one, and Stromberg had one. But they were all built for their own use. They weren't built for general sale. We made one for general sale.

Nebeker:

And did you make money on that?

Gueldenpfennig:

Yes. And after a while, of course, people saw there was a market and there were competitors who popped up. Naturally. You know they'll never be far behind.

Nebeker:

Are there other examples of testing measurement devices in switching?

Gueldenpfennig:

Yes, but we didn't get involved. We didn't get involved in transmission at all. That's basically a main device.

Nebeker:

I see. Transmission...

Gueldenpfennig:

We have some transmission measuring capabilities built in against some default values. To say, "Okay, we've made that call and it's correct. We put ten thousand calls in the switch before we let it out of here and there had better be no error in the setup, or in the transmission quality of the call."

Nebeker:

But as far as a product for testing measurement, your company has produced the traffic generator.

Gueldenpfennig:

Yes, we've sold thousands of them.

Nebeker:

Thousands?

Gueldenpfennig:

And every time we had figured originally, "Well, how many installation crews do you have? And how many manufacturers do you have? And who could use them in their factory fields?" So we came up with 500 to 1,000, but every time we think we're done, some new niche of it shows up and here goes the next batch.

Nebeker:

And you're heading for a succession?

Gueldenpfennig:

Yes, we have successions of them and they become even more sophisticated. We put all the features on the Japanese market. So we delivered a lot of them to NTT. That is still going on.

Nebeker:

Oh, I see. And this is still an important product for you?

Gueldenpfennig: Still an important product. It comes in spurts. That's the funny thing, you know. it's not a product which you continuously sell so many a week, or so many a month. It comes all of a sudden that you sell fifty or sixty of them. It is quiet for a while and then all of a sudden, somebody needs five or ten. And so on. They run anywhere, from a sales point of view, depending on the configuration, on equipment, from $25,000 to $50,000 apiece. And nobody ever plans ahead. That's the other problem, you know. When they need them, they need them now. Since you don't know what they need, it is very difficult to guess how to build them. Which ones to build? Which pieces to have ready and which not?

Programmable switching systems

Gueldenpfennig:

So you have your marketing guess-game, but that's true with pretty much all areas. We pioneered what we refer to now as "dumb switches." I don't consider them dumb; I consider them highly intelligent. But these are switching systems which are fully programmable by the customer. In other words, it doesn't have a call-processing type function in it. It has, instead, an operating system much like a PC. You can tell it to do anything and everything you want it to do.

Nebeker:

You can get standard software for doing the standard switching?

Gueldenpfennig:

Yes, and we don't have anything to do with the final development of the systems function. That would be someone else. That's how people use them in laboratory environments, and in the implementation of ISDN and a special type of structure.

Nebeker:

ISDN?

Gueldenpfennig:

The Integrated Services Digital Network.

PAXs for railroads and utilities

Nebeker:

Can you sketch any other product you had in the...?

Gueldenpfennig:

Well, in the central office we've got the special switching nodes, or switching platforms programmable. These are the programmable switches. Then we've got PAXs. But fairly sophisticated PAXs. Not your old ones. It will do the job of serving an office, or a bureau or something like that. But there is a lot more to it. Our supply is mainly for the uniqueness of the railroads, of the power utilities.

Nebeker:

So you produce one, specifically, say, for the railroads?

Gueldenpfennig:

With a lot of extra features in it, a lot of extra capabilities. They are unique, you know. We have quite a few networks in China where the switches are in the substations of the power networks. They use a radio device to use it as a transmission means to high-tension wires. So they're all tandem switches but they have special override capabilities to get through an emergency there.

Nebeker:

Have you produced things to their specifications?

Gueldenpfennig:

Well, we develop them. Ultimately people started adopting what our general specifications are.

Standards and the IEEE

Nebeker:

I wanted to ask about the standards in this area. How much of it is de facto standards by key players? How much of it is it worked out by...

Gueldenpfennig:

<flashmp3>162 - guldenpfennig - clip 2.mp3</flashmp3>

There's a lot of standards, a lot of standards. They are all related, one way or another. It depends very often on how much a country or an institution, or a government entity, if you wish, wants to maintain its very narrow standards. Very often, standards are used as deterrents to getting in the market. There is no doubt about it. The Japanese do it greatly and then say, "Well, even if they had them, nobody would buy them." They say, "It's just because people don't want to buy them." Right. You get all the answers in the world. Well, if they want it, they'll get it. I find again and again that if you have something somebody else needs, they'll all of a sudden become flexible. They'll try to first push their standards on you, but if they can't get their standards then they'll settle for the next best thing. So you want to be a niche player; you need to find out where is it you have something they don't have and how to get around the standards.

I have to tell many people, when they ask, "Well, here's my spec. Please answer the spec." There are pages and pages where you say, "Well, we don't do it this way' we do it that way. We don't do this, we do that." And so on. You go through this. Invariably. I am not going to design my switch to somebody else's bid papers. Impossible. So I'd have to take exception and say, "But here's what it is". In a lot of areas, that's quite acceptable. They want to know what the difference is. Standards used over the last fifty to a hundred years have developed pending on who was in the country, who has done something. You go into Australia and in the South Pacific and mostly you find Ericsson standards. The company was there. The company put its equipment in there. De facto, that became the standard, and they have maybe modified it a little bit here and there and so made it maybe unique. Although the Japanese, initially after World War II, took a lot of North American equipment and modified it slightly. That became their standards. Anyone who comes in now has to adapt to what is there and that becomes the standard. You go to Iran and it's mostly Germans there. Siemens and so forth.

Nebeker:

What has been the rule of IEEE and standards in this area?

Gueldenpfennig:

IEEE has tried to back into the standards retroactively, at least IEEE here. You know, the CCITT in Europe has written a lot of so called recommendations. Every time I hear it, they are talking about CCITT standards. Yet, when you find out that it isn't really quite that way, then they tell you "Well, it really was a recommendation." Therefore, a slight modification is available . We have a standards committee on the IEEE for telecommunications, too. A lot of the players who are on the switching committee, the transmission committees, and others are also involved in the standards committee. They are trying to write standards, but it's a North American standard. We are trying to look at the others and trying to consider the European standards and somehow bring the world together. But you don't undo what's done. We make some equipment we call I-GATE International Gateway Transit Exchange. We can convert from the North American standards to the European standards in the same piece of equipment. Side by side. I can't go over there and say, "The hell with your standard. Throw it out and use mine." Well, you can't change the whole network. It's been built up over fifty to one hundred years. So you have to adapt in one form or another. Even in this country, it depended on who installed the equipment. If it was Stromberg, it had slightly different operational characteristics than if it was Northern or from AT&T or GTE.

Nebeker:

So, even within this country, the IEEE standards have been more looking back?

Gueldenpfennig:

Yes, it's sort of looking back. There is no such thing as writing the standards up front and then seeing how you can fill in. For general terms, yes. But for the unique ones, a new component comes in the market. It's the "cat's meow.” Let's dream a little bit; it's the "cat's meow" in terms of building the switching systems now. In the future they will take in maybe 20% of what they take in now. You can reduce it that far down or almost get as small as a connector block. You don't need any parts but in order to do that, you have to make some allowances in some interface areas. In some other way, you have to do something different. What do you think is going to happen? Somebody is going to put it in somewhere. It is going to get in and start to grow. If it's also economically done, it'll grow more and more and more. Eventually somebody will say, "Wait a minute. This is not according to the standards!" Well, great. How do we shoehorn this in to the standards? Invariably, it's going to happen that we are going to make some allowances for that. I look at standards, and supposedly the REA writes a spec. Well, they have to change. Their paragraphs about control architectures were written twenty to thirty years ago. They are totally inadequate for what is today. Do we want to forgo new technologies because we have a standard which prohibits the new technology? That's what is so difficult about standards. You know, if we just say in terms of standards what it should do and refrain from any kind of implementation as to how to do it, then we would be all right. But that's a difficult thing for engineers to do.

Nebeker:

I also imagine there are a lot of practical reasons for specifying exactly how you do it.

Gueldenpfennig:

That's right. You know, some technology does not subscribe or is not compatible with that standard. Now do you want to therefore throw a new development overboard if it is good or if it can be done and it accomplishes the same thing? No. You must be open. Otherwise you stifle the development.

Nebeker:

What about international bodies trying to teach international standards?

Gueldenpfennig:

Oh, there are international standards and there are representatives on the respective committees. When you look at the IEEE and you look at the Telecommunications Conferences, we have these international conferences and most of the committees have international members on it. For instance, the Switching Committee has Italians in it, French, Japanese. A lot of people on there with some input into the various things if they want to. They could obviously sit through the conferences and take the minutes and walk away, but then between the committees there are liaisons who are representatives on other committees. One of us was on the Standards Committee and he took the Standards Committee Report back to other committees. And so on.

Nebeker:

Does it seem to serve a useful purpose?

Gueldenpfennig:

Yes, it serves a good purpose because a lot of things in technical papers are written and sometimes technical papers are co-sponsored by two different committees. Maybe there is a paper which deals in software or in control architecture but has a profound impact on transmission. Well, maybe the Transmission Committee will co-sponsor that. It's a discussion point. You know, people read these, go through them, and they could be international or domestic papers. They work in their respective fields. It's quite a bit of exchange going on.

Historical trends in telecommunications

Influence of transistors

Nebeker:

What about if one looks at it through the years that you have been in the field, purely from the technical capabilities side? I mean, you obviously have earlier developments, but the transistors are coming in and...

Gueldenpfennig:

I remember the old days when the transistor was literally a little can with three legs, out there. You had to borrow that thing from the library and return it, and you couldn't let any harm come to it, and if you or a breeze blew on it the parameters moved away. We would take a heat gun and warm it up a little bit and see how the collector current changed and things like that. I mean, that ancient. You know, that dates me, unfortunately.

Nebeker:

I am glad. I'm trying to get the history so I am glad to hear that.

Gueldenpfennig:

I tell you, it is unbelievable. We're talking about old steam radio and stuff like that. But you know, I watched all this to go on. Obviously the transistor moved into two areas. It moved into the radio technology and it also moved into the digital. It seemed to have moved from an audio point of view much earlier into the radio area, then into the telecommunications area. Yet, when it came to the control and total systems point, the software control aspect of it, the processor functions of it or the switching areas, the telecommunications areas, it moved a hell of a lot faster. Only recently do you see digital radios, processor-controlled radios, programmable radios and all of that. These things are getting very sophisticated, but only in the last maybe ten years, while in telecommunications it's been twenty to twenty-five years. So it seems to be shifting position.

Nebeker:

In the telecommunications area, was it initially used for replacing tubes? Were tubes being used?

Gueldenpfennig:

No, we used relays. There were some tubes mainly in transmission areas and yes, there were some old flip-flops built out of some tubes, with some old triodes. But by and large it was relay-exchanges, because in the telephone world you had to take DC instructions for signaling. They are wired so you had to signal the directions where you had to go, and if the radio or your hair moved, or if you went to some frequencies, your modulation was not compatible. The air is not a clash point of interfacing. The wires you go into may be. Many things can happen to the wires. The old cables were paper-insulated and they got wet and the shorts were in there and your resistances were not right. Now we don't have to worry about that. Your antenna is out of tune but you can do something about it. Your standing-wave ratio went to hell, you know. Or your dipole fell down. We are now using radios to complement the network. Obviously how do you get from here to there? Well, you go up to the satellite and you come down. There are many different ways to do it, you know. You've got transponders but yet at the same time you've got demand access control-type functions to occupy the spectrum efficiently because it's very expensive to take fixed satellite transmission band waves. All these things are interrelated these days in the switching field. At the end of the routing station where does it tie into? It ties into the network. What is sitting right after the network? Re-routing mechanisms. They are the switching systems. Where do we go from there? We don't broadcast; we're dedicated to going into specific directions. That's the big difference, you know. With radio, you broadcast, and the guy who picks it up at the other end is the one you may want to talk to, or maybe not. The telephone world is different. You have me and only the dedicated channel.

Frequency and bandwidth; digital converters

Nebeker:

I'm just trying to understand this. Do you have in the telephone world the people who are concerned with the audio signal? Principally?

Gueldenpfennig:

Oh, yes.

Nebeker:

Of course. And then you have people who are principally concerned with the switching?

Gueldenpfennig:

Of course. All audio, you know, for communication principles is a 4 K band frequency. That's what we use; that's the channel bandwidth. 64 K bits gets you there. Basically, that's the digital implementation of it. At an 8 kHz sample rate that's what you get for kHz. But now we're getting more and more, because there's data transmission and a tremendous amount of info which is always required. It's an information world. We're looking at wider and wider bandwidths.

Nebeker:

So that's one thing. I'm looking for those generalizations or trends or whatever. One would be that people are in the last decade or two trying to accommodate more data over the same lines...

Gueldenpfennig:

Right, like ISDN. You have 2B+D which means two channels of 64 K bits plus a D channel which is a 16 kHz. So generally you're looking at 144 K-bit transmission bandwidth. Well, that whole thing is a lot of systems got designed towards that. Now we've changed. Now we're talking broad-band ISDN. Now we want to go broader and broader because the info we want to transmit not only is a voice but it also becomes visual. You want to try and transmit your TV pieces, your movies, your own interactions, your video-conferencing. All that requires band-width and all of that has got to be done directly. It's not a broadcast situation.

Nebeker:

On the one band, though, you might think that this is a separate matter. One thing is mostly control, switching and all of that. You needn't think about maybe...

Gueldenpfennig:

Well, it gets a little less simple. I mean it can be done and it is done, but it's not as easy to switch broad-band as it is to switch narrow-band.

Nebeker:

So they're a whole new...

Gueldenpfennig:

If you want to use fiber optics, you can put tremendous bandwidths on that. Can you switch it? Not so easily. You've got to reduce it back down, and I've seen some devices, playing with prisms, where they break down the light beam and so on and then say, "Okay, this part of it on the infrared end will go this way; the blue end will go that way," and use that as a switching mechanism by applying some kind of force to the brake of the class so it can route the call. Crazy things. Now, people are talking about photonic switching.

Nebeker:

Okay, but for practical purposes, still today all switching has to be electronic?

Gueldenpfennig:

Yes, and the general standard band was the A channel. A channel is 64 K bits. That's all over the world now. There is a standard. In North America we combined twenty-four channels, plus we borrow every frame a bit, in order to do some signaling. The Europeans use thirty channels and have two dedicated channels for it. So that's thirty plus two.

Nebeker:

Does that kind of a difference mean that it's difficult to produce some device for both markets?

Gueldenpfennig:

Well, no, we can. We do, actually. On top of it, to make it more difficult, we have different combining ratios. We have different low-end behavior in the frequencies than the Europeans have. That's the famous µ (mu) 250 vs. µ 100. The European combining ratio is A-law, and the North American is µ- law. The only way you can combine it is from a linear function, so it's not directly compatible. What you literally would have to do is to convert from one to the other, to linearize an 8-bit signal. Take the combining out of it, linearize it, then re-combine it with the other behavior. And we can do it. We can update them with some other things inside a switching matrix, so it's a little unique. I can convert from twenty-four channels to thirty-two channels, from A-law to µ- law. The mix is 120 channels, if you wanted to have it match. You know, the smallest denominator. It's 4/30 and 5/24, so you build a little switch like that.

Nebeker:

I see. Have you produced such things?

Gueldenpfennig:

Oh, yes. We've already achieved a number of them.

Nebeker:

I see.

Gueldenpfennig:

You know, they happened to have come in very handy when dealing with today's new satellites. There are American satellites, and there are European satellites. When they get digital, some work one way, some work the other. So at the entry point it depends on what you're getting. Are you getting North American? Are you getting European? What is it that you're going into? That may be the other one. Then you go across certain borders or so, or other communications links through cables even. You could have the same point where the entry point is. There may be a difference and you have to match it. Now, in the old days you took the digital signal and went back to analog. Then you had channel by channel analog go in there, and reconverted it to digital. Well, you do that six times and your signal to noise ratio is a little affected.

Nebeker:

So that's the reason for digital converters?

Gueldenpfennig:

You want to stay digital.

Nebeker:

I see.

Gueldenpfennig:

Digital has the great advantage of that you rebuild the signal at the other end based on the information that it's clean. You can wipe out a lot of that noise.

Influence of digitization on switching technology

Nebeker:

Another one of these very general questions. When you look back, what difference does digitization of the signal make to switching technology?

Gueldenpfennig:

Well, when we started out, everybody ran around and said, "Oh, it is more reliable, it is a lot faster, it is much cheaper, and it will use a hell of a lot less power." Well, there are very few things which are right in that statement. In retrospect, yes, it uses for instance less power, less peak power. But it uses more average power. You know, selectors, relays, and all of that only use power when you operate them. And that happens when you have traffic. Well, in digital systems you always use power whether you have traffic or not. Much higher basic power and then some incremental power. So there's already a difference. If you're talking about more reliable, well, the way I look at it, reliability is inversely proportional to the amount of components you've got in the box. And it doesn't matter who builds it. Right? The more you've got, the more that can fail and the less reliable it is. You're building them better, and better, and it seems to be that overall it needs less maintenance. No doubt about it. But it doesn't really mean that it is more reliable. Then we've got to separate catastrophic failures from service-affecting failures. It's a huge ball of wax, if you want to compare and contrast everything. It's another subject.

Nebeker:

But what we're hoping for is someone who can summarize in two sentences the difference in it.

Gueldenpfennig:

Yes, there is a difference.

Nebeker:

And I appreciate it. It's not easy.

Gueldenpfennig:

Well, there is a big difference between the catastrophic failure and a failure which reduces your capability but does not take you totally out. In other words, if you have to jump off a cliff, you want know where the nearest parachute is. That might help. You know you don't have another chance. I say that instead of having five channels, I could probably get by with three. I could still communicate. But if all of it is gone, then I've got a problem. So which device, which system, which control technology, allows me one versus the other? No matter how good it is, if the system has a catastrophic failure, then what do you do? Would you like to go on to the Space Shuttle and would you like to have reduced capability? Or would you like to entertain catastrophic failures? It depends on the application entirely.

Nebeker:

Yes.

Gueldenpfennig:

Otherwise, the game is over. So, that applies to a larger extent in the philosophy of switching technology too. That's been basically my idea, telling people the way we built it as a distributive system; the more you've got, the less likely you are to have a catastrophic failure. Worse is what the common control system is, and we pioneered that and we keep repeating it. Of course a lot of people tried to jump onto the bandwagon, but that's what it is all about.

Nebeker:

Well, I certainly appreciate your time.

Gueldenpfennig:

I talk a mile a minute!

End of Interview