THE simplest way to understand the Cunningham crossbar and switch and to see why Cunningham was ideally suited to develop it is to consider its evolution.
In 1946 a young electrical engineer, Andrew W. Vincent, left his job with Stromberg-Carlson in Rochester in order to devote himself to perfecting a small dial telephone system of his own devising. The heart of this system was the crossbar switch, and as Vincent worked he became more and more interested in the possibilities of the switch, until at last he was spending as much time considering them as he was on the system itself.
The switch that Vincent considered was and still is the ultimate development of electro-mechanical telephone switching. Early telephone systems had relied upon an operator at a switchboard to make connections between callers. As the number of subscribers mounted and as telephone systems were linked from town to town, the role of the operator became more complex, but her essential function did not change; it was simply to find and hold a path of communication through a network of wires. Various switching devices were called into being to assist the operator in this function, but she continued to play the primary role.
The crossbar switch changed that by performing this role automatically. When you dial a number nowadays you are, in effect, instructing a machine instead of a human operator to connect you with whatever telephone you are calling. In a nation that has eighty million telephones automatic switching is not merely labor-saving; it is a practical necessity.
It is worth remarking that the crossbar was not the first automatic switching device. Very early in the history of the telephone inventors had granted switching systems designed to replace human operators; some of these were actually adopted by telephone companies and are still in use, but they had mechanical and electrical drawbacks that the crossbar switch does not suffer from. Novelty of invention is often less significant than the subsequent refinement of what has been invented. The point has a bearing on what follows.
Vincent, trying to produce an improved dial telephone system, was only one of many engineers working with the crossbar switch; but in one respect he was unique. Instead of trying to build on the developments that had already been made, he went back to the fundamentals of the switch and considered them. To any competent electrical engineer these appeared obvious, but because Vincent did not take them for granted he discovered hitherto unsuspected potentialities in the crossbar. In the words of Alfred North Whitehead, "It requires a very unusual mind to undertake an analysis of the obvious."
The ultimate result of Vincent's analysis was to be a crossbar switch unlike any other. Its qualitative differences would give it a reliability and a range of uses far beyond those of the ordinary crossbar. Most important, many of these uses would be outside the field of telephony.
But this is to anticipate and to run the risk of missing the point of the story. When Vincent, in 1950, was ready to look for a manufacturer, his intention was simply to produce an improved dial telephone system, and his improvements in the switch were incidental to that. The model of his crossbar switch that he brought to Cunningham was incorporated in a model of the dial system.
Francis Cunningham, the younger brother and co-partner of Augustine Cunningham was receptive, and the brothers began to consider seriously the possibility of adding telephone equipment to their products. Their decision was made in 1951; the firm took over Vincent's initial designs and patent applications and hired him as a consultant.
Development work began in 1952, and as it got under way Andrew Vincent and Peter Cunningham, the son of Francis Cunningham, traveled through the United States interviewing potential customers. At this point the course of events took a characteristic, which is to say unexpected, direction. Manufacturers of telephone equipment, on whom the firm had looked as its likeliest customers, were polite but noncommittal. "Come back and see us when you have a complete line." Companies outside the field of telephone equipment: men in the new television industry and makers of automatic machine tools and machines for gathering and processing information were far more encouraging. However, what aroused their interest was the crossbar switch rather than the dial system.
Their response gave a more specific direction to the work in the Rochester plant. Emphasis was put on developing the switch, and Vincent's title when he joined the firm shortly afterward was Chief Engineer of the Crossbar Switch Department.
The first deliveries were made in 1953. In effect, if not in intention, the work thus far had been a pilot operation, and it revealed practical shortcomings. The sophisticated uses for which the Cunningham crossbar was potentially suited demanded refinement in design and technique of production.
Accordingly, at the end of 1953 Cunningham decided to reduce its production to a minimum in order to concentrate on designing and re-tooling. It was not an easy decision to make or adhere to. Re-tooling, especially, was expensive, and there was always the temptation to abandon the entire project on the grounds that it was too ambitious for a small firm; but against that were the tradition of persistence and the recognition that the Cunningham crossbar was unique.
It took three years to solve the design and manufacturing problems that had occasioned the setback, but during these years the firm was not simply catching up. In its restricted production it was creating prototypes of switching devices that it would later develop in more complex form.
By November 1956 Cunningham was ready to resume full-scale production, and this time it was altogether successful. The switch consistently met its customers' high standards of reliability and durability, and the list of these customers began to resemble a roster of America's leading business firms and scientific agencies. Throughout 1957 the volume of orders increased, at first steadily, then at an accelerated pace, and it has continued to increase ever since.
Seeing a Cunningham crossbar an expert would be struck by its versatility. In his language, it is unique in its ability to switch electrical information from low-level DC signals to 100 megacycles, and it does this reliably, at high speeds. He would be impressed, too, by the precision with which it is constructed and its consequent long life. A layman, ignorant of the uses to which it could be put and aware only of a glittering complexity of wires and metal, would nevertheless sense a certain elegance in the product. In the words of an electrical engineer too young to remember Cunningham carriages and automobiles, it is "the Rolls-Royce of crossbars."
CROSSBAR SWITCHING CONTROL FOR NASA TELEMETRY CENTER,
The Cunninghams of the third generation, Augustine and Francis, lived just long enough to see the first successful results of their struggle to maintain the firm. Augustine Cunningham died in 1957, Francis in 1958.
Brought up in an era of tranquility, when the prosperity of the firm had been taken for granted, and having successfully made the transition from carriages to automobiles, in middle age they had suddenly been confronted by adversity, and they had maintained the company in spite of it. Moreover, they had maintained it through a series of bewildering technological and economic changes and dislocations. James Cunningham had experienced change, but it was gradual and nearly always to his advantage; and his setbacks were short-lived. Joseph Cunningham and Rufus Dryer had lived in a monotony of fair weather: their lives coincided almost exactly with the most peaceful era civilization has ever known. If life had not always been easy for them, at least the world in their time had been predictable. In the world of Augustine and Francis Cunningham almost nothing was predictable except change itself.
With the death of Francis Cunningham in 1958, his son, Peter Cunningham, became the fourth president of the firm.
He had had that training that seems most suited to manage an enterprise in an era of change; that is, a training that confers experience in dealing quickly with a multiplicity of situations. Between the time of his graduation from Harvard in 1939 and his joining the firm after the second World War, he had worked on a newspaper, in an architect's office, in a forge, for a tool company, and at Bausch and Lomb, as a trainee in their export sales department. During the war he became station manager of the Leopoldville airport; in those days it was a key point on the route the Air Transport Command had established across Africa to by-pass the Axis powers. From Leopoldville he went on to Brussels, in 1944, to take over Melsbroek airport for the A. T. C. As a member of the firm after the war he had lived through its years of making garden machines and trailer equipment.
But if he had gained the basic practical experience that
manufacturers have always needed, when the firm decided to make crossbar switches he was confronted with the necessity of extending his education in a way that is so characteristic of modern industry that it is worth dwelling upon. He had to learn a new and complex language, that of a highly specialized technology. His decisions and plans, ultimately his success as a producer, would depend on his skill in using it.
AUTOMATIC TESTING SYSTEM FOR MINUTEMAN MISSILE
For James Cunningham the problem had hardly existed. He was his own engineer; in fact, he had begun his career in the role of technician and picked up the language of commerce as he progressed. In the second generation, Joseph Cunningham and Rufus Dryer, having served an apprenticeship under James Cunningham, were altogether competent to make decisions based on a knowledge of the fine points of carriage-making. It was not until the coming of the automobile that the specialization of the functions of management and the engineer began to create difficulties in communication; and these were by no means insuperable. But as technology became more complex its language became more remote from everyday experiences; moreover, it was now a language not only of the engineer but that of the customer, for the customer was no longer a private citizen who wanted a fine carriage or automobile. He had become a technologist himself, and he spoke for a corporation, in the language of technology.
The problem was not simply to learn to discuss the crossbar switch in its own language. As a manufacturer, Peter Cunningham had also to be able to translate back from that language into the ordinary language of business, to conduct a dialogue in two languages at once. This was what he was learning in the course of his travels with Andrew Vincent.
The skill was essential not only to the commercial success of the enterprise but to the development of the switch itself. The Cunningham crossbar was not first invented, as a static thing, then produced. It evolved in the course of a long dialogue between engineers and producers and users, a dialogue that still continues.
One early result of this dialogue was especially significant. In 1957, as part of an attempt to arouse wider interest in its crossbar switches by demonstrating some of their applications, the firm produced for the first time a crossbar system; that is to say, an automated machine using a crossbar for its switching. The function of this particular system was to read and record the computations of an analog computer. It was successful, and the firm began to make more systems, with a diversity of uses.
The need to communicate had further consequences. One of them was the creation within the plant of a group of applications engineers, who serve as liaison between the firm and its customers. These have made Cunningham known as a solver of switching problems. Another was the publication of technical bulletins and a small technical journal, Select and Hold, in which the nature and developing uses of crossbar switching are explained. And as a result of the success of its systems, the firm organized a group of systems engineers, more or less distinct from a group that concentrates on the switch itself. The switch is still the heart of the enterprise, but like most successful products it has continued to react upon and develop the purposes of its makers.
The reader who has stayed with the story this far will be aware of a certain irony. An old firm that appeared to be doomed in an era of mass-production because of its tradition of quality-production survived and prospered in a later era, that of automation, precisely by virtue of this tradition.
How can we account for it? The reason seems to lie in the nature of automation. The essence of automation is feedback: an automated machine responds to and even modifies its circumstances. Clearly a mechanism that does this must be more refined, more reliable, more of a "quality product" than one whose activities can constantly be guided and corrected by a human operator. It was not accidental that the Cunningham crossbar switch was developed by a small firm with a tradition of quality production.
SCANNEX SWITCHING SYSTEM FOR TRANSDUCER DATA ACQUISITION
Conceivably a large firm might have done it. It might have transformed the ordinary crossbar switch into something so superior in quality and so adjusted to the various demands of its users that the net result was a new and unique product; but the odds would have been against it. What was needed was something more than the application of funds and engineering skill. It was the psychological commitment of the whole firm to the enterprise and an ability on the part of management as well as its engineers to enter directly into the dialogue between producer and consumer.