Other articles in the series:
- History of the relay
- History of electronic computers
- History of the transistor
- Internet history
worked one on one, connecting one pair of stations. But already in 1877 imagined a universal connected system. Bell wrote in an advertisement for potential investors that, just as municipal gas and water networks connect homes and offices in large cities to distribution centers,
One can imagine how cables of telephone wires will be laid underground or suspended above, and their branches will go to private houses, country estates, shops, factories, etc., etc., combining them with a main cable with the central office, where the wires can be connected in any way, establishing a direct connection between any two places in the city. Moreover, I believe that in the future, wires will connect the Telephone Company headquarters in different cities, and a person from one part of the country will be able to communicate with another person from a remote place.
But neither he nor his contemporaries had the technical ability to realize these predictions. It will take decades and the application of a great deal of ingenious invention and hard work to transform the telephone into the most extensive and intricate machine known to mankind, which will cross continents, and eventually oceans, to connect any telephone exchange in the world to any other.
This transformation was made possible thanks, among other things, to the development of the switchboard, a central office with equipment capable of redirecting a call from the calling line to the called line. The automation of switches led to a significant increase in the complexity of relay circuits, which greatly affected computers.
First switches
In the early days of telephones, no one could say for sure why they were needed. The transmission of recorded messages over long distances has already been mastered and has shown its usefulness in commercial and military applications. But there were no precedents for the transmission of sound over long distances. Was it a business tool like the telegraph? A device for social communication? A means of entertainment and moralizing, like broadcasting music and political speeches?
Gardiner Green Hubbard, one of Bell's main sponsors, found one useful analogy. Telegraph entrepreneurs had built many local telegraph companies over the previous decades. Rich people or small businesses rented a dedicated telegraph line linking them to the company's central office. After sending a telegram, they could call a taxi, send a courier with a message to a client or friend, call the police. Hubbard believed that in such matters the telephone could replace the telegraph. It is much easier to use, and the ability to maintain voice contact speeds up the service and reduces misunderstandings. So he encouraged the creation of just such a company, offering to rent telephones associated with local telephone companies, both newly formed and converted from telegraph stations.
The manager of one of these telephone companies might notice that he needs twenty phones to talk to twenty customers. And in some cases, one customer wanted to send a message to another—for example, a doctor sending a prescription to a pharmacist. Why not just let them talk to each other?
Bell himself may have come up with such an idea. He spent most of 1877 on lecture tours promoting the telephone. George Coy attended one of these lectures in New Haven, Connecticut, when Bell was ranting about his vision for a central telephone office. Coy was inspired by the idea, organized the New Haven District Telephone Company, acquired a license from the Bell Company and found the first subscribers. By January 1878, he had connected 21 subscribers with the first public telephone exchange, fashioned from used wires and handles from kettle lids.
Within a year, the same makeshift devices for connecting local telephone subscribers began to appear throughout the country. A speculative societal pattern of telephone use began to crystallize around these hubs of local communication—between salespeople and suppliers, businessmen and customers, doctors and pharmacists. Even between friends and pals who were rich enough to afford such a luxury. Alternative methods of using the phone (for example, as a means of broadcasting) began to gradually fall away.
Within a few years, telephone offices agreed on a common layout of switch equipment that would be stable for decades to come: an array of jacks that the operator could connect via plug-in wires. They also agreed on an ideal field for the operator. At first, the telephone companies, many of which grew out of telegraph companies, hired people from the available labor force - young clerks and messengers. But customers complained about their rudeness, and managers suffered from their violent behavior. Pretty soon they were replaced by polite, decent girls.
Further development of these central exchanges will define a contest for telephony dominance between the Goliath-class Bell company and emerging independent competitors.
Bell and independent companies
The American Bell Telephone Company, holding Bell's 1876 patent number 174 for "improvements to the telegraph", was in an extremely advantageous position due to the rather broad scope of the patent. The court ruled that this patent included not only certain instruments described in it, but also the principle of transmitting sound through a wave current, as a result giving Bell a monopoly on telephony in the United States until 465, when the 1893-year patent expired.
Management companies wisely used this period. Of particular note is the President и . Forbes was a Boston aristocrat and top of the list of investors who took control of the company when Bell's early associates ran out of money. Vail, great-nephew of partner Samuel Morse, , was president of the most important of the Bell companies, Metropolitan Telephone, located in New York, and was the general manager of American Bell. Vail showed his managerial character as head of the Railroad Mail Service, sorting mail in wagons on the way to its destination, considered one of the most impressive logistical feats of its time.
Forbes and Vail focused on having Bell in every major city in the country, and having all those cities connected by long-distance communication lines. Since the company's greatest asset was its existing subscriber base, they believed that unparalleled access to existing Bell network customers would give them an insurmountable competitive advantage in recruiting new customers once the patent expired.
Bell entered new cities not under the American Bell name, but by licensing its set of patents to a local operator and buying a majority stake in that company in the deal. To further advance and expand lines connecting city offices, they founded another company, American Telephone and Telegraph (AT&T) in 1885. Vail added to his impressive list of positions the presidency of this company. But perhaps the most important addition to the company's portfolio was the acquisition in 1881 of a controlling interest in the Chicago electrical company Western Electric. Initially founded by Bell rival Elisha Gray, it then became Western Union's primary equipment supplier to eventually become a manufacturer at Bell.
It was not until the early 1890s, near the end of Bell's legal monopoly, that independent telephone companies began to crawl out of the corners that Bell had driven them into with the club U.S. Patent No. 174. Over the next twenty years, independent telephone companies posed a serious threat to Bell, and both the parties expanded rapidly in the struggle for territory and subscribers. To spur expansion, Bell turned its organizational structure inside out in a magician's gesture, transforming AT&T from a private company into a holding company. American Bell was issued according to the laws of the piece. Massachusetts, which followed the old concept of a corporation as a limited public charter - so American Bell had to ask the state legislature to enter a new city. But AT&T, organized under New York's liberal corporate laws, had no such need.
AT&T expanded networks and founded or bought companies to consolidate and protect its claims to major urban centers, extending its ever-growing long-distance network across the country. Independent companies were moving into new territory as fast as they could, especially in small towns where AT&T hadn't yet reached.
During this intense competition, the number of phones in use grew at an astonishing rate. By 1900 there were 1,4 million telephones in the US, compared to 800 in Europe and 000 in the rest of the world. For every 100 Americans, there was one device. In addition to the United States, only Sweden and Switzerland have come close to this density. Of the 000 million phone lines, 60 belonged to Bell subscribers, and the rest to independent companies. In just three years, those numbers rose to 1,4 million and 800 million, respectively, and the number of switches approached the tens of thousands.
Number of switches, approx. 1910
The growing number of switches put even more pressure on central telephone exchanges. In response, the telephone industry developed a new technology for switching that branched into two main parts: one favored by Bell was served by operators. Another, adopted by independent companies, used electromechanical devices to completely eliminate operators.
For convenience, we'll call this the split between manual and automatic switching. But don't let this terminology fool you. Just as with "automated" checkouts in supermarkets, electromechanical switches, especially their early versions, put additional strain on customers. From the telephone company's point of view, automation reduced the cost of labor, but from a systems point of view, they shifted the paid labor of the operator to the user.
Operator on standby
During this era of competition, Chicago was the Bell System's primary center of innovation. Angus Hibbard, CEO of Chicago Telephone, was pushing the boundaries of telephony to increase opportunities for a wider user base—and AT&T headquarters didn't like it. But since there wasn't too much of a connection between AT&T and the operating companies, she couldn't control him directly—just stare and wince.
By then, most of Bell's customers were merchants, business leaders, doctors, or lawyers who paid a flat fee for unlimited telephone use. Few people could still afford to pay $125 a year, the equivalent of several thousand dollars today. To expand service to more customers, Chicago Telephone introduced three new offerings in the 1890s that offered both lower cost and reduced levels of service. First, there was a service with a time meter on the line with access for several persons, the cost of which consisted of a per-minute and very small subscription fee (due to the division of one line between several users). The operator recorded the client's use of time on paper - the first automatic counter in Chicago did not appear until after the First World War. Then there was a service for local switches, with unlimited calls for several blocks around, but with a reduced number of operators per client (and therefore with increased connection time). And, finally, there was also a pay phone installed at home or in the client's office. A nickel was enough to make a call lasting up to five minutes to any place in the city. It was the first telephone service available to the middle class, and by 1906, 40 of Chicago's 000 telephones were pay phones.
To keep up with his rapidly growing subscriber base, Hibbard worked closely with Western Electric, whose main factory was also located in Chicago, and specifically with Charles Scribner, its chief engineer. Now no one knows about Scribner, but then he, the author of several hundred patents, was considered a famous inventor and engineer. Among his first achievements was the development of a standard switch for the Bell system, including a connector for the operator's wire, called the "jackknife" for its resemblance to a folding pocket knife [jackknife]. Subsequently, this name was shortened to "jack".
Scribner, Hibbard and their teams redesigned the central switchboard layout to increase operator efficiency. Busy and howler signals (indicating that the handset was off the hook) freed operators from the burden of informing callers of an error. Small electric lamps indicating active bells replaced the latches that the operator had to put in place each time. The operator's greeting "hello", inviting to the conversation, was replaced with "number, please", suggesting only one answer. With these changes, the average call time for local calls in Chicago decreased from 45 seconds in 1887 to 6,2 seconds in 1900.
Typical switchboard with operators, ca. 1910
While Chicago Telephone, Western Electric, and other Bell tentacles worked to make carrier communications fast and efficient, others tried to get rid of carriers altogether.
Elmon Brown Strowger
Devices for connecting telephones without human intervention have been patented, demonstrated and put into operation since 1879 by inventors from the USA, France, Britain, Sweden, Italy, Russia and Hungary. In the United States alone, by 1889, 27 patents for an automatic telephone exchange were registered. But, as has so often happened throughout our history, the credit for inventing the automatic switch unfairly went to one man: Elmon Strowger. It's not entirely wrong, because before him people built disposable devices, treated them as fun things, couldn't get out of the small and slow growing phone markets, or simply couldn't use the idea well. The Strowger machine was the first to be introduced on an industrial scale. But it’s also impossible to call it “Strowger’s machine”, because he himself never built it.
Strowger, a 50-year-old Kansas City schoolteacher-turned-entrepreneur, looked little like an innovator in an era of ever-increasing technical specialization. The stories of his invention of the switch have been told many times, and they seem to belong to the realm of myths, not hard facts. But they all stem from Strowger's displeasure that his local telephone exchange operators were redirecting customers to his competitor. It is no longer possible to know whether such a conspiracy actually existed, and whether Strowger was a victim of it. Most likely, he himself was not such a good entrepreneur as he considered himself. In any case, from this situation, the idea of a phone "without girls" was born.
His 1889 patent described the appearance of a device in which a rigid metal hand replaced the delicate handle of a telephone operator. Instead of a wire with a jack, she held a metal contact that could move in an arc and select one of 100 different client lines (either in one plane, or, in the “two-motor” version, in ten planes with ten lines each).
The caller controlled his hand with two telegraph keys, one for tens, the second for units. To connect with subscriber 57, the caller pressed the tens key five times to move the hand to the desired group of ten customers, then pressed the ones key seven times to reach the desired subscriber in the group, then pressed the final key to connect. On a telephone with an operator, the caller simply had to pick up the phone, wait for the operator to answer, say "57" and wait for the connection.
The system was not only tedious to use, but also required redundant equipment: five wires from the subscriber to the switch and two batteries for the phone (one to control the switch, the other to talk). By that time, Bell was already transitioning to a centralized battery system, and their newest stations had no batteries and only one pair of wires.
Strowger is said to have built the first model switch from pins stuck into a stack of starched collars. To implement a practical device, he needed the financial and technical assistance of several important partners: in particular, businessman Joseph Harris and engineer Alexander Keith. Harris provided funding for Strowger and oversaw the creation of the Strowger Automatic Telephone Exchange Company, which made the switches. He wisely decided not to locate the company in Kansas City, but in his home in Chicago. Because of its presence, Western Electric was at the center of telephone engineering. Among the first engineers hired was Keith, who moved to the company from the world of electricity generation and became the CTO of Strowger Automatic. With the help of other experienced engineers, he turned Strowger's crude concept into a precision instrument ready for mass production and use, and directed all major technical improvements to this instrument over the next 20 years.
Of this series of improvements, two were particularly important. The first was the replacement of many keys with a single dial that automatically generated both the pulses that moved the switch to the desired position and the connection signal. This greatly simplified subscriber equipment and became the default mechanism for controlling automatic switches until Bell introduced tone dialing to the world in the 1960s. The automatic telephone has become synonymous with the rotary telephone. The second was the development of a two-way switching system that allowed first 1000 and then 10 users to connect to each other by dialing 000 or 3 digits. The first level switch selected one of ten or one hundred second level switches, and that switch selected the desired one from 4 subscribers. This allowed automatic switchboards to become competitive in large cities with thousands of subscribers.
Strowger Automatic installed the first commercial switch in Laporte, Indiana in 1892, serving eighty subscribers of the independent Cushman Telephone Company. The city's former Bell affiliate successfully exited by losing a patent dispute with AT&T, giving Cushman and Strowger an excellent opportunity to take his place and poach his clients. Five years later, Keith led the first installation of a two-layer switch in August, Georgia, serving 900 lines.
By that time, Strowger had retired and lived in Florida, where he died a few years later. His name was dropped from the Automatic Telephone Company and it became known as Autelco. Autelco has been a major supplier of electromechanical switches in the US and much of Europe. By 1910, automatic exchanges served 200 American subscribers at 000 telephone exchanges, almost all of which were built by Autelco. Each was owned by an independent telephone company. But 131 were a small fraction of America's millions of telephone subscribers. Even most independent companies followed in the footsteps of Bell, and Bell itself had not yet seriously considered replacing its operators.
General management
Opponents of the Bell system have tried to attribute the company's commitment to using operators to some malicious motive, but their accusations are hard to believe. There were several good reasons for this, and one that seemed reasonable at the time, but in retrospect looks wrong.
Bell needed to develop its own switch first. AT&T was not going to pay Autelco for its telephone exchanges. Fortunately, in 1903 she acquired a patent for a device designed by the Lorimer brothers of Brantford, Ontario. It was in this city that Alexander Bell's parents settled after leaving Scotland, and where the idea of the telephone first came to his mind when he was visiting there in 1874. Unlike the Strowger switch, the Lorimer's device used reverse pulses to move the selector lever - that is, electrical pulses came from the switch, each switching a relay in the subscriber's equipment, causing it to count down from the number set by the subscriber on the lever to zero.
In 1906, Western Electric loaded two separate teams into developing switches based on the Lorimers' idea, and the systems they created—panel and rotary—formed the second generation of automatic switches. Both of them replaced the lever with a conventional dialer, moving the pulse receiver inside the central station.
More importantly for us, the mechanics of Western Electric's switching equipment - meticulously described in great detail by telephone historians - were the relay circuits used to control the switching. And historians have mentioned this only in passing.
And this is a pity, because the appearance of control relay circuits has two important consequences for our history. In the long run, they inspired the idea that combinations of switches could be used to represent arbitrary arithmetic and logical operations. The implementation of these ideas will be the subject of the next article. And first, they got around the last major engineering problem with automatic switches—the ability to scale to serve large metropolitan areas where Bell had thousands of subscribers.
The Strowger switch scaling method used by Alexander Keith to switch between 10 lines could not be scaled too much. If we continued to increase the number of levels, then each call required too much equipment to be dedicated. Bell engineers named the alternative scaling mechanism sender. It stored the number dialed by the caller in a register, then translated this number into arbitrary (usually non-numeric) codes that controlled the switches. This allowed much more flexibility in setting up switching - for example, calls between switches could be redirected through the central station (which did not correspond to any digit in the dialed number), instead of connecting every switch in the city to all the others.
Apparently, , a research engineer in the AT&T Traffic Division, was the first to come up with the "sender". Molina was noted for innovative research that applied mathematical probability to the study of telephone traffic. These studies led him around 1905 to the idea that if call forwarding could be decoupled from the decimal number dialed by the user, the machines could use the lines much more efficiently.
Molina mathematically demonstrated that spreading calls over larger groups of lines allowed the switch to handle more calls while keeping the probability of a busy signal at the same level. But Strowger's switches were limited to a hundred lines, selected with two digits. Switches with 1000 lines based on three digits were found to be inefficient. But the selector movements controlled by the sender did not have to match the numbers dialed by the caller. Such a selector could choose from 200 or 500 lines available to rotary and panel systems, respectively. Molina had proposed a design for a call-recording and transferring device built from a mixture of relays and ratchets, but when AT&T was ready to implement panel and rotary systems, other engineers had already come up with faster "senders" based on relays alone.
Molina Call Transfer Device Patent No. 1 (Sent 083, Approved 456)
From the "sender" to the combined control, there was a small step. The teams at Western Electric realized they didn't need to fence the sender for every subscriber, or even for every active call. A small number of control devices could be shared between all lines. When a call comes in, the sender turns on for a while and writes down the dialed numbers, works with the switch to redirect the call, and then turns off and waits for the next one. With panel switch, sender, and co-management, AT&T has a flexible and scalable system capable of meeting even the massive network needs of New York and Chicago.
Relay in panel switch
But even though the company's engineers had cast aside all technical objections to unattended telephony, AT&T executives were still dubious. They weren't sure that users could handle the six- and seven-digit dialing required for automatic switching in large cities. At that time, callers dialed subscribers of local switches, giving the operator two details - the name of the desired switch and (usually) a four-digit number. For example, a client in Pasadena could call a friend in Burbank by saying "Burbank, 5553." Bell management believed that replacing "Burbank" with a random two- or three-digit code would lead to a large number of incorrect dialing, user frustration, and poor service quality.
In 1917, William Blowwell, an AT&T worker, proposed a method that eliminated these problems. Western Electric may have printed two or three letters next to each digit on the dial when making a subscriber machine. The telephone directory would show the first few letters of each switch, corresponding to its numeric year, in capital letters. Instead of memorizing a random numeric code for the correct switch, the caller would simply spell the number: BUR-5553 (for Burbank).
Rotary dialer of a 1939 Bell telephone with the number for Lakewood 2697, i.e. 52-2697.
But even when there were no objections to moving to automatic switches, AT&T still had no technical or operational reason to move away from the successful method of connecting calls. She was pushed to this only by the war. The gigantic increase in the demand for industrial goods constantly raised the cost of labor for workers: in the United States it almost doubled from 1914 to 1919, which led to higher wages in other areas. All of a sudden, the key point in comparing operator-controlled or automated switches is not technical or operational, but financial. Given the rising cost of paying operators, by 1920 AT&T decided that mechanization could no longer be resisted and ordered the installation of automatic systems.
The first such panel-switched system in Omaha, Nebraska went into operation in 1921. It was followed by the New York switchboard in October 1922. By 1928, 20% of AT&T's switches were automatic; by 1934 - 50%, by 1960 - 97%. Bell closed the last telephone exchange with operators in Maine in 1978. But operators were still needed to organize calls over long distances, and in this position they began to be replaced only after the end of the Second World War.
Based on the stories of technology and business that are popular in our culture, it would be easy to assume that the clumsy AT&T narrowly escaped destruction by nimble small independent companies, eventually moving to an apparently more advanced technology that was first tried by small enterprises. But in fact, AT&T paid the price for the threat posed by independent companies ten years before the automation of the telephone exchanges began.
Triumph Bell
Two events in the first decade of the XNUMXth century convinced much of the business community that no one could beat the Bell System. The first was the failure of the United States Independent Telephone Company of Rochester of New York. The United States Independent ventured for the first time to build a competitive long-distance communications network. But they failed to enter the critical New York market and went bankrupt. The second was the collapse of the independent Illinois Telephone and Telegraph, which was trying to enter the Chicago market. Other companies not only couldn't compete with AT&T's long-distance communications, they also seemed to be unable to compete with it in large metropolitan markets.
Moreover, the endorsement by Chicago officials of the Bell operating company (Hibbard's Chicago Telephone) in 1907 made it clear that city governments would not try to foster competition in the telephone business. A new economic concept of natural monopoly has emerged - the belief that, for certain types of public services, pooling them under one supplier was a beneficial and natural result of market development. According to this theory, the correct response to a monopoly was its public regulation, not forced competition.
«1913 confirmed the rights received from the federal government to operate the Bell company. At first it seemed that the progressive administration skeptical of massive corporate mash-ups could shatter the Bell System or otherwise curtail its dominance. That's what everyone thought when Wilson's Attorney General, James McReynolds, reopened the lawsuit against Bell filed under the first antitrust , and set aside on the table by his predecessor. But AT&T and the government soon came to an agreement signed by the company's vice president, Nathan Kingsbury. AT&T agreed to sell Western Union (which it had bought a majority stake in several years earlier), stop buying independent telephone companies, and connect independent companies through its long-distance network at reasonable rates.
It seemed that AT&T had experienced a serious blow to its ambitions. But the result of Kingsbury's commitment only confirmed her power in national telephony. Cities and states have already made it clear that they will not try to forcefully limit the monopoly on telephony, and now the federal government has joined them. Moreover, the fact that independent companies gained access to the long-distance network ensured that this network would remain the only network of its kind in the US until the advent of microwave networks half a century later.
Independent companies became part of a huge machine, at the center of which was Bell. The prohibition against acquiring independent companies was lifted in 1921 because it was the large number of such companies seeking to sell themselves to AT&T that the government had requested. But many independent companies survived and even prospered, notably General Telephone & Electric (GTE), which bought Autelco as a competitor to Western Electric and had its own collection of local companies. But they all felt the gravitational pull of the Bell star they orbited.
Despite the comfortable conditions, the Bell directors were not going to sit still. To promote innovation in telephony that would ensure continued dominance of the industry, AT&T President Walter Gifford formed the Bell Telephone Laboratories in 1925 with 4000 employees. Bell also soon developed third-generation stepper-finder automatic switches controlled by the most complex relay circuits known at the time. These two developments will bring two people, и to the study of interesting analogies between commutator circuits and systems of mathematical logic and computation.
In the following series:
• [translation by Mail.ru] •
Source: habr.com
