GROWTH AND CONSOLIDATION

Expansive growth followed by consolidation in the telephone industry occurred initially between around 1910 and some would contend, 1982. Telephone service in the United States was almost solely the domain of AT&T, or The American Telephone and Telegraph Company. AT&T was a large, highly regulated monopoly. In other countries telephone service was mostly a government utility, usually part of the post office, thus the acronym PTT for Postal, Telephone, and Telegraph service.

An example of successful consolidation and growth in the telephone industry involves the consolidation of the original Bell Telephone Company and New England Telephone Company into American Bell, Inc. On December 30, 1899, American Bell became a wholly owned subsidiary of American Telephone and Telegraph Corporation, a New York Long Distance Company, and a wholly owned subsidiary of American Telephone and Telegraph Company.

AT&T was incorporated in 1885 to manage and expand the long distance business of American Bell Telephone Company and its licensees. In 1899, it assumed the business and property of American Bell and became the parent company of the Bell System, yet another informal name.

The company grew and consolidated over many years as a legal regulated monopoly. It literally built a public telephone system in the United States with connections and relationships outside the United States that were the envy of the world for many years. The company was organized across three lines of business: local and long distance, (including international), telephone service and equipment manufacturing (Figure 1). Regulation of the company was centered on the prices it could charge for it’s local and long distance service. The US Federal Communications Commission (FCC) dealt with interstate and international pricing. State Public Utility Commissions (PUCs) regulated intra-state pricing.

Figure 1: AT&T Business Entities Before 1982

Figure 1: AT&T Business Entities Before 1982

Overall economics of the company turned on revenue produced by the local operating companies and the long distance business. The local operating companies built and maintained the local exchange networks. There were two basic parts to the local services business. Local telephone service typically included both the ability for one subscriber to call another in a common service area and service across a wider geographic area that was not deemed long distance. The second service or capability was also the point where long distance calls were handled, thus the name access network. Together, these two local service entities would later become the basis for creation of the local access and transport area (LATA) in a consent decree with the US Department of Justice.

Because the local services business was regulated by the PUC in each state, these companies typically were organized and operated locally with unique names and local management. Examples include New Jersey Bell, Chesapeake and Potomac Telephone, Nevada Bell, Michigan Bell, and so on.

The long distance business operated under the name of ‘‘AT&T Long Lines.’’ AT&T Long Lines, essentially a domestic US business, was responsible for building and maintaining the interstate long distance network and switching gateways to the international network. The long distance business was the first to deploy digital transmission and switching technology. The basic operation of the long distance network was simply to take a call from one local service network, connect it to a peer local company, and terminate the call. The originating company kept track of the date, time, and duration of the call and billed it as a separate item or ‘‘Long Distance Call.’’ On a periodic basis, the entities separated the revenue, paid their suppliers, employees, and invested in new plant and equipment, all under the watchful eyes of the FCC and state PUCs.

Two other important, but less well-known entities were Western Electric Manufacturing and Bell Labs. Western Electric was the company’s equipment manufacturing business. Essentially, this entity manufactured equipment and in later years developed software to run it. Its output was switching and transmission products and systems.

AT&T survived, grew, and prospered as the Bell System until 1984 when it was broken up by the consent decree. One corporate entity was converted overnight into eight. The long distance and equipment manufacturing business retained the AT&T brand; the seven new entities got the Bell name. The seven new entities Regional Bell Operating Companies—RBOC, for short—were given the rights to the famous Bell Logo, while AT&T got the rights to the word Bell as in Bell Labs.

On January 8, 1982, AT&T announced that it had agreed to break up the Bell System in response to Justice Department demands. This news was very unsettling to the company’s 992,000 employees and welcome news to competitors, many of whom had been chipping away at its business for many years. Why? Different people will interpret the facts in various ways. However, most agree that the company’s once unchallenged position had been seriously curtailed. Regulatory and judicial rulings had whittled away its ability to protect its business, starting with the Carterfone decision in 1968.

At the time, the most lucrative part of AT&T’s business was the long distance business. For many years, the operating companies or local exchange business had used their share of long distance revenues to make up for losses incurred in providing local service. Long distance competitors using Bell telephones as terminal points set up service at reduced pricing. AT&T long distance rates were set on average cost and included the subsidy paid to local exchange companies. Long distance competitors picked high traffic routes, leaving the lower traffic, higher cost routes to AT&T. Such action was referred to as cream skimming or cherry picking.

On another front, Western Electric was restricted from selling its output to any domestic customer except the Bell Operating Companies. Bell Labs, where the transistor was invented, had to make its patents available to one and all and was not allowed to use them in products other manufacturing businesses were making and selling. Competition was a one-way street. Other companies could use Bell facilities and discoveries to benefit their business, but AT&T was forbidden to start any new enterprise that might be construed as taking advantage of its size, skill, and knowledge.

Big and powerful though it was, the 22 Bell operating companies were not the only ones in the business. At the time, almost 1500 independent companies provided service to some 35 million subscribers. The 1982 Consent Decree did not apply to these companies, but they were impacted in many ways because of working relationships with various parts of the Bell System. 1983 was a year of planning and preparing for January 1, 1984. This was the date of start of business for the Bell Systems’ 22 operating companies. Each entity was organized under one of the seven newly created regional Bell operating companies (RBOCs).

Assets, liabilities, and equity of AT&T were partitioned off into eight new corporate entities. One share of stock was replaced with eight shares of the new entities. A single listing on the New York Stock Exchange turned into eight. AT&T stock now represented equipment manufacturing and long distance businesses. Seven new stock listings including Ameritech, Bell Atlantic, BellSouth, Nynex, Pacific Telesis, Southwestern Bell, and US West represented the local service business.

Two other actions worth noting involved the famous Bell Labs. In addition to splitting the Bell name and logo, the assets associated with Bell Labs were divided between the new AT&T and the seven regional operating companies. All research and development remained with AT&T’s equipment manufacturing arm. A ninth entity was created and named Bellcore. Bellcore was essentially the part of the former Bell Labs responsible for standards development and important things, such as the North American numbering plan. Bellcore was set up as an independent entity with its own set of books; however, ownership was held by the RBOCs and managed by representatives of each of the RBOCs. This entity would later be sold off to a third party and renamed Telcordia.

The new AT&T consisted of administrative staff, AT&T Communications , AT&T Technologies, (Western Electric and Bell Labs) and AT&T International (Figure 2).

Figure 2: AT&T Post 1984

Figure 2: AT&T Post 1984

The RBOCs lost all their share of long distance revenue except charges for intra-LATA calls not included in fixed monthly service charges, but retained the directory services business. They were permitted to sell equipment in competition with AT&T Information Systems, but restricted from designing and manufacturing equipment.

The seven RBOCs included Ameritech, Bell Atlantic, BellSouth, NYNEX, Pacific Telesis, Southwestern Bell, and US West.

Economically, the RBOCs were viewed as being in trouble. Charges for monthly service covering calls within a local access and transport area or LATA belonged entirely to the BOCs. The resulting revenue didn’t cover expenses. Loss of so much of the long distance revenue made them unprofitable. Previously, state PUCs had refused to allow increases in pricing for local service. Now that long distance revenue was gone, something had to be devised to make up for the loss in revenue. So the FCC created an access charge to be added to all long distance charges. The charge showed up on bills sent to long distance carriers and on subscriber monthly bills.

The RBOCs are permitted to sell new customer premises equipment that is not of AT&T Technologies sourcing. They cannot manufacture equipment but are permitted to enter other lines of business with restrictions.

Calls between LATAs or across LATA boundaries belong to the long distance carriers.

EXPERIMENTATION | Internet and Telecom: A Brief History

Experimentation is at the root of technological success. Technological success forges reality from fantasy or the fantasy remains just that— an idea with little or no value other than personal satisfaction.

Experimentation with telephony and radio proceeded almost in parallel. Bell’s invention of the telephone took place around 1875 to 1876 when he built models and demonstrated technical feasibility. His early working samples enabled people to talk to each other over wire conductors across distances that far exceeded the range of direct human speech and hearing. Experimental development would continue throughout the late 1800s and into the 20th century as patents were granted and the first services became available.

From the start, the telephone industry was dependent on lines, simply two pair of wires connecting two telephone instruments. Over time, it became obvious that if all the lines were connected to a single, centralized location where any telephone user could be connected to any other user on demand, service and usefulness would be greatly improved. With this capability, the central office (CO) came into being.

Ten to twelve years after Bell’s initial experiments, Guglielmo Marconi read about and began experimenting with Hertz’ work with electromagnetic waves. Marconi believed that magnetic waves could free telegraphy from the constraints of wire and cable. After significant development work in the 1880s, and a convincing demonstration of sending signals over water between the shore and an island in Bristol Channel 8.7 miles distant, he changed the name of his company, ‘‘Wireless Telegraph and Signal, Ltd.’’ to ‘‘The Marconi Wireless Telegraph Co. Ltd.’’ He received an English patent on the wireless in 1896.

Farnsworth and Zworykin conceived television in the early 1900s. However, neither knew of the others work, and it was well into the 1930s before serious experimentation was started on today’s so-called ‘‘analog’’ television system. Radio had become a commercial success both in terms of broadcasting and receiver manufacturing. In addition, the art and science of broadcasting audio or speech using electromagnetic waves was well known. Most of the experimentation was about getting pictures—‘‘video’’ over the same medium.

Early telephone companies built and installed lines by stringing wires on poles or other convenient supporting structures where the right to use could be arranged. The poles or rights to use other supporting structure might be owned by the same company or by another entity. Over time, a collection of groups of private lines grew across the city and landscape. Ownership and rights to use these assets were traded, bought, sold, and bartered. Companies went into and got out of the business. Some succeeded while others failed.

Telephone lines were similar to the lines used by telegraph operators to transmit messages using a code conceived by Samuel B. Morse in the early 1800s.

Telegrams—written messages—were the end result of a process that started with a spoken or written message given to a telegraph operator who encoded it into dots and dashes and sent it to another telegraph operator using a telegraph key. The second operator received the dots and dashes, decoded it, wrote it out on paper, and delivered or had it delivered by a third party to the person or entity it was intended for.

Telegraphy was adopted by the railroads. Their rights-of-way easily and conveniently supported cross-country, intrastate, and interstate lines. Command and control of trains as well as switching trains to other tracks with speed and efficiency couldn’t be accomplished without telegraphy. Can you imagine what it might be like to dispatch and control trains with messages delivered by the Pony Express? Over time, this capability migrated to public use in exchange for money and became a commercial business. In similar fashion, the telegraph key and typewriter were motorized and morphed into the Teletype machine and eliminated the need for the skilled Morse Operator. The Teletype machine spawned Telex (international) and TWX (Domestic) services that lasted well into the 1980s when they were largely replaced by facsimile technology standardized by the CCIT (later renamed ITU).

Internet and Telecom: A Brief History

Alexander Graham Bell is credited with inventing the telephone sometime after Samuel B. Morse came up with the telegraph key and code, making smoke signals obsolete technology. Somewhere along the way, in the more recent past, computers learned to talk to one another over telephone lines. And then along came the Internet.

That’s almost enough history for practitioners. However, a perspective on the past is well worth a quick read because understanding some of the history, especially since 1982, provides insight into the changes the Telecom industry has undergone and how that has, or will impact media and entertainment industry operations in the future.

Keeping our focus on practical considerations it makes sense to start somewhere in the early part of the past century. After all, radio and telephones came from similar inventive roots and had electrical or ‘‘electronics’’ in common. It’s also instructive to observe that the two parted ways when digital electronics went solid state. With digital electronics—initially the switching transistor—devices could count and keep track of items or service transactions, calculate, or measure interesting, valuable operational and accounting characteristics of the business.

Someone figured out how to convert an analog signal into digital form, and the telephone world went for it with a vengeance. On the other hand, radio and television receivers didn’t warm to digital techniques until well after integrated circuits with divide and multiply capability became cost effective for use in tuners. Other similar events along the way could be mentioned, but suffice it to say that broadcasting and consumer electronics lagged the telephone industry in adoption of digital technology by many years and did so, one painful step at a time.

The telephone industry was digital in nature from the start. For example, the basic business transaction unit, or service function, was and still is, a phone call. The line was in use or it was free. It could be used or one had to wait. One person could talk to one person at a time (until party lines were figured out). One operator could connect two parties to make conversation, later capitalized by the industry in the form of a telephone call.

Almon B. Strowger, a Kansas City undertaker, invented the stepping switch to allow customers to decide between his establishment and a competitor without undue influence of the wife of his competitor who just happened to be the local telephone operator. The stepping switch enabled the industry to continue in a mechanized and automated fashion, without operators, one telephone call at a time, the staple of long distance telephone bills.

Audio—the staple of radio—was analog and stayed that way. In 2003, radio broadcasters began upgrading transmission facilities to digital transmission.

Video might be characterized as digital in nature—at the very least it’s highly structured. Pioneer inventors Philo Farnsworth and Vladimir Zworykin both conceived a fixed, repetitive scan structure. Farnsworth reportedly conceived his version while plowing successive rows in a field.

Historically, telephony, radio, and television share a common business life-cycle behavior. Still in relative infancy, the Internet seems to be evolving in similar fashion. The life-cycle includes three distinct phases starting with experimentation, moving into growth and consolidation, and finally into a mature state where the technology continues to evolve and the business segment makes a continuing contribution to the economy over a long time.

Quality of Service (QoS)

Voice quality is a well-known hot button in the world of telephony. Public telephone networks traditionally strove to provide a uniform level of voice quality that, while surprisingly limited in the dimension of bandwidth (the nominal 4 kHz being less than 25 percent of a normal young adult’s perceptible frequency range) was overall quite good. By the 1960s, the standard for toll (long-distance) connections in the United States was that 95 percent of users should rate them good or better on a subjective poor-fair-good-excellent scale with respect to key impairments such as circuit noise. With the rapid deployment of long-haul digital fiber systems in the 1980s, users became accustomed to essentially noiseless telephone calls, even to overseas points. Echo cancelers employing digital signal processing cleaned up echoes on long circuits without introducing the other impairments characteristic of cruder echo suppression techniques. After bad user reaction to the 500-ms round trip delay, satellite circuits were banished to those parts of the international network too lightly loaded to justify undersea fiber cables. Most fundamentally, the basic technologies of circuit switching combined with pulse-code modulation (PCM) encoding essentially guarantee that whatever waveform enters one end of the network comes out the other side with very little modification.

Users have shown a definite willingness to tolerate lower voice quality in exchange for convenience—hands-free operation with speakerphones, mobility with cordless phones and cellular services. However, as long as wireline telephone companies manage to continue delivering standard quality in combination with attractive calling plans, there does not seem to be much margin for services with lower quality and no compensating convenience factors. It remains to be seen whether the ability to talk into your computer will turn out to be viewed as a convenience factor.

So, the best advice is to be very cautious about specifying or allowing lower voice quality, except perhaps in specific enterprise network situations where there is a well-quantified economic benefit and the business function being supported by the service does not involve, for example, customer contact.

Vendors may state that they support toll quality voice or make some quantitative statements about quality. However, given the crucial role of voice quality in voice networks, rather than trying to interpret these statements by themselves, you should also do the following:

1.     Be sure you understand the voice coding scheme being employed. If it includes compression, you are making a trade-off against quality; and, depending on how aggressive a compression technique is employed, the quality difference may be perceptible under certain conditions.

2.     Review one of the many published reports of bake-off events, in which voice quality is always one of the key system properties tested.

Quality of service in telecommunications networks in fact has many dimensions other than voice quality per se. Other commonly specified parameters include dial tone delay, postdial call setup time, and percent blocking. Standard objectives for these parameters for public networks are published by ITU and by Telcordia (formerly Bellcore, and at one time the joint R&D arm of the U.S. RBOCs). The ITU standards tend to be on the liberal side to allow for the significant variation in public network standards among various countries of the world.

Other quality of service parameters apply to traditional data networks. These include the probability of dropped, duplicated, or misrouted packets, as well as the statistics of packet delay. In an integrated system employing Internet technology to deliver telecommunications applications, these parameters may not be directly observable by the users, but will instead influence other quality parameters such as voice quality, call setup time, and blocking.

An important fact to keep in mind about quality is that in general there is a trade-off against cost. If you want a higher-quality system, you may expect to pay for that. If you are looking for the ultimate in low cost, you may need to compromise on various quality parameters.

Class 5 Switches with IP Trunks

The number 5 in reference to telecommunications switches remains from predivestiture times when the U.S. telephone network had a hierarchical model in which five classes (according to their respective functions) of switches were employed. Specifically, class 5 was reserved for the central office switches. Although the name endures, the actual capabilities of some of today’s large class 5 switches allow them to be configured for tandem, government, and other specialized applications. Furthermore, those class 5 switches that are equipped with appropriate signaling gateways are sometimes used as gateways to foreign networks. On the other end of the spectrum are smaller class 5 switches that are sometimes employed by large enterprises as PBXs.

Large class 5 switches (with about 200,000 access lines and 100,000 trunks) support metropolitan and urban areas; smaller ones (with about 40,000 access lines and 20,000 trunks) support rural areas; and the smallest ones (with about 20,000 access lines and 10,000 trunks) typically support small towns. Advanced switches provide ISDN (PRI and BRI) and ADSL access, and they invariably support SS No. 7 signaling and its applications (such as Intelligent Network) as well as the interfaces to telecommunications network traffic management platforms, billing systems, and other operations systems.

There are very few vendors of telecommunications switches; their products are so complex and so dissimilar—insofar as their architecture, hardware, and software are concerned—that each warrants a separate book. We do not attempt to review the existing hardware and software solutions, nor do we describe the resulting IP services or mechanisms that support them. Our review of VoIP gateways, gatekeepers, remote access servers, and SS7 gateways, whose functions are combined in the new switches, covers the IP services and mechanisms. We do report on the emerging solutions for IP connectivity (often combined with support for sending voice over ATM and frame relay networks—thus justifying the name multiservice switching used in the industry to describe these capabilities), which effectively make the telephony switches into packet switches.

Such switching solutions unite traditional call processing with the support of modem pooling and remote access, thus allowing ISPs (or enterprises) to terminate calls directly at the multiservice switch. All the features of remote access servers described in the preceding text are supported. In addition, the VoIP gateways and gatekeepers are also part of the multiservice switch, so the PSTN voice service is converted to voice over IP as close to the source as possible. Note that because class 5 switches already support SS No. 7 and its applications, support of the PSTN signaling and, in particular, use of Intelligent Network—which immediately solves the issues of supporting services and number portability, as well as the use of the end-to-end PSTN operations and management—come automatically with the product. Again, all the relevant IP telephony and remote access capabilities (including out-of-band signaling over IP) described earlier in this chapter are supported, but since they are supported at the switch itself, no front-end boxes for performing interconnections and conversions are needed. Telephone companies or enterprises that plan to replace their existing switches should replace them with the emerging multiservice switches; in many cases the costs of upgrade and integration of all the products needed to provide such functions are below the cost of an appropriate multiservice switch.

Typically, the features supported by the multiservice switches (in addition to over 3000 voice-related features inherent from the PSTN) come in feature groups called bundles in PSTN parlance. To date, the following bundles are available:

§  Residential

§  SOHO

§  Large enterprise

§  Multienterprise

§  Virtual telco

Bundling eliminates the need to buy unnecessary features; each bundle costs less than an all-or-nothing package.

One more application of IP trunking is offloading the signaling network. Messages between the SS No. 7 endpoints (network elements) can be tunneled through the IP network. The IETF sigtran working group is presently developing the transport layer standard protocol for such tunneling; for the moment, there are several proprietary solutions. Here again an important and interesting duality can be observed: While the IP endpoints (such as SS7 gateways) adapt to the PSTN in order to connect to its signaling network, the PSTN switches use IP to tunnel the PSTN signaling.

This concludes the description of the available types of products. We are ready to proceed with the criteria for choosing those that best suit your applications.