Woo circles

Voicemail (or voice mail, voice-mail, vmail or VMS, sometimes called messagebank) is a centralized system of managing telephone messages for a large group of people.

Features

In its simplest form it mimics the functions of an answering machine, uses a standard telephone handset for the user interface, and uses a centralized, computerized system rather than equipment at the individual telephone. Voicemail systems are much more sophisticated than answering machines in that they can:

• answer many phones at the same time
• store incoming voice messages in personalized mailboxes associated with the user’s phone number
• enable users to forward received messages to another voice mailbox
• send messages to one or more other user voice mailboxes
• add a voice introduction to a forwarded message
• store voice messages for future delivery
• make calls to a telephone or paging service to notify the user a message has arrived in his/her mailbox
• transfer callers to another phone number for personal assistance
• play different message greetings to different callers.

Storage

Voicemail messages are stored on hard disk drives, media generally used by computers to store other forms of data. Messages are recorded in digitized natural human voice similar to how music is stored on a CD. To retrieve messages, a user calls the system from any phone, logs on using Touch-tones (clearing security), and his messages can be retrieved immediately. Many users can retrieve or store messages at the same time on the same voicemail system.

Many voicemail systems also offer an automated attendant facility. Automated attendants enable callers to a “main” business number to access directory service or self-route the call to various places such as a specific department, an extension number, or to an informational recording in a voice mailbox, etc.

History of voicemail

Voicemail systems are often associated with office telephone systems or PBXs. They may also be associated with public telephone network services such as residential phones or cellular phones. Mobile phones generally have voicemail as a standard network feature. The most modern implementations of voicemail support fax delivery to personal voice mailboxes and retrieval via printers, are integrated into e-mail systems for shared directories and shared message storage (also called Unified Messaging), and use touch tone voice user interfaces (VUI), speech technologies, and/or visual, screen-based graphical user interfaces (GUI) user interfaces.

The need for voicemail

In the 1970s and early 1980s, the cost of making a phone call decreased and more business communication was done by phone. As corporations grew and labor rates increased, the ratio of secretaries to employees decreased. With multiple time zones, fewer secretaries and more communication by phone, real-time phone communications were hampered by callers being unable to reach people. Some early studies showed that only 1 in 4 phone calls resulted in a completed call and half the calls were one-way in nature (that is, they did not require a conversation). This happened because people were either not at work (due to time zone differences, being away on business, etc.), or if they were at work, they were on the phone, away from their desks in meetings, on breaks, etc. This bottleneck hindered the effectiveness of business activities and decreased both individual and group productivity. It also wasted the caller’s time and created delays in resolving time-critical issues.

First solutions did not work

Neither email messaging nor cellular phones were widespread in the 1970s and 80s, and did not really begin to flourish until the mid-1990s. The initial solution to the phone communication problem for businesses was the “message center.” A message center or “message desk” was a centralized, manual answering service inside a company manned by a few people answering everyone’s phones. Extensions that were busy or rang “no answer” would forward to the message center onto a device called a “call director”. The call director had a button for each extension in the company which would flash when that person’s extension forwarded to the message center. A little label next to the button told the operator whose extension it was.

While it was an improvement over earlier systems, the message center had many disadvantages. Operators were busy, and volumes of calls would come in simulataneously at peak periods, such as lunch time. This left message attendants with little time to take each message accurately. Often, they also weren’t familiar with employees' names or how to spell or pronounce them. Messages were written on pink slips and distributed by the internal mail system. The messages often arrived at people’s desks after lengthy delays, contained little content other than the caller’s name and number, and were often inaccurate, with misspelled names and wrong phone numbers.

Tape-based telephone answering machines had come into the residential telephone market, but they weren’t used much in the corporate environment due to physical limitations of the technology. One answering machine was needed for each telephone; messages couldn’t be recorded if the user was on his phone; messages had to be retrieved in sequential order; and messages couldn’t be retrieved remotely, selectively discarded, saved, or forwarded to others. Further, the manufacturers of PBXs (private branch exchanges — the name for corporate phone systems) used proprietary digital phone sets in order to increase the functionality and value of the PBX. These phone sets were, by design, incompatible with answering machines.

How voicemail solved the business communication problem

Voice mail’s introduction enabled people to leave lengthy, secure and detailed messages in natural voice, working hand-in-hand with corporate phone systems. The adoption of voicemail in corporations improved the flow of communications and saved huge amounts of money. GE, one of the pioneer adopters of voicemail in all of its offices around the world, claimed that voicemail saved, on average, over US$1,100 per year per employee.

Voicemail has two main modes of operation: telephone answering and voice messaging. Telephone answering mode answers outside calls and takes a message from any outside caller (either because the extension was busy or rang no-answer). Voice messaging enables any subscriber (someone with a mailbox number) to send messages directly to any or many subscribers’ mailboxes without first calling them. Both of these modes are described below.

Telephone answering mode

One of the advantages of a PBX is its ability to forward calls. If a person is using his phone or does not answer it, calls to his extension are forwarded automatically by the PBX to another extension, presumably someone (like a secretary) who can answer the call and take a message. With a voicemail system installed, the PBX is programmed to forward busy or unanswered extensions to a machine — the voicemail system.

Suppose an outside caller, Willma, calls someone in a company, Fred. If Fred’s phone rings "no answer" or "busy", the PBX will forward the call to the voicemail system. Somehow the PBX needs to tell the voicemail system that Fred’s phone is the one that the call is being forwarded to so that the voicemail system can answer with Fred’s personal greeting. Without this information, the voicemail system would have no idea whose phone it was answering. Once a message is left, the voicemail system illuminates the message waiting light on Fred’s phone. It does this by sending a signal to the PBX to tell it which light to light. When Fred returns to his desk and calls the voicemail system (or calls in remotely) he is presented only with the messages in his personal mailbox even though thousands of messages belonging to other people are stored on the same system. Once the messages are played, the voicemail system signals the PBX to turn off the message waiting light on Fred’s phone.

Early voicemail systems (notably those made by IBM and VMX) could not answer outside calls — that is, they could not automatically answer a call originally destined to an extension on the PBX which rang busy or was not answered. As subsequent voicemail systems emerged (notably ROLM and Octel which later merged with Boston Technology), the systems could answer outside calls. However, most PBX’s did not provide signaling to tell the voicemail system which extension it was forwarding, nor did they support telephones with message waiting lights. This signaling would come later, but until it did it created a major challenge for voicemail systems for many years.

Voice messaging

This mode is to phones what email is to computers. Messages are sent to other users by calling the voicemail system rather than the user’s phone. For example, suppose two employees, Fred and Mary, are working on a project. Fred has some information that Mary should have, but does not want to phone her and talk to her — he just wants to give her the information. Rather than phone her, Fred calls the voicemail system, logs on with his number and password, and records a message to Mary in his own voice. He tells the voicemail system to send it to Mary by keying in her mailbox number (same as her extension) or spelling her name using Touch-tone keys. The message is immediately put in Mary’s voice mailbox without her phone ever ringing. The message waiting light on her phone immediately comes on telling her there is a message. Fred can send this message just to Mary, to Mary and any number of additional employees, or to group lists which contain any number of pre-programmed names and numbers. The same message can be sent to thousands of people. Additional features are available, like marking a message urgent, private or asking for notification when the message has been picked up.

Voice messaging does not always have to be sent between individuals on the same voicemail system. In VOIP (Voice Over Internet Protocol), technology allows individuals to record a voice message into a computer system and then the computer dials a specified phone number or numbers and plays the message. Like email, this method of delivering voice messages can be subject to abuse such as spam or vishing. There are Federal and State laws and regulations designed curb these abuses, such as the United States National Do Not Call Registry.

History of corporate voicemail

There is often some discussion about who invented voicemail. Invention of a device is normally defined as who first created a viable design for a product and reduced it to practice, not who first had the seed of an idea (since that is impossible to prove). Beyond invention is the concept of who brought the product to the world most effectively, or, in other words, who was the first to make it commercially viable and widely used. In those terms, voicemail was invented simultaneously at IBM by Dr. Steven J. Boies (c.1975) and at the Xerox Palo Alto Research Center (PARC) by Stan Kugell and Edward McCreight. Both implementations employed front end systems to manage telephone and user interfaces (IBM System /7 and Xerox Alto), with larger file services attached (IBM's VM370 and Xerox's Network File System and Ethernet). Despite promising internal deployments, neither company successfully commercialized voicemail.

Voicemail was broadly commercialized by Octel Communications (founded in 1982 by Bob Cohn and Peter Olson). ROLM Corporation (founded in 1969 by Gene Richeson, Ken Oshman, Walter Loewenstern and Robert Maxfield and later owned by IBM before IBM sold it to Siemens) was the first PBX manufacturer to offer integrated voicemail with its PhoneMail system, and also played a major role commercializing voicemail.

IBM’s product, initially called the SFS (Speech Filing System) was developed as an intensive research project at the IBM Thomas J. Watson Research Center. It was meant to mimic the concept of email, but using the telephone as the input device and the human voice as the medium for the message. Work on the system began in 1973 and the first operational prototype was made available to users in 1975. Four people could use it at once. From 1975-1981, about 750 IBM executives, mainly in the U.S., used various SFS prototypes in their daily work. Those prototypes ran on an IBM System /7 computer attached to an IBM VM370 for additional storage. The prototype was converted to run on a Series /1 computer in 1978. In September, 1981, IBM announced this product as the “Audio Distribution System” (ADS) with the first customer installation being February, 1982. It was marketed directly by IBM and for a short while by AT&T. IBM’s ADS required special attention as a computer (special room, special power, air conditioning, etc.) ADS was richly featured for voice messaging, the result of IBM’s enormous research in human factors and observing SFS in real operational use. However, ADS had major limitations which resulted in its failure as a commercial product: for example, it was physically large, expensive, limited to 1,000 users, had no telephone answering mode (could not answer outside calls), and had to be taken out of service to make administrative changes to the user data base (called “MAC”, for “moves, adds and changes”).

Another company, Delphi Communications of California, deserves some partial credit for invention of voicemail. Under the leadership of Jay Stoffer, Delphi developed a proprietary system (called Delta 1) that picked up incoming calls directly from the telephone company. Stoffer presented the Delphi concept publicly to the association of Telephone Answering Services around 1973 and the prototype system was launched in San Francisco in 1976 by a Delphi company called VoiceBank. Delphi developed Delta 1 as a purely service-oriented voice messaging system to answer subscriber telephones for businesses and professionals. Delta 1 required human intervention for message deposit. While three machines were built, only one machine was put into operational service. The completely automated voice messaging system (Delta 2) was developed for initial operational use in Los Angeles in 1981. Apparently Delta 2 was built, installed and operational for a short while, but unfortunately Delphi’s major early investor, Exxon Enterprises, abruptly shut down Delphi in July, 1982. Nothing further was done with Delphi’s technology. A patent was applied for and issued for Delphi’s Automated Telephone Voice Service System. The patent, U.S. Patent No. 4,625,081, was issued after Delphi’s closure, but Delphi’s assets (and the patent) were transferred to another Exxon company, Gilbarco, which made equipment for gas pumps at filling stations. Gilbarco is now owned by GEC in the United Kingdom.

In 1979, five years after IBM’s SFS (ADS) system and three years after Delphi’s Delta 1 system were first operational, a company was founded in Texas by Gordon Matthews called ECS Communications (the name was later changed to VMX). According to Jay Stoffer, founder of Delphi Communications, Gordon Matthews learned about Delphi’s voicemail prior to his founding VMX. Regardless of how he was inspired, Matthews eventually founded VMX which developed a 3,000-user voice messaging system called the VMX/64. VMX was arguably the first company to offer voicemail for sale commercially for corporate use. Matthews was able to sell his system to several notable large corporations, such as 3M, Kodak, American Express, Intel, Hoffman La Roche, Corning Glass, Arco, Shell Canada and Westinghouse. This impressive list of early adopters started the ball rolling on corporate voicemail. While some claim that VMX and Gordon Matthews invented voicemail, this claim is not true. The first inventor of record was Stephen Boies of IBM years before VMX was founded.

While VMX began with a good start, it failed at developing the market, and the company was not a commercial success. It took many years before its products could answer outside calls (and then only under certain circumstances), they were physically enormous, expensive, light on important user features and had serious reliability issues. In addition, the user interface was cumbersome, requiring the users to remember non-intuitive multi-digit Touch-tone commands. Matthews, a prolific entrepreneur and patenter, applied for and was granted a patent on voicemail (patent number 4,371,752) which issued in February, 1983. The patent was promoted as the pioneering patent for voicemail.

Shortly after the development of the first voicemail systems, several companies sprang up to develop their own systems including Wang computers, ROLM, Opcom, Octel, Centigram, Genesis, and many others. Wang Computers, under the leadership of Dr. Larry Bergeron, developed a voicemail system modeled after the IBM system. Wang called its system the DVX. It too could not answer outside calls but was smaller and less expensive than the IBM system.

Matthews was quite astute at the way he used his patent. Matthews tried to assert his patent with IBM, AT&T and then Wang, but all three companies reportedly would have been able to invalidate the Matthews patent because of prior art. Matthews cleverly achieved a settlement where the patent was let stand, not challenged in court and IBM, Wang and AT&T (in separate settlements) received royalty-free licenses to all VMX patents. Wang, the last of the majors to get such a license, essentially paid $20,000 and cross licensed a few patent applications (not issued patents). IBM and AT&T also cross-licensed a number of patents to VMX, most of which were obsolete or outdated. VMX could claim that several major companies licensed the patent (even though they paid almost nothing to VMX for the rights), but that part wasn’t disclosed. The patent was never challenged in court and VMX then continued to assert (incorrectly) that it had invented voicemail and that Matthews was the father of voicemail. Following the settlement with Wang, VMX settled with Octel. In exchange for a small payment and Octel’s agreeing not to litigate any VMX patent, Octel received a paid-up, royalty-free license on all existing and future VMX patents.

ROLM (one of the first makers of digital PBX’s) was the first company to offer integrated voicemail through its product called PhoneMail, which name is a registered trademark. PhoneMail offered impressive recording quality of its digitized messages. ROLM’s digital PBX (called a CBX, for Computerized Branch eXchange) was the first PBX to provide signaling to indicate which extension was being forwarded to a voicemail system (the first PBX to do so). However, the signaling was proprietary and intended only for use by its voicemail product, PhoneMail. ROLM’s CBX also provided signaling to enable PhoneMail to illuminate a message waiting light on ROLM’s electronic phones and later standard phones equipped with message waiting lights (also a studder dialtone is used with analog and digital phones). PhoneMail worked with most but not all models of ROLM’s CBXs, would work with some other brands of PBXs such as Nortel's Option Meridian (with adaptors and loss of some features), and was heavily promoted by ROLM. PhoneMail is still a commercial success. Siemens still offers PhoneMail in various configurations/sizes (including a micro-sized version) and its unified messaging successor, Xpressions 470; along with the same pleasing female voice most ROLM techs have nicknamed, Silicon Sally.

Opcom, a company started by David Ladd, was another maker of voicemail which also pioneered and patented the feature of automated attendant (U.S. Patent numbers 4,747,124 and 4,783,796 both of which issued in 1988). Opcom developed a voicemail system primarily marketed to smaller enterprises. Automated attendant enables callers to direct calls by pressing single digit keys. For example, “If you are making domestic reservations, press ‘1’; for international reservations, press ‘2’; for frequent flier information, press ‘3’, etc.” Automated attendant is not technically voicemail, but all the features to enable automated attendant are already part of a voicemail system so it is a natural feature to add to it. Opcom was an innovative company and also pioneered the concept of Unified Messaging (to be discussed later in this article). Opcom’s voicemail product was a commercial success with smaller companies and some large ones. Around 1991, VMX was on the verge of bankruptcy and was acquired by Opcom. Since Opcom was private and VMX was public, the transaction was done as a reverse merger and the surviving company was called VMX. Little of the original VMX Company was retained. Within a few years, VMX was acquired by Octel and David Ladd became Octel’s Chief Technology Officer.

Octel Communications Corporation was founded in Silicon Valley in 1982 by Bob Cohn and Peter Olson. Octel’s voicemail system (developed during the period from 1982-1984 and first sold in 1984), became the clear market leader fairly quickly. While Octel benefited from the work and experiments of others, it was the first stand-alone voicemail company to build a strong business and strategy to win at this important market. In addition, Octel innovated substantially new technology which contributed heavily to its success. Octel’s differentiated hardware and software architecture enabled its systems to be physically smaller, faster, more reliable, and much less costly to build than any other vendor. These features, many of which were patented, gave Octel market leadership:

• User-friendly user-interface (other systems were not intuitive and had no help prompts).
• Error-free Touch-tone detection (other systems falsely detected a Touch- tone out of human voice, or didn’t detect Touch-tones when users pressed the buttons).
• Scrambled messages so no one could hear anyone else’s messages (other systems could accidentally get other people’s messages if the system failed at the right time).
• Telephone answering, voice messaging and automated attendant.
• Moves, adds and changes could be done while the system was running.
• Large amounts of message storage.
• Physically small size (about the size of a 2-drawer filing cabinet, compared to ROLM’s original PhoneMail being about 5' × 5' × 5' and VMX’s system filling a computer room). No requirement for special environment.
• Locatable anywhere. Octel systems could be located in any office environment and they were not susceptible to electrical shocks (often common on carpeted floors in offices, especially during winter).
• High reliability (being the first voicemail system to achieve up-time of 99.9% with its first system).
• Compatible with virtually all brands of PBX (voicemail offered by PBX vendors could only work with that vendor’s PBX system).
• Telephone answering with all PBXs, even those which had no method of providing caller ID.
• Message notification (phoning subscribers at various locations pre-programmed by the subscriber, when messages were received).
• Range of capacities. Small, medium, large and extra large capacity systems that addressed the needs of major companies (For example, Octel’s systems had 50% greater port capacity than VMX’s largest system). Small systems went in branch offices, medium systems went in district offices, large systems went in regional offices, and extra large systems could handle large corporate headquarters with over 10,000 people.
• Networking between voicemail systems so companies could have their voicemail systems operate as one large virtual network.

Octel’s strategy addressed needs of major accounts which other vendors did not until much later: advanced training, customer service, sales and market education. Octel’s system could identify the extension number of calls being forwarded to it and light message-waiting lights on most PBXs. This was possible because Octel’s engineers reverse engineered the major brands of PBX (legally) and often figured out ways to communicate with the PBX in ways the PBX manufacturer had not. Eventually most makers of PBX chose to work cooperatively with Octel. Octel integrated with almost 100 brands of PBX worldwide. As a result of Octel’s worldwide leadership, its user interface (which was done in more than 75 languages and dialects) became the most widely known in the world.

Toward the late 1990s, Octel introduced the concept of Visual Mailbox and Unified Messaging. Visual Mailbox enabled users to manage their voice mailboxes through their PCs, although the messages were still stored on the Octel system. Unified Messaging integrated voicemail into Microsoft Exchange, the corporate email system made by Microsoft. Unified Messaging had actually been invented by Roberta Cohen, Kenneth Huber and Deborah Mill at AT&T Bell Labs. The patent for Unified Messaging was received in June, 1989 (Patent number 4,837,798).

         File:Unified Messenger Screen Shot.png

The figure above shows a screen shot from an early Unified Messaging system (GUI). Emails are identified with the icon of an envelope; voicemails are identified with the icon of a phone handset. This system fully integrated voicemail into Microsoft Exchange so both voicemails and emails could be displayed and accessed via Microsoft Outlook. Users could also call into the system by phone and hear both voicemails and machine-read emails (TUI).

Unified Messaging: With Unified Messaging, users could access voice and email messages using either the graphical user interface (GUI) on their PC, or using the telephone user interface (TUI) with any telephone in the world. On the PC, users could see voicemails and emails mixed together in their email inbox. Voice mails had a little telephone icon next to them and emails had a little envelope icon next to them (see figure below). For voicemail, they’d see the “header information” (sender, date sent, size, and subject). Users could double-click a voicemail from their email inbox and hear the message through their PC or a phone next to their desk. Using any phone in the world, users could listen to voice messages like they normally did, plus have emails read to them (in synthesized voice). Voice messages could be sent using email or telephone addressing schemes, and the data networking infrastructure was used to send messages between locations rather than the public switched telephone network. Unified Messaging was not a commercial success at the time because in the late 1990s email did not enjoy a huge market share, email servers were not very reliable, internet connections were slow (voice messages were large files) and most PCs did not have speakers or microphones.

History of voicemail services for small businesses, residences, and cellular services

Until 1988, telephone companies and the newly formed cellular phone companies were barred by law from offering voicemail to their subscribers. This was done by the FCC to protect the telephone answering businesses around the country. This prohibition continued with the decree which broke up AT&T in 1984. A subsequent ruling by Judge Harold H. Greene on March 7, 1988, reversed this barrier. Phone companies were allowed to offer voicemail as a service, but they were barred from designing or manufacturing the machines that could provide the service.

VMX’s large system was used by a few carriers (telephone companies), but severe reliability and cost issues prevented VMX from expansion to the carrier market. Octel already had very high capacity systems for corporate use and by 1988 all seven Regional Bell Operating companies were using Octel for internal use. Octel first adapted its largest system for the carriers, which enabled them to offer reliable voicemail to their subscribers. Within a year, Octel launched a new generation of its large system specifically designed for carriers which was compliant with “NEBS standards,” the tight standard required by phone companies for any equipment located in their central offices. A few other manufacturers entered the voicemail market for carriers including Unisys, Boston Technology (founded by Greg Carr and Scott Jones), and Comverse Technology (an Israeli based company founded by Kobi Alexander). These vendors did not offer voicemail to corporations but they focused on the potentially large and lucrative carrier market. Unisys secured PacBell’s residential voicemail services, and Boston Technology was the mainstay of Bell Atlantic’s residential voicemail offering. None of the other corporate voicemail manufacturers had notable success with the carrier market because their systems’ capacities were too small and the equipment wasn’t reliable enough. Selling to carriers also required a different method of sales and marketing than selling to the corporate market, and only Octel succeeded at both.

Perhaps the first cellular carrier in North America to offer voicemail successfully to its subscribers was Bell Cellular, the Canadian carrier serving Ontario and Quebec (Bell Cellular later changed its name to Bell Mobility). Bell Cellular’s success with voicemail caught on, and cellular voicemail spread throughout Canada and then to the US and overseas. Within a few years, 100% of Canadian cellular companies ultimately used Octel voicemail, followed by virtually all of the major US wireless carriers (including the seven RBOCs, AT&T Wireless and McCaw) and a large percent of the GSM carriers around the world. Comverse Technology was very successful in the GSM market outside the US. The Octel user interface became the most common in the world with carriers, but each carrier made minor variations on the interface.

Other interesting markets developed from the carrier market including a concept called “virtual telephony.” Virtual Telephony, developed by Octel, used voicemail to provide phone service rapidly in emerging countries without wiring for telephones. The problem this solved was that emerging countries did not have many telephones. Wiring for telephones was very expensive, and many poorer citizens didn’t have homes to wire. The economies of emerging countries were held back partly because people couldn’t communicate beyond the area where they could walk or ride a bicycle. Giving them phones was one way to help their economies, but there wasn’t a practical way to do it. In some countries, the wait for a phone was several years and the cost was in the thousands of dollars. Cellular phones weren’t an option at the time because they were extremely expensive (thousands of dollars per handset) and the infrastructure to install cell sites was also costly.

With virtual telephony, each person could be given a phone number (just the number, not the phone) and a voice mailbox. The citizen would also be given a pager. If someone called the phone number, it never rang on an actual phone, but would be routed immediately to a central voicemail system. The voicemail system answered the call and the caller could leave a long, detailed message. As soon as the message was received, the voicemail system would trigger the citizen’s pager. When the page was received, the citizen would find a pay phone and call in to pick up the message. This concept was used successfully in South America and South Africa.

Consolidation

In the early ’80s there were over 30 companies vying for the corporate voicemail market including many companies no longer in business today. Among the many contenders were IBM, VMX, Wang, Octel, ROLM, AT&T, Northern Telecom, Delphi Communications, Voice and Data Systems, Opcom, Commterm, Genesis, Brook Trout, Glenayre, BBL, AVT, AVST, Digital Sound, Centigram, Voicemail International, Active Voice, and many others. Virtually all contenders in the corporate voicemail market were based in the United States.

By the mid 1990s, IBM and Wang exited the voicemail market because they couldn’t get enough traction. ROLM was purchased by IBM in the mid 1980s (which was a financial disaster for the profitable ROLM, as IBM clearly could not grasp the laid back, "think outside the box" attitude of ROLM, which was the #2 PBX supplier in the US from the mid 70s to late 80s), then sold half interest to the German company, Siemens. In 1992, Siemens bought ROLM entirely from IBM and the original ROLM product line was done for, except for PhoneMail (the only product Siemens did not destroy). VMX suffered from poor product and ineffective management and was about to fold when Opcom merged with it. The surviving company was called VMX, but VMX was all but erased by Opcom except for its name and patent portfolio. In 1994, Octel bought VMX. By the early 90s, AT&T/Lucent created its version of voicemail for the corporate market (called Audix) but it would only work on AT&T/Lucent PBXs. Nortel developed Meridian Mail and followed the same strategy as AT&T in that Meridian Mail only worked with Northern Telecom PBXs. As a result, neither company achieved much market share with large national or multi-national accounts (because few major companies, if any, used only one brand of PBX, though Nortel had been the major leader since the late 1970s with ROLM a close second and poised to overtake Nortel until IBM bought ROLM). AT&T spun off its equipment business into a company called Lucent Technologies, and Northern Telecom changed its name to Nortel. Several small companies offering voicemail folded because of inadequate product or management.

By the mid-1990s, Octel had become the number one supplier of voicemail both to corporations and to carriers. It had about a 60% market share in the U.S., Canada, Europe and Japan (for large corporations) and between a 30% and 100% of the carrier market, depending on the country. By 1997 Octel’s biggest competitors were Audix, made by Lucent, and Meridian Mail, made by Nortel. In July 1997, Octel was purchased by Lucent Technology. Lucent’s AUDIX division was merged into Octel to form the Octel Messaging Division. In the same year, Boston Technology was acquired by Comverse Technology making it the second largest supplier to carriers after Octel. In a few years Comverse became the largest supplier to carriers with Lucent holding its leadership in the corporate market and second place with carriers. By 2000, some estimate that there were over 150,000,000 active users of corporate and carrier voicemail made by the Octel Messaging Division. Shortly thereafter, Lucent spun off its corporate business, including the Octel Messaging Division, into a company known as Avaya. Comverse today retains its leadership of voicemail systems sold to carriers around the world.

Voicemail today and tomorrow

By the year 2000, voicemail had become a ubiquitous feature on phone systems serving companies, cellular and residential subscribers. Cellular and residential voicemail continue today in their previous form, primarily simple telephone answering. Email became the prevalent messaging system, email servers and software became quite reliable, and virtually all office workers were equipped with multimedia desktop PCs.

Instant messaging in voice: The next development in messaging was in making text messaging real-time, rather than just asynchronous store-and-forward delivery into a mailbox. It started with Internet service provider America Online (AOL) as a public Internet-based free text “chat” service for consumers, but soon was being used by business people as well. It introduced the concept of Internet Protocol “presence management” or being able to detect device connectivity to the Internet and contact recipient “availability” status to exchange real-time messages, as well as personalized “Buddy list” directories to allow only people you knew to find out your status and initiate a real-time text messaging exchange with you. Presence and Instant Messaging has since evolved into more than short text messages, but now can include the exchange of data files (documents, pictures) and the escalation of the contact into a voice conversational connection.

Voice messaging with mobile devices

The increase in wireless mobility, originally through cellular services and today through IP-based Wi-Fi, was also a driver for messaging convergence with mobile telephony. Today it is not only fostering the use of speech user interfaces for message management, but increasing the demand for retrieval of voice messages integrated with email. It also enables people to reply to both voice and email messages in voice rather than text. New services, such as GotVoice and SpinVox, are helping to blur the boundaries between voicemail and text by delivering voicemails to mobile phones as SMS text messages.

Unified messaging with voip/ip telephony

Corporate voicemail, however, did not change much until the advent of Voice over IP (VoIP — voice being transmitted over the internet) and the development of IP telephony applications to replace legacy PBX telephony (called TDM technologies). IP (Internet Protocol) telephony changed the style and technology of PBXs and the way voicemail systems integrated with them. This, in turn, facilitated a new generation of Unified Messaging, which is now likely to catch on widely. The flexibility, manageability, lower costs, reliability, speed, and user convenience for messaging convergence is now possible where it wasn’t before. This might include intra- and inter-enterprise contacts, mobile contacts, proactive application information delivery, and customer contact applications.

The corporate IP telephony-based voicemail CPE market is served by several vendors including Avaya, Cisco systems, Adomo, Interactive Intelligence[2], Nortel, Mitel, 3Com, and AVST. Their marketing strategy will have to address the need to support a variety of legacy PBXs as well as new Voice over IP as enterprises migrate towards converging IP-based telecommunications. A similar situation exists for the carrier market for voicemail servers, currently dominated by Comverse Technology, with some share still held by Lucent Technologies.

VoIP and IP telephony enable centralized, shared servers, with remote administration and usage management for corporate (enterprise) customers. In the past, carriers lost this business because it was far too expensive and inflexible to have remote managed facilities by the phone company. With VoIP, remote administration is far more economical. This technology has re-opened opportunities for carriers to offer hosted, shared services for all forms of converged IP telecommunications, including IP-PBX and voicemail services. Because of the convergence of wired and wireless communications, such services may also include support of a variety of multi-modal handheld and desktop end user devices.

There are a few technologies which have made a directly dramatic impact on people’s lives. Some of these rose from total obscurity to ubiquity relatively quickly, such as computers, telephones, cellular phones, personal computer, photocopying, voicemail, e-mail, integrated circuits, to name a few. Voicemail has become a standard part of everyone’s life and now is taken for granted. It is everywhere, both as a simple telephone answering system and as a more complex unified messaging system. Voicemail has touched everyone’s life differently: it has enabled businesses to operate more efficiently, propagated humor, advanced romantic relationships, saved lives, and enabled commerce to blossom in the poorest areas of the worlds. It went from nothing to ubiquity in less than 15 years. As long as people use their voice to communicate, some form of voicemail will live on for many years to come.

How voicemail systems work

This section describes how the original style, standalone, voicemail system worked with a corporate PBX. The principle is the same with Central Office Switches (CO Switches) or Mobile Telephone Switching Systems (MTSOs). More modern voicemail systems work on the same principle, but some of the components may be shared with other systems, such as email systems.

Voicemail systems contain several elements shown in the figure below:

• A central processor (CPU) which runs the operating system and a program (software) that gives the system the look-and-feel of a voicemail system. This software includes thousands of pre-recorded prompts that “speak” to the users as they interact with the system;
• Disk controller and multiple disk drives for message storage;
• System disks which not only include the software above, but also contain a complete directory of all users with pertinent data about each (name, extension number, voicemail preferences, and pointers to each of the messages stored on the message disk that belong to them);
• Telephone interface system that enables many phone lines to be connected to it.

File:Voice Mail Block Diagram.jpg

The drawing below shows how the voicemail system interacts with the PBX. Suppose an outside caller is calling Fred’s extension 2345. The incoming call comes in from the public network (A) and comes into the PBX. The call is routed to Fred’s extension (B), but Fred doesn’t answer. After a certain number of rings, the PBX stops ringing Fred’s extension and forwards the call to an extension connected to the voicemail system (C). It does this because PBXs are generally programmed to forward busy or unanswered calls to another extension. Simultaneously the PBX tells the voicemail system (through signaling link D) that the call it is forwarding to voicemail is for Fred at extension 2345. In this way, the voicemail system can answer the call with Fred’s greeting.

File:Voice Mail-PBX Block Diagram.jpg

There are many microprocessors throughout the system since the system must handle large amounts of data and it’s unacceptable to have any wait times (for example, when the system is recording or playing your message, it’s unacceptable if the system stops recording momentarily like computers often do while accessing large files).

When Fred’s extension forwards to the voicemail system, the Telephone Interface detects ringing. It signals to the Central Processor (CPU) that a call is coming in. The CPU simultaneously receives a signal on the PBX-Voicemail Data Link (D) telling it that extension 2345 is being forwarded on ring-no-answer to the specific extension that is now ringing. The CPU directs the Telephone Interface (which controls the line interface cards) to answer the call. The CPU’s program realizes that it’s a call for Fred so it looks up Fred’s greeting immediately and directs the Disk Controller to start playing it to the caller. It also plays some system prompts instructing the caller what comes next (for example, “When you have finished recording, you may hang up or press ‘#’ for more options”). All “talking” to the caller is done through prompts that are selected by the CPU according to the program stored in the voicemail system. The CPU selects the prompts in response to the keys the caller presses.

The caller’s message is digitized by the Telephone Interface system and transmitted to the Disk Controller for storage onto the Message Disks. Some voicemail systems will scramble the message for further security. The CPU then stores the location of that message in the System Disk inside Fred’s mailbox directory entry. After the caller hangs up and the message has been stored, the CPU sends a signal to the PBX through the link (D) instructing the PBX to turn on the message waiting light on Fred’s phone.

When Fred comes back to his desk and sees the light on his phone, he calls a designated extension number for the voicemail system (an actual extension number assigned to the lines in “C” in the figure above).

Again the Telephone Interface alerts the CPU that a call is coming in on a particular line, but this time the signaling from the PBX-Voicemail Data Link (D) indicates that Fred is calling directly, not being forwarded. The CPU directs the Telephone Interface to answer the call.

Since the CPU “knows” it is Fred (from the signaling on the Data Link D), it looks up Fred’s information on the System Disk, specifically his password. The CPU then directs Disk Controller to play a log-on prompt to the user: “Please enter your password.” Once the password is entered (via Touch-tones), the CPU compares it to the correct one and, if entered correctly, allows Fred to continue.

The CPU then determines (from Fred’s directory entry) that Fred has a new message. The CPU then presents Fred his options (e.g., “You have a new message. To listen to your new message, press 1; to record a message, press 2” etc.) The options are presented by the CPU directing the Disk Controller to play prompts, and the CPU listens for Touch-tones from Fred. This interaction of playing prompts and responding with Touch-tones enables Fred to interact with the voicemail system easily.

If Fred presses 1 to listen to his message, the CPU looks up the location of Fred’s new message in his mailbox directory (on the System Disk), and directs the Disk Controller to play that message. The Disk Controller finds the message on the Message Disks, and sends the data stream directly to the Telephone Interface. The Telephone Interface then converts the data stream to sound and plays it to Fred through the Line Interface Card which Fred is connected to.

Playback controls (like rewind, pause, fast forward, changing volume, etc) are all input via Touch-tones, are “read” by the CPU, and the appropriate actions are taken based on the stored program in the system. For example, if Fred wants to pause message playback, he might press 2. Since the CPU is constantly listening for Touch-tones from Fred, his command causes the CPU to direct the Disk Controller to stop playing the message. A variety of playback controls and options are available on most sophisticated voicemail systems so that users can control message playback, store messages in archives, send messages to groups, change their preferences, etc.

The better designed voicemail systems have a user-friendly interface with clear and meaningful prompts so the interaction with the voicemail system is quick and easy.

References

1. Touch-tones are the tones generated by pushing buttons on a telephone with push-button dialing. The term Touch-tone was coined by AT&T (Western Electric). The official name for the tones that are generated are DTMF, or dual-tone multi-frequency. The buttons collectively are referred to as the “Touch-tone dial” or the “Touch-tone keypad”. Each button generates two tones.
2. “A Reactive Telephone Message Network for the Office of the Future”, Business Communications Review, July-Aug 1980; “Voice Mail Arrives in the Office”, Business Week, June 9, 1980, p. 81.
3. “The Case for Voice Mail: Confirmed.” GE Corporate Telecommunications publication, May, 1989, Constance C. Kelly, editor.
4. “IBM Audio Distribution System”, IBM publication GX60-0075-0
5. “Toward Competitive Provision of Public Record Message Services”, Experimental Technology Incentives Program, National Bureau of Standards, Washington, DC. Publication NBX-GCR-ETIP-81-97 October, 1991.
6. “Speech Filing System Reference Manual”, 1975, by J. W. Schoonard and S. J. Boies, IBM Research Center, Yorktown Heights, NY, 10598.
7. “How to Shoulder Aside the Titans”, Gene Bylinsky, Fortune, May 18, 1992; “Octel Keeps Bringing You Voice Mail”, Global Telecoms Business (UK), February/March 1996, pp. 22-24
8. “Human Factors Challenges In Creating A Principal Support Office System — The Speech Filing System Approach”, by John D. Gould and Stephen J. Boies, IBM Tomas J. Watson Research Center, as quoted in a paper presented to the Association for Computer Machinery. See ACM Transactions on Office Information Systems, Vol. 1, No. 4, October 1983, pp. 273-298.
9. “Speech Filing — An Office System For Principals”, by J.D. Gould and S.J. Boies, IBM Systems Journal, Volume 23, Number 1, 1984, p. 65.
10. “IBM Audio Distribution System Subscriber’s Guide” and “IBM Audio Distribution System, Administrator’s Guide”. IBM Publications SC34-0400-3 and SC34-0400-1
11. Correspondence with Jay Stoffer, March 26, 2006: “… As to Gordon Matthews, I was introduced to him by a Venture Capitalist that later invested in Delphi. I met with Gordon and his wife at their home in Dallas with the objective of ascertaining if he could contribute to our product planning process. I concluded that he would not be likely to add value in that activity but that my colleagues should evaluate his potential contribution to Engineering. To that end, we flew Gordon back to LA where he was interviewed by members of our technical team. It was at this time (1973/1974) that he would have seen a demonstration of the voice application. He had definitely been thinking about a voicemail system prior to this visit but he had definitely not established his company or raised the capital to do so. Furthermore, his product plan was still very much in the formative stage and never reached the sophistication of the Delphi offering.” Needless to say, there was no employment offer made to Matthews by Delphi.
12. Transcriptions of various seminars sponsored by Probe Research, Inc., September, 1982:
    • “Voice Message Service,” Proceedings of Voice Processing Seminar, September 15, 1982;
    • “BBL Industries, Inc.,” Proceedings of Voice Processing Seminar, September 15, 1982;
    • “Wang Laboratories,” Proceedings of Voice Processing Seminar, September 16, 1982;
    • “American Telephone and Telegraph, Inc.,” Proceedings of Voice Processing Seminar, September 16, 1982;
    • “Commterm, Inc.,” Proceedings of Voice Processing Seminar, Sep. 16, 1982.
13. “Voice Store and Forward for the Automated Office”, a presentation by Lawrence E. Bergeron, Dennis B. Howell and Dean Osborne, Wang Laboratories, Inc., Lowell, Mass., transcribed in “Computer Controlled Voice Message Systems and the Office of the Future”, Professional Program Session Record (10), Wescon/81, Electronic Show and Convention, September 15-17 September, 1981, section 2, pp. 1-8.
14. “The PhoneMail System for the ROLM CBX”, publication by ROLM Corporation.
15. “Octel Emerges as Rising Star in Voice Messaging Systems”, Peninsula Times Tribune, November 7, 1988, page C-1; “Investors Waking Up to Octel’s Leadership”, Investor’s Daily, February 17, 1989; “Octel’s Stock Gamble Has Paid Off”, USA Today, Friday, February 24, 1989, page 3B.
16. “Octel Communications Corporation”, filings with the Securities and Exchange Commission for its prospectus for secondary public offering, August 15, 1989; Various internal manuals and publications from Octel Communications Corporation.
17. “All Your Messages in One Place”, Michael H. Martin, Fortune, May 12, 1997, p. 172.
18. “Toward Competitive Provision of Public Record Message Services”, ETIP (Experimental Technology Incentives Program”, National Bureau of Standards, Washington, D.C., October, 1981; “domestic Public Message Services”, FCC publication 71FCC 2d 471; “Telecommunications Competition and Deregulation Act of 1981” (FCC Computer Inquiry II), Docket 20828, December 30, 1980; “Denial of AT&T Petition for Waiver of Section 64.702 of the Commission Rules and Regulations”, October 7, 1981, Federal Communications Reports 88FCC 2d.
19. United States of America (Plaintiff) v. Western Electric Company, Inc., et al (Defendants). Civil action no. 82-0192, Section VII pp. 51-65: “The judge on review considers the threat to possible competition in the voice mail and storage business to be less real than the opportunities lost to the public welfare by these services not being broadly available. Hence, the BOCs should be able to provide voice mail.”
20. AT&T Wireless ultimately bought McCaw Cellular. The combined company was eventually bought by Cingular.
21. GSM (Global System for Mobile Carriers) is one of the various cellular technologies which include TDMA, CDMA, iDEN and others. GSM is currently the technology used by Cingular in the US and is the prevalent technology in over 100 countries around the world.
22. Investor’s Business Daily, February 1, 1996. “Octel’s Robert Cohn: CEO of Voice-Messaging Firm Puts Premium on Speed”, by Kathleen Doler.
23. “Lucent Is Set To Buy Leader In Voice Mail”, Seth Schiesel, New York Times, July 18, 1997, Page C1.