Past Entry of Zippy's Telecom Blog
More Previous Columns
December 17, 2010
Computer Telephony Still Extant
Years ago, Newton’s Telecom Dictionary (still as popular today as it was then) first defined CT something along the lines of: “the discipline of applying computer-based intelligence to the making and receiving of phone calls, fax communications and other complex messaging and transactions involving public, private and Internet networks.”
That definition left out reverse situations where the telephone system must control some of the functionality of the computer system, a reminder of the increasingly symbiotic relationship between computer intelligence communications. That definition was also written before the rise of IP Telephony and more general IP Communications (including video). Computer telephony is thought of as the circuit-switched predecessor of today’s IP Communications, but in fact it underlies all modern communications applications that call upon computer intelligence, everything from call/contact centers to individuals to telepresence to individuals using Skype. That’s partly because, aside from Newton’s definition, there was no generally agreed-upon definition of what exactly comprises computer telephony. That’s because telecom kept changing—it continues to be a moving target for those who would like to place it neatly in a pigeon hole and move on to other matters.
As I wrote in 2000 in the Computer Telephony Encyclopedia: “In the late 1960s the then-unnamed field of CT consisted of a few IBM 360 mainframes connected to their IBM 2750 PBXs, receiving over the public network in Germany order fulfillment-related messages from bookstores. In the 1970s it was the automatic distribution of calls to agents in large call centers with Rockwell equipment, and the Delphi Corporation’s Delta 1 system used by telephone answering services in San Francisco. In the 1980s it was VMX voicemail, Periphonics IVR, and predictive dialers.”
Following the dismantlement of the AT&T empire starting in 1984, a tsunami of PBX makers appeared with slightly less expensive equipment offerings, each believing that their devices — proprietary card cages accompanied by so-called OAIs (Open Application Interfaces, a misnomer if there ever was one)—would continue to expand in capability. Indeed, many experts thought that a powerful, computerized PBX would somehow become the enterprise “office mainframe” that would control the office, computing data and moving it around between departments, as well as provide interesting new forms of call control. Thus, these proposed office mainframe PBXs would drive many of the services that we now take for granted in an IP environment.
A few phone systems did appear that could actually allow data to travel over the phone system wiring (increasingly rare modems for dial-up service are a simple example of how to send data over voice). IBM / Rohm placed computer-like phones on the desktop, but a combination of complex functions, expense and limited utility kept such systems from taking over the enterprise. (Still, one wonders: If IP had never happened, would we have had to invent something similar, or could miniaturized, inexpensive 21st century electronics allow circuit-switched technology to handle sophisticated applications as flexibly and inexpensively as packet-switched technology does today?)
What did survive from that era was the concept of applying computer intelligence to telephone calls. This time, however, the computer “smarts” would be running on PCs. As early as 1983, Nick Zwick, one of the founders of the Dialogic Corporation, began designing the first, open multiline voice board for PCs. The following year, Hank Magnuski developed the first computer-based fax board and founded GammaLink, Inc., which ultimately became part of Dialogic.
In 1985 an employee at the Digital Equipment Corporation (DEC) named Carl Strathmeyer, developed the first computer-phone system link at DEC (which later became Dialogic’s Computer-Telephone Division). He used the term Computer Integrated Telephony (CIT). IBM became interested in the concept, but did not want to use a term devised by their chief competitor, DEC. Jim Burton, then at IBM, made the CIT more palatable to Big Blue by switching some letters of the initialism around to get CTI. Ironically, Burton has joked many times since then about how he’s always liked Strathmeyer’s term more than his own.
As telecom-related applications scaled up, coordinating the activities of many computing and telecom resources scattered about became increasingly important, not to mention difficult. Historically, the first major problem was simply coordinating what was happening among the plug-in boards on a backplane and how a PC could talk to a PBX. Just a year after Carl Strathmeyer at DEC concocted the first CTI-PBX link in 1985, Dialogic introduced the first telephony bus for resource sharing, the Analog Expansion Bus (AEB). The “resources” connected to AEB where in most cases Dialogic boards that fit into a PC AT-expansion slot, such as the DTI/124, D/4x and AMX line.
In 1989 Dialogic introduced the first 12 channel DSP-based voice processing board, the first T-1 interface board for voice processing, and the first digital TDM bus for resource sharing (PEB). Other companies, such as Natural MicroSystems, Brooktrout (now Cantata), Amtelco, PIKA, NewVoice (now part of AudioCodes) and others followed suit, joining the fray with their own “CT boards”.
During this time (the end of the 1980s and beginning of the 1990s) Harry Newton and his partner, Gerry Friesen, were running a monthly telecommunications magazine called Teleconnect. But one area they didn’t have covered (as regards advertising) was computer telephony. Marc Ostrofsky (and then Advanstar whom Ostrofsky sold his properties to) had a bunch of trade shows and a magazine entitled Voice. They were doing very well. Newton needed to kill off the competing Voice shows and magazines, so he coined the term “Computer Telephony” and told anyone who would listen that computer telephony was “bigger” than “Voice” and that limiting your participation to Voice would limit your sales and your profits. It was a brilliant story. Many people listened to Newton and believed him. He and Friesen put out Computer Telephony magazine (the test issue was in October 1993, the regular issues began in January 1994) which was listed in the February, 2000 Folio magazine as one of the 20 hot start up magazines of the 1990s. Also in 1994 they changed the name of their “Telecom Developers” Expo to “Computer Telephony Expo.” Computer Telephony and the Computer Telephony Conference and Exposition effectively killed Advanstar’s Voice shows and magazine, which continued for a while in England. In September, 1997, Newton and Friesen then sold their whole empire of successful magazines and trade shows—Call Center, Computer Telephony, Imaging, Teleconnect and Computer Telephony Conference and Exposition (CT Expo) to Miller Freeman, Inc. (later CMP) for a smidgen over $130 million. Folio magazine listed it as the fourth biggest magazine deal of 1997.
In response, those who had grown accustomed to using the term CTI had by this time recognized that expression “computer telephone integration” was somewhat restrictive, suggesting technology relating solely to physical computer-telephone connectivity. Still, they tried to hold onto the acronym by formulating and popularizing the “hybrid” term Computer Telephony Integration. Both “CTI” and “computer telephony” were then used interchangeably—for several years Computer Telephony magazine’s only competitor was CTI magazine, published by Nadji and Richard Tehrani of the Technology Marketing Corporation, who also publish C@LL CENTER Solutions magazine, Internet Telephony magazine, and sponsor the biannual CTI EXPO and Internet Telephony EXPOs.
In the March 1997 issue of Computer Telephony magazine, Harry Newton published an article entitled “Why CTI is not Computer Telephony” in which he explained how CTI is just one small part of the total CT picture. People must have believed him. In January 2000 CTI magazine was relaunched with a new name, Communications Solutions.
In any case, from the early 1990s on computer telephony was not just about big computers any more, but little ones—microcomputers and PCs that promised to bring new and “open” programmable capabilities to both a staid telecom industry and to affordable, scalable business phone systems. Industry standardization and the availability of low-cost CT links gave rise to a new generation of CT applications that linked LAN servers, public phone switches, private PBXs and desktop PCs to phone lines. With the successful melding of microcomputers and telephony, all kinds of amazing new business opportunities appeared in the 1990s to tempt developers and integrators: IVR systems, fax servers, debit card systems, international callback systems, inexpensive PC-based phone systems (called “CT Servers” or “unPBXs”, a term coined by Ed Margulies, now at Cisco) with voicemail and automatic call distribution, and call centers using PC-based predictive dialers and PC-based databases furnishing “screen pops” of customer data to the PC screens of customer service representatives.
In those days the two great names in computing where Novell and Microsoft, engendering such popular adages as “Novell owns the network and Microsoft owns the desktop.” Ironically, these two companies’ perceived personae would carry over into their first attempts to impose a standard computer telephony API model. In 1992 AT&T and Novell deployed the Telephony Services Applications Programming Interface (TSAPI) which only ran on NetWare, was too expensive, but provided true client/server functionality for use in third-party call control applications, meaning that the server is intelligent and can establish and route calls independently of the user and can do such things as control a distant PBX on a WAN. Intel and Microsoft followed on TSAPI’s heels with the Telephony Application Programming Interface (TAPI), a free piece of telephony software bundled with Windows that could telephony-enable applications running at the desktop (first party call control), but would not have full blown third-party client/server call control until version 3.0. This did not stop several ingenious early vendors from working around this problem by creating TAPI service providers which tricked the desktop PC into seeing a TAPI “device” inside the PC when in fact there is none, so that TAPI function calls could be redirected over a LAN/WAN connection to a PBX.
In the late 1990s, computer telephony had to be redefined yet again as the traditional circuit-switched telephone system converged with packetized technologies and networks such as IP and the Internet. At this point it became evident that one could “take the computer out of computer telephony” leaving only an abstract model of call control and messaging intelligence that was threatening to disappear into the larger machinery of customer and enterprise relationship management. In an era where the concept of a platform on which to build shifted from operating systems to browsers, TAPI, having just emerged victorious from the “TAPI TSAPI war” faced in 1996 a totally new kinds of competitors such as the Java Telephony API, or JTAPI.
Not only has computer telephony rapidly evolved during the 1990s, but it has continually enlarged in scope. After all, computer telephony was never a single application, but a collection of them: Voice mail systems, fax servers, IVR, ACDs, intelligent e-mail systems, unified messaging, etc., all of which somehow had to work together to make life easier and more profitable for those daring enough to build such systems. This melange of technologies that comprise computer telephony goes on getting larger and larger and more diffuse, with no signs of stopping, like some great puff of smoke that expands and ultimately blends into its surroundings.
But if there is a single leitmotiv that runs through CT, it is the primacy of voice communications. Early promoters of CT defined it as a confluence of computer, telecom and voice processing systems. Indeed, when I first went to work for the flamboyant Harry Newton and his then-fledgling Computer Telephony magazine in 1994, we were chatting in an elevator when he suddenly turned to me and said: “You know, Richard, this field used to be just a small part of voice processing before I began to promote it into bigger and better things.” (At that Newton was obviously successful, having sold his company in 1997 for more than $130 million). Even later, when Computer Telephony magazine at the end of 1999 decided to revamp its tagline “The Magazine for Computer and Telephone Integration” to acknowledge the convergence of the circuit-switched and packet-switched networks, we came up with “The Voice of Converging Communications.” We used the word “voice” to emphasize not only the fact that CTM was the leading publication in its field, but also to recognize the almost archetypal nature and predominance of the human voice in communications.
After all, the telephone, be it connected to a wire or wireless, has been with us for over a century and remains the most widely used business and personal communications tool. One of computer telephony’s many tricks is to convert telephones into virtual “terminals,” expanding their capabilities so that all sorts of messages can be sent and received with them, saving time and money.
Because computer telephony embraces so many things, we tend to lose sight of it as a distinct entity. Without realizing it, we use some form of CT every day: We check our voice mail several times a day, get our bank balances by phone, call a phone number to hear the weather forecast or call 777-FILM and to hear audiotex telling us about the movies playing downtown, make phone calls with a PC, request technical and product information via fax using the phone’s keypad or even by voice command, use touchtones to get information, hear e-mail read to us over the phone via text-to-speech, instruct our phone system find us wherever we happen to be, do fax broadcasts to our customers, review all of our voice, e-mail and fax messages in one list on one screen, and use the Internet for phone calls over wireline or wireless networks. Researchers and pollsters use CT to collect and analyze information. Corporate and Financial information is provided to employees or stockholders via CT. Even when your phone rings while you’re eating dinner and someone tries to sell you something, that call has been placed by a telemarketer using a predictive dialing system, yet another technological denizen that lurks in the computer telephony realm.
Thus, computer telephony is not a “single anything” but a technological enabling environment that brings intelligence and choices to the communications process, be it circuit-switched or packet-switched in nature. It solves no particular business problem unless someone builds a suitable application upon its foundations, using the many tools that it provides. And indeed some of the new tools available to computer telephony developers, such as application generators and customizable turnkey systems, allow non-technically oriented people to easily shape their communications environment to their own vision. It’s a far cry from the early days when even CT’s principal tool—raw computing power—was so unsophisticated and expensive that only major entities such as AT&T, IBM / ROLM, Tandem, etc. could implement such systems.
Not only is there confusion over just what computer telephony is, there’s also some bafflement concerning just “where” computer telephony resides. Similar CT applications can and have been deployed both in Customer Premise Equipment (CPE) and as enhanced services originated by telcos (or originated by third parties and resold by telcos) and offered by subscription. Further confusion results from the fact that both telecommunications and computer / data communications industries have separate ideas over what computer telephony is and where it should live in the network, which has lead to perplexity over who sells, installs and maintains CT solutions. The bright side of all this is that computer telephony is fostering the convergence of voice and data, driving the merger of packet and circuit-switched technologies for personal and business communications, and will continue to do so until it leads to some future, simplified set of communications platforms based solely on IP.
In a ghostwritten September 2007 interview I did for the now defunct VON Magazine (probably the best magazine Yours Truly ever helped to start up), I mentioned to Michael Thurk, then the group vice president, Enterprise Communications Group, for Avaya, that he had spent 14 years dealing with enterprise network and telecommunications-related businesses at DEC and that DEC employee Carl Strathmeyer had coined the term CIT in 1985. I then asked him what he thought were the “killer apps” for the Enterprise today and how are they different than they were in the era of CIT/CTI.
Thurk’s response: “The world of communications for enterprises is a different business today. CIT was, essentially, the very first step toward allowing telecommunications voice switching platforms to interoperate with data infrastructures. Today, they are one and the same. We are now seeing enterprises integrate voice into their business applications and processes to create intelligent communications, enabling companies to implement new forms of customer care, increase their employees’ efficiency, and lower their infrastructure costs by combining both voice and data on the same infrastructure.”
Interesting how he correctly saw the merging of telecom, computing and data, making them “one and the same.”
So, Computer Telephony isn’t really an obsolete term. Granted, it probably should have been called “Computer-Controlled Communications” (CCC) or something like that, but it actually underlies the ubiquitous world of IP Communications. Computer Telephony, both the word and the methodology, is still extant and, in some form or other, will be with us forever.