The TMC Labs engineers spent many weeks picking the most important and influential
products of 1997: You should keep this Products Of The Year issue as a resource and use it
as a guide for all your purchasing decisions in the coming months. During the selection
process, it occurred to all of us at CTI magazine that, aside from products that
were introduced, a host of technologies were either invented, or came to the forefront
during 1997 technologies that will be instrumental in shaping the future of most
every product in the CTI market.
These technologies will certainly all be seen in products as time goes on. Consider
this column an early warning system for the technologies to look for in future product
acquisitions. Developers, integrators, and technology analysts can use this Outlook as a
hot list of technologies that they should be familiar with potentially examining or
implementing them now or in the near future.
INTERNET TELEPHONY
Internet telephony has been around for more than a year. In fact, as I understand it,
rudimentary tinkering in this field began over 20 years ago. 1997 became the year that it
rocketed to become a mainstream term with dramatic implications. Of course, this
technology is equally (if not more) suitable on an intranet due to the fact that reliable
Quality of Service (QoS) is not yet available on the Internet.
To understand the implications of this technology, it is helpful to analyze the current
state of telecommunications. The telecommunications infrastructure of the world was
developed many decades ago. In the 1960s, equipment capable of transmitting voice over the
telephone network was considered state-of-the-art, and the transmission speed of the
telephone network was linked to the capabilities of the hardware and software at that
time. Over the past thirty years, computational power has increased at an extremely rapid
rate, translating into many orders of magnitude. However, our telephone network speed is
still mated to state-of-theart technology from decades past.
TCP/IP is the underlying network protocol of not only UNIX, but the Internet as well,
and has been around for decades. As recently as five years ago, many competing network
standards were supported and maintained by different hardware and software vendors. The
success of the Internet has made it clear that TCP/IP will be the standard network
protocol of the present and near future.
CTI components from Dialogic, Natural MicroSystems, Brooktrout, and others have become
extremely powerful in the past few years through the use of extremely powerful Digital
Signal Processors (DSPs). These DSPs are capable of converting analog voice to compressed
digital data and back. Using the telephone networks existing cabling, augmented with
strategically placed Internet telephony gateways, allows at least 8 conversations to
travel over the same circuit that currently can transmit only one conversation. The
circuit, of course, would transport compressed voice through Internet Protocol (IP)
packets as opposed to uncompressed, circuit-switched voice. This is one of the reasons the
entire telecommunications infrastructure will be revolutionized by Internet or IP
telephony.
It is clear that any data network can now become a telephone network that is orders of
magnitudes more efficient than the current telephone network. The fact is, the telephone
companies have an infrastructure in place that supports either. They will eventually
transport their traditional circuit-switched calls onto IP packet-switched networks. ISPs
and cable companies have been looking for new money-making opportunities and have
discovered that they, too, can become telephone companies, using Internet telephony as a
highly efficient transport mechanism. Corporate intranets can now pay for themselves and
still have money left over by piggybacking voice, video, and fax on existing data lines.
These examples are but a brief taste of what Internet telephony can and will do for us.
For more in-depth information on this topic you will certainly want to subscribe and
encourage others who can benefit from this technology to subscribe to our new publication,
Internet Telephony magazine, focusing exclusively on this technology. Free
subscriptions are available on the Web at www.itmag.com
COMPACTPCI
CompactPCI is a PC-based I/O bus specification that brings features of the most rugged
industrialized computers into the realm of the PC. The standard was developed by the PCI Industrial Computers Manufacturers Group, a
consortium of over 190 vendors. The standard will bring PCs and PC-based CTI into the
central office and other service organizations that require near-zero downtime. In fact,
any product that currently runs on the standard PCI bus should be software compat-ible
with this new hardware specification.
You may recall the miniaturization of components prevalent in NASA R&D trickled
down to the masses in the form of the microprocessor and other computer technology.
Similarly, service providers will roll CompactPCI out en masse, and end users will enjoy
the benefits of this industrial grade technology at a reasonable price in the
future.
Compared to standard desktop PCI, CompactPCI supports twice as many PCI slots (8 versus
4) and offers a packaging scheme better suited to industrial applications. For example,
CompactPCI cards are designed for front-loading and removal from a card cage. The cards
are firmly held in position by their connector, card guides on both sides, and a
faceplate, which solidly screws into the card cage. Cards are mounted vertically, allowing
for natural or forced air convection for cooling. Finally, the pin-and-socket con-nector
of the CompactPCI card is signifi-cantly more reliable and has better shock and vibration
characteristics than the card edge connector of standard PCI cards. The power and signal
pins on the CompactPCI connector are staged so as to allow the specification in the future
to support hot swapping, a feature that is very important for fault tolerant systems and
which is not possible on standard PCI.
Also, a special flavor of CompactPCI known as 6U CompactPCI supports 3 additional 2-mm
connectors with a total of 315 pins. These can be used for secondary buses such as SCSA or
MVIP. They may also be used as bridges to other buses like VME or SCSI, or for user I/O.
User I/O can be routed out of the back of a 6U card and out of the back of the chassis, a
practice popular in the telecommunications industry. For more information on CompactPCI,
please refer to the article entitled CompactPCI Next-Generation Bus
Architecture For Embedded Applications in this issue.
It is important to note that the telecom-munications market is very slow to change.
Telcos have specialized equip-ment that requires tedious, time-consum-ing programming. The
demand for more and varied services by customers forces service providers to explore new
plat-forms that allow rapid development and customization. The PC will be the plat-form of
the future in the Telco, and CompactPCI will help bring it there.
VIRTUAL SECOND LINE
This is one of those technologies that fits into the category of solving a problem we all
knew we had but thought there could be no solution to. Virtual second line technology
manifests itself in a few different ways but the concept is similar in each application.
It is use-ful in situations where a user is con-nected to the Internet and someone is
trying to call the user on the same phone line. Using a busy call forward service from the
local Telco, the call is routed to the ISP. The call is forwarded with the caller ID
information of the busy number, which is the line connect-ed to the Internet. The ISP
looks up the IP address that matches this telephone number that is logged onto their
sys-tem. It sends a signal to a browser plug-in, which was previously downloaded by the
user. The plug-in activates and informs the user of the telephone call they have waiting.
They either accept or reject the call. If they accept, the call is sent from the
ISPs point of presence (POP) to the connected computer using IP telephony.
Since the connection is direct and encounters no routers, the quality is as good as any
telephone call you have ever heard (assuming the ISP has ade-quate bandwidth and computing
power). The required bandwidth per line is as low as around 2 Kbps, but 48 Kbps is
more realistic when quality is an issue. This concept can easily be extended to allow any
ISP to, in effect, become your telephone company. A browser or Java interface can easily
be developed to allow you to make outbound calls through your ISP. Since IP telephony
allows telephony to travel highly com-pressed, you could have multiple lines handled by a
single connection to the ISP. These calls can be conferenced together or transferred as
you wish. The beauty of this is that you can have a graphical call control application on
the desktop and control of multiple lines through a single connection to the ISP.
VPIM
If you have ever wished you could forward or respond to a voice mail message to someone
not physically located at your company, the way you do with an e-mail message, VPIM (Voice
Protocol for Internet Mail) is the solution you have been waiting for. VPIM has
established itself as an internationally accepted standard to allow the interexchange of
voice and fax mes-sages between voice messaging sys-tems. VPIM also allows interexchange
with non-voice messaging such as MIME (Multipurpose Internet Mail Extensions)-compatible
e-mail systems.
Most e-mail systems already conform to many standards such as MIME and Simple Mail
Transfer Protocol (SMTP). VPIM works with these standards as well as a newer standard
known as Extended SMTP (ESMTP). VPIM also establishes a directory service that maps e-mail
addresses to telephone numbers and vice versa. The transport mechanism for the messages is
any TCP/IP network such as an intranet or the Internet. The standard voice encoding scheme
is G.721. Voice is encoded at 4 Kbps, which translates into roughly 256K per minute of
speech.
H.100
Standards that are accepted by competing companies propel industries. Examples often
overlooked include gasoline for automobiles, 110-Volt outlets in our offices, and
PostScript in our printers. The computer industry has a history of competing standards,
and the CTI hardware and software industry is no different. There are two popular CTI bus
schemes that developers must consider. Developers who developed for one CTI bus would have
to develop again to run on a second vendors bus. H.100 has changed this.
H.100 is a PCI bus standard devel-oped by the Enterprise
Computer Telephony Forum. Prior to H.100, manufacturers would have to decide whether
to support the SCbus or the MVIP bus. If you developed for one bus, you could not easily
have your products installed on systems that were of the other bus type. If your products
worked in both bus environments, you not only increased your development costs, you
reduced your time to market and increased your support costs as well.
System integrators can now smoothly manage their migration from existing ISA SCbus and
MVIP90 to mixed ISA/PCI to 100-percent PCI, while cleanly mixing the various bus types
together along the way. This is truly an industry breakthrough in computer telephony
integration.
S.100
The ECTF is also working on soft-ware interoperability through the S.100 specification.
Just as a common hardware bus allows hardware interoperability, a common software API
helps with software interoperability. S.100 defines a client/server model in which
applications use a collection of services to allocate, configure, and operate hard-ware
resources. It abstracts implemen-tation details of call processing hard-ware and switch
fabrics to enable portable applications to be written. It furnishes these services via an
operat-ing system independent API that may be extended to support custom APIs. S.100
compatibility allows end users to purchase off-the-shelf, best-in-class computer telephony
applications as well as media servers, mixing and matching to provide the perfect
solu-tion to meet their needs. End users will also be able to purchase multi-vendor
applications and be assured they will all run on their S.100 platform, share com-mon
resources, and interoperate through standards-based call hand-off.
A switching application can now share the same resources as a faxback appli-cation.
TAPI allows software abstrac-tion of call control. S.100 allows soft-ware abstraction of
media processing. S.100 allows client/server computer telephony. Applications are isolated
from the underlying hardware compo-nents through a resource manager. The resource manager
allows your applica-tions to be extensible just as Windows supplies a generic API
that application vendors write to, allowing any software to work with any printer without
requiring separate drivers. S.100 allows developers to write to a single API to take
advantage of speech recognition that will work with all speech recognition engines.
TAPI 3.0
Internet telephony or IP telephony is the most important technology to ever enter the
telecommunications market. Microsoft realizes the implications of Internet telephony not
only within the telecommunications infrastructure and corporate intranets but within the
LAN environment as well.
The TAPI 3.0 white paper outlines a plan to facilitate telephony application
development and call routing trans-parent of the underlying infrastructure (be it
an IP network or the telephone network). Telephony and IP telephony will be absorbed into
the infrastructure of future Microsoft operating systems. TAPI 3.0 is an evolution of the
TAPI 2.1 API to the COM model, allowing TAPI applications to be written in any language,
such as Java, C/C++, and the Microsoft Visual Basic programming system. COM is a component
technolo-gy that provides a common object model for both local and network soft-ware
integration and delivers a single, widely implemented standard allowing applications from
multiple vendors to integrate seamlessly over the Internet and corporate networks.
Microsoft will integrate the Active Directory service of NT to provide user-to-IP
address mapping. The direc-tory is continuously updated with cur-rent information. There
is a great deal of likelihood that NT servers with Active Directory services will be the
universal white pages of the future, allowing any Internet telephony user to
seamlessly connect to any other user. NT 5.0 gateway server presence within ISPs will
provide the infrastructure to help make this a reality.
ADVANCED INTELLIGENT NETWORK (AIN) CTI
The (Advanced) Intelligent Network refers to the hardware infrastructure and the signaling
protocols that allow equipment from different telephone companies to work together
seamlessly. AIN has already empowered individual users with a wealth of features.
Three-way calling, area-wide Centrex, *69, call waiting with caller ID: These are all
examples of AIN services. AIN CTI (Network CTI) extends the power of CTI into the network
by inte-grating into the central office. Once installed, the local service provider can
charge for CTI applications much the same way they charge for Centrex or caller ID. At the
moment, these ser-vices are targeted primarily to the call center market.
The goal of Network CTI is to pro-vide customers with all the power, increased
productivity, and enhanced features of CTI without purchasing costly equipment.
The implications of putting CTI into the public network are immense. Network CTI
affords subscribers the ability to control telephony across the telephone network as if
the public net-work were a local PBX. Calls can be routed, conferenced, and diverted based
on database information located anywhere on the network. Call centers will have
opportunities available to them that were heretofore unimagin-able. For more information
on Network CTI , please contact Genesys Telecommunications Laboratories at
www.genesyslabs.com.
E-MAIL ACD
The immense popularity of the Web caught everyone by surprise, and that includes customer
service departments of firms that receive huge volumes of e-mail questions. Enter the
e-mail ACD. Automatic call distribution (ACD) technology was originally invented for the
call center market as a way to route callers to the agent most qualified to answer
questions and handle the inquiry. During this process, the better ACD applications were
able to give customers feedback such as estimated wait time, while offering information
that helped make the wait more pleasant.
In a similar fashion, e-mail ACDs allow e-mail to be routed to the appro-priate person
based upon a set of prede-termined, customizable rules. Whats really terrific about
this technology is the fact that as a corporation receives messages, the messages can be
assigned a tracking number and entered into a database. A predetermined reply can be sent
based upon content in the subject or message body. Customers now get an immediate response
to their e-mail with a tracking number that allows them to follow up on the problem later.
As com-panies continue to receive an endless stream of e-mail to generic mailboxes, this
type of system is a must-have to keep up with the incoming barrage. For more information
on e-mail ACDs, please contact Mustang Software at www.mustang.com.
|