|
When I travel or review my e-mail messages, I'm often confronted by questions about barriers to progress
in the CTI industry. What's missing? What could really make CTI take off? My stock reply: "Standards.
Better standards will help us progress at an even faster rate." Although people are usually satisfied with
that answer, I've grown less willing to settle for it myself.
I've decided that a lack of standards isn't a problem after all. In fact, we may have too many standards.
The problem isn't the lack of a "killer app," either. There are plenty of useful applications, including
unified messaging applications. So, what's missing? In a word, bandwidth.
STANDARDS: LOOKING GOOD
We've already come a long way from the early days of CTI. The first users of CTI, which were usually
large call centers, got started by connecting expensive, proprietary telephone switches to mainframes or
UNIX-based minicomputers. Usually the computer manufacturers would provide the links to large
telephone switch vendors. The result? Once the call center solution was up and running, the user would
find himself or herself locked into a particular computer vendor's hardware.
In the last few years, Microsoft, Novell, Lucent, and Sun have begun to push truly open network operating
system (NOS)-level standards that enable the NOS to take control of any PBX. Although a single standard
is preferable, it is encouraging to see the largest NOS and hardware companies taking an interest in CTI.
These standards, though new, are extremely robust. Furthermore, they are evolving rapidly.
Organizations have sprouted up to facilitate the creation and acceptance of open standards: VERSIT (an
alliance between AT&T, Siemens Business Communications, IBM, and Apple), GO-MVIP (Global
Organization for the Multi-Vendor Integration Protocol), and the ECTF (Enterprise Computer Telephony
Forum). These organizations have already made progress toward defining a single standard that will
appeal to all manufacturers of CTI hardware and software.
You could make the case that this industry began when NOS companies started to take CTI seriously.
Accordingly, you could also say that the CTI industry is only a few years old. In this short time, we have
done an outstanding job of addressing many types of PBXs, allowing them to be "taken over" or
controlled by shrinkwrapped software. Further standards integration will soon allow all PBXs and ACDs
to offer plug-and-play functionality.
USEFUL APPLICATIONS: LOOKING GOOD
Another important component of CTI is the availability of useful applications. I'll describe just one of
many: unified messaging. This application allows one mailbox to contain your e-mail, faxes, voice mail,
and video mail.
Many manufacturers are selling products that support all of the above media and allow you to prioritize
all of your messages and view the most important ones first. By taking advantage of the TAPI and TSAPI
standards, these vendors seamlessly unite all your messaging while giving you full Windows-based
graphical control over all message types. Many vendors use Microsoft Exchange or some other e-mail
package that is MAPI-compliant.
The point is, these applications have immediate utility. Nothing is holding up the progress of CTI
here.
BANDWIDTH: NOT LOOKING SO GOOD
The biggest issue challenging this industry at the moment is bandwidth. Specifically, LAN bandwidth and
Internet bandwidth.
Bandwidth
Most desktops have both a network connection and a telephone connection. This amounts to a duplication
of effort to string both LAN and telephone lines to each desktop. Why haven't we combined these
connections? This is no idle musing. I would venture to say at least one billion dollars are spent
duplicating both LAN and telephone connections to the desktop each year. (This estimate accounts for
equipment, staffing, service charges, and so on).
The most prevalent LAN type, 10 Base T, affords more than enough bandwidth for voice traffic, light file
transfer, printing, and client/server database access. The drawback is that speed decreases as more nodes
are added to the network in most LAN environments. All data traffic must travel over a single shared 10
Mb/s data pipe.
As the reliability of the LAN improves and desktop operating systems become more stable, it's inevitable
that we will see the telephone hanging off the corporate LAN connected to a PC. I've heard that many of
the major network vendors have already figured this out and will soon be working with PBX vendors or
by themselves to develop LAN-based telephone systems.
If you've read our reviews of Internet telephony programs, you already know that we have used many of
these programs on our own corporate LAN. Since we have a lot of traffic on our network, the voice quality
was surprisingly poor.
Guaranteed bandwidth is what is really needed to carry voice over our corporate network. The most
popular method of providing guaranteed bandwidth is either isoEthernet, a switched isochronous network
topology (please refer to our premier issue, page 26 for details), or a generic switched network
technology. A switched network hub allows each connection to the hub to have guaranteed bandwidth. A
switched 10 Base T hub would guarantee that each computer has a 10 Mb/s pipe to transmit voice and
data to the hub.
Switched hubs have been relatively expensive compared to traditional hub technology. As the prices drop,
we should begin to see expanded use of network telephony products.
Internet Bandwidth
Internet bandwidth is continuously increasing, but the amount of traffic that travels over it has been
increasing at an even faster rate. There's no doubt that Internet telephony is being hampered by the
latency of Internet connections. Much the same way that we can use the LAN to transport telephony, we
should be able to use the Internet as well. This opens up the possibility of Internet PBXs that allow us to
transfer calls and set up conference calls with anyone who has an Internet connection.
When your telephone call becomes data over a network, it can be saved on disk like any other data type.
A salesman can save his best sales call and study it over and over. He can e-mail the call to other
salespeople in his company. He can even save his best phone calls on disk and include them as part of his
resume. Forwarding conversations instead of having to describe a conversation you just had to someone
else will save office workers hours of reiterating conversations, freeing them up to do more productive
things. I have spent hours in conversation to prepare people to perform tasks I wished to delegate to them.
Now, all I need to do is e-mail a call with a brief introduction such as "Please handle" to someone else in
my office. This is a great productivity booster.
CONCLUSION
Guaranteed network bandwidth will allow all telephone users to treat their telephone conversations the
way they treat e-mail today. All calls can be archived and searched in the future. Using voice recognition,
I can search all my voice mail for a word or phrase - just as I do with e-mail today.
With the necessary bandwidth in place, CTI technology will come into its own. It will allow us to treat
voice conversations just like we treat every other type of data in our organizations. It will allow us to
become more efficient and productive. No one will be able to do without CTI.
Sincerely,
Rich Tehrani
Publisher, CTI magazine
| 
TMCnet LOGIN
Webinars
